[Federal Register Volume 64, Number 7 (Tuesday, January 12, 1999)]
[Rules and Regulations]
[Pages 2038-2061]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 99-446]


-----------------------------------------------------------------------

DEPARTMENT OF TRANSPORTATION

Federal Aviation Administration

14 CFR Part 39

[Docket No. 97-NM-80-AD; Amendment 39-10963; AD 98-26-20]
RIN 2120-AA64


Airworthiness Directives; Boeing Model 727 Series Airplanes 
Modified in Accordance With Supplemental Type Certificate ST00015AT

AGENCY: Federal Aviation Administration, DOT.

ACTION: Final rule; technical public meeting.

-----------------------------------------------------------------------

SUMMARY: This amendment adopts a new airworthiness directive (AD), 
applicable to certain Boeing Model 727 series airplanes that have been 
converted from a passenger to a cargo-carrying (``freighter'') 
configuration, that requires limiting the payload on the main cargo 
deck by revising the Limitations Sections of all Airplane Flight 
Manuals (AFM), AFM Supplements, and Airplane Weight and Balance 
Supplements for these airplanes. This amendment also provides for the 
submission of data and analyses that substantiate the strength of the 
main cargo deck, or modification of the main cargo deck, as optional 
terminating action for these payload restrictions. This amendment is 
prompted by the FAA's determination that under certain conditions 
unreinforced floor structure of the main cargo deck is not strong 
enough to enable the airplane to safely carry the maximum payload that 
is currently allowed in this area. The actions specified by this AD are 
intended to prevent failure of the floor structure, which could lead to 
loss of the airplane.

DATES: Effective February 16, 1999.
    The public meeting will be held January 20, 1999, at 9:00 a.m., in 
Seattle, Washington. Registration will begin at 8:30 a.m. on the day of 
the meeting.

ADDRESSES: Information concerning this amendment may be obtained from 
or examined at the Federal Aviation Administration (FAA), Transport 
Airplane Directorate, Rules Docket, 1601 Lind Avenue, SW., Renton, 
Washington, by appointment only between the hours of 8:00 a.m. and 2:00 
p.m.
    The public meeting will be held at the following location: The 
Radisson Hotel, 17001 Pacific Highway South, Seattle, Washington 98188; 
telephone (206) 244-6000.

FOR FURTHER INFORMATION CONTACT: Questions concerning the airworthiness

[[Page 2039]]

directive should be directed to Michael O'Neil, Aerospace Engineer, 
Airframe Branch, ANM-120L, FAA, Transport Airplane Directorate, Los 
Angeles Aircraft Certification Office, 3960 Paramount Boulevard, 
Lakewood, California 90712; telephone (562) 627-5320; fax (562) 627-
5210.
    Requests to present a statement at the public meeting regarding the 
logistics of the meeting should be directed to Mike Zielinski, Federal 
Aviation Administration, Northwest Mountain Region, Transport Airplane 
Directorate, ANM-113, 1601 Lind Avenue, SW, Renton, Washington 98055-
4056; telephone (425) 227-2279; fax (425) 227-1149.

SUPPLEMENTARY INFORMATION: A proposal to amend part 39 of the Federal 
Aviation Regulations (14 CFR part 39) to include an airworthiness 
directive (AD) that is applicable to certain Boeing Model 727 series 
airplanes that have been converted from a passenger to a cargo-carrying 
(``freighter'') configuration was published in the Federal Register on 
July 15, 1997 (62 FR 37788). At the same time, the FAA issued three 
other similar notices of proposed rulemaking (NPRM's) to address 
airplanes similarly converted in accordance with STC's held by FedEx, 
Aeronautical Engineers, Inc., and Pemco. That action proposed to 
require limiting the payload on the main cargo deck by revising the 
Limitations Sections of all Airplane Flight Manuals (AFM), AFM 
Supplements, and Airplane Weight and Balance Supplements for these 
airplanes. That action also proposed to provide for the submission of 
data and analyses that substantiate the strength of the main cargo 
deck, or modification of the main cargo deck, as optional terminating 
action for these payload restrictions.
    On February 4, 1998, in order to obtain additional public 
participation in these NPRM's, the FAA reopened the comment period for 
a period of 90 days and scheduled two sets of public meetings, which 
were held in Seattle, Washington, on February 18 and 19, 1998, and 
April 1 and 2, 1998. In addition to the comments submitted during the 
original comment period, the comments that were provided at the public 
meetings and submitted to the Rules Dockets during the reopened comment 
period also are discussed below.

Comments

    Interested persons have been afforded an opportunity to participate 
in the making of this amendment. Due consideration has been given to 
the comments received.
    The FAA has received comments in response to the four NPRM's 
discussed previously (i.e., Docket No.'s 97-NM-09-AD, 97-NM-79-AD, 97-
NM-80-AD, and 97-NM-81-AD). Some of these comments addressed only one 
NPRM, while others addressed all four. For example, although the 
comments submitted by FedEx address only the NPRM applicable to its 
STC's (i.e., Docket No. 97-NM-09-AD), other commenters referenced 
FedEx's comments and requested that those comments be considered in the 
context of the other three NPRM's, as well. Because in most cases the 
issues raised by the commenters are generally relevant to all four 
NPRM's, each final rule includes a discussion of all comments received.

Existence of Unsafe Condition

    Several commenters disagree with the FAA's finding of an unsafe 
condition and refer to the following statement in the NPRM's, ``[a] 
design which does not meet [certification] standards is presumed to be 
unsafe.'' The commenters contend that, while this statement is 
``convenient,'' the FAA is still obliged to issue the AD in accordance 
with 14 CFR part 39. In accordance with part 39, prior to the issuance 
of an AD, the FAA must establish that an unsafe condition exists in a 
product and that this condition is likely to exist in other products of 
the same type design.
    From this comment, the FAA infers that the commenters believe the 
proposed AD is merely a consequence of non-compliance with Civil Air 
Regulations (CAR) part 4b, which are the design standards to which the 
Model 727 was certificated, and that the unsafe condition has not been 
substantiated. The FAA does not concur. The context of the quoted 
statement in the NPRM's was an explanation of the FAA's method used in 
the design review that led to issuance of the NPRM's. Initially, the 
FAA had identified the potential non-compliance based on observation 
and review of original certification data. Since, in accordance with 
the Federal Aviation Act, CAR part 4b standards establish the minimum 
level of safety, the FAA considered that further evaluation was 
necessary and appropriate to determine whether this potential non-
compliance created an unsafe condition warranting an AD. As explained 
in the NPRM's, the FAA determined not only that the design was non-
compliant, but that the degree of non-compliance was highly 
significant, and resulted in substantial negative structural margins of 
safety. The FAA's analysis addressed the ``up'' load case, which was 
considered to be the most likely critical load case, in the sense that 
it was likely to be the load case that would present the most serious 
negative margins of safety. The analysis verified these negative 
margins and confirmed the FAA's concerns that serious negative margins 
may exist for other load cases, as well. The effect of these 
substantial negative margins is that the likelihood of catastrophic 
failure of the floor structure is unacceptably high. The FAA's finding 
of unsafe condition arises from this determination rather than from a 
finding of non-compliance with CAR part 4b.

Risk From Actual Operations

    Several commenters state that the FAA's finding of an unsafe 
condition in the NPRM's is incorrect because, based on the way the 
airplanes are actually loaded and operated, the likelihood of 
encountering conditions specified in CAR part 4b that would exceed the 
strength of the floor structure is extremely improbable.
    The FAA does not concur. The FAA's evaluation was based on the 
potential for a catastrophic event occurring as a result of an airplane 
encountering severe gust conditions while transporting containers 
loaded with maximum allowable payloads. (Unless otherwise stated, 
throughout the preamble of this AD the FAA uses the term ``container'' 
to refer to all unit load devices, including pallets.) The fact that 
operators may transport containers with maximum payloads only for a 
small percentage of their operations does not diminish the seriousness 
of the unsafe condition when they do transport such containers. (It 
should be noted that one commenter stated that its operations with even 
one container at maximum allowable payload are only a small percentage 
of its total operations, but also stated that it engages in such 
operations daily.)
    In addition, the FAA disagrees with the commenters' conclusions 
regarding the probability of catastrophic events. The events that may 
cause a catastrophic failure occur randomly and, thus, cannot be 
reliably predicted and avoided for any particular operation. Although 
the probability of large gusts or excessive maneuvers (as specified in 
CAR part 4b) is low (approximately once in the lifetime of an airplane 
for a large gust), because of the large negative margins of safety 
associated with these unreinforced floor structure designs (discussed 
in the NPRM's), less severe events (i.e., lower gusts or milder 
maneuvers) also could result in catastrophic failure. Therefore, 
because the likelihood of encountering

[[Page 2040]]

less severe events is significantly greater than the likelihood of 
encountering the events contemplated by CAR part 4b standards, and 
because the consequences of such encounters may be catastrophic, the 
FAA considers that the risk is unacceptable.
    During the public meetings, several commenters suggested using 
analytical methods developed to show compliance with 14 CFR 25.1309 in 
assessing risks from gust loads. Their position was that if such 
analysis were performed, it would demonstrate that the unsafe condition 
addressed by the proposed AD is ``extremely improbable;'' therefore, an 
AD is unnecessary to address it.
    The FAA does not concur. The purpose of section 25.1309 is to 
require that type certificate applicants demonstrate the robustness of 
the airplane systems and equipment. Therefore, it is not applicable to 
the assessment of the seriousness of an unsafe condition associated 
with identified structural deficiencies. Nevertheless, assuming that it 
is appropriate, section 25.1309(a) states that the airplane systems, 
equipment, and installations ``must be designed to ensure that they 
perform their intended functions under any foreseeable operating 
condition.'' This means that the airplane must function properly if it 
is being operated within its approved operating and environmental 
conditions. As discussed in the NPRM's, the FAA's analysis demonstrates 
that the affected airplanes, when operated with allowable payload 
weights and distributions (which is foreseeable), could experience 
catastrophic failure if they encounter gust conditions that are also 
foreseeable. Therefore, applying the analytical methods of section 
25.1309(a), these STC designs would be found not to comply.
    In addition, section 25.1309(b) requires that any system failure 
condition that would result in a catastrophic event be shown to be 
extremely improbable, even if the system failure occurred concurrently 
with environmental conditions that would reduce the capability of the 
airplane or the ability of the crew to cope with the system failure. 
Probabilistic analyses are used to demonstrate compliance with section 
25.1309(b) by estimating the probability of random system and equipment 
failures occurring on the airplane. The consequences of failures that 
are more probable must be shown to be relatively minor; failures with 
more serious consequences must be shown to have lower probabilities. 
However, in providing guidance for compliance with this requirement, 
Advisory Circular (AC) No. 25.1309-1A advises: ``In any system or 
subsystem, the failure of any single element, component or connection 
during any one flight * * * should be assumed, regardless of 
probability. Such single failures should not prevent continued safe 
flight and landing * * *.''
    Applying this analytical method to the circumstances of this AD, if 
the failure of the floor beam is assumed, the consequences are likely 
to be catastrophic, preventing continued safe flight and landing. 
Therefore, under the analytical approaches of either section 25.1309(a) 
or (b), the operations with understrength floors without limitations is 
unacceptable.
    During the reopened comment period, FedEx submitted a risk 
assessment from which it concluded that, even assuming the NPRM 
identified a potential unsafe condition, the probability of occurrence 
was sufficiently small (i.e., once every 300 years) so that AD action 
should be postponed until additional testing and analysis has been 
completed. Other commenters referenced this analysis and supported 
FedEx's conclusion.
    The FAA has evaluated the risk assessment submitted to Rules Docket 
No. 97-NM-09-AD, and does not concur with the commenters' conclusion. 
Regarding the general relevance of the kind of risk assessment 
submitted by the commenter, it should be noted that the probability of 
the limit gust event has already been considered when establishing the 
gust intensities specified in CAR section 4b.211(b). CAR part 4b 
requires that all airplanes be capable of structurally withstanding a 
gust of the intensities specified therein, as such a gust is expected 
to occur at some time in the airplane's operating life.
    Regarding the specific data presented in the FedEx risk assessment, 
the FAA does not concur with the assumption that extreme gusts will be 
encountered by a cargo carrying Boeing Model 727 airplane only once in 
5 million flight hours. As its basis for this assumption, the commenter 
states that ``FAA data indicate that, in approximately 50 million 
flight-hours of experience among US domestic 727s, there have been five 
pilot reports of extreme gusts that exceeded federal thresholds for 
danger.'' The commenter states that this equates to a rate of 
occurrence of approximately once every 10 million flights. The 
commenter also states that due to potential errors, it would be 
conservative to double this rate to 10 total events, and use an 
estimate of 1 occurrence per 5 million hours.
    The FAA does not concur with the commenter's statement that FAA 
data show that only five cases of extreme gust have been encountered by 
the U.S. 727 fleet. Turbulence events must be reported only if they 
result in detected airplane damage or passenger injuries. During 
certain gust events, the gust loads encountered in the cockpit are 
substantially less severe than those encountered in the aft portion of 
the airplane. Therefore, some large gust encounters may not ``feel'' 
very severe to the flight crew. As a result, the FAA recognizes that 
not all severe turbulence events are reported. Further, in the NPRM's, 
the FAA provided five cases of turbulence as examples, to illustrate 
that turbulence is a real occurrence, and not merely theoretical. These 
five examples were obtained from data showing 87 reported severe 
turbulence events, which resulted in passenger injuries, on the Boeing 
727 from 1966 to March 1997. The FAA selected the five reports because 
the airplane operators had reported the magnitude of the turbulence 
event after obtaining this information from the flight data recorder. 
Operators are not required to obtain data regarding the magnitude of 
the turbulence event, and therefore it is rarely reported.
    During the public meeting held on Thursday, February 19, 1998, the 
FAA explained that these turbulence cases were just examples and had 
been selected because the reports included information regarding event 
magnitude. The FAA further explained at that meeting that it was 
inappropriate to use these data in a probabilistic analysis. The 
commenter's risk assessment provides no information to change the FAA's 
views.
    A section of the commenter's report states, ``Detailed equations 
that combine empirical evidence and physical theory estimate how 
frequently gusts of different magnitudes arise at different 
altitudes.'' The commenter states that its calculations indicate that 
gusts with intensities that equal or exceed 50 feet per second are 
encountered once per 50 million flight hours at 35,000 feet. The report 
does not provide the equations themselves, does not describe the 
methodology used to determine the 1 in 50 million flight hours 
probability value, and does not specifically identify the referenced 
source data. Therefore, the FAA cannot assess the validity of the 
commenter's conclusions.
    The commenter also refers to graphs contained in a 1988 American 
Institute of Aeronautics and Astronautics (AIAA) publication by 
Frederic M. Hoblit that the commenter states indicate even lower 
encounter rates for gusts during

[[Page 2041]]

climb and descent. The FAA has examined this publication, and does not 
concur with the commenter's statements regarding these data. First, the 
commenter appears to be incorrectly referencing the graphs, which 
represent continuous turbulence, and not discrete gusts, as provided in 
CAR 4b. The two types of atmospheric disturbances are different, and to 
reference these graphs is inappropriate. Secondly, the commenter's risk 
assessment only addresses gusts ``that exceed the Federal threshold'' 
(which the FAA infers to mean limit load gusts) in combination with 
cargo loads with two adjacent containers having a total weight that 
equals or exceeds 9,600 lbs. This approach is unconservative. As 
discussed in the NPRM, the cargo floor has a high negative margin of 
safety, and the risk of structural collapse exists at gust intensities 
well below the limit gust load when carrying currently allowed payloads 
above 9,600 lbs. The greater the weight being carried in the container, 
the lower the gust needed to cause catastrophic failure of the floor. 
The lower the gust intensity, the more common the gust occurrence 
becomes.
    Based on the foregoing, the FAA has determined that the risk 
assessment submitted by FedEx does not provide a basis for delaying the 
final rule.
    One group of commenters, identifying themselves as airmen for one 
of the affected operators, supports issuance of the final rule, as 
proposed. The commenters state that they do not have procedures to 
avoid clear air turbulence, and based on their knowledge, if any of 
them had encountered a similar wind condition to that experienced by a 
Boeing 747 in January 1998, their airplane would ``come apart, in-
flight.''
    The FAA concurs that there is no reliable means to forecast or to 
avoid clear air turbulence. The flight conditions encountered by the 
referenced 747 could be very hazardous to one of the affected airplanes 
if encountered while critically loaded with heavy containers.

Change in Applicable Standards

    Several commenters state that the NPRM's reflect a radical change 
in the assumptions that certificate holders are permitted to use to 
substantiate the main deck floor structure. The FAA does not concur. As 
discussed below, the FAA's analysis is consistent with the applicable 
CAR part 4b standards, which became effective in 1953.

``Infinitesimal Probability''

    One commenter states that the proposed AD would impose unnecessary 
costs which would then be passed to its customers, for what the FAA's 
Director of Aircraft Certification Service has stated is an 
``infinitesimal probability of a safety related happening.'' The 
referenced comment is contained in an article in the April 15, 1997, 
issue of ``Commercial Aviation Report.''
    From this comment, the FAA infers that the commenter believes the 
reference to ``infinitesimal probability'' belies the need for an AD. 
The commenter has taken the remark out of context. The actual quote is, 
``What is the probability of it [catastrophe] happening in the next 
month? Infinitesimal.'' This remark was made in response to a question 
regarding why the FAA was issuing an NPRM rather than an emergency AD. 
The Director of the Aircraft Certification Service was explaining that, 
although the FAA had determined that the unsafe condition must be 
addressed by issuance of an AD, the urgency of the issue was not so 
great as to preclude the normal legally required process of providing 
public notice and opportunity to comment.

Accident Data

    One commenter states that the fact that no crashes have occurred 
with the affected airplanes has nothing whatsoever to do with these 
airplanes being of a safe design. They merely have had the good fortune 
to have not yet encountered a critical condition. The FAA concurs.

``Erroneous Certification''

    One commenter states that it counted on the competence of the FAA 
when obtaining the affected airplanes, as the cargo modifications were 
FAA-approved. The commenter further states that the FAA's error in 
issuing these approvals is going to severely hurt small operators of 
these airplanes, who are neither culpable nor negligent. While the FAA 
understands that the impact of this AD may be significant for some 
operators, the FAA cannot ignore the fact that an unsafe condition 
exists that requires action to ensure the continued operational safety 
of the fleet. If the FAA had been aware of these deficiencies at the 
time of the original STC issuance, the FAA would not have issued the 
STC's.
    One commenter points out that the FAA design review team observed 
that the original passenger floor beams had not been structurally 
reinforced, and that this fact is immediately apparent from the 
technical drawings associated with the STC. The commenter questions why 
the FAA has not expressed any concern or noticed these facts earlier.
    The applicant for any design approval is responsible for compliance 
with all applicable FAA regulations. The FAA has the discretion to 
review or otherwise evaluate the applicant's compliance to the degree 
the FAA considers appropriate in the interest of safety. The normal 
certification process allows for the review and approval of data by FAA 
designees. Consequently, the FAA office responsible for the 
certification of an airplane or modification to an airplane or an 
aeronautical appliance may not review all details regarding compliance 
with the appropriate regulations. Also, the fact that the cargo floor 
structure was unmodified does not necessarily lead to the conclusion 
that the floors are structurally deficient. As explained in the NPRM, 
the understrength floors on certain 747 airplanes converted to 
freighters caused the FAA to question the adequacy of all STC-converted 
passenger-to-freighter cargo floor structures. This AD arises from this 
evaluation.

An FAA/Industry Team

    Several commenters request that the FAA establish an industry team 
comprised of the FAA, STC holders, and operators before issuing an AD 
to establish the requirements and a corrective action plan to resolve 
the problems with the STC's in a logical manner. One commenter states 
that ``too much time has been spent going in different directions to 
resolve common problems for all STC's,'' and that ``the FAA has not 
been sufficiently clear in their requirements for the re-design.''
    The FAA does not concur that issuance of the AD should be delayed. 
An unsafe condition has been identified, and the FAA must take action 
to ensure an acceptable level of safety of the affected fleet of 
airplanes. The STC holders and operators are certainly free to form an 
industry team to find common solutions, and the FAA is willing to 
participate in such efforts. The FAA also does not concur that the 
requirements for re-design are unclear; as the FAA has stated 
repeatedly, the standards for evaluating proposed corrective actions 
are the original certification basis for the airplane, CAR part 4b. Any 
non-compliance with CAR part 4b would have to be shown to provide an 
acceptable level of long-term safety.

FAA/Industry Communication

    One commenter states that there has been ``virtually no opportunity 
for technical exchange'' and, therefore, the FAA should delay issuance 
of the final rule until such an exchange has taken

[[Page 2042]]

place. The FAA does not concur. Since as early as November 1996, the 
STC holders have been made aware of the FAA's concerns regarding the 
cargo floor structure. More specifically, meetings were held with each 
of the affected STC holders in January 1997 to discuss further details 
regarding FAA concerns.
    On February 14, 1997, the FAA again discussed its concerns with the 
affected industry and again requested that industry provide the FAA 
with valid data to address those FAA concerns. Subsequently, over the 
course of the next four months as the FAA prepared the NPRM's, only one 
STC holder provided any data relative to the merits of the proposed 
AD's, and that data did not alleviate the FAA's concerns. In response 
to the NPRM's first comment period, three of the affected STC holders 
did not submit technical data and, for reasons discussed below, the 
data submitted by the fourth STC holder (FedEx) did not alleviate the 
FAA's concerns. During the reopened comment period, the FAA engaged in 
further extensive discussion with the affected industry and those 
discussions continue in the context of on-going efforts to identify 
necessary actions to address the unsafe condition. Based on this 
history, the FAA considers that sufficient opportunity for technical 
exchange has been provided and that further delay is unwarranted and 
unnecessarily jeopardizes public safety.

Delay Issuance

    Two commenters state that additional time is necessary so that the 
airplanes would be removed from service only once to incorporate all 
needed corrective actions (i.e., not only for the floors, but also for 
other problems identified in the NPRM) due to the high cost of 
incorporating partial solutions to the overall problem. One commenter 
requests that all problems associated with the STC's be identified, 
solutions provided, and methods for accomplishment of the solutions be 
agreed upon prior to the issuance of any AD. The FAA does not concur. 
In light of the seriousness of the unsafe condition, the FAA has 
determined that it would first address the strength of the cargo floor 
structure. All of the remaining issues will be addressed in future 
rulemaking efforts. Even though this AD addresses only the cargo floor 
structure, it should not inhibit industry from taking corrective action 
with regard to the remaining issues. In fact, in order to minimize the 
inefficiencies identified by the commenter, the FAA is committed to 
working with industry to identify as expeditiously as possible 
necessary corrective actions for all of the problems discussed in the 
NPRM.
    The Cargo Airline Association (CAA) requests that the FAA not adopt 
an AD imposing interim limits. Since the CAA believes that the risk of 
a catastrophic failure is ``virtually nonexistent,'' and since several 
potential STC holders with varying solutions to issues raised are in 
the process of working with FAA, scarce resources should be devoted to 
ensuring expeditious approval of these proposals.
    Another commenter requests that the FAA delay issuance of the final 
rules until industry solutions are approved [estimating an additional 
60 to 90 days for Israel Aircraft Industries (IAI) to complete its 
analysis, as it has only recently had access to Boeing drawings]. The 
commenter also states that the FAA rulemaking process has caused 
industry to make significant progress and aggressively pursue solutions 
that will likely meet with relatively prompt FAA approvals. The 
commenter also states that although these approvals will result in a 25 
percent reduction in allowable payload, it is willing to operate with 
that limitation. This commenter, and several other commenters reference 
the FedEx risk assessment, which purports to demonstrate a low 
probability of catastrophic failure, as a basis for delaying the final 
rules.
    Another commenter requests 4 to 6 months for completion of certain 
industry tests and risk analysis, as the 3-month timetable for the 
reopened comment period was not adequate, due to the highly complex and 
time-consuming nature of testing and evaluation procedures.
    For the reasons discussed above under the heading ``Risk From 
Actual Operations,'' the FAA does not agree that the risk assessment 
submitted by FedEx warrants delaying this rulemaking. Furthermore, the 
FAA does not agree that correction of the unsafe condition can be 
assured within 60 to 90 days, or 4 to 6 months without this final rule. 
The STC holders and many operators have been aware of this issue since 
the fall of 1996. The FAA anticipates that, with the adoption of this 
AD, industry will continue recent significant progress in addressing 
these issues, which will result in timely implementation of appropriate 
corrective action.

Extension of Interim Operational Period

    Several commenters state that the proposed 120-day interim 
allowances must have been determined to be safe by the FAA, with 
positive margins of safety. Therefore, the commenters request that the 
interim time limits be extended. Some of the commenters request that 
the extension coincide with regularly scheduled heavy maintenance. The 
CAA requests that the interim limits should be allowed to continue for 
however long it takes to modify the airplanes to bring them up to the 
original design limits. This commenter states that under normal 
operations, there is no risk of floor beam failure, and also states 
that the FedEx risk assessment shows that the likelihood of 
encountering conditions set forth in the NPRM are virtually 
nonexistent.
    As discussed above under the heading ``Risk from Actual 
Operations,'' the FAA does not concur that the information provided in 
the FedEx risk assessment provides a basis for an extension of the 
interim period. However, for other reasons, the FAA concurs that the 
interim operational period can be extended.
    In the NPRM, the FAA stated, ``because the determination of the 
effects of operational limitations on payload is based on 
approximations, the resulting payload limits may be unconservative.'' 
The 120-day interim limit was based on this potential unconservatism. 
Since issuance of the NPRM, the FAA has received data (Reports DFE-
72701 and DFE-72702, submitted during the initial comment period as 
Appendices 5 and 6 to FedEx's comments to the NPRM) that partially 
confirm these approximations. In addition, although some progress has 
been made by industry in developing corrective actions, neither 
industry's proposal (as discussed in the NPRM) nor the FAA's 
expectations have been fulfilled. Based on current information 
regarding the status of various efforts to develop corrective actions, 
the FAA estimates that the entire affected fleet can incorporate 
corrective actions during scheduled heavy maintenance within 28 months 
after the effective date of this AD. In light of this new information, 
the FAA has reassessed the proposed interim period of 120 days and 
concluded that the period should be extended to 28 months. Therefore, 
the FAA has revised the final rule accordingly.
    The FAA's decision to extend the interim limitations does not imply 
that the cargo floor structure has been determined by the FAA to be 
safe for an indefinite period, or in compliance with CAR part 4b 
requirements. As stated in the NPRM, the FAA's analysis considered only 
the most likely critical load case, and the proposed interim 
limitations were based on that analysis. The confirming data referenced 
above still does not address other potential critical load cases or all 
locations within

[[Page 2043]]

the airplane. Nevertheless, in light of the balance of the safety and 
economic factors discussed above, the FAA considers that the level of 
safety provided by the interim limitations is adequate for the time 
period of 28 months. However, it is less than the level of safety 
provided by demonstrated compliance with CAR part 4b standards, and the 
FAA considers that compliance with those standards is a necessary 
objective to ensure the long term safety of the affected fleet. The 
balancing that the FAA has considered in establishing this interim 
compliance period is typical of the balancing that occurs in all AD's 
establishing interim requirements and is fully consistent with the 
FAA's obligation to consider economic impacts, such as those imposed by 
Executive Order 12866.

Increased Interim Payload Limits

    Several commenters also request that, due to ``highly 
conservative'' methodologies used by FAA, the proposed interim weight 
limit should be expanded to allow an average maximum container weight 
of 6,000 lbs. The FAA does not concur that its methodologies are highly 
conservative. As discussed in the NPRM and in more detail below, the 
FAA's analytical methods are typical of industry practice, and the 
commenters have not demonstrated how these methods are highly 
conservative. The FAA has not been provided with any acceptable data to 
support the allowance for 6,000-lb. containers, except as discussed 
below under the heading ``Position-by-Position Limitations.'' A 
commenter requests that the FAA maximize the interim limits. The FAA 
concurs that the interim limits should be maximized to the extent that 
they are consistent with the necessity of addressing the unsafe 
condition. The FAA considers that the interim limits established in the 
final rule meet this objective; however, as discussed below, the FAA 
will continue to work to approve higher limitations, once their safety 
is substantiated.
    Federal Express submitted report 98-026 ``Substantiation of Side 
Vertical Cargo Restraint Installation Using Static Test Results,'' 
Revision A, during the reopened comment period. FedEx states that this 
report ``proves conclusively that the side restraint installation is 
adequate to restrain the applied container loads due to vertical 
gust.'' The FAA concurs, and has changed the final rule (Rules Docket 
No. 97-NM-09-AD) applicable to the FedEx STC's to allow the higher 
interim limits with the FedEx side restraints installed.

Position-by-Position Limitations

    The CAA requests that the FAA consider ``position-by-position'' 
limitations, which would establish individual weight limits for each 
container position on the airplane, based on the strength of the floor 
structure at that location. The CAA states that this would allow a 
higher total payload, while addressing the unsafe condition. The FAA 
concurs with the concept of position-by-position limitations, and will 
consider any such proposal when presented with supporting data.
    For example, one commenter, Amerijet, has submitted a position-by-
position proposal, which includes analysis providing for increased 
weights for certain container positions relative to those determined by 
the FAA for the interim period. This proposal also contained lower 
limits for other container positions and presupposes the installation 
of sidelocks. The commenter stated at the April 2 public meeting that 
it intends to install vertical side restraints [sidelocks], but has not 
submitted any data to the FAA on a sidelock installation. The FAA has 
determined that this proposal would provide an acceptable level of 
safety for the 28-month interim period, when the affected airplanes are 
equipped with approved sidelocks. The commenter's proposal would not be 
acceptable to the FAA for indefinite operations, however, as the 
analysis did not consider other issues such as CAR part 4b emergency 
landing loads. The FAA will continue to work with the commenter, or any 
other interested parties, to refine these proposals so that they may be 
approved under paragraph (f) or (g) of the final rule.
    FedEx also submitted a position-by-position proposal, which also 
contained both higher and lower limits as compared to the FAA's 
proposed interim limits. FedEx's proposal also is promising, however, 
its analysis is based on assumptions which the FAA has determined to be 
inaccurate, given the limitations of the weight and balance manual. For 
example, FedEx's assumption for the percentage of the load distributed 
to the sidelocks (40 percent) was derived from its ``Inverted Container 
Test.'' As discussed below under the heading ``FedEx's Tests,'' the FAA 
considers this assumption to be unconservative. The FAA also will 
continue to work with FedEx to refine its proposal, so that it may be 
approved under paragraph (f) or (g) of the final rule.
    The CAA also submitted a finite element analysis (FEA) and, based 
on this analysis, requested that the final rule allow interim container 
payload limitations (regardless of whether sidelocks are installed) of 
approximately 3,500 lbs. in the most forward and aft positions, and 
8,000 lbs. over the wing and wheel well. All other positions would be 
limited to 4,800 lbs. per container position with no sidelocks 
installed, and 5,000 lbs. with sidelocks installed. The CAA also 
requested that, after unspecified frame modifications are incorporated 
and sidelocks installed, interim limitations of 6,000 lbs. per 
container be allowed. Three other commenters submitted similar 
proposals.
    As stated previously, the FAA is willing to work with commenters to 
establish interim limits other than those established in the final 
rule. However, the data submitted with the comment do not establish 
that the model used in CAA's FEA accurately represents the airplane. 
The CAA states that the model was made using the Boeing Structural 
Repair Manual (SRM) and various unspecified measurements of the 
airplane, but without access to the type design data that define the 
airplane configuration. It is, therefore, based on numerous assumptions 
regarding the configuration, which have not been validated. 
Furthermore, the model purports only to represent a 120-inch long 
section of the fuselage. The model does not account for the numerous 
fuselage cutouts for cargo and passenger doors, which affect the way 
the floor structure reacts to loads. Also, the model does not address 
the different structural design of the wing box or wheel well areas.
    Even if it were assumed that the model is accurate for some 
airplanes, it is based on the cargo container locations used by FedEx, 
which are different from those of the other affected airplanes. The 
positions of the containers and locks determine the loads introduced 
into the floor beams. Therefore, using the FedEx container layout 
produces a result which, even if valid, would be only applicable to the 
FedEx airplanes. Based on the foregoing, the FAA does not consider that 
the model provides a sufficient basis for revising the interim limits.
    Several commenters state that the FAA's findings of negative 
margins of safety are too conservative over the wing box and wheel 
well, as these areas are capable of supporting higher container 
payloads due to their stronger design. The FAA concurs partially. The 
FAA has determined that an unsafe condition exists by analyzing the 
basic floor structure rather than the much more complex wheel well or 
wing box

[[Page 2044]]

structure. These areas are capable of supporting greater loads, but the 
commenters have submitted insufficient data to determine what loads may 
be safe in these areas.
    However, the FAA has issued STC's which substantiate the wing box 
and wheel well areas for payload capabilities equivalent to the 
carriage of 6,000- to 10,000-lb. containers, depending on the 
individual airplane's structural capability, which has increased as the 
727's type design has evolved. The FAA notes that, although no 
structural reinforcement was added to the wing box and wheel well for 
these STC's, limitations were sometimes imposed in consideration of the 
individual airplane's structural capability.
    The FAA has considered the greater strength of the wing box and 
wheel well and has determined that an acceptable level of safety will 
be achieved by allowing a total payload of 12,000 lbs. for any two 
adjacent containers in this area, without other limitations, for the 
28-month interim period. To eliminate potential ambiguity as to the 
containers to which this limitation applies, the final rule specifies 
that this alternative limitation applies to containers located 
completely or partially between body stations (BS) 740 to 950. However, 
the FAA does not consider that it is acceptable to allow combined 
payloads above 12,000 lbs. for this interim period, or to allow 12,000-
lb. combined payloads indefinitely, because the FAA does not have the 
detailed information or resources necessary to determine the 
appropriate payload and operational limitations for all configurations 
of the affected airplanes. Operators who desire further increased 
loading in this area are invited to submit their requests and 
supporting data to the FAA in accordance with paragraph (f) or (g) of 
this AD.
    Paragraph (a) of the NPRM did include a limited position-by-
position proposal, in that it specified a reduced payload limitation in 
the area of the cargo door (BS 440 to BS 660). As with the wing box and 
wheel well area, to eliminate potential ambiguity as to the containers 
to which this limitation applies, the final rule specifies that this 
limitation applies to containers located completely or partially 
between BS 440 and BS 660.

Extension of Initial Compliance Time

    One commenter states that the NPRM's will ``wreak havoc'' on the 
express industry and shipping public. The commenter states that it has 
no way of knowing when the effective date of the AD will be. The 48-
hour implementation of the load limits will inevitably result in 
serious disruption to cargo already booked or in transit when the final 
AD's are issued. Several other commenters requested 120 days after AD 
issuance for interim limits to become effective, as this time was 
necessary to alter manuals, provide personnel training, and generally 
prepare for a significantly different loading procedure. The FAA 
concurs partially. The FAA has changed the final rule to extend the 
compliance time from 48 hours to 90 days. The AD becomes effective 35 
days after the date of publication in the Federal Register. As 
requested by the commenters, this allows a total of 125 days for 
operators to make necessary changes to the FAA-approved Airplane Flight 
Manual and cargo loading procedures.

All Container Types

    Several commenters state that the proposed AD should address the 
use of all possible containers, pallets, and the intermixing of pallets 
and containers. Other commenters followed with similar statements about 
pallets, bulk loading, oversized cargo, and combi configurations (i.e., 
configurations with provisions for passenger seating and cargo on the 
main deck). One of the commenters requests that the wording of the 
proposed AD be changed to contain generalized wording that would 
address all container sizes, using a ratio of the length and width of 
other containers to the 88- by 125-inch container specified in the 
proposed AD as a means to determine the container payload limit. The 
commenter further states that this could help the implementation of the 
rule. The commenters request these changes to avoid the disruption that 
might result from having to obtain individual approvals for each of the 
types of containers.
    The FAA concurs partially. In light of the administrative burden of 
approving individual container types, the FAA has reassessed this 
proposed requirement. The FAA recognizes that, except for half-size 
containers (discussed below), the FAA analysis used to establish the 
payload limits for containers measuring 88 by 125 inches also is 
applicable to any container within the same floor area. The reasons are 
that the analysis considered the effect of the container weight on the 
floor structure supporting the container, and that the differences in 
the stresses in the floor structure associated with the different 
container types are not sufficient to warrant different limits. 
Therefore, the FAA has revised the final rule to specify the same 
limitations for container size codes ``A,'' ``B,'' and ``C,'' as 
defined in National Aerospace Standard (NAS) 3610, which is the 
specification referenced in FAA's Technical Standard Order (TSO) C90c 
for cargo unit load devices (containers).
    For half-size containers (i.e., size code ``D'' or ``E'' of NAS 
3610, or the FedEx ``Demi'' container), the final rule specifies 
payload limits that are one-half those for other containers. Since 
these half-size containers are designed to be placed side-by-side 
across the fuselage, this separate limit is necessary to ensure proper 
load distribution within the area. It should be noted that paragraph 
(g) of the final rule allows operators to establish different container 
payload limits from those specified in the rule by substantiating that 
those limits provide an acceptable level of safety.
    For oversize cargo, operators may apply for approval of alternative 
methods of compliance in accordance with paragraph (f) or (g) of the AD 
by proposing appropriate limitations for such cargo.

Service History

    One commenter claims that, for the converted 727 freighters, 
``successful flight history is direct evidence which supports [the 
commenter's] analysis showing the airplanes to be safe.'' The commenter 
references CAR sections 4b.202, 4b.270, and 4b.300 to show that service 
history is a reliable indicator ``to support or define a substantiation 
methodology.''
    The FAA does not concur. The requirements of CAR part 4b that the 
commenter references are related to the determination of the fatigue 
strength of structure, where it is acceptable to utilize the service 
history of airplanes of similar structural design. However, the unsafe 
condition addressed in this AD is not related to fatigue, but is the 
result of the existing floor structure being significantly 
understrength. The only conclusion that can be drawn analytically from 
the accumulated flight history of the converted 727 freighters is that 
these airplanes have yet to encounter a sufficiently severe gust 
condition when critically loaded with an allowable payload 
configuration to cause failure of the floor structure.

Deflection of Floor Beams

    One commenter states that the FAA did not provide a reasoned 
explanation of the NPRM claim that ``even if the floor beams of the 
main cargo deck only become deformed, the results could be 
catastrophic.'' The commenter compares this statement to McDonnell 
Douglas Report MDC-J5568, applicable to Model DC-10 series airplanes, 
which was approved by the FAA and showed

[[Page 2045]]

significant and permanent deformation of the wing.
    From this comment, the FAA infers that the commenter believes that, 
if the wing can bend safely and even deform permanently when it has 
cables/fuel lines, etc., passing through the structure, then the floor 
beams also must be capable of safely deforming or bending.
    The FAA does not concur. The NPRM states why deformation of the 
floor beams could be catastrophic. For the ``up'' load case analyzed by 
the FAA, which consisted of ``up'' loads applied to the containers due 
to a down gust on the airplane, the floor beams common to the forward 
and aft locks of a container bend upward due to the applied upward 
load. The adjacent floor beams underneath the containers that are not 
attached to the container do not bend. If this deflection relative to 
the adjacent floor beams is excessive, this could result in the bending 
and stretching of all control cables and fuel lines passing through the 
floor beams. Such bending and stretching could result in uncommanded 
flight control inputs at a critical time when the airplane is subject 
to severe gust conditions. In addition, the fuel lines located in the 
floor beams are not designed to flex in the same manner as fuel lines 
located in the wing structure of an airplane and, therefore, may crack, 
bend, or rupture.
    The occurrence of either an uncommanded flight control input during 
critical flight conditions or the rupture of a fuel line can be 
catastrophic. The McDonnell Douglas report referenced by the commenter 
is not applicable to the floor beam deflections of a 727 converted 
freighter since the fuel lines and control cables located in the wing 
of Model DC-10 series airplanes are specifically designed to 
accommodate large wing deflections and are in compliance with the 
applicable regulations.

Safety Factor

    One commenter states that the use of a safety factor as small as 
1.5 presupposes very accurate analysis, knowledge of loads and material 
properties, and sound engineering practices. Structure with negative 
margins of safety of -0.63 clearly indicates that some or all of these 
suppositions have not been achieved. In addition, some operating 
conditions, such as gusts, are beyond human control. The safety factor 
of 1.5, as required by CAR part 4b, is necessary to maintain the safety 
of the airplanes. The FAA concurs with the commenter, but notes that 
the finding of unsafe condition in this AD is based on the FAA's 
determination that the risk of catastrophic failure of the 
understrength floor structure is unacceptably high, rather than on a 
simple finding of non-compliance with CAR part 4b.

Fore and Aft Center Of Gravity Shifts

    Several commenters objected to the FAA's analytical use of the 
trapezoidal method for evaluating shifts in the center of gravity (cg) 
within a container. One commenter, FedEx, states that the FAA's use of 
the trapezoidal shift results in impracticable--if not impossible--
circumstances that exceed the requirements of CAR section 4b.210.
    In order to gain a better understanding of this and other FedEx 
comments, the FAA met with FedEx on September 19, 1997, having first 
provided FedEx with a series of questions to be discussed at the 
meeting. (The minutes of this meeting are included in Rules Docket No. 
97-NM-09-AD.) At this meeting, FedEx reported that it had only recently 
obtained a scale that would allow it, for the first time, to determine 
the actual locations of the cg's inside its containers. FedEx stated 
that it had weighed and determined the cg location on a sampling of 
1,500 containers, but did not provide any data to the FAA at the 
meeting. In any case, the FAA does not consider it appropriate to 
evaluate only an operator's average container payload when establishing 
the safety of the affected airplanes. The unsafe condition determined 
by the FAA's analysis is based on the payload weight and distribution 
with which these airplanes are currently allowed to operate.
    In addition, in a letter dated November 4, 1997, to the FAA (a copy 
of which has been placed in Rules Docket No. 97-NM-09-AD), FedEx states 
that ``A review of container weights, quadrant weights, and cg's for 
the `SAA' (88- by 125-inch) container finds no containers in the 4,000 
to 8,000 lb. range with a cg offset greater than 8.67%.'' However, 
FedEx did not provide data (e.g., the numbers and types of containers 
reviewed; the percentage of cg shift for different container weights) 
to substantiate the value of 8.67 percent. Therefore, the FAA is unable 
to determine the significance of this comment.
    FedEx states that it chose to use a ``stair step'' or ``box'' 
method to evaluate the effects of cg shifts within a container. FedEx 
also states that the FAA rejected this method for use on the 727 
converted freighters without a reasoned explanation.
    The FAA does not concur with the comments regarding the FAA's 
methodology. As stated in the NPRM, the large negative margins of 
safety calculated using the FAA's analysis included consideration of 
the effect of a horizontal cg shift of 10 percent within the container 
(e.g., 8.8 inches from the geometric center of the base of the 
container for the forward and aft direction). Shifts in cg are 
particularly important in considering the ``up'' load case because the 
container loads are applied primarily to the floor beams at the forward 
and aft edges of the container where the container locks are located. 
The effect of the cg shift is to increase the loading on the beam in 
the direction of the cg shift. For example, if the cg is shifted aft, 
the applied loads will be increased on the floor beam located at the 
aft edge of the container.
    In analyzing the effects of forward or aft cg shifts, the FAA 
employed a ``trapezoidal method.'' The trapezoidal method is well 
accepted and used by both Type Certificate (TC) and STC holders. The 
trapezoidal method is analogous to shifting sand in a box. With no cg 
shift, the weight of the cargo is uniformly distributed across the base 
of the container. As the cg is shifted, the load or ``sand'' is taken 
from one side and applied to the other side. This results in a sloping 
load distribution, with a load ``peak'' on one end of the container, 
and a load ``valley'' on the other end. Another acceptable method for 
considering forward or aft cg shifts is the ``box'' or ``stair step'' 
method. In this method, rather than sloping, the load ``steps'' up from 
a low level on one end, to a high level on the other.
    The FAA does not concur that the trapezoidal shift used in the 
FAA's analysis exceeds the requirements of CAR section 4b.210. For 
``up'' loads on the container, and a forward or aft cg shift (which the 
FAA has identified as the most likely critical case), if the airplane 
is not equipped with side vertical restraints (sidelocks), the results 
of the loads analysis are the same regardless of whether the stair step 
or trapezoidal method is used. Since all loads are carried by the floor 
beams that support the forward and aft container locks, the loads on 
the beams will be identical for any method that shifts the cg a 
particular percentage within the container. It is the percentage of cg 
shift that is important, not how that cg shift was achieved. This 
represents the majority of the airplanes affected by these four AD's. 
For those airplanes equipped with sidelocks, there is a maximum 
difference of 14 percent in the two methods for ``up'' loads, at the 
``peak'' of the trapezoid. In consideration of the varying locations of 
sidelocks and the manner in which loads are actually distributed among 
all locks, this difference does not

[[Page 2046]]

significantly affect the FAA's analysis or alter the finding of the 
unsafe condition.
    The FAA considered 10 percent as the appropriate amount to shift 
the cg within the container, as it is realistic and typical of cg shift 
limitations contained in operator weight and balance manuals. 
Consideration of a 10 percent cg shift also represents an industry 
standard as evidenced by NAS 3610 (contained in the Rules Dockets). The 
vast majority of containers used by operators comply with this 
standard. FedEx has not provided any data that indicate that a 10 
percent cg shift is unreasonable, or that show that the FAA's use of a 
trapezoidal shift is unrealistic. The data that FedEx provided (average 
container densities ranging from 7 to 18 lb./cubic foot) concern only 
the average weight of a container used in its operations and assumes 
the weight to be equally distributed throughout the container.
    FedEx also states that the trapezoidal method results in load 
distributions that greatly exceed the 90 lb./inch ``running load'' 
(freight payload per inch of airplane floor length) limitation 
specified in the FedEx weight and balance manual. FedEx states that the 
trapezoidal shift method will result in possible freight densities of 
40 lb./cubic foot in approximately 1/4 of the container volume. FedEx 
states that this equates to an average value of over 200 lb./inch 
running load in this area of the container. FedEx reports that its 
daily average operational load density is approximately 7 to 7.5 lb./
cubic foot, and on rare occasions may have reached the 18 lb./cubic 
foot range; therefore, the FAA's analysis bears no relationship to 
operational reality. (An average density of 18 lb./cubic foot over the 
entire volume for the full-size FedEx container equates approximately 
to a 7,920-lb. container, or about 90 lb./inch running load.)
    The FAA acknowledges that, in its analysis described in the NPRM, 
it was not constrained by the 90 lb./running inch limitation specified 
in the FedEx weight and balance manual. However, the FAA does not 
concur that this results in inaccurate weight limits. The FAA notes 
that, for a FedEx container at the maximum permitted payload of 8,000 
lbs., the running load limit is exceeded even with no shift in the 
container cg (88-inch container width times 90 lbs. per inch equals 
7,920 lbs.). For any forward/aft cg shift within the container, using 
either the trapezoidal or ``box'' method, the degree to which the limit 
is exceeded increases in direct relation to the magnitude of the cg 
shift.
    In addition, the FAA reviewed FedEx's loading procedures during a 
visit to its flight line at Sea-Tac International Airport, Seattle, 
Washington, on February 5, 1997. During this review, the FAA became 
aware that FedEx neither determines the actual cg location of the cargo 
within each container nor has the necessary equipment at all of its 
loading facilities to determine that it is operating within the cg and 
running load limitations of its weight and balance manual.
    Based on other comments received in response to the NPRM, it 
appears that FedEx's practice is not unusual even though it is 
inconsistent with its weight and balance manuals. In light of the fact 
that, to the FAA's knowledge, no operators are measuring the cg's for 
all containers, and that a recent sampling accomplished by FedEx shows 
cg shifts as high as 8.67 percent, the FAA concludes that use of 10 
percent cg shift in its analysis is not only an appropriate reflection 
of industry cargo loading practice, but may actually be unconservative.
    Finally, the FAA does not concur that it has rejected the use of 
the ``box'' method proposed by FedEx. FedEx did not consider a cg shift 
effect in the original substantiation documentation for its original 
STC design, but later proposed to employ a ``box'' method used by 
McDonnell Douglas for the certification of a DC-10 freighter (submitted 
by FedEx as a comment during the first comment period in Appendix 2, 
Report 97-028, Revision I/R, dated April 1, 1997). After review of this 
method, the FAA accepted it in a meeting with FedEx on April 29, 1997. 
The basis for this acceptance is that it provides an acceptable level 
of conservatism in the absence of more rational data to predict the cg 
within a container. As discussed above, the use of the ``box'' method 
does not significantly affect the FAA's analysis or alter its finding 
of an unsafe condition.

FAA's Methodology

    Boeing states that the FAA's analysis is similar to that used by 
Boeing for initial certification of Model 727 series airplanes. 
However, Boeing also states that while the analysis is conventional, 
some of the assumptions made are not typical of industry practice for 
the floor beam analysis and are conservative relative to the original 
certification practice of Boeing, with respect to trapezoidal loading 
and credit for pressurization. Boeing states that, when it evaluates cg 
offsets in containers, it uses the stepped rectangular or ``box'' 
method to determine cg shifts.
    The FAA concurs partially. As explained previously, the trapezoidal 
loading assumption is nominally more conservative than the stepped 
rectangular or ``box method.'' For the ``up'' load case, this nominal 
difference only affects those airplanes with sidelocks. In any case, 
this difference does not significantly affect the FAA's analysis or 
alter its finding of an unsafe condition.
    The FAA does not concur that its analysis is inappropriately 
conservative because it considered zero fuselage pressurization. 
Fuselage pressurization tends to provide an increase in floor beam load 
carrying capability because the pressurized fuselage, to which the ends 
of the floor beams are attached, pulls outward on the ends of the floor 
beams, which makes the floor beams act stiffer. Severe gust conditions, 
such as microbursts, may be encountered at low altitudes when the 
fuselage is not pressurized; therefore, it is realistic to consider 
those conditions. Even with credit for fuselage pressurization, the 
FAA's conclusion would be unchanged because the pressurization effects 
do not significantly affect the substantial negative margins of safety 
found as a result of the analysis. Furthermore, CAR section 
4b.216(c)(1) requires that ``The airplane structure shall have 
sufficient strength to withstand the flight loads combined with 
pressure differential loads from zero up to the maximum relief valve 
setting.''
    Another commenter, FedEx, states that the FAA's analytical 
techniques are too conservative and, therefore, result in artificially 
low payload numbers (container weights) for the 727 converted 
freighters. The FAA does not concur. The FAA reviewed the 
substantiating data submitted for the original certification of FedEx's 
727 freighter conversion STC and found that this data package lacked 
any stress analysis substantiating the floor structure. Lacking this 
data, the FAA reviewed the analytical methods used by others in 
industry. The FAA determined that other industry analytical methods for 
cargo systems used conservative overlapping assumptions to ensure that 
the design resulted in a safe product that complied with CAR part 4b. 
The FAA's decision to use these methods to perform an analysis of the 
floor structure of the affected 727 converted freighters is consistent 
with industry standard practices.
    One commenter expresses concern over the methods utilized in the 
structural substantiation of floor beam loads in the documentation 
contained in these Rule Dockets, although the commenter did not 
identify a basis for the concern. The commenter states that over the 
course of the last two decades

[[Page 2047]]

it has developed stringent methods for accurately predicting cargo 
induced loads in airplane structure. The commenter requests that the 
FAA consider these methods in performing its evaluations. The commenter 
submitted data regarding its analytical methodology used in development 
of numerous STC approvals of cargo handling systems.
    The FAA has reviewed the commenter's methods and considers that 
this methodology utilized conservative, overlapping assumptions to 
``bracket'' unknown variables and utilized a trapezoidal distribution 
of cargo in defining its cg offsets. The FAA agrees that these are 
appropriate methods for determining loads for cargo floor structure and 
are consistent with those employed by the FAA. These methods result in 
conclusions that are consistent with the FAA's findings that the floor 
structure addressed by these AD's presents an unsafe condition. 
Further, the FAA notes that these conclusions are consistent with those 
derived from other methods commonly used in industry.
    Boeing addresses the statement in the FAA's analysis of the floor 
beam allowables (contained in the Rules Dockets) that the analysis is 
``partial'' and ``unconservative.'' Boeing states that, for the 
``down'' load case (i.e., ``down'' loads applied to the container), the 
FAA's analysis is sufficiently conservative for the following reasons: 
(1) the critical section selected for analysis reflects the worst case 
hole-out situation; (2) all significant [down] load cases were dealt 
with; (3) the critical section analyzed would have no degradation of 
[safety] margins because of secondary bending effects; and (4) the 
critical section analyzed has no shear on it by first principles and, 
therefore, any shear interaction effects should be small.
    The FAA concurs with the commenter's statement; however, the FAA 
notes that this statement was carefully limited to apply to ``the down 
load case being considered'' and does not address all load cases, the 
actual strength of the floor, or the floor beam as a whole.
    The FAA does not concur that the commenter's statement is valid for 
all load cases and all floor beam structure. The FAA's statement that 
the analysis is ``partial'' and ``unconservative'' relates to the fact 
that there are many floor beams, several with differing applied loads, 
load carrying capabilities, and critical cross-sections. As a result, 
the FAA's analysis could not be considered complete (therefore 
partial), nor could the FAA state that it had accounted for all 
effects, which may result in yet higher stress levels and larger 
negative margins of safety (therefore unconservative).
    One commenter states that the standard being pursued by the FAA for 
the converted 727 freighter includes all known theoretical 
possibilities, plus an additional safety factor of indeterminate size. 
The commenter refers to a statement in the NPRM that ``airplanes may 
encounter severe turbulence that exerts wind gust forces beyond the 
critical case forces of CAR part 4b * * *'' as implying that the FAA is 
imposing standards beyond that of CAR part 4b.
    The FAA does not concur. The FAA's analysis of the converted 727 
freighter floor beams was accomplished using the standards identified 
in CAR part 4b. No new standard is being applied to these airplanes. 
The commenter has taken the NPRM statement out of context. The FAA's 
reference to gusts that exceed CAR part 4b critical load cases is in a 
portion of the NPRM that addresses the basis for the retention of the 
1.5 factor of safety, which is required by CAR section 4b.200(a). This 
factor is used to protect the airplane from failure when experiencing 
limit load, the highest expected actual in-flight loading, and other 
unknown situations.
    As stated in the NPRM, interested parties had requested that the 
FAA eliminate the safety factor during preparation of the NPRM, which 
would allow higher payloads. The statement that the commenter 
characterizes as implying ``new standards,'' and a safety factor of 
``indeterminate size,'' was simply a discussion of the existing level 
of safety established by the CAR part 4b standards (this airplane was 
originally certificated to those standards over 30 years ago).
    One commenter quotes from CAR section 4b.210 that the analysis must 
be conducted using ``any practicable distribution of disposable 
loads.'' The commenter states that the loading scenarios the FAA uses 
are much higher than the maximum [loading] experienced in actual 
service. Several other commenters characterize the FAA's assumptions 
and analysis as ``ultra conservative.''
    The commenters appear to have misinterpreted the referenced CAR 
section 4b.210. The word ``practicable,'' which means possible to put 
into practice, appears to be read as ``practical.'' Subpart C of CAR 
part 4b requires that analysis be conducted for conditions (e.g., 
critical altitude, critical load, or maximum/minimum weight) that are 
possible; Subpart C is not restricted to normal, average, or practical 
conditions. Designing airplanes to withstand only average loads would 
result in a greater potential for catastrophic failures whenever those 
loads are exceeded.

Boeing Data

    FedEx states that none of Boeing's analysis for the affected 727 
airplanes provides any baseline for comparison of the unit load device 
(ULD) cg shifts, container load distribution, or other key 
methodologies. The FAA does not concur. As a check to verify that its 
analysis was generally correct, the FAA examined some of the type 
certification data that Boeing had submitted prior to certification of 
727 passenger and freighter airplanes. The Boeing data verified the 
FAA's analysis in the following two significant respects:
    1. Boeing's stress analysis that established allowable floor beam 
strength for the passenger version was entirely consistent with the 
FAA's stress analysis; and
    2. Boeing's loads analysis for the freighter version, while using a 
different methodology from that used by the FAA, would result in 
substantial negative margins of safety for passenger floor structure 
when carrying 8,000-lb. containers.
    In accordance with CAR part 4b, Boeing's analysis of the 727 
freighter considered all aspects of cargo loading, including cg 
offsets, load distribution, and multiple other facets. It should be 
noted that Boeing found it necessary to substantially strengthen the 
floor structure for its freighter version in order to carry the same 
payloads currently allowed by the subject STC's and remain in full 
compliance with CAR part 4b.

FedEx's Analysis

    In support of its position that there is no unsafe condition, FedEx 
states that it has used a rational, conservative analytical approach 
for determining that the cargo floor structure is safe, which has not 
been accepted by the FAA. Specifically, FedEx references individual 
floor beam analysis and tests conducted with combinations of loads, 
offsets, container positioning, airplane weight, and flight maneuvers 
that create conditions exceeding any that statistically will occur.
    The FAA does not concur. Except for the lateral floor beams over 
the 80-inch long wheel well area, which is discussed below under the 
heading ``Data Showing Floors to be Safe,'' FedEx has not yet submitted 
a complete analysis of the floor structure, or of a single floor beam. 
The tests that have been run to date are of limited relevance

[[Page 2048]]

as discussed under the heading ``FedEx's Tests.'' Further, as discussed 
previously, the FAA also does not concur that the unsafe condition is 
so improbable that it should not be addressed.
    FedEx states that the statement in the NPRM that the FAA used 
commonly accepted analytical methods in its structural analysis is 
misleading because it fails to address other ``commonly accepted 
analytical methods.'' In particular, FedEx references the FAA's use of 
a pinned end column fixity coefficient (``c'') of 1.0, and in contrast 
points out that a ``c'' of 2.58 is used in an example problem contained 
in ``Analysis and Design of Flight Vehicle Structures'' by E.F. Bruhn. 
FedEx considers this example problem to be analogous to a floor beam 
lower cap analysis. FedEx states that other alternative analytical 
methods (such as Bruhn) result in a significant increase in allowable 
loads for the floor beams (therefore potentially higher allowable 
container weights), but these methods have been rejected by the FAA as 
inapplicable to the converted 727 freighters, even though they have 
been accepted previously by the FAA on other certification efforts.
    The FAA does not concur. The selection of this coefficient can have 
a significant effect on the determination of the allowable payloads. A 
low column fixity coefficient of 1.0 means that the ends of the beam 
are ``pinned'' (i.e., free to rotate or move like a hinge). A column 
fixity coefficient of 4.0 means that the ends of the beam are fully 
``fixed'' (i.e., unable to rotate or move for any applied load). The 
FAA's analysis uses a ``pin end coefficient'' because it represents the 
airplane structure. As stated previously, the FAA's analysis considered 
the ``up'' load case to be the most likely critical case. For this load 
case, the lower horizontal member or ``chord'' of the ``I'' shaped 
floor beam will be in compression and, therefore, will behave in the 
same manner as a column under compression. It will be free to rotate or 
move like a hinge, not fixed as a higher fixity coefficient would 
suggest.
    FedEx's proposed ``c'' coefficient of 2.58 does not appear in any 
of its analysis in support of its comments to the NPRM. At the 
September 19 meeting, FedEx stated that it did not use the 2.58 value 
in any of its analyses submitted in its comments. FedEx also stated at 
the meeting that the 2.58 value was merely an illustration of a fixity 
coefficient that could be found in the Bruhn handbook for a similar 
problem. Nevertheless, FedEx maintained at that meeting that it 
estimates the true value of ``c'' is in excess of 1.2, and may be as 
high as 2.58, although FedEx did not provide any data to the FAA to 
show that a ``c'' of 2.58 would be representative of the structure.
    In addition, in FedEx's analysis submitted to the NPRM, FedEx used 
a ``c'' value of 1.2. (Document 97-021, initial release, dated February 
28, 1997, submitted to the NPRM (Rules Docket No. 97-NM-09-AD) as 
Appendix 1 during the first comment period). However, in a later 
version of the same document, FedEx also used a ``c'' coefficient of 
1.01 (Document 97-021, dated March 24, 1997, but designated as the 
initial release of the document, as well), submitted to the FAA for 
review on April 7, 1997. The FAA has determined that there is 
essentially no difference between 1.00 and 1.01 for a column end fixity 
coefficient. Therefore, the FAA concludes that the more recent data 
submitted by FedEx is consistent with the value of 1.0 for the column 
fixity coefficient used in the FAA's analysis.
    FedEx states that it has submitted reports to the Seattle Aircraft 
Certification Office (ACO) that employ assumptions that were used by 
Douglas Aircraft Company and were accepted by the Los Angeles ACO for 
the original certification of the Model DC-10 airplane. FedEx also 
states that the Los Angeles ACO's earlier approval of the assumptions 
used in the Model DC-10 analysis affirms that it is using an 
appropriate method to substantiate the integrity of its converted 727 
freighters. FedEx states that the FAA has not explained how the 
methodology can be accepted by the Los Angeles ACO and not accepted by 
the Seattle ACO.
    The FAA acknowledges that use of the particular assumption(s) 
referenced in the DC-10 analysis, if applicable to FedEx's 727 
analysis, may allow higher container weights than those specified in 
the proposed AD.
    The FAA does not concur with the commenter's statements. For many 
certification projects, it has been acceptable to use a particular 
assumption which may not be conservative, provided that there are other 
quantifiable assumptions used which account for the lack of 
conservatism and result in the overall design being conservative and in 
compliance with CAR part 4b. Therefore, an unconservative assumption 
used as part of a particular approved methodology is not equally 
acceptable for another methodology without ensuring that the lack of 
conservatism is accounted for elsewhere in the methodology and that the 
overall design is conservative.
    At the July 24, 1997, meeting with FedEx, an FAA representative 
from the Los Angeles ACO stated that it was the responsibility of FedEx 
to demonstrate that the analytical assumptions and methodologies used 
on the DC-10 were conservative for the Boeing 727. To date, FedEx has 
not made that demonstration. During the September 19 meeting with 
FedEx, the FAA asked FedEx if it had used the entire analytical 
methodology that was used for the DC-10. FedEx replied that it had not. 
Therefore, the FAA does not agree that the two ACO's have been 
inconsistent.
    FedEx states that neither it nor the FAA has a complete, accurate 
model which objectively demonstrates the actual performance of the vast 
array of the TSO and STC ULD's in any one of the hundreds of individual 
airplane cargo positions and latch configurations of in-service 
airplanes. The FAA concurs that there is no accurate model which 
demonstrates the actual loads input into the structure of the 727 
converted freighters for the myriad of possible configurations. 
However, an analysis using conservative overlapping (or enveloping) 
assumptions can be performed to show the design is safe for the 
proposed usage and is in compliance with CAR section 4b.200(c). This 
approach has been successfully used by aerospace companies for many 
years and is acceptable to the FAA.

FedEx's Tests

    FedEx states that three tests (descriptions follow) indicate that 
the floor structure of the existing main cargo deck is in compliance 
with CAR part 4b when supporting existing weight limits of the weight 
and balance manual.
    1. Inverted Container Test. FedEx states that it has conducted an 
inverted container test that demonstrates that its existing sidelocks 
are effective in carrying 35 to 40 percent of the container load. The 
test report is contained in Appendix 9 (Report 97-048, Revision I/R, 
dated May 5, 1997) of FedEx's comments to the NPRM (Rules Docket No. 
97-NM-09-AD) during the initial comment period. FedEx also states that 
these results show that the FAA's estimation that the sidelocks carry 
20 percent of the container load is far too conservative.
    The FAA infers that FedEx considers that the FAA's estimation that 
20 percent of the total container load is carried by all sidelocks (10 
percent per side) is conservatively low since this results in 80 
percent of the total load being carried by the locks attached to the 
main deck floor beams. Because FedEx's inverted container test showed 
that 35 to 40 percent of the container

[[Page 2049]]

load was carried by the sidelocks (approximately 20 percent per side), 
60 to 65 percent of the total load would be carried by the locks 
attached to the main deck floor beams.
    FedEx states that this test indicates that the floor structure of 
the existing main cargo deck is in compliance with CAR part 4b when 
supporting existing weight limits. The FAA does not concur that FedEx's 
testing has shown that sidelocks are 35 to 40 percent effective because 
the testing does not address all container types, cg shifts, and all 
container positions on the airplane. The FAA estimated that the 
sidelocks are 20 percent effective based on current industry methods, 
as used in TC and STC programs. To date, industry, with the exception 
of this test by FedEx, has little or no data showing the exact 
distributions of actual sidelock load percentages. Therefore, 
enveloping assumptions and/or conservative analytical methodologies 
have been consistently used by various manufacturers to show compliance 
with CAR sections 4b.200(c), 4b.210, and 4b.359, to which these STC's 
also were certified. This approach has previously obviated the need to 
determine the exact load distributions to each lock for the various 
container types used by operators.
    Several commenters point out that there is a vast array of 
different types of containers and other ULD's used by the affected 
operators. This includes a wide range of construction, shapes, and 
materials. Some ULD's look like boxes; others look like flat pallets or 
``cookie sheets.'' These differences significantly affect the 
distribution of loads to all locks when subjected to ``up'' loads on 
the container. Although FedEx's airplanes that have been modified in 
accordance with the affected STC's predominantly haul the full-size or 
``SAA'' container, and the half-size or ``Demi'' container, FedEx 
reported at the September 19 meeting with the FAA that its modified 
727's haul other kinds of containers, such as flat pallets, when 
necessary.
    For these reasons, the FAA's analysis used to determine the maximum 
safe payload limits for operations must conservatively account for any 
of the currently permitted container types.
    CAR section 4b.359 requires that ``each cargo and baggage 
compartment be designed for the placarded maximum weight of contents 
and the critical load at the appropriate maximum load factors 
corresponding to all specified flight * * * conditions * * *.'' CAR 
section 4b.210 requires that ``flight load requirements shall be 
complied with * * * at all weights from the design minimum weight to 
the maximum weight appropriate to each particular flight condition, 
with any practicable distribution of disposable load (mass load) within 
the prescribed operating limitations stated in the Airplane Flight 
Manual.'' CAR section 4b.200(c) requires that ``all loads [force loads] 
shall be distributed in a manner closely approximating, or 
conservatively representing actual conditions.''
    Therefore, in order to show compliance with the applicable 
regulations, either the distribution of the container loads to latches 
used to analyze the floor beam structure must be accurately determined 
for all container types used, or conservative assumptions must be used 
considering all practicable distribution of cargo loads. Finally, the 
floor structure must be strong enough to carry the maximum weight at 
the critical cargo load distribution at the appropriate maximum applied 
loads.
    As stated previously, the FAA's analysis in the NPRM's identifies 
one of several possible critical load cases--that of a large gust 
pushing the airplane down, which causes ``up'' loads on two adjacent 
containers. On all of the affected STC's, adjacent containers share the 
same set of container locks at the forward and aft edges, and these 
locks are attached to the floor structure. This condition results in 
the loads for both containers being concentrated on isolated floor 
beam(s) at the location of the locks.
    A ``typical'' full-size (88-by 125-inch) container is an enclosed 
box with two sides curved to match the rounded contour of the airplane 
fuselage, a fully or partially removable front side (i.e., a door), and 
a fixed or rigid back wall. Because of the design of a typical 
container, the back wall tends to carry the majority of the load (the 
curved sides and removable front are not as effective in supporting an 
``up'' load as the rigid back wall). A different type of ULD, a flat 
pallet, with netting to restrain the cargo, distributes the loads to 
the container locks very differently than the 88- by 125-inch 
container. The net tends to distribute the load more uniformly around 
the pallet edges.
    The rational basis for the FAA's analysis is illustrated by the 
following two examples of container/ULD arrangements that result in 
load distributions to the floor beams which approach or exceed the 80 
percent estimate used by the FAA (i.e., the converse of the estimate 
that 20 percent of the load is carried by the sidelocks). These two 
examples assume maximum allowable ULD payloads of 8,000 lbs. using 
configurations that are permitted for all of these STC's.

    Example 1: Back-to-Back Containers. Based on the data from 
FedEx's inverted container test with an ``SAA'' container facing 
(door side) forward, 43 percent of the total load was carried by the 
locks on the back side of the container. If two containers of equal 
weight are placed back to back, the equivalent of 86 percent of the 
total load of one container would be placed on the floor beam(s) at 
the interface (43 percent plus 43 percent).
    Example 2: Container and Flat Pallet. Using the test data for 
the inverted container test, 43 percent of the load would be carried 
by the back wall. A flat pallet (``cookie sheet'') placed just aft 
of this container in a cargo position, which has four sidelocks on 
each side, will place approximately 28 percent of the total load on 
the front side of the ``cookie sheet'' [as discussed previously, the 
net on the flat pallet tends to distribute the load equally to all 
sides of the sheet, and since there are five locks each on the floor 
beam(s) supporting the front and back side of the sheet, and four on 
each side, 5/18 (or 28 percent of the total load) will be on the 
front side]. This results in a total of 71 percent (43 percent plus 
28 percent) of the maximum ULD payload, being placed on the floor 
beam(s) between these two ULD's.

    These two examples of the many possible loading configurations 
illustrate the reasonableness of the FAA's estimation that 80 percent 
of the maximum allowable container payload could be concentrated on the 
floor beam(s) at the interface between two adjacent containers.
    In addition, the FAA has other concerns with FedEx's inverted 
container test. First, the effects of a critical cg shift within the 
container were not tested. As tested by FedEx, the back wall of the 
container carried 43 percent of the load with a zero percent cg shift 
(i.e., the cg of the container was at its geometric center). As 
discussed previously, this is impractical to achieve in actual 
operations. If the cg had been shifted towards the back wall of the 
container, the load at the back wall of the container would have been 
higher than the 43 percent noted previously.
    It should be noted that the FedEx test plan submitted to the FAA in 
May 1997 (Appendix 4 of FedEx's comment to Rules Docket No. 97-NM-09-AD 
submitted during the initial comment period; Document 97-034, dated May 
6, 1997) listed aft cg shift load cases on page 9 of that plan. 
However, these critical load cases were not tested because the actual 
test (described in Appendix 9) had taken place in accordance with an 
earlier test plan, Document 97-023 (which is referenced in Appendix 9). 
This was confirmed by FedEx at the September 19 meeting.
    A second concern with the FedEx inverted container test is that the 
container was tested in a fixture in

[[Page 2050]]

which the lock locations were representative of only one cargo position 
on the airplane. There are typically a maximum of 8 to 12 containers 
that may be carried on the main deck, depending on the configuration of 
the airplane. Sidelocks are evenly spaced along the fuselage, and 
different cargo container positions result in either four or five 
sidelocks along the container side edges. For these reasons, a variety 
of locations should be tested to determine the critical load case for 
the floor beams.
    A third concern is that FedEx tested cargo position 5 on the 727-
200 with the door of the container on the aft side of the cargo 
position. This orientation is opposite of how FedEx reports that the 
``SAA'' containers are usually placed in its airplanes. This 
orientation of the container in the test fixture resulted in a sidelock 
being within 4 inches of the back wall of the container. The distance 
from the front wall of the container to the nearest sidelock was 23.5 
inches. Due to this large distance, or ``overhang,'' and the 
flexibility of the ``SAA'' container, the nearest sidelock to the front 
wall on each side of the container together carried 32 percent of the 
total test load. If the container had been placed in the fixture with 
the door on the front side of the cargo position, such that the back 
wall of the container had a 23.5-inch ``overhang,'' or was in one of 
the several other cargo positions possible which have greater than a 4-
inch ``overhang'' to the backwall of the container, the loads on the 
container back wall (which are carried by the floor beams) would have 
been significantly higher.
    Finally, it is important to note that FedEx has provided no 
analysis of the floor beam structure showing that the large negative 
margins of safety are resolved based on its assertion that 35 to 40 
percent of the container load is distributed to the sidelocks. The load 
distribution is only part of the answer; the load distribution must be 
used in a stress analysis to develop data identifying stresses in the 
structural members.
    The FAA concurs that, in principal, testing of containers using a 
fixture such as that used by FedEx, if it represents the most adverse 
case of ``overhang'' for the back wall for all applicable cargo 
positions, and if it shifts the container cg to the most adverse 
position, will produce conservative results for the latches common to 
the floor beams, for the container type tested. The results will be 
conservative because of the flexibility of the floor beams, relative to 
the stiff behavior of the test fixture. The degree of conservatism is 
unknown to the FAA and has not been demonstrated by FedEx.
    FedEx, in its test, did not consider all practicable load 
distributions nor establish the critical case considering an adverse 
aft cg shift and sidelock location. FedEx tested only those containers 
or ULD's that it predominantly uses, but not all the types that it 
actually uses in service; therefore, it is impossible to draw broad 
conclusions about the behavior of many different container types, 
applicable to all cargo positions, or the degree of conservatism 
introduced by floor beam flexibility from its limited testing.
    Therefore, the FAA concludes that the 35 to 40 percent distribution 
of the ``up'' load to the sidelocks used by FedEx is artificially high. 
The FAA does not concur that the data ``Container Test,'' documented in 
Appendix 9, demonstrate that the commenter's existing sidelocks, in 
general, are effective in reacting 35 to 40 percent of the container 
load, or that the tests ``indicate that the floor structure of the 
existing main cargo deck is in compliance with the requirements of CAR 
part 4b when supporting existing weight limits.'' The test also does 
not demonstrate that the FAA's finding of unsafe condition is 
incorrect.
    2. Single ``I'' Beam Test. FedEx states that it performed a floor 
beam test on a conservative representation of an unmodified passenger 
floor beam. This test is documented in Appendix 8 of FedEx's submittal 
to Rules Docket No. 97-NM-09-AD (FedEx Engineering Report 97-049, 
Revision I/R, dated August 15, 1997), and the additional data is 
contained in Appendices 10 (FedEx Floor Beam Test, Wyle Lab) and 11 
(FedEx Floor Beam Test Videotapes).
    FedEx also states that this test showed a lower floor beam chord 
compression allowable in excess of 60 ksi (60,000 lbs. per square inch) 
just prior to failure of the floor beam. FedEx states that this value 
controverts the FAA's calculation of 40.6 ksi in the FAA's analysis. In 
addition, FedEx states that the floor beam was tested in a fixture 
designed to replicate the airplane floor support structure, and that 
the test results are conservative due to the interaction of other floor 
beams, seat tracks, and floor panels in the airplane; the benefits of 
which were not addressed during this test. FedEx states that this test 
indicates that the floor structure of the existing main cargo deck is 
in compliance with CAR part 4b when supporting existing weight limits.
    The FAA does not concur that FedEx's measurement of 60 ksi 
compressive stress is relevant to the actual strength of the floor 
beam. In the FedEx test, the 60 ksi measurement was taken just before 
the floor beam fractured in tension (i.e., stretching of the floor beam 
to the point of failure). The FAA considers that the critical failure 
mode (i.e., the failure mode that would cause collapse of the floor 
structure in actual operation) is buckling of the floor beam. Buckling 
occurs when the floor beam warps or twists under applied loads. As 
discussed below, the test data indicate that the actual compressive 
stress at which the floor beam buckled was approximately 18 ksi.
    Although the floor beam buckled during the test, the floor beam did 
not collapse, in part because the test fixture substantially and 
artificially limited the amount of warping of the beam. The test 
fixture used a rigid ``I'' beam to support the ends of the floor beam. 
This kept the ends of the floor beam from moving inward during the 
test. In contrast, on an actual airplane, the ends of the floor beam 
can move inward because they are attached to the fuselage frames, which 
are much more flexible than the rigid ``I'' beam used in the test 
fixture. The result of this artificial restraint was that the floor 
beam buckled and began to deflect. Instead of collapsing, as would be 
expected on an airplane, the floor beam behaved more like a cable, 
suspended from two rigid ends, with very little bending strength, but 
significant axial strength. This behavior was ultimately demonstrated 
by the catastrophic failure of the beam in tension, similar to a cable 
failure. If the beam had been supported as it is in the airplane, it is 
likely that the floor beam would have collapsed at the onset of 
buckling.
    For example, if a horizontal beam is supported at each end, and 
vertical loads are placed on the beam, as the beam deflects the ends 
will pull inward. Restraining the beam ends will limit the bending 
deflection and stiffen the beam, preventing collapse of the beam as it 
buckles. This artificial restraint does not affect the buckling 
capability of the beam, but it causes the beam to appear to have higher 
load carrying capability than it actually has. FedEx acknowledged the 
effect of this axial restraint in a November 4, 1997, letter to the 
FAA. FedEx stated that ``It is conceivable that the bending deformation 
of the beam * * * would be influenced by restraining the ends of the 
floor beam from translating * * *.''
    As stated previously, the critical compression buckling stress of 
the floor beam tested was approximately 18 ksi. (This occurred at the 
load step entitled ``0.6g.'') At this point the beam buckled as a 
column in the forward/aft direction.

[[Page 2051]]

Beyond this load factor, at the spanwise location left buttock line 
(LBL) 11, the beam began bending in the forward and aft direction, as 
evidenced by the detailed test data for load case number 5, 2.8 g (2.8 
times the force exerted by gravity at sea level) ``up'' load in 
Appendix 8. Forward and aft bending of the beam clearly indicates that 
the beam has buckled, and can be seen by observing the FedEx videotapes 
contained in Appendix 11. This buckling failure occurred prior to 40.6 
ksi as predicted by the FAA, and before the 49.1 ksi value predicted 
analytically by FedEx in Appendix 1.
    The occurrence of buckling at 18 ksi rather than approximately 40 
ksi can be explained by the ineffectiveness of the stability straps in 
the test fixture. Over most of the airplane, the floor beams extend 
from one side of the airplane to the other. A stability strap is a 
long, thin strip of metal, running perpendicular to the floor beam, and 
attached to the lower surface of several beams, at intervals ranging 
from 17 to 24.75 inches along the lower surface of the floor beam. The 
purpose of the stability straps is to support or stabilize the lower 
chord to strengthen the floor beam. This is accomplished by reducing 
the ``effective length'' of the lower chord of the beam from one long 
column (the entire length) by splitting it into a series of shorter, 
stiffer columns that are equal in length to the distance between the 
stability straps. The stability straps in the test model were 
ineffective because the portion of the test fixture to which the straps 
were attached was not stiff enough to allow the straps to fully 
stabilize the floor beam. (This is exactly the opposite problem from 
that described above with respect to the excessive rigidity of the test 
fixture where the floor beam ends were attached.)
    By graphing the results obtained from the test, the FAA determined 
that the stability straps were not fully effective at the location 
where the beam buckled. This graphing demonstrated that the ``effective 
length'' of the floor beam lower chord at the point of buckling was 
40.4 inches [between LBL 32.6 and right buttock line (RBL) 7.8], rather 
than the ``effective length'' of 24.75 inches used in the analyses 
conducted by FedEx and the FAA. Since the ``effective length'' was 
longer for the tested beam due to the ineffectiveness of the stability 
straps, the resulting column was weaker and buckled at a lower stress 
than would occur on the affected airplanes.
    The FAA subsequently used the same analytical techniques used in 
its previous analysis to confirm that the buckling strength of the beam 
is approximately 20 ksi based on the effective column length of 40.4 
inches demonstrated by the FedEx tests. This correlates well with the 
stress at buckling of 18 ksi measured in the tests and confirms the 
validity of the FAA's analysis.
    During the September 19, 1997, meeting, and at the February 18, 
1998, public meeting, FedEx concurred with the FAA that the stability 
straps buckled during the test, and were largely ineffective, as the 
straps could not provide stability to the lower chord.
    At the public meeting on February 18, 1998, two FedEx consultants 
made presentations regarding this test. Both consultants agreed that, 
although the test was properly performed in accordance with the test 
protocol, the test fixture was not representative of the airplane. As a 
result, one of the consultants (Dr. Foster of Auburn University) stated 
that it would be inappropriate to draw conclusions from this test for 
the airplane floor beam.
    Based on the discussion above, the FAA concludes that FedEx's 
``Single I Beam Test'' does not demonstrate a lower chord stress 
capability greater than that calculated by the FAA, or that the 
existing main cargo deck is in compliance with the requirements of CAR 
part 4b when supporting existing weight limits. The test also does not 
demonstrate that the FAA's finding of unsafe condition is incorrect.
    3. ``On-Aircraft'' Test. FedEx states that an ``on-aircraft'' test 
was conducted (Appendix 12, Report 97-052, Revision I/R, dated August 
27, 1997), and that this test demonstrated that the container/airplane 
combination withstood an applied ``up'' load of approximately 20,000 
lbs. FedEx states that this test indicates that the floor structure of 
the existing main cargo deck is in compliance with the requirements of 
CAR part 4b when supporting existing weight limits. FedEx also states 
in Section 6 of Report 97-051, also in Appendix 12, that a margin of 
safety of 2.1 was demonstrated with a 10,700-lb. container.
    The FAA does not concur that this test demonstrates that the 
airplane is safe and in compliance with CAR part 4b. The test also does 
not demonstrate that the FAA's finding of unsafe condition is 
incorrect. The ``on-aircraft'' test consisted of FedEx's ``SAA'' or 
full-size container, situated on the main cargo deck of a 727, 
restrained vertically by the forward and aft pallet locks (attached to 
the floor beams), and side vertical restraints (sidelocks). The 
container was modified to place four ``I'' shaped beams running 
lengthwise through the container. Four hydraulic jacks were positioned 
underneath the ``I'' beams on either side of the container and attached 
to jacking platforms on the main deck floor. The jacks were used to 
apply ``up'' loads to the container, as is shown in Figure 2.1 of 
FedEx's Report 97-051 (Appendix 12 of FedEx's submittal to Rules Docket 
No. 97-NM-09-AD). To transmit the loads applied to the ``I'' beams to 
the container, a rigid structure made of seventy-two 4- by 4-inch thick 
wood beam spacers, and thirty-eight \3/4\-inch thick plywood sheet 
formers curved at the edges to match the contour of the container, were 
fastened with screws to the 0.063-inch thick aluminum skin of the 
container. This structure, weighing approximately 1,400 lbs., provided 
a rigid platform for the ``I'' beams to lift the container (details of 
the plywood structure and its estimated weight are provided in Figure 
2.3 of Report 97-051, Appendix 12).
    The FAA has determined that the ``I'' beams and rigid structure 
used to introduce ``up'' load into the container artificially limited 
the distortion of the container under load and forced most of the 
applied load to the sidelocks and away from the floor beams. This is 
unconservative for the floor beams because it results in the test not 
representing how an actual loaded container or other ULD would affect 
the loads on the floor beams.
    During the September 19 meeting, FedEx agreed that in the ``up'' 
load case, if the container is loaded and not restrained by the rigid 
structure, it attempts to deform to a catenary (arched) shape at the 
front of the container where the door is located. This effect is 
demonstrated by FedEx's inverted container test described in Appendix 
9. FedEx also stated, however, that this would have no effect on the 
test results, although it was considering the use of airbags or 
hydraulic bags instead of the rigid structure to allow the ``SAA'' 
container to behave as it did in the test documented in Appendix 9. 
FedEx also stated in the meeting that it believed that testing to 2.5 
g's, or 20,000 lbs. of ``up'' load, helps to account for the load being 
``beamed'' or forced to the sidelocks.
    The test results indicated that over 80 percent of the load was 
directed to the sidewalls of the container and, therefore, to the 
sidelocks rather than the floor beams. The FAA finds that this effect 
results from the rigid structure used to introduce the load into the 
container, and that this renders the test unrepresentative of the 
actual loading of the floor beam and significantly unconservative.

[[Page 2052]]

    Even though the FAA determined that the results of the inverted 
container test (Appendix 9 of FedEx's comment) were unconservative, it 
showed that the percentage of the load carried by the back wall of the 
container was approximately three times greater than that determined by 
the ``on-aircraft'' test. The loads carried by the rigid back wall are 
largely carried by floor beam(s) locks, not the sidelocks. These 
results also contradict FedEx's conclusion that the ``on-aircraft'' 
test demonstrates that the floor structure is safe. The ``on-aircraft'' 
test provides confidence in the strength of FedEx's sidelocks. However, 
because of the artificial shifting of the loads from the floor beams to 
the sidelocks, the test fails to demonstrate that the floor structure 
is safe. Further, the ``on-aircraft'' testing to 2.5 g's did not result 
in the application of significant loading to the floor beams. 
Therefore, the results of the testing to 2.5 g's is of little 
significance when addressing the unsafe condition of the floor beams.
    In Appendix 1 of FedEx's April 30, 1998, submission to Rules Docket 
No. 97-NM-09-AD during the reopened comment period, FedEx appears to 
now recognize the effect of the rigid plywood formers in forcing the 
load to the sidelocks and away from the floor beams. In this Appendix, 
on page 2 of the FedEx Engineering Report 98-026, Revision A, FedEx 
states ``Measured loads for the container perimeter latch locations 
indicate that 40% of the applied load was reacted on each side by the 
side latches (see Reference 3). This is due to the fact that the rigid 
formers did not allow the top of the container to deform as it would 
during actual conditions and thereby forced more load outboard than 
what would be typically encountered during flight.''
    In summary, based on the previous discussion, the FAA does not 
concur that this test demonstrates that the airplane is safe and in 
compliance with CAR part 4b. The test also does not demonstrate that 
the FAA's finding of unsafe condition is incorrect. One commenter 
states that he participated in FedEx's ``on-aircraft'' test. He states 
that the data from the latch load cells were inconclusive for the 
tests, and although he considered the test to be a reasonable 
representation of airplane conditions, he suggests that FedEx improve 
the latch load cell installation and data acquisition system and 
investigate whether the plywood formers used to apply the test load to 
the container roof could influence the latch load distribution. As 
discussed previously, the FAA does not concur that the ``on-aircraft'' 
test was representative of the airplane, but concurs that the plywood 
formers influenced the load distribution.

First Container Facing Aft

    Two commenters state that positioning the first container aft of 
the 9g cargo barrier with the door facing forward is not optimum from a 
crashworthiness perspective and request that the AD specify that this 
container be facing aft instead. The FAA concurs. Paragraphs (a) and 
(b) of the final rule have been revised to allow the first container 
aft of the bulkhead to face aft, with all other containers facing 
forward.

Increased Running Load

    One commenter states that the following statement in the NPRM is 
factually inaccurate: ``This running load of 90 pounds per inch is a 
safety concern, as it is approximately 2.6 times higher than the 
maximum running load of 34.5 pounds per inch allowed on these same 
floor beams when the airplane was in a passenger configuration.'' The 
commenter states that in a negative gust (``up'' load) situation the 
passenger floor beams must act to restrain upper deck loads and lower 
deck cargo loads simultaneously and, as a result, must react 81.0-lbs. 
per inch, not just the 34.5 figure as the NPRM indicates. The commenter 
maintains that if reduced loads are necessary to maintain the safety of 
cargo airplanes, then passenger airplanes should be similarly 
restricted.
    The FAA does not concur that the passenger and cargo airplanes 
present similar safety concerns. The NPRM statement quoted by the 
commenter appeared in the section of the NPRM that described the FAA's 
reasons for undertaking the detailed design review which led to the 
conclusion that there is an unsafe condition. The statement in the NPRM 
is factually accurate for the running loads and the ``down'' load case 
and contributed to the FAA's concern with the strength of an 
unreinforced cargo floor.
    The FAA subsequently determined that the ``up'' load case is the 
most likely critical case. The FAA agrees that, for the ``up'' load 
case, the running load figures identified in the comment are accurate. 
However, the passenger compartment is designed to uniformly distribute 
passenger loads such that every floor beam is active in carrying these 
loads. In contrast, the freighter floor loads are applied differently. 
Instead of the main deck loads being applied uniformly, each 88-inch 
deep container spans several floor beams. As discussed previously, the 
result of this is that only floor beams located at the edges of 
containers are active in carrying the ``up'' loads. Hence, as the FAA 
determined in its detailed design review, the effect on the airplane is 
that the 90 lbs. per inch cargo container loading is much more critical 
than the uniformly applied upper and lower deck loads of the passenger 
configuration and is, in fact, a safety concern.
    One commenter states that the interim weight reduction is too 
restrictive considering that the passenger 727 can carry in excess of 
6,800 lbs. in the same zone. The 3,000-lb. limitation imposed in the 
NPRM is unjustified. The FAA does not concur. As discussed previously, 
the loading on the floor is significantly different depending on 
whether it is loaded by the carriage of passengers or containers. The 
3,000-lb. limitation specified for the carriage of cargo in the NPRM is 
justified by the FAA's analysis provided in the Rules Dockets.

Netted Lower Lobe Cargo

    One commenter states that if the lower lobe cargo is assumed to be 
netted (restrained), it would not have any relevance in a down gust 
situation. The FAA infers that the commenter believes that, as the 
cargo would be restrained to the belly of the airplane, it would not 
load the underside of the floor beams in a negative ``g'' environment 
due to a down gust.
    Another commenter states that the NPRM should be changed to allow 
lower lobe weights to be subtracted from the main deck limits if the 
load is properly tied down. The FAA concurs partially. If the lower 
lobe cargo is properly tied down, it will be restrained by the 
structure differently than represented in the FAA analysis. While the 
FAA is not currently aware of configurations that restrain lower lobe 
cargo, paragraphs (f) and (g) of this AD allow for approval of this 
type of configuration as an alternative method of compliance with the 
final rule.

Airplane Weight Increases

    One commenter states that the FAA should reconsider the present 
policy of withholding approval of maximum take-off weight (MTOW) and 
maximum landing weight (MLW) increases for 727 freighter modified 
airplanes. The rationale for this is that the resulting higher weights 
would allow greater fuel loads for remote region operators, and also 
would increase the safety margin of the airplane's modified fuselage 
structure, which is the FAA's prime concern addressed by the NPRM's. 
The FAA infers that the commenter believes

[[Page 2053]]

that the proposed AD should be changed to reflect this.
    The FAA concurs partially. The FAA concurs that maintaining a 
minimum in-flight weight reduces the loads resulting from vertical 
gusts, unless this additional weight is carried in body fuel tanks that 
are suspended from floor beams. Additional loads to the floor beams 
exacerbate the unsafe condition. This issue is addressed appropriately 
in the context of type certification and is not addressed in this AD. 
Therefore, the FAA has determined that no change to the final rule is 
necessary.

Operators' Ability To Determine Container CG's

    One commenter states that there is no means to measure or comply 
with the requirement that the container cg's be within +/- 10 percent 
of the geometric center of the container. Two commenters state that the 
wording in the proposed AD should be changed to allow those operators 
having a loading procedure that maintains the container cg within +/- 
10 percent to be considered compliant with this requirement. The FAA 
does not concur that the cg of the container cannot be determined, or 
that the requirement to maintain the cg within 10 percent of the 
horizontal cg cannot be complied with. For example, FedEx has recently 
acquired equipment for this purpose. Because the cg location within the 
container has a major effect on the loads imposed on the floor beams, 
the FAA considers that this limitation is necessary to address the 
unsafe condition. It should be noted that the vast majority of cargo 
containers are certificated to TSO C90c, which specifies a maximum cg 
shift of 10 percent. Therefore, operators should always have been 
ensuring that the cg shift did not exceed this limitation in the TSO.
    One commenter submitted data to the Rules Dockets that the 
commenter states will allow an operator with a properly designed or 
modified scale to accurately determine, display, and record the 
container cg. The FAA did not evaluate the technical accuracy of the 
submission, as no change to the proposed AD was requested by the 
commenter.

Airplanes With Apparent Increased Floor Capability

    One commenter states that one of its 727-200 airplanes has a 
greater running load allowable than its other two airplanes (37.5 lbs. 
per running inch versus 34 lbs. per running inch) and asks why this 
airplane is limited by the same restriction.
    The FAA infers that the commenter believes that its airplane should 
have higher allowable container loads, based on this apparent increased 
capability, and that the AD should be changed to reflect this. The FAA 
does not concur. From its analysis, the design review team determined 
that the 727 main cargo decks are capable of supporting a maximum 
payload of approximately 3,000 lbs. per container. Paragraphs (f) and 
(g) of the AD allow for an applicant to propose new payloads along with 
substantiating data and analysis. No change to the final rule is 
necessary.

Inconsistent Limitations

    One commenter states that the FAA's determination that these 
airplanes are capable of supporting only 3,000 lbs. per container is 
entirely inconsistent with the FAA's interim proposal, which would 
allow an 8,000-lb. pallet in any position where the entire load would 
be carried by one set of container locks. The commenter does not see 
any rational or consistent approach in the NPRM's. The FAA does not 
concur. The analysis that resulted in the 3,000 lb. per container limit 
was based on the current operational limits of the airplane. As 
discussed in the NPRM, the FAA determined that, if more restrictive 
operational limits are imposed, a higher payload could be allowed on an 
interim basis. The FAA has estimated that the airplane gust loads will 
be reduced with limitations on in-flight weight and maximum operating 
airspeed to the extent that the 3,000-lb. limit per container can be 
raised to 4,000 lbs. for the interim period.
    For the ``up'' load case, two 4,000-lb. containers placed back-to-
back, without side vertical restraints, impose approximately the same 
amount of load on the floor structure as a single 8,000-lb. container 
with the adjacent cargo positions carrying no payload. Because of this, 
for the interim period, the operator would have the flexibility to 
carry an 8,000-lb. container, provided the containers on either side 
are empty.
    If side vertical restraints acceptable to the FAA are installed, 
then the interim payload is not to exceed a total weight of 9,600 lbs. 
for any two adjacent containers. In this case, as stated in paragraph 
(b) of the AD, the 8,000-lb. limit per container would still apply. 
Many of the different containers and flat pallets or ``cookie sheets'' 
used by operators require side vertical restraints, as specified in TSO 
C90c.

Irrelevancy of Model 747 Problems

    One commenter states that the FAA only proposed payload reduction 
because of the incidents occurring on 747's, but the FAA has no reason 
to believe the problems found on the 747's will occur on the 727's. The 
FAA does not concur. The FAA did, in fact, look into the 727 
conversions because those conversions had been performed by some of the 
same companies and with similar procedures and design methods as some 
747's which had been found to be unsafe. The unsafe condition that is 
the subject of this AD, however, is specific to the 727 and has been 
documented in the Rules Dockets.

Applicability of 14 CFR 25.1529

    One commenter states that the NPRM statement indicating that STC 
holders are required to issue Instructions for Continued Airworthiness 
in accordance with 14 CFR 25.1529 does not apply to its STC's because 
the applicable airworthiness standards for the 727 are CAR part 4b, 
rather than 14 CFR part 25. The FAA does not concur. Since January 28, 
1981, 14 CFR 21.50(b) has required that the holder of an STC for which 
application was made after that date shall furnish the Instructions for 
Continued Airworthiness prepared in accordance with 14 CFR 25.1529. 
This requirement is effective regardless of the specific certification 
basis of the airplane.

Fatigue Cracks as Evidence of Unsafe Condition

    FedEx states that, if the FAA's report of huge negative margins of 
safety at ultimate load are true, then the ``typical daily operating 
conditions would still impose substantial loads on the structure,'' and 
result in wear and cracking of the floor structure. FedEx's review of 
the FAA service difficulty report data generated only two reports of 
cracks on the converted 727 freighters, and no other damage was found 
that could be attributed to the 727 cargo conversion modification.
    The FAA does not concur that a low number of in-service difficulty 
reports indicates that the FAA's finding of unsafe condition is 
unfounded. FedEx has reported that its average cargo load density is 
approximately 7.5 lbs. per cubic foot, which equates to an average 
cargo payload of approximately 3,300 lbs. per container. This results 
in stress levels that on average would be similar to those of a 
passenger 727. Therefore, it is not expected that fatigue cracks would 
develop in only 11,008 total flight cycles, which is the highest number 
of cycles accumulated (as of August 27, 1998) by any FedEx 727 airplane 
since conversion to a freighter configuration. As discussed previously, 
the unsafe condition addressed in these AD's is not a result of 
fatigue, but is the

[[Page 2054]]

result of the existing floor structure not being able to support the 
allowable payloads and distributions for the critical gust conditions.

Data Showing Floors To Be Safe

    FedEx states that the NPRM is inaccurate in stating that the FAA 
design review team was unable to find any data which showed that the 
floors were safe for the heavier (than passenger loading) freight 
payloads. FedEx states that the FAA has received and accepted data 
verifying the safety of the floor structure. FedEx also states that the 
FAA has failed to provide ``reasoned explanation'' for not approving 
various documents.
    The FAA does not concur. In performing its own analysis, the FAA 
was careful to use only methodologies that were commonly employed in 
industry. One of the ways that the reasonableness of the FAA analysis 
contained in the Rules Dockets was checked was to compare the results 
with results of the STC holders' analyses, where possible. In this 
case, several analysis documents (Dee Howard Reports R90-2, R90-4, and 
R90-6) were used by FedEx to analyze the main deck floor beams in 
support of its STC for half-size containers (SA7447SW). However, these 
documents do not ``verify that the unreinforced floor structure of the 
main cargo deck can safely support the heavier freighter payloads.'' 
Also, they do not address all of the critical load cases or 
configurations, nor do they address the effect of cg shifts.
    Recognizing these limitations, the FAA used FedEx's methodology to 
verify that the FAA analysis yielded similar results for a similar load 
case. In doing this, the FAA used the load case which placed ``down'' 
loads on the containers, as provided in FedEx's analysis, as its 
analysis did not contain an ``up'' load case (as required by CAR part 
4b standards). Using the applied loads from FedEx's ``down'' load case, 
the FAA calculated the margins of safety for the floor beams using the 
FAA's documented methodology. The results for the mid-span of the floor 
beam matched very closely to those documented in FedEx's STC analysis 
for the half-size containers, which verifies that the FAA's and FedEx's 
analytical methodologies were quite similar for the same load case.
    However, because FedEx's (Dee Howard) documents do not address all 
the critical load cases, locations on the floor beam, or 
configurations, nor do they address the effects of cg shifts, they do 
not ``verify the safety of the floor structure.''
    In addition, of the ten documents related to the floor beam 
analysis testing that FedEx submitted in its comments, three documents 
(Appendices 1, 2, and 3) describe analytical methodologies and do not 
(and are not intended to) ``show the floor structure can safely support 
the heavier payloads.'' Regarding the decompression methodology 
document submitted in Appendix 3, FedEx acknowledged at the September 
19, 1997, meeting that it had not yet revised the document following 
comments received from the FAA at a meeting held between FedEx and the 
FAA on July 24, 1997.
    Three other documents (Appendices 4, 8, and 9) are test plans or 
results that have been discussed previously and also do not ``show the 
floor structure can safely support the heavier payloads.''
    The two external loads documents (Appendices 5 and 6) have been 
approved by the FAA prior to FedEx's comment submittal (FAA letter 97-
120S-534, dated August 21, 1997) and are considered appropriate as a 
starting point for an analysis of the floor structure. However, these 
documents by themselves do not ``verify the safety of the floor 
structure.''
    Appendix 12 includes a document containing an incomplete analysis 
of one floor beam, a test report which was discussed previously, and 
two videotapes of that test, none of which ``verify the safety of the 
floor structure.'' Finally, FedEx's Document ER 97-035 I/R, dated July 
20, 1997 (Appendix 7), which was approved by FedEx on August 13, 1997, 
had not been submitted to the FAA prior to its inclusion in FedEx's 
comment submittal. In reviewing this document, the FAA has determined 
that because the area addressed is shorter than an 88-inch container, 
this document alone does not substantiate higher container loads. The 
floor under the rest of the container also would need to be 
substantiated to warrant a change to the AD limits.
    The FAA does not concur that it has received and accepted data 
verifying the safety of the floor structure, or that the FAA design 
review team was in possession of any data which showed that the floors 
were safe for the heavier (than passenger loading) freight payloads. 
Finally, the FAA does not concur that it has failed to provide FedEx 
with a ``reasoned explanation'' for not approving various documents. 
FedEx is aware of the current status of all the above mentioned 
documents.
    FedEx also states that a Boeing letter (Appendix 41) indicated that 
the floor beams were safe for a passenger to freighter airplane 
conversion at (container) weights of 8,000 lbs. The FAA does not 
concur. The referenced letter was part of an initial budget quote for a 
zero fuel weight increase that estimated potential weight increases 
that might be applicable to airplanes converted from passenger to 
freighter configurations. Simplifying assumptions were used by Boeing 
in order to allow FedEx to quickly establish, as a rough approximation, 
the financial feasibility of converting an airplane. Any necessary 
changes to the floor beams in estimating the weight of the airplane 
following conversion were not addressed.

FedEx's Finite Element Model

    FedEx states that the FAA misused FedEx's finite element model 
(contained in Engineering Report 8504), which identifies negative 
margins of safety in the fuselage monocoque, to substantiate its 
finding of unsafe condition. FedEx also states that the NPRM was 
inaccurate in stating that the report was used for certification. The 
FAA does not concur. The FAA did not use FedEx's Engineering Report 
8504 to validate its analysis. Rather, as discussed previously, the FAA 
used the floor beam analysis documents submitted as part of the 
substantiation for FedEx's STC for half-size containers (SA7447SW) to 
validate its analysis. The NPRM did state that the original STC 
certification data contained documented negative margins of safety. The 
FAA does not concur that this statement is incorrect. At the meeting 
held September 19, 1997, FedEx stated that the document was used to 
support original STC issuance, and that no other document was 
submitted.

Critical Loading on Floor Beams

    FedEx states that, contrary to a statement in the NPRM, the FAA has 
not established that floor beams at the forward and aft edges of the 
container are more critically loaded. In its August 28, 1997, submittal 
to Rules Docket No. 97-NM-09-AD, FedEx cited its ``on-aircraft'' test 
as proof that the sidelocks are more critically loaded. FedEx appears 
to have mistakenly inferred that this statement addresses the 
effectiveness of FedEx's sidelocks. This inference is incorrect. In 
context, this statement simply points out that, for the ``up'' load 
case, ``the floor beams at the forward or aft edges of the containers 
would be more critically loaded'' than the floor beams under the center 
of the container. The reason for this is that a full-size container is 
restrained against vertical movement by the container locks attached to 
the floor beams at container edges and there are no

[[Page 2055]]

container locks in the center of the container.

Communications With FAA

    FedEx's comments included a number of disagreements with 
documentation of various communications prepared by the FAA and placed 
in Rules Docket No. 97-NM-09-AD. Because these comments do not relate 
to the merits of this AD, they are not addressed in this final rule. 
However, the FAA has provided a response to these comments in that 
Rules Docket.

Interim Limitations Already Observed

    One commenter states that the interim operating limitations are not 
necessary because the commenter does not know of a 727 freighter STC 
that allows operation higher than 350 knots indicated airspeed (KIAS) 
and, for practical reasons, 727-200 airplanes almost never operate at 
weights below 100,000 lbs. The FAA does not concur. While many of the 
affected airplanes are subject to a maximum operational speed 
limitation of approximately 350 KIAS, other affected airplanes are not 
subject to such limitations and do operate at higher speeds. In 
addition, while operation at weights below 100,000 lbs. is not likely 
for most 727-200 converted freighters, such operation is permitted and 
may occur. Such operation is even more likely for the lighter weight 
727-100, which also is subject to this AD.

Alternatives to Limitations in the AD

    Several commenters asked about alternatives to the proposed rule 
and suggested increased inspections, such as those in other AD's. The 
FAA does not concur. The unsafe condition identified in the AD is not 
based on loads imposed on the floor structure on an average flight 
(i.e., fatigue-type loading). The unsafe condition is caused by loads 
experienced on the airplane due to a large gust while carrying certain 
cargo payloads and distributions. In this case, a floor beam failure or 
excessive deflection would likely result in the loss of the airplane. 
Because such a failure would not necessarily be preceded by cracking, 
inspections of the airplane would not prevent the failure. The only 
means for preventing a catastrophic event is to limit the flight 
operation of the airplane and/or the container payloads.
    One commenter proposes a statistical approach to study the unsafe 
condition by requiring certain inspections over the next year while 
imposing certain operational limitations. The FAA does not concur. 
Because the unsafe condition is a collapse of the floor caused by large 
gusts, increased inspections in the areas of concern will not serve to 
lessen the likelihood of loss of the airplane.
    One commenter proposes that the FAA revise the proposed AD to 
further limit the maximum operational speed to 280 KIAS as an 
alternative to payload limitations. The FAA does not concur with the 
commenter's proposal to reduce the maximum operational speed to 280 
KIAS. Reducing the maximum operational speed levels below 350 KIAS does 
reduce the gust loads on the airplane. However, speed restrictions 
below 350 KIAS that permit safe operation of the airplane do not affect 
the maneuver loads, which at these speeds become more critical than the 
gust loads.

``Mode B''

    One commenter requests that, for the interim limitations, the FAA 
also allows operation at ``Mode B'' [350 knots equivalent airspeed 
(KEAS)] for the maximum operating airspeed (Vmo). The commenter states 
that operations at ``Mode B'' would be more convenient than the 350 
KIAS limitation specified in the proposed AD. The FAA concurs. The FAA 
has revised the interim limitations of the final rule accordingly.

Release of Proprietary Data

    Several commenters state that the FAA must divulge all data used to 
make its finding of an unsafe condition; the commenters cited various 
legal cases.
    The FAA infers that commenters are insisting that the FAA release 
relevant proprietary data that was considered by the FAA during this 
rulemaking. The FAA does not concur for two reasons. First, the Trade 
Secret Act (18 U.S.C. 1905) prohibits the disclosure of such data, and 
this prohibition is not overridden by the requirements of the 
Administrative Procedure Act (APA). The cases cited by the commenters, 
while generally stating that agencies must release all information on 
which they rely during rulemaking, do not address the prohibition 
against the release of trade secret data.
    Because AD's address unsafe conditions associated with aeronautical 
products, the FAA routinely evaluates proprietary design data in 
determining whether AD's are necessary. In determining whether such 
material should be placed in the Rules Docket, the FAA applies the 
standards developed under the Freedom of Information Act (FOIA; 5 
U.S.C. 552) in the application of Exemption 4 [Sec. 552(B)(4)], which 
protects ``trade secrets and commercial or financial information 
obtained from a person and privileged or confidential.'' If data are 
determined to meet those standards, they are not placed in the Rules 
Docket, but are retained in separate files that are not released to the 
public. Apart from violation of the Trade Secret Act, if the FAA were 
to release such data, it would be much more difficult for the FAA to 
obtain the data on which its findings of unsafe conditions are 
necessarily based.
    Second, the APA generally has been interpreted as requiring that 
agencies provide the public with a meaningful opportunity to comment on 
proposed rules. In this rulemaking, the FAA has fully complied with 
this requirement, even without releasing trade secret data. In 
developing the NPRM, the FAA used proprietary Boeing loads data in its 
analysis, from which the FAA identified the existence of the unsafe 
condition. Although Boeing has not consented to releasing these data, 
FedEx has submitted comparable loads data (discussed previously under 
the heading, ``Extension of Interim Operational Period'') which, when 
used in the FAA analysis (which has been placed in the Rules Dockets), 
also demonstrate the existence of the unsafe condition. FedEx did 
consent to the release of these data. In fact, at the first public 
meeting on February 18, 1998, the FAA used these data in its 
presentation explaining its analysis. The analysis and the presentation 
are fully documented in the Rules Dockets, and have been available for 
review by commenters. The FAA also has referenced other proprietary 
data, which have been submitted by applicants seeking approval for 
modifications to correct the unsafe condition, as confirming the FAA's 
analysis. Although these data are relevant to the rulemaking, they do 
not provide the basis for the FAA's action, and their release would not 
significantly increase the meaningfulness of the public's opportunity 
to comment on the FAA's proposal.
    One commenter requests copies of three recently updated Boeing 
computer programs which it believes were utilized by the FAA in 
determining the container payload limits specified in the NPRM. The 
commenter states that those programs are entitled: (1) ``Vertical Gust 
Load Factors 'Gs;'' (2) ``727 Movement (sic) of Inertia Model;'' and 
(3) ``Operating Empty Weight Plus Payload Distribution.'' The FAA is 
not aware of the referenced programs, does not have them, and did not 
use them in its analysis.

Economic Analysis

    Several commenters state that the FAA underestimated the cost to 
modify the airplane floor structure into

[[Page 2056]]

compliance to CAR part 4b, citing a Pemco estimate of $400,000, as 
opposed to the $100,000 estimate contained in the NPRM. Several 
commenters also state that the FAA had underestimated (1) the loss in 
revenue due to the reduced allowable payloads, and (2) the amount of 
time necessary to get all airplanes modified due to the short 120-day 
interim period, a lack of FAA-approved fixes, and the limited 
availability of facilities to install the modifications within the 120-
day period proposed by the NPRM.
    The FAA concurs. The FAA used data supplied by industry to conduct 
its cost and regulatory flexibility analysis used in the NPRM and has 
considered the data supplied by commenters during the comment period to 
conduct the cost and regulatory flexibility analysis used for the final 
rule.

Cost-Benefit Analysis

    One commenter states that the FAA must undertake a thorough cost-
benefit analysis and economic impact assessment in conjunction with its 
consideration of the remedial actions at issue in this rulemaking. The 
commenter states that the FAA has thus far failed to conduct an 
adequate cost-benefit analysis. The commenter states that a cost-
benefit analysis and economic impact assessment are required by the 
provisions of the Regulatory Flexibility Act.
    The FAA does not concur. As discussed below under the heading 
``Regulatory Evaluation Summary,'' the FAA has performed an extensive 
analysis of the costs and benefits of this AD and has fulfilled the 
requirements of the Regulatory Flexibility Act.

Combi Airplanes

    One commenter states that the NPRM has not considered those 
operators that operate airplanes in a combi mode (a combi airplane has 
provisions for passengers and cargo on the main deck in separate 
compartments). The commenter also states that it assumes that the load 
restrictions would not apply to the floor structure which is used to 
carry passengers and that the original manufacturer's limitations are 
applicable. The FAA concurs. Although the commenter is correct with 
respect to floor structure carrying passengers, combi airplanes 
transporting containers on the main deck must be in compliance with the 
limitations specified in this AD.

Applicability of Proposal

    FedEx points out that the wording of the applicability in the AD 
could easily be misconstrued as also applying to airplanes manufactured 
as freighters by the original equipment manufacturer. The FAA concurs 
and has revised the applicability of the final rule to read ``Model 727 
series airplanes that have been converted from a passenger to a cargo-
carrying (``freighter'') configuration in accordance with Supplemental 
Type Certificate ST00015AT; certificated in any category.''

Other Cargo Lock Devices

    One commenter requests that the proposed AD be revised to add a 
paragraph discussing a ``special load-alleviating cargo container 
lock'' for which the commenter has applied for an STC at the FAA, Los 
Angeles ACO. The commenter reports that this lock will allow for the 
carriage of 16,000 lbs. rather than 8,000 lbs. in two adjacent 
containers, as specified in the proposed AD, but to be conservative, 
the commenter requests that the rule allow 12,000 lbs. for two adjacent 
containers for the interim period. During the reopened comment period, 
this commenter submitted additional information in support of its 
original comment.
    The FAA does not concur. The information submitted is not 
sufficient to substantiate the safety of the airplane with the locks 
installed. This lock is the subject of an STC application and is not 
currently FAA-approved. Paragraphs (f) and (g) of the AD provide for 
approval of alternative methods of compliance to address potentially 
alleviating devices for the unsafe condition. The commenter may obtain 
such an approval upon submission of data substantiating that the 
referenced device provides an acceptable level of safety. Therefore, no 
change to the final rule is necessary.

``Fine Tune'' the AD

    The CAA and others request that the AD should be ``fine tuned'' 
after issuance, as new data become available. The FAA does not concur 
that ``fine tuning'' of the AD is necessary. Paragraphs (f) and (g) of 
the AD allow for approval of alternative methods of addressing the 
unsafe condition when substantiated properly. As with any AD, if new 
information indicates that changes to the AD itself are needed, the FAA 
has the authority to revise or supersede this AD.

Request for Clarification

    One commenter requests clarification of the procedures that will be 
used to obtain future FAA approvals with respect to this rulemaking and 
to inform the public of those approvals.
    As stated in the final rule, all submissions should be made to the 
Atlanta ACO. The Transport Airplane Directorate has established a team 
consisting of members from several ACO's to review all requests in 
accordance with paragraphs (e) and (f) of this AD. In all other 
respects, the process for approvals under this AD will be similar to 
that followed for all AD's. For example, in order to protect 
applicants' proprietary data, the FAA will notify only the applicant 
for an approval of the FAA's decision; while the FAA will disclose 
whether approvals have been granted, requests for approved data would 
be handled under normal FOIA procedures.

Other Safety Improvements

    One commenter states that, because this AD will necessitate large 
expenditures and does not address an unsafe condition, requiring 
compliance with it will prevent the affected airlines from adopting 
other less costly and more effective safety enhancements, such as 
updating flight deck equipment. The FAA does not concur. As discussed 
previously, this AD addresses a serious unsafe condition. Although 
correcting this condition may be expensive, the FAA has determined that 
it must be corrected to ensure an acceptable level of safety.

Petitions for Reconsideration

    In addition to their comments, several commenters also filed 
``Petitions for Reconsideration'' in accordance with 14 CFR 11.93. 
Because these petitions were filed prematurely, the FAA considered them 
as comments to the Rules Docket. However, because the substance of the 
petitions is repetitious of the more extensive comments submitted by 
FedEx and others discussed above, the petitions are not discussed 
separately in this final rule.

Explanation of Change of Aircraft Certification Office Contact

    The FAA has changed the point of contact for obtaining further 
information, for obtaining FAA approval of certain actions, and for 
submitting substantiating data and analyses in accordance with the 
provisions of this AD, due to relocation of certain STC holders.

Conclusion

    After careful review of the available data, including the comments 
noted above, the FAA has determined that air safety and the public 
interest require the adoption of the rule with the changes previously 
described. The FAA has determined that these changes will

[[Page 2057]]

neither increase the economic burden on any operator nor increase the 
scope of the AD.

Participation at the Public Meeting on the Final Rule

    Requests from persons who wish to present oral statements at the 
public meeting should be received by the FAA no later than 5 days prior 
to the meeting. Such requests should be submitted to Mike Zielinski as 
listed in the section titled FOR FURTHER INFORMATION CONTACT above, and 
should include a written summary of oral remarks to be presented, and 
an estimate of time needed for the presentation. Requests received 
after the date specified above will be scheduled if there is time 
available during the meeting; however, the names of those individuals 
may not appear on the written agenda. The FAA will prepare an agenda of 
speakers that will be available at the meeting. To accommodate as many 
speakers as possible, the amount of time allocated to each speaker may 
be less than the amount of time requested. Those persons desiring to 
have available audiovisual equipment should notify the FAA when 
requesting to be placed on the agenda.

Purpose of Public Meeting

    Because of the high degree of public interest in this AD, the FAA 
has scheduled a public meeting to discuss its content and issues 
relating to compliance. The FAA's objective is to ensure that all 
affected operators and design approval holders have a full 
understanding of the issues addressed in the AD and of the actions 
necessary to comply with it. The FAA anticipates that, following this 
meeting, there will continue to be extensive discussions between the 
affected parties and the FAA for the purpose of identifying and 
implementing the most timely and cost-effective means to eliminate the 
unsafe condition addressed in this AD.

Public Meeting Procedures

    Persons who plan to attend the public meeting should be aware of 
the following procedures that have been established for this meeting:
    1. There will be no admission fee or other charge to attend or to 
participate in the public meeting. The meeting will be open to all 
persons who have requested in advance to present statements, or who 
register on the day of the meeting (between 8:30 a.m. and 9:00 a.m.) 
subject to availability of space in the meeting room.
    2. Representatives from the FAA will conduct the public meeting. A 
technical panel of FAA experts will be established to discuss 
information presented by participants.
    3. The FAA will try to accommodate all speakers; therefore, it may 
be necessary to limit the time available for an individual or group. If 
necessary, the public meeting may be extended to evenings or additional 
days. If practicable, the meeting may be accelerated to enable 
adjournment in less than the time scheduled.
    4. Sign and oral interpretation can be made available at the public 
meeting, as well assistive listening device, if requested 5 calendar 
days before the meeting.
    5. The public meeting will be recorded by a court reporter. Any 
person who is interested in purchasing a copy of the transcript should 
contact the court reporter directly. This information will be available 
at the meeting.
    6. The FAA requests that persons participating in the public 
meeting provide 10 copies of all materials to be presented for 
distribution to the panel members; other copies may be provided to the 
audience at the discretion of the participant.

Regulatory Evaluation Summary

    The regulations adopted herein will not have substantial direct 
effects on the States, on the relationship between the national 
government and the States, or on the distribution of power and 
responsibilities among the various levels of government. Therefore, in 
accordance with Executive Order 12612, it is determined that this final 
rule does not have sufficient federalism implications to warrant the 
preparation of a Federalism Assessment.
    The FAA conducted a Cost Analysis and Final Regulatory Flexibility 
Analysis to determine the regulatory impacts of this and three other 
AD's to operators of all 244 U.S.-registered Boeing Model 727 passenger 
airplanes that have been converted to cargo-carrying configurations 
under 10 STC's held by four companies. This analysis is included in the 
Rules Docket for each AD. The FAA has determined that approximately 
four 727-200's operated by one carrier were converted under Pemco 
STC's. This carrier also operates nine converted 727's affected by 
other STC's. [There were 15 727's for which the FAA could not identify 
the STC holder. It is possible that these airplanes were also converted 
under a Kitty Hawk STC (STC previously held by ATAZ, Inc.) Their costs 
are not included here.]
    Assuming that the operator of affected airplanes converted under 
Kitty Hawk STC's will comply with the restricted interim operating 
conditions set forth in the AD, the FAA estimates that this operator 
will not lose revenues during the 28-month interim period after the 
effective date of the AD. During the interim period, these airplanes 
will be limited to a total of 8,000 lbs. per pair of adjacent 
containers (a total of 36,000 to 48,000 lbs., depending on the number 
of pallets) because none of the Kitty Hawk-converted 727's have 
approved side restraints. Assuming typical payloads ranging from 34,835 
lbs. for a 727-100 with nine pallets to 47,820 lbs. for a 727-200 with 
12 pallets, none of the operators of Kitty Hawk-converted airplanes 
will lose revenues during this interim period.
    The Cost Analysis and Final Regulatory Flexibility Analysis, 
completed by the FAA and included in the Rules Dockets, estimates that 
affected airplanes can be modified at a cost of $385,000 per airplane 
to carry the maximum payloads currently allowed, or a total of $1.5 
million for the four Kitty Hawk 727's. The FAA expects that the 
operator will modify its airplanes during the 28-month interim period, 
scheduling the modifications to coincide with periodic maintenance. A 
modification will require that the airplane be removed from service for 
a period of 17 days; the FAA conservatively estimates that scheduling a 
modification during periodic maintenance will reduce the net time out 
of service by two days. The FAA estimates the lost revenue during this 
15-day period will be $14,829 per day, per 727-100, and $23,405 per 
day, per 727-200. The total down-time lost revenue for the operator 
will be $1.4 million. This estimate conservatively assumes that cargo 
is not shifted from airplanes being modified to other airplanes. Such 
cargo shifting is typical industry practice and would reduce the costs 
attributable to lost revenues. Incremental fuel costs to carry the 
additional weight of the floor modification will be $17,000 over the 
28-month period, as airplanes are modified. When all affected Kitty 
Hawk 727's are modified, additional fuel costs will be about $1,100 per 
month.
    The total cost, therefore, to modify the fleet of affected 727's 
that were originally modified to the Kitty Hawk STC's, including lost 
revenues while the airplanes are out of service and modification costs, 
is $3.0 million, or $2.7 million discounted at seven percent.
    The Regulatory Flexibility Act of 1980 (RFA), as amended by the 
Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA), 
was enacted by Congress to ensure that small

[[Page 2058]]

entities are not unnecessarily or disproportionately burdened by 
government regulations. The RFA requires a Regulatory Flexibility 
Analysis if a rule would have a significant economic impact, either 
detrimental or beneficial, on a substantial number of small entities. 
The purpose of this analysis is to ensure that the agency has 
considered all reasonable regulatory alternatives that will minimize 
the rule's economic burdens for affected small entities, while 
achieving its safety objectives. Under section 63(b) of the RFA, the 
analysis must address:
    1. Reasons Why the Agency Is Promulgating the Rule
    2. The objectives and legal basis for the rule;
    3. The kind and number of small entities to which the rule will 
apply;
    4. The projected reporting, recordkeeping, and other compliance 
requirements of the rule; and
    5. All federal rules that may duplicate, overlap, or conflict with 
the rule. These elements of the RFA are addressed below.

A. Reasons Why the Agency Is Promulgating the Rule

    The FAA has determined that the unreinforced floor structure of the 
main cargo deck of converted 727's is not strong enough to enable the 
airplane to safely carry the maximum payload that is currently allowed 
in this area. The actions specified in this AD are intended to prevent 
failure of the floor structure, which could lead to loss of the 
airplane.

B. Statement of Objective and Legal Basis

     Under the United States Code (U.S.C.), the FAA Administrator is 
required to consider the following matter, among others, as being in 
the public interest: assigning, maintaining, and enhancing safety and 
security as the highest priorities in air commerce. [See 49 U.S.C. 
44101(d).] Accordingly, this AD amends Title 14 of the CFR's to require 
operators of Boeing 727 airplanes that have been converted from a 
passenger to a cargo-carrying (``freighter'') configuration to comply 
with certain payload limitations, substantiate data showing other 
acceptable limits, or show an alternative method of compliance (AMOC).

C. Regulatory Flexibility Determination

    Under the RFA, the FAA must determine whether or not a rule 
significantly affects a substantial number of small entities. This 
determination is typically based on small entity size and cost 
thresholds that vary depending on the affected industry. The entity 
affected by this rule operates four U.S.-registered converted Boeing 
727 airplanes that have been converted under Kitty Hawk STC's as well 
as nine other affected airplanes converted under other STC's. The FAA 
has prepared an analysis of cost impacts and has examined possible 
regulatory alternatives.

D. Projected Reporting, Recordkeeping, and Other Compliance 
Requirements

    With two minor exceptions, the rule will not mandate additional 
reporting or recordkeeping. First, there will be a negligible one-time 
cost to operators to revise their AFM's and Supplements. Second, 
operators will be required to keep records of the modifications to 
their airplanes. This requirement is common to all maintenance, 
preventive maintenance, and alterations under Sec. 91.417, Maintenance 
records.

E. Overlapping, Duplicative, or Conflicting Federal Rules

    The rule will not overlap, duplicate, or conflict with existing 
Federal rules.

F. Analysis of Alternatives

    This AD will impose a financial requirement on small entities that 
operate 727's that were converted under Kitty Hawk STC's. The FAA 
examined potential alternatives to the AD's requirements to minimize 
the rule's economic burden for small entities while achieving its 
safety objectives. The alternatives are:
     Exclude small entities;
     Extend the compliance deadline for small entities; and
     Establish higher payload limits for small entities.
    The FAA has determined that the option to exclude small entities 
from the requirements of the rule is not justified. The unsafe 
condition that exists on an affected 727 operated by a small entity is 
as potentially catastrophic as that on an affected 727 operated by a 
large entity. In fact, the average payloads carried by small entities 
may exceed the average payloads carried by large operators, resulting 
in a higher probability of a catastrophic event.
    The FAA also considered options to extend the compliance period for 
small operators. The proposed rule established a final compliance date 
of 120 days after the effective date of the rule. During this 120-day 
period, operators could comply with interim operating conditions that 
would enable them to carry higher payloads than those permitted after 
that interim period. When the proposed rule was published, the FAA had 
information that indicated that a portion of engineering data from an 
FAA-approved STC for a floor modification that could be used as an AMOC 
would be available within a few months of the proposed rule's 
publication. In addition, the FAA estimated that operators would be 
able to modify their airplanes within the 120-day interim period.
    Hamilton Aviation has received letters of approval for work towards 
obtaining an STC for strengthening the floor beams aft of Station 700 
and expects to be able to submit additional data in the Fall of 1998 
that will provide the basis for an STC for the entire floor. Pemco 
World Air Services expects to be able to use Hamilton's engineering 
data to modify the floors of the 727's it has converted. The FAA is 
confident, therefore, that there will be AMOC's for operators of all 
affected airplanes when this final rule is published.
    Several commenters to the Rules Dockets for the proposed AD's 
rejected the FAA's claim that their airplanes could be modified within 
the 120-day interim period. Their arguments were based on the 
unavailability of an approved STC that could be used as an AMOC (or, at 
that time, even letters of approval toward an STC). Operators also 
stated that modification of all 244 U.S.-registered airplanes would be 
impossible within a 120-day time frame.
    The FAA agrees 120 days is unrealistic and would have severe 
economic consequences as operators would be required to reduce their 
payloads substantially at the end of the interim period. In the final 
rule, therefore, the FAA extends the interim period to 28 months. This 
will permit operators to modify their airplanes during regularly 
scheduled maintenance, minimizing down time and associated lost 
revenues. This change will be especially beneficial to small entities 
that may find it difficult to find alternative means of carrying cargo.
    Finally, the FAA rejects the compliance alternative that would 
reduce payloads from those currently required but would establish 
higher payload limits than those for larger entities. This alternative 
is unacceptable because the unsafe condition is dependent on the size 
of the payload, not the size of the entity. The FAA cannot permit a 
small entity to operate under an unsafe condition.
    Title II of the Unfunded Mandates Reform Act of 1995 (the Act), 
enacted as Pub. L. 104-4 on March 22, 1995, requires each Federal 
agency, to the extent permitted by law, to prepare a written assessment 
of the effects of any

[[Page 2059]]

Federal mandate in a proposed or final agency rule that may result in 
the expenditure by State, local, and tribal governments, in the 
aggregate, or by the private sector, of $100 million or more (adjusted 
annually for inflation) in any one year. Section 204(a) of the Act, 2 
U.S.C. 1534(a), requires the Federal agency to develop an effective 
process to permit timely input by elected officers (or their designees) 
of State, local, and tribal governments on a proposed ``significant 
intergovernmental mandate.'' A ``significant intergovernmental 
mandate'' under the Act is any provision in a Federal agency regulation 
that would impose an enforceable duty upon State, local, and tribal 
governments, in the aggregate, of $100 million (adjusted annually for 
inflation) in any one year. Section 203 of the Act, 2 U.S.C. 1533, 
which supplements section 204(a), provides that before establishing any 
regulatory requirements that might significantly or uniquely affect 
small governments, the agency shall have developed a plan that, among 
other things, provides for notice to potentially affected small 
governments, if any, and for a meaningful and timely opportunity to 
provide input in the development of regulatory proposals.
    This AD does not contain any Federal intergovernmental or private 
sector mandate. Therefore, the requirements of Title II of the Unfunded 
Mandates Reform Act of 1995 do not apply.

List of Subjects in 14 CFR Part 39

    Air transportation, Aircraft, Aviation safety, Safety.

Adoption of the Amendment

    Accordingly, pursuant to the authority delegated to me by the 
Administrator, the Federal Aviation Administration amends part 39 of 
the Federal Aviation Regulations (14 CFR part 39) as follows:

PART 39--AIRWORTHINESS DIRECTIVES

    1. The authority citation for part 39 continues to read as follows:

    Authority: 49 U.S.C. 106(g), 40113, 44701.

Sec. 39.13  [Amended]

    2. Section 39.13 is amended by adding the following new 
airworthiness directive:

98-26-20  Boeing: Amendment 39-10963. Docket 97-NM-80-AD.

    Applicability: Model 727 series airplanes that have been 
converted from a passenger to a cargo-carrying (``freighter'') 
configuration in accordance with Supplemental Type Certificate 
ST00015AT; certificated in any category.

    Note 1: This AD applies to each airplane identified in the 
preceding applicability provision, regardless of whether it has been 
otherwise modified, altered, or repaired in the area subject to the 
requirements of this AD. For airplanes that have been modified, 
altered, or repaired so that the performance of the requirements of 
this AD is affected, the owner/operator must request approval for an 
alternative method of compliance in accordance with paragraph (g) of 
this AD. The request should include an assessment of the effect of 
the modification, alteration, or repair on the unsafe condition 
addressed by this AD; and, if the unsafe condition has not been 
eliminated, the request should include specific proposed actions to 
address it.
    Compliance: Required as indicated, unless accomplished 
previously.

    Note 2: The payload limitations specified in this AD are in 
addition to payload limitations that are otherwise applicable and do 
not allow for increases in payloads beyond those specified in such 
limitations.

    To prevent structural failure of the floor beams of the main cargo 
deck, which could lead to loss of the airplane, accomplish the 
following:
    (a) Except as provided in paragraphs (b) and (c) of this AD, within 
90 days after the effective date of this AD, accomplish the 
requirements of paragraph (a)(1) or (a)(2) of this AD, as applicable.
    (1) For airplanes that transport containers or pallets that have 
been manufactured in accordance with National Aerospace Standard 
(NAS) 3610 Size Codes ``A,'' ``B,'' ``C,'' ``D,'' or ``E,'' 
containers: Revise the Limitations Section of all FAA-approved 
Airplane Flight Manuals (AFM) and AFM Supplements, and the 
Limitations Section of all FAA-approved Airplane Weight and Balance 
Supplements to include the following information. This may be 
accomplished by inserting a copy of this AD in all AFM's, AFM 
Supplements, and Weight and Balance Supplements.

``LIMITATIONS

    All containers with one door must be oriented with the door side 
of the container facing forward, except the door of the first 
container aft of the cargo barrier may face aft.
    The location of the horizontal center of gravity for the total 
payload within each container or pallet shall not vary more than 10 
percent (8.8 inches) from the geometric center of the base of the 
container or pallet for the forward and aft direction, and 10 
percent of the width from the geometric center of the base of the 
container or pallet for the left or right direction.''

``PAYLOAD LIMITATIONS

    For containers or pallets that have been manufactured in 
accordance with National Aerospace Standard (NAS) 3610 Size Code 
``A'' (88 by 125 inches), ``B'' (88 by 108 inches), or ``C'' (88 by 
118 inches):
    Do not exceed a total weight of 3,000 pounds per container or 
pallet on the main cargo deck, except in the area adjacent to the 
side cargo door. In the side cargo door area, for all containers or 
pallets completely or partially located between Body Station 440 and 
Body Station 660, those containers or pallets are restricted to a 
maximum payload of 2,700 pounds per container or pallet. The 3,000 
and 2,700 pound payload limits include the payload in the lower lobe 
cargo compartments and any other load applied to the bottom of the 
floor beams of the main cargo deck for the same body station 
location as the container or pallet on the main cargo deck.
    For containers or pallets that have been manufactured in 
accordance with NAS 3610 Size Code ``D'' (88 by 54 inches) or ``E'' 
(88 by 53 inches) containers:
    Do not exceed a total weight of 1,500 pounds per container or 
pallet on the main cargo deck, except in the area adjacent to the 
side cargo door. In the side cargo door area, for all containers or 
pallets completely or partially located between Body Station 440 and 
Body Station 660, those containers or pallets are restricted to a 
maximum payload of 1,350 pounds per container or pallet. The 1,500 
and 1,350 pound payload limits include the payload in the lower lobe 
cargo compartments and any other load applied to the bottom of the 
floor beams of the main cargo deck for the same body station 
location as the container or pallet on the main cargo deck.''

    (2) For airplanes on which any other containers or pallets are 
transported: Revise the Limitations Section of all FAA-approved 
AFM's and AFM Supplements, and the Limitations Section of all FAA-
approved Airplane Weight and Balance Supplements, in accordance with 
a method approved by the Manager, Standardization Branch, ANM-113, 
FAA, Transport Airplane Directorate.

    Note 3: The weight restrictions to be approved under paragraph 
(a)(2) will be consistent with the limitations specified in 
paragraph (a)(1) of this AD.

    (b) For airplanes that ARE equipped with side vertical cargo 
container restraints that have been approved by the Manager, 
Standardization Branch, ANM-113: As an optional alternative to 
compliance with paragraph (a) of this AD, within 90 days after the 
effective date of this AD, accomplish the requirements of paragraph 
(b)(1) or (b)(2) of this AD, as applicable. This alternative may be 
used only during the period ending 28 months after the effective 
date of this AD.

    Note 4: To be eligible for compliance with this paragraph, the 
side vertical cargo container restraints must be approved by the 
Manager, Standardization Branch, ANM-113, regardless of whether they 
have been previously FAA approved.

    (1) For airplanes on which containers complying with NAS 3610 
Size Codes ``A,'' ``B,'' ``C,'' ``D,'' or ``E,'' are transported: 
Revise the Limitations Section of all FAA-approved AFM's and AFM 
Supplements, and the Limitations Section of all FAA-approved 
Airplane Weight and Balance Supplements to include the following 
limitations. This may be accomplished by inserting a copy of this AD 
in all AFM's, AFM Supplements, and Weight and Balance Supplements.

[[Page 2060]]

``LIMITATIONS

    Maximum Operating Airspeed of Vmo equals 350 knots indicated 
airspeed (KIAS), or Mode ``B'' [350 knots equivalent airspeed 
(KEAS)].
    Minimum operating weight: 100,000 pounds.
    All containers with one door must be oriented with the door side 
of the container facing forward, except the door of the first 
container aft of the cargo barrier may face aft.
    The location of the horizontal center of gravity for the total 
payload within each container shall not vary more than 10 percent 
(8.8 inches) from the geometric center of the base of the container 
for the forward and aft direction and 10 percent of the width from 
the geometric center of the base of the container for the left or 
right direction.''

``PAYLOAD LIMITATIONS

    For airplanes that transport containers or pallets that have 
been manufactured in accordance with National Aerospace Standard 
(NAS) 3610 Size Code ``A'' (88 by 125 inches), ``B'' (88 by 108 
inches), or ``C'' (88 by 118 inches):
    Except as provided below for Body Station 740 to Body Station 
950, do not exceed a total weight of 9,600 pounds for any two 
adjacent containers or pallets and a total weight of 8,000 pounds 
for any single container or pallet.
    For those containers or pallets which are completely or 
partially located within Body Station 740 to Body Station 950 (the 
region of the wing box and main landing gear wheel well): Do not 
exceed a total weight of 12,000 pounds for any two adjacent 
containers or pallets and a total weight of 8,000 pounds for any 
single container or pallet.
    These container payload limits include the payload in the lower 
lobe cargo compartments and any other load applied to the bottom of 
the floor beams of the main cargo deck for the same body station 
location as the container or pallet on the main cargo deck; and
    For containers or pallets that have been manufactured in 
accordance with NAS 3610 Size Code ``D'' (88 by 54 inches) or ``E'' 
(88 by 53 inches) containers:
    Except as provided below for Body Station 740 to Body Station 
950, do not exceed a total weight of 4,800 pounds for any two 
adjacent (in the forward and aft direction) containers or pallets 
and a total weight of 4,000 pounds for any single container or 
pallet.
    For those containers or pallets which are completely or 
partially contained within Body Station 740 to Body Station 950 (the 
region of the wing box and main landing gear wheel well): Do not 
exceed a total weight of 6,000 pounds for any two adjacent (in the 
forward and aft direction) containers or pallets and a total weight 
of 4,000 pounds for any single container or pallet.
    These payload limits include the payload in the lower lobe cargo 
compartments and any other load applied to the bottom of the floor 
beams of the main cargo deck for the same body station location as 
the container or pallet on the main cargo deck.''

    (2) For airplanes on which pallets or containers other than 
those specified in paragraph (b)(1) of this AD, are transported: 
Revise the Limitations Section of all FAA-approved AFM's and AFM 
Supplements, and the Limitations Section of all FAA-approved 
Airplane Weight and Balance Supplements, in accordance with a method 
approved by the Manager, Standardization Branch, ANM-113.

    Note 5: The weight restrictions to be approved under paragraph 
(b)(2) will be consistent with the limitations specified in 
paragraph (b)(1) of this AD.

    (c) For airplanes that are NOT equipped with side vertical cargo 
container restraints that have been approved by the Manager, 
Standardization Branch, ANM-113: As an optional alternative to 
compliance with paragraph (a) of this AD, within 90 days after the 
effective date of this AD, accomplish the requirements of paragraph 
(c)(1) or (c)(2) of this AD, as applicable. This alternative may be 
used only during the period ending 28 months after the effective 
date of this AD.
    (1) For airplanes on which containers complying with NAS 3610 
Size Codes ``A,'' ``B,'' ``C,'' ``D,'' or ``E,'' are transported: 
Revise the Limitations Section of all FAA-approved AFM's and AFM 
Supplements, and the Limitations Section of all FAA-approved 
Airplane Weight and Balance Supplements to include the following 
limitations. This may be accomplished by inserting a copy of this AD 
in all AFM's, AFM Supplements, and Weight and Balance Supplements.

``LIMITATIONS

    Maximum Operating Airspeed of Vmo equals 350 knots indicated 
airspeed (KIAS), or Mode ``B'' [350 knots equivalent airspeed 
(KEAS)].
    Minimum operating weight: 100,000 pounds.
    All containers with one door must be oriented with the door side 
of the container facing forward, except the door of the first 
container aft of the cargo barrier may face aft.
    The location of the horizontal center of gravity for the total 
payload within each container shall not vary more than 10 percent 
(8.8 inches) from the geometric center of the base of the container 
for the forward and aft direction and 10 percent of the width from 
the geometric center of the base of the container for the left or 
right direction.''

``PAYLOAD LIMITATIONS

    For airplanes that transport containers or pallets that have 
been manufactured in accordance with National Aerospace Standard 
(NAS) 3610 Size Code ``A'' (88 by 125 inches), ``B'' (88 by 108 
inches), or ``C'' (88 by 118 inches):
    Except as provided below for Body Station 740 to Body Station 
950, do not exceed a total weight of 8,000 pounds for any two 
adjacent containers or pallets and a total weight of 8,000 pounds 
for any single container or pallet.
    For those cargo pallets which are completely or partially 
contained within Body Station 740 to Body Station 950 (the region of 
the wing box and main landing gear wheel well): Do not exceed a 
total weight of 12,000 pounds for any two adjacent containers or 
pallets and a total weight of 8,000 pounds for any single container 
or pallet.
    These payload limits include the payload in the lower lobe cargo 
compartments and any other load applied to the bottom of the floor 
beams of the main cargo deck for the same body station location as 
the container or pallet on the main cargo deck.
    For containers or pallets that have been manufactured in 
accordance with NAS 3610 Size Code ``D'' (88 by 54 inches) or ``E'' 
(88 by 53 inches) containers:
    Except as provided below for Body Station 740 to Body Station 
950, do not exceed a total weight of 4,000 pounds for any two 
adjacent (in the forward and aft direction) containers or pallets 
and a total weight of 4,000 pounds for any single container or 
pallet.
    For those cargo pallets which are completely or partially 
contained within Body Station 740 to Body Station 950 (the region of 
the wing box and main landing gear wheel well): Do not exceed a 
total weight of 6,000 pounds for any two adjacent containers or 
pallets and a total weight of 4,000 pounds for any single container 
or pallet.
    These payload limits include the payload in the lower lobe cargo 
compartments and any other load applied to the bottom of the floor 
beams of the main cargo deck for the same body station location as 
the container or pallet on the main cargo deck.''

    (2) For airplanes on which pallets or containers other than 
those specified in paragraph (c)(1) of this AD, are transported: 
Revise the Limitations Section of all FAA-approved AFM's and AFM 
Supplements, and the Limitations Section of all FAA-approved 
Airplane Weight and Balance Supplements, in accordance with a method 
approved by the Manager, Standardization Branch, ANM-113.

    Note 6: The weight restrictions to be approved under paragraph 
(c)(2) will be consistent with the limitations specified in 
paragraph (c)(1) of this AD.

    (d) For airplanes complying with paragraph (b) or (c) of this 
AD, within 28 months after the effective date of this AD, accomplish 
the requirements of paragraph (a) of this AD.
    (e) For airplanes that operate under the 350 KIAS limitations 
specified in paragraph (b) or (c) of this AD: A maximum operating 
airspeed limitation placard must be installed adjacent to the 
airspeed indicator and in full view of both pilots. This placard 
must state: ``Limit Vmo to 350 KIAS.''
    (f) As an alternative to compliance with paragraphs (a), (b), 
(c), (d), and (e) of this AD: An applicant may propose to modify the 
floor structure or propose differing payloads and other limits by 
submitting substantiating data and analyses to the Manager, Los 
Angeles Aircraft Certification Office (ACO), FAA, Transport Airplane 
Directorate, 3960 Paramount Boulevard, Lakewood, California 90712. 
The Manager of the Los Angeles ACO will coordinate the review of the 
submittal with the Manager of the Standardization Branch, ANM-113, 
in accordance with the procedures of paragraph (g) of this AD. If 
the FAA determines that the proposal is in compliance with the 
requirements of Civil Air Regulations (CAR) part 4b and is 
applicable to the specific airplane being analyzed and approves the 
proposed limits,

[[Page 2061]]

prior to flight under these new limits, the operator must revise the 
Limitations Section of all FAA-approved AFM's and AFM Supplements, 
and the Limitations Section of all FAA-approved Airplane Weight and 
Balance Supplements, in accordance with a method approved by the 
Manager, Standardization Branch, ANM-113. Accomplishment of these 
revisions in accordance with the requirements of this paragraph 
constitutes terminating action for the requirements of this AD.
    (g) An alternative method of compliance or adjustment of the 
compliance time that provides an acceptable level of safety may be 
used if approved by the Manager, Standardization Branch, ANM-113. 
Operators shall submit their requests through an appropriate FAA 
Principal Maintenance Inspector, who may add comments and then send 
it to the Manager, Los Angeles ACO, who will coordinate the approval 
with the Manager of the Standardization Branch, ANM-113.

    Note 7: Information concerning the existence of approved 
alternative methods of compliance with this AD, if any, may be 
obtained from the Standardization Branch, ANM-113.

    (h) Special flight permits may be issued in accordance with 
sections 21.197 and 21.199 of the Federal Aviation Regulations (14 
CFR 21.197 and 21.199) to operate the airplane to a location where 
the requirements of this AD can be accomplished.
    (i) This amendment becomes effective on February 16, 1999.

    Issued in Renton, Washington, on December 16, 1998.
Ronald T. Wojnar,
Manager, Transport Airplane Directorate, Aircraft Certification 
Service.
[FR Doc. 99-446 Filed 1-11-99; 8:45 am]
BILLING CODE 4910-13-P