Preliminary report disaster site investigation of manufactured homes

[From the U.S. Government Printing Office, www.gpo.gov]

NCS BCS

National Conference of States on Building Codes and Standards, Inc.

481 Carlisle Drive, Herndon, Virginia 22070
TA
656.5
.P73 COASTAL ZCNE-
1989
INPCIELMATION

f TAbrary DEPARTMENT OF COMMERCE NOAA
COASTAL SERVICES CENTER
2234 SOUTH HOBSON AVENUE
CHARLESTON, SC 29405-2413

Preliminary Report

Disaster Site Investigation of

Manufactured Homes

December 28, 1989

Prepared for:
The Department of Housing and Urban Development
Office of Manufactured Housing and
Regulatory Functions
Manufactured Housing and Construction Division
451 Seventh Street, S.W.
Washington, DC 20410

Prepared by:

The National Conference of States on
Building Codes and Standards
481 Carlisle Drive
Herndon, VA 11070

TABLE OF CONTENTS

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . I

Preliminary Conclusions and Recommendations . . . . . . . . . . . 3

Tie-Down/Anchoring . . . . . . . . . . . . . . . . . . . . . 3
Negative Wind Pressure . . . . . . . . . . . . . . . . . . . 4
Metal Roofs . . . . . . . . . . . . . . . . . . . . . . . . . 5
Hurricane Zone . . . . . . . . . . . . . . . . . . . . . . . 6
Earthquake Zone . . . . . . . . . . . . . . . . . . . . . . . 7

Background/Summary of Findings . . . . . . . . . . . . . . . . . . 9

Tornadoes in Central Texas . . . . . . . . . . . . . . . . . 9
Hurricane Hugo . . . . . . . . . . . . . . . . . . . . . . . 11
Hurricane Jerry . . . . . . . . . . . . . . . . . . . . . . . 14
California Earthquake . . . . . . . . . . . . . . . . . . . . 16

Appendices

A - Manufactured Home Disaster Damage Inspection Report . . . 18
B - Incorrect Method of Installation . . . . . . . . . . . . 19
C - Correct Method of Installation . . . . . . . . . . . . . 20
D - Texas Tornado Photos . . . . . . . . . . . . . . . . . . 21
E - Hurricane Hugo Photos . . . . . . . . . . . . . . . . . . 22
F - Hurricane Jerry Photos . . . . . . . . . . . . . . . . . 23
G - California Earthquake Photos . . . . . . . . . . . . . . 24
H National Profile of Tie-down Regulations . . . . . . . . 25
I NCSBCS Tie-Down Recommendations . . . . . . . . . . . . . 26

INTRODUCTION

In December 1988, NCSBCS contracted with the U.S. Department of Housing
and Urban Development (HUD) to investigate the cause of failures,
sustained by manufactured homes following natural disasters.
Historically, it has appeared that the number of manufactured homes and
the severity of damage caused to those homes have been disproportionate
to site built homes which have been exposed to the same disasters. This
report covers four separate investigations conducted by NCSBCS at
disaster sites involving manufactured homes. The purpose of the
investigations was to determine which structural components of
manufactured homes are most likely to fail in disaster situations and to
recommend possible remedial actions to the Department. Data was
collected at disaster sites as an attempt to identify consistent patterns
of component failures. Findings from these investigations are intended
to identify not only the cause of failures but also which systems are
most apt to fail.

Several areas will be examined when determining the cause of failures,
such as: 1) whether the design met the standard, 2) manufacturer's
installation instructions, 3) proper field installation of units, and 4)
the degree to which the construction standards appropriately address the
structure and its components. The information obtained from these
investigations can be utilized to prevent future damage to manufactured
homes to the greatest extent possible, by recommending appropriate
changes to the above listed items. Such recommendations would prevent
similar nonconformances from being constructed in future units and would
possibly serve to identify the affected units in the field and allow for
their repair prior to a disaster.

Between May and October of 1989, four disaster site investigations were
conducted. The types of natural disasters which damaged or destroyed
manufactured homes at these sites are as follows: 1) Tornadoes in
central Texas; 2) Hurricane Hugo in South Carolina; 3) Hurricane Jerry in
Galveston, Texas; and 4) a California earthquake in the Santa Cruz area.

Shortly following each disaster, a NCSBCS engineer investigated the
affected sites, utilizing an inspection form developed by NCSBCS (see
Appendix A). This form identifies the critical points of a manufactured
home which are subject to failure. The form describes the environmental
conditions in which the home is located, such as soil conditions and
terrain. The form also provides_details of the disaster. such as wind
speeds, and flood levels. Photographs were taken at each disaster site.
Additionally, camcorder tapes of the damage caused by hurricanes Hugo and
Jerry were also made.

A background of each disaster and the findings from the investigations
are described as follows.

PRELIMINARY CONCLUSIONS AND RECOMMENDATIONS

As a result of the investigations conducted at the four disaster sites,
several weaknesses in manufactured homes have been identified. Failures
witnessed in the wind related disasters were of a different nature than
those discovered after the earthquake, although some weaknesses were
generic in nature to both failures that occurred with the manufactured
home supports. Based on the structural failures found at these disaster
sites, NCSBCS offers the following preliminary conclusions and
recommendations:

1. Tie-Down/Anchoring

The most severe damage observed at the disaster sites resulted
from homes shifting off their piers or overturning. This was
generally caused by failure of the ground anchors due to
saturated soils or improper installation of the anchoring
devices.

Improper Installation

Several.examples of improper installations were witnessed at the
disaster sites and are described in this report. The two most
common of these being ground anchors that were installed at
angles contrary to installation instructions and the improper
attachment of anchor straps to I-beams. Since DAPIA approved
installation instructions are provided with all manufactured
homes, proper set up and securement of homes can be enforced by
state and/or local jurisdictions.

There are approximately twenty one (21) states that have not
adopted any tie-down regulations (see Appendix H). Until
adoption and enforcement of tie-down regulations is achieved
nationwide, manufactured homes will continue to be destroyed and
damaged by disasters regardless of the structural integrity of
the home.

Anchor Failure

Hurricanes and tornadoes are usually accompanied by heavy rains
and occasional flooding. Water saturated soil provides little
resistance to uplift forces and the anchors are easily pulled
from the ground (see Appendix E, photos 23 and 24). The HUD
Standards require anchoring equipment to resist an allowable
working load of 4725 pounds. However, most installation
instructions provided by the anchor manufacturers do not
adequately address the performance of ground anchors in various
types of soil, particularly when the soil is saturated.

As a result of the tie-down failures witnessed at the disaster
sites, it is NCSBCS' recommendation that HUD consider
implementing the recommendations contained in the manufactured
home installation report, prepared by NCSBCS and submitted to
HUD in January of 1989 (see Appendix D. As a supplement to
that report, NCSBCS proposes to research literature in the area
of soil resistive capacities as related to ground anchors. This
information would serve to assist HUD and the applicable ASTM
committees in the development of future standards and guidelines.

2. Negative Wind Pressure

The most common damage observed after Hurricanes Hugo and Jerry
was the loss of siding at the ends of homes. This damage may
have been the result of the construction of the manufactured
homes which are designed according to the Standards to resist
only an inward wind pressure on the windward side of the unit.
When air flows over and around an object, it exterts inward as
well as outward pressures on the object. The inward pressure is
on windward side and all other sides have outward pressure. On
the leeward side of the object there is usually a suction
effect, consisting of outward pressure on the surface of the
object in comparison to the direction of pressure on the
windward side. This is known as negative pressure.

Whenever wind is a major problem, the projected area method
which considers negative pressure. All major model building
codes, SBC, BOCA and UBC, consider negative pressure, as a
viable tool for regulating wind problems. Additionally, the
most up-to-date and complex standards for wind design, ASCE 7-88
(ANSI A58.1-1982) addresses negative pressure. Negative
pressure is not addressed by the HUD Standards.

NCSBCS recommends the following:

0 NCSBCS provides a report to HUD justifying a revision to
the standards for negative pressure.

0 NCSBCS proposes to conduct calculations of exterior wall
siding connections based on other model building codes and
compare them with HUD standards.

3. Metal Roofs

Extensive damage to metal roofs on manufactured homes occurred
in all three wind-related disaster sites. Roofs were either
completely or partially lifted off the homes and scattered
around the surrounding area. The metal roof connections to the
side walls or trusses were the most frequent failure points.
Design of these connections required dynamic analysis for
vibration and flutter, which are not accountable when using the
equivalent static load method. In general, stiffening, bracing,
and tightening the metal roof would have helped to minimize the
extensive damage.

NCSBCS recommends the following:

NCSBCS conducts DAPIA package reviews of all manufacturers
with metal roof designs and submit recommendations for
suitable metal roof connections.

NCSBCS performs a true dynamic analysis or a wind tunnel
test to assure the adequacy of the metal roof. the
connections and the capacity to withstand the HUD
recommended loads.

4. Re-evaluate the Design Criteria for Wind Loading of Structures
in Hurricane Zone II

The HUD standards designate two wind zones, a standard zone
(Zone 1) and a hurricane zone (Zone ID.- The standards require
that homes located in Zone II be designed to resist horizontal
wind loads of 25 psf and net uplift of 15 psf.

The most widely used standard for calculating wind loads for
structures in this country is ASCE 7-88, formerly ANSI A58.1
(minimum design loads for buildings and other structures). This
standard is referenced by all three model building codes, BOCA,
UBC, and SBC. Depending on the wind exposure of a manufactured
home, it can be shown, per ANSI A58.1, that the current design
criteria of 25 psf/15 psf does not apply to wind conditions in
excess of 95 mph. In areas where homes are not afforded wind
protection by hills, trees etc., this figure would drop to
80 mph. When calculating wind loads, using ANSI A58.1, many
variables affect the design value assigned to a structure, i.e.,
wind exposure, terrain, dimension of home etc. The current
design values from the HUD standards do not address any of these
variables.

Considering that a category one hurricane (74-95 mph) can have
wind gusts in excess of 125 mph, as was the case with hurricane
Jerry, and that category four hurricane Hugo sustained winds of
135 mph, it is apparent that the current Zone II design values
are not appropriate for hurricane conditions.

NCSBCS recommends the following:

0 The standards be revised to include earthquake zones.

0 The manufacturer be required to provide one acceptable
method for secondary bracing of the foundation for homes
located in Nearthquake zonesu. This method should be
approved by a registered professional engineer or architect.

To assist HUD in further evaluating this recommendation, NCSBCS
proposes to-provide a report that will include:

0 Information on which HUD may pursue the establishment of
Nearthquake zonesu in the Federal Standards.

0 Overview of information on earthquake braces, jacks and
other means of protection that are used on manufactured
homes in potential earthquake areas.

0 Recommended language for inclusion in the Standards.

Section 3280-305(c)(2)(ii) of the Standards allows the
department to establish more stringent requirements in those
areas which experience winds of 125 mph or greater. This
section however is vague and does not provide any guidance on
how these "more stringent requirements" would be derived.

Therefore, NCSBCS recommends that HUD establish a hurricane Zone
III and assign correlating design values for homes located in
those areas. Further, NCSBCS proposes the following:

To gather data documenting those areas which experience
re-occurring high winds that exceed the current Zone II
design values.

The data used to substantiate the new zone will be collected
from sources such as the National Weather Service, Federal
Emergency Management Agency (FEMA) National Oceanic and
Atmospheric Administration (NOAA) etc.

To develop design criteria for those homes which are to be
located in Zone III areas.

This criteria will be derived from ASCE 7-88 (ANSI A58.1).

5. The primary damage observed in manufactured homes affected by
earthquakes was caused by the support piers penetrating the
bottom board and flooring after the homes had shifted.

All model codes contain provisions for protecting structures
from earthquake loads. Similarly, it is recommended that the
Federal Standards provide protection of manufactured homes.
Although many manufactured home foundations are currently
installed with means of earthquake protection, requirements for
such protection and the adequacy of such protection is not
governed by the Standards.

BACKGROUND

Tornadoes in Central Texas

On Tuesday night, May 16, 1989, a strong storm front consisting of hail,
heavy rains and tornadoes swept across central Texas. NCSBCS
investigated the damage at White's Mobile Home Park located thirty miles
south of Ft. Worth, in the town of Cleyburne, which was among those areas
hardest hit by the storm. Approximately eight manufactured homes in the
park sustained damage from the tornadoes and associated high winds.
Damage to the homes varied from minor shifts off of foundations, loss of
roofs, and overturning of units to total destruction of units.

SUMMARY OF FINDINGS

Tornadoes in Central Texas

Due to the small number of homes left in tact following the storm and
since very few structures, whether they be manufactured homes or site
built, can withstand a direct hit by a tornado, the data collected during
the first inspection was limited. However, two consistent weaknesses
appeared in all of the homes that were inspected.

1. Improper Anchoring of Units: The installation of ground anchors
was not coaxial with the straps. (Anchors were angled in
towards the home, see Appendix B.) This method of anchor
installation, while not uncommon, is contrary to the
installation instructions of all anchor manufacturers. When
anchors are installed at this angle there is a tendency for the
anchor to slice through the soil thereby creating slack in the
tie-down strap which allows the unit to shift off its
foundation. (See Appendix C for installation guidelines.)
Also, the soil in which the units were anchored did not appear

to be suited for the anchoring method. The soil was a medium
dense sand to firm silt mixture. The anchors were screw augers
installed at a depth of 2'-0". During the storm the soil was
saturated and provided little resistance from the vertical and
lateral forces that the winds imposed on the homes. It was the
opinion of the inspector that the condition of the soil was such
that "two men could have pulled an anchor from the ground."

2. Improper Attachment of Anchor Straps to the I-beams: The straps
were hooked to the bottom of the I-beams, on all of the homes
inspected, as opposed to being wrapped around the I-beams or
attached to the top of the I-beams as recommended by
installation manuals. This unapproved method of attachment
allows the clip to fall off the beam when there is slack in the
strap, such as when the home is rocked by strong winds.

The anchoring and attachment described above resulted in the anchors
being pulled from the ground on the windward side of all the homes and
the straps dropping off the I-beams on the leeward sides.

See Appendix D for photos of disaster site.

BACKGROUN

Hurricane Hugo, South Carolina

On Friday, September 22, 1989, Hurricane Hugo struck Charleston,
South Carolina leaving over 20 people dead and tens of thousands
homeless. Hugo was a category 4 hurricane, on a scale of 5, with winds
of 135 mph and a tidal surge of 12 to 17 feet. Hurricane force winds
extended 140 miles from its center. Although South Carolina was the
hardest hit, damage from Hugo was reported from as far north as Maine and
as far inland as western Pennsylvania, where 55 mph wind gusts were
reported. (See Appendix E.)

The investigation of damaged homes was concentrated in the Charleston
area, which was hardest hit. Homes in Georgetown, SC, approximately
60 miles north of Charleston and in Columbia, SC, which lies 115 miles
inland were also inspected. Wind speeds in Georgetown and Columbia were
reported to be between 100 and 110 mph.

SUMMARY OF FINDINGS

Hurricane Hugo, South Carolina

Three manufactured home parks in the Charleston area, containing a total
of approximately 125 homes, were inspected on September 25 and 26
following the storm. The primary damage observed at these parks is as
follows.

1. Tie-down/anchors

Approximately 70 percent of the homes had shifted from their
piers to varying degrees. One-half of those had totally fallen
off their supports. (See Appendix E, photos 7, 17, 18 and 21.)
The reasons for the shifting are as follows:

e Anchor failure occurred after heavy rains had saturated the
ground and greatly reduced the anchors resistance. Many
anchors were totally pulled out from the ground or had sliced
through the soil and were bent (see Appendix E, photos 19, 23
and 24).

9 Strap failure was not an uncommon observation. In some
instances the tie down straps had broken when subjected to
the excessive wind pressures (see Appendix E, photo 17).

e The anchor straps were improperly attached to the I-beams.
The strap clips on many homes were clipped to the bottom
flange of the I-beams as opposed to being wraped around the
beams as required by installation instructions. (See
Appendix E, photo 22.) When installed in this mannor, minor
rocking of the home allows the clips to fall off the I-beams.

e Ground anchors were installed at wrong angles (see
Appendix B). Ground anchors should generally be coaxial to
the tie-down straps (see Appendix C).

2. Metal Roofs

Approximately 15 percent of the inspected homes were missing
between 20 to 100 percent of their metal roof covering. Several
also suffered minor truss damage. (See Appendix E, photos 3, 4,
7 and 8.) Homes with shingled roofs experienced damage to a
lesser extent, generally missing a dozen or less shingles.
Typically, individual sections of roofing were completely torn
from the homes exposing the units to the massive amounts of rain
that accompanied the hurricane and created further damage. In
all cases, when the roof covering was torn from the home, the
staples fastening the roof to the edge rail were also torn out
of the edge rail.

3. Siding

Approximately 15 percent of the homes suffered damage to siding
and/or soffit covering. The primary damage was concentrated at
the end of the units and affected only homes with metal and
vinyl siding (see Appendix E, photos 14, 15 and 16). The
primary reason for the loss of siding is the leeward suction
forces at edges and ends of homes. When air flows over and
around a stationary object, it exerts both inward and suction
pressure on the surface of the object.

4. Demolition

In addition to the above mentioned damage, approximately
8 percent of the homes were totally destroyed (see Appendix E,
photos 1, 2, 5, 6, 9, 10, 11, 12 and 13). It was observed that
the homes which had overturned, the anchor and strap failed.
Other homes were destroyed as a result of wind pressures that
severed the entire roof from the home. Hence, the exterior
walls collapsed when the roof trusses no longer helped stablize
the wall. The floor systems remained intact in all of the
demolished homes.

Damage to homes in Georgetown and Columbia where wind velocity
was reported at 100 to 110 mph, was similar to that found in
Charleston, except to a lesser extent. Approximately 55 percent
of the homes had shifted either totally or partially from their
piers. One to two percent of the homes were destroyed. Siding
and roof damage was observed in approximately 15 percent of the
homes. Although slightly fewer homes were damaged in these
areas, the type of damage observed was identical to that
experienced in Charleston.

BACKGROUND

Hurricane Jerry. Galveston. Texas

On Sunday October 15, 1989, Hurricane Jerry swept across Galveston Island
from the Gulf of Mexico with sustained northerly winds of 80 mph, with
gusts of more than 125 mph and several reported tornadoes (see
Appendix F). Although Jerry was a category one hurricane, the weakest on
a scale of five, it left three people dead and caused an estimated
8 million dollars in property damage. Galveston is located in hurricane
Zone II, as defined by the Manufactured Home Construction and Safety
Standards. Galveston is a 32 mile long island located in the Gulf of
Mexico just off the Texas coast. The damage caused by the hurricane was
concentrated on a path of approximately 4 miles wide across the island.
This tourist resort island is primarily occupied by apartments,
condominiums and hotels. It contains only a few manufactured homes. One
of the manufactured home parks on the island, the Villa Del Sol, was
located in that 4 mile wide path. The Villa Del Sol is currently located
approximately 1/4 mile from the Gulf and is virtually protected from the
wind by an existing seawall. Approximately 8 of the 25 homes in the park
were damaged to varying degrees.

The following observations were noted during a site investigation
conducted by NCSBCS on October 17 and 18, 1989.

SUMMARY OF FINDINGS

Hurricane Jerry, Galveston. Texas

All of the damaged manufactured homes were produced prior to HUD's
adoption of the Manufactured Home Standard on June 1976. The average
home was 14 to 15 years old. The majority of homes in the park were
pre-HUD, single wide units. Only one home was completely destroyed,
which was reportedly hit by a tornado. (See Appendix F, photos I
and 2.) The primary damage sustained by the manufactured homes was
1) loss of roof and soffit coverings and 2) loss of siding. (See
Appendix F, Photos 3, 4, 7, 8, 11 and 12.)

Unlike other wind related disaster sites, none of the homes had
overturned or shifted from their pier supports, including the home struck
by the tornado. Most homes were anchored with frame and over the roof
ties. However, several units were not properly tied down and had an
insufficient number of anchors. (See Appendix F, photos 5 and 6.) The
fact that none of the homes had shifted may partially be due to the
protection afforded the park by the seawall. The roofs of the homes were
approximately 5'-0" lower than the roadway which separates the park from
the beach.

1. Roof Damage

A total of six homes were inspected in the Villa Del Sol park
and immediate vicinity. Five of those homes lost 20 to 90
percent of their roof coverings. All homes had metal roofs. In
all cases, entire sections of roofing were blown off, from one
sidewall to the other. (See Appendix F, photos 7 and 8.) With
the exception of the tornado struck unit, all homes had either
the staples fastening the roof to the perimeter roof rail or the
staples fastening the perimeter rail to the top plate pulled out
of the wood members to which they were fastened. On five out of
six homes, roof trusses remained intact and sustained only minor
damage. The unit hit by the tornado lost 75 percent of its
trusses. (See Appendix F, photos 1 and 2.)

2. Siding Damage

Siding and/or soffit covering on five of the six units was
pulled loose or torn completely off (see Appendix F, photos
9 and 10). All of the affected units had metal siding with
damage occurring at the ends (see Appendix F, photos 9, 10, 11
and 12). (See hurricane Hugo for explanation.) Similar to the
roofing situation, the staples which fastened the siding were
pulled away from the wood members.

BACKGROUN

California Earthquake

On Tuesday October 17, 1989 an earthquake, measuring 7.1 on the Richter
scale, struck northern California killing over 50 people and causing over
a billion dollars in damage. The epicenter of the quake was located 10
miles north of Santa Cruz, CA. (See Appendix G.)

Investigations of earthquake damage to homes were conducted by NCSBCS on
October 23, at four manufactured home parks in-Santa Cruz County, CA.
Two of the parks were located in Santa Cruz, CA, approximately 10 miles
from the epicenter, and two parks were in the town of Watsonville, CA,
approximately five miles from the epicenter. The average age of the
damaged single and double wide units ranged from 10 to 15 years.

SUMMARY OF FINDINGS

California Earthquake

Minor damage was observed in both Santa Cruz parks. Although a majority
of the homes showed no signs of physical damage, several homes were
slightly damaged when a few of the piers had either been loosened from
the I-beams or had broken. This was evidenced by the sagging ends of
these homes. It was observed that one home had completely shifted off
its piers.

The first of the two parks investigated in Watsonville was Meadows
Manor. Approximately 10 to 20 percent of the homes in Meadows Manor
had completely fallen from their piers (see Appendix G, photos I and 2)
while other homes, like those in Santa Cruz, had only partially shifted.
The second park, investigated in Watsonville was Rancho Cerritos.
Approximately 75 to 85 percent of the homes had fallen from their piers.

One of the homes in Rancho Cerritos was totally destroyed by fire
reportedly caused by the gas service line rupturing when the home shifted
(see Appendix G, photo 5). The primary damage to the homes was caused by
the piers penetrating the bottom board and, in several instances,
penetrating the floor decking (see Appendix G. photos 7 and 8). The
skirting around the homes was also damaged, as was service entrance
fixtures (see Appendix G, photo 6).

Most of the homes had no means of anchoring, however, those homes that
were anchored still shifted from their piers (see Appendix G. photo 9).
A possible reason why the homes, both anchored and unanchored, fell is
that the majority of piers used in these parks had a small surface
bearing area (4 inch x 4 inch) for the I-beams. (See Attachment G,
photo 10.) Therefore, the slightest movement would allow the beam to
slip off the pier.

Several homes in the Watsonville park, which had shifted off their piers,
were equipped with earthquake braces or jacks. These particular homes
sustained no damage to the underbelly, because the jacks prevented the
homes from falling onto the piers. (See Appendix G, photo 11 and 12.)
Homes which still had axles and wheels attached to their frames sustained
less damage than those in which the axles had been removed. The wheels
provided temporary support for the homes, prevented floor penetration by
piers and allowed the homes to be jacked up and reset more easily. (See
Appendix G, photos 1 and 2.)

I
I
I
I
I
I
I
I Appendix A - manufactured Home Disaster Damage Inspection Report
I
I
I
I
I
I
I
I
I
I
1 18 @

MANUFACTURED HOME DISASTER DAMAGE INSPECTION gEpog

INSPECTOR: DATE:

ADDRESS:

CITY/COUNTY: STATE/ZIP:

BATE OF DISASTER

DISASTER TYPE (WIND) HIGH WINDS HURRICANE TORNADO OTHER

WIND SPEED: WIND ZONE: - ZONE I' ZONE 2

COMMENTS:

DISASTER TYPE (FLOOD) FLASH FLOOD GRADUAL FLOOD FLOOD PLAIN

KATER DEPTH ABOVE GROUND SURFACE OR FLOOR LEVEL OF UNIT (SPECIFY):

COMMENTS:

2. SIZE OF UNIT: - SINGLE WIDE DOUBLE WIDE FIELD INSTALLED ADDITIONS

DIMENSION OF UNIT:

UNIT TYPE: MANUFACTURER/MODEL/SERIAL NOJAGE/COMMENTS:

3. TERRAIN: - FLAT HILLY MOUNTAINOUS WOODED OPEN FIELD TYPE

COMMENTS:

SOIL TYPE: (SAND, CLAY, SILT, GRAVEL ETC.)

POSITION OF UNIT W - - E (SKETCH IN UNIT AS ANCHORED)

PREVAILING WIND DIRECTION:

4. FOUNDATION/PIERS: PERIMETER FOUNDATION PIERS ONLY COMBINATION

MATERIAL:

DIMENSION: HEIGHT:

SPACING: REINFORCED

SHIMMING METHOD AND MATERIAL:

SKIRTING TYPE:

FOOTING: TYPE (CONCRETE, GRAVEL, PRECAST)

DEPTH: DIMENSION:

COMMENTS:

5. TIE DOWN METHOD: FRAME TIES (SPACING)
COMMENTS' OVER-THE-TOP TIES (SPACING)

STRAP CONNECTION TO I-BEAM: WRAP AROUND BEAM OTHER

COMMENTS:

I-BEAM DIMENSION (DEPTH) I-BEAM SPACING:

6. ANCHORING METHOD: SCREW AUGERS DIMENSION:

EXPANDING ANCHORS DIMENSION:

CONCRETE DEAD MEN DIMENSION:
COMMENTS' OTHER DIMENSION:

ANCHOR INCLINATION (ANGLE, DEPTH)

COMMENTS:

7. FLOOR TO CHASSIS ATTACHMENT: - OUTRIGGER ATTACHMENT I-BEAM ATTACHMENT

LAG SCREW SIZE: LAG SCREWS PER OUTRIGGER:

I-BEAM CLIP ATTACHMENT SPACING OF CLIPS:

WELDED CLIPS PRESSURE FITTED CLIPS

COMMENTS:

8. FLOOR TO WALL ATTACHMENT: UPLIFT STRAPS

SIZE AND SPACING OF STRAPS:

SIZE AND NUMBER OF FASTENERS PER STRAP:

COMMENTS:

9. ROOF TO WALL ATTACHMENT: - UPLIFT STRAPS SIZE & SPACING OF STRAPS:

SIZE AND NUMBER OF FASTENERS PER STRAP:

SHEARWALLS - NUMBER AND SPACING:

COMMENTS:

10. UNIT CONDITION: SAFE FOR OCCUPANCY HABITABLE, REPAIRS NECESSARY

UNINHABITABLE DEMOLISHED

SHIFTED OFF FOUNDATION/ OVERTURNED
PIERS

ANCHORS FAILED CHASSIS INTACT - DAMAGE TO SHELL

COMMENTS:

11. STRUCTURAL SEPARATION AT: CHASSI S TO FLOOR:

FLOOR TO WALL:

WALL TO ROOF:

12. FLOOR CONDITION: STRUCTURAL DAMAGE DESCRIBE:

BOTTOM BOARD DAMAGE DESCRIBE:

BOTTOM BOARD MANUFACTURER:

COMMENTS:

13. EXTERIOR WALL CONDITION: TYPE AND MANUFACTURER OF SIDING:

SIDING DAMAGE:

WINDOW/DOOR DAMAGE:

WALL DAMAGE:

14. ROOF TYPE OF ROOF: - GABLE BOW OTHER

TYPE OF FRAMING: SCISSOR TRUSS ARCH TRUSS TRADITIONAL TRUSS

2 x RAFTERS OTHER

SPACING: COMMENTS:

ROOF COVERING: ASPHALT SHINGLE METAL OTHER

OVER HANG/EAVE: - NONE PARTIAL PERIMETER

DEPTH: COMMENTS:

ROOF VENTILATION DESCRIBE:

ROOF CONDITION: NO DAMAGE SHINGLE DAMAGE - DESCRIBE:

SHEATHING DAMAGE - TYPE OF FASTENERS AND SPACING:

DESCRIBE DAMAGE:

RAFTER DAMAGE - SPECIFY RAFTER SIZE, SPAN, RATING, FASTENING:

DESCRIBE DAMAGE:

COMMENTS:

15. DOUBLE WIDE UNITS: SIZE OF INTERIOR WALL OPENINGS AT MATELINE:

TYPE, SIZE AND QUANTITY OF FASTENERS USED TO CONNECT HALVES AT:

ROOF:

FLOOR:

SIDEWALLS:

GENERAL COMMENTS:

I
I
I
I
I
I
I
Appendix B - Incorrect Method of Installation
I
I
I
I
I
I
I
I
I
I
i
1 19

INCORRECT METHOD OF INSTAIJATION

Manufactured Home

Steel: Strap not placed around bean,
Beam per manufacturerts instructions.

Over Rztension
of metal stand

-Footing of insufficient Ground anchors not
size suitable for the soil
-placed typeg Vith insufficient
on soil containing depth
organic material or L
disturbed uncompacted (;round anchw not
soil -Footings above parallel to strap,
frost penetration
,@@Mmfactur @Hom

I
I
I
I
I
I
I
Appendix C - Correct Method of Installation
I
I
I
I
I
I
I
I
I
I
I
.1 20

CORRECT MEMO OF IN TALUT-IQN

0
C3 Q
OE:110:3

AWMAmatOy 6(r

WTTH INCUNED
"HOM

Manufactured Bone strap spacing and angle per
Approved set-up manmal. Strap
bxkle located close to pier
Steel Beam (vithin 6* - 12" of pier)

Approved Approved Tension Head
getal Stand

IWI Angle of Ancbos
to be vItbIn 5
)dof strip anchor
Depth
//,--@factured @Bow*
;@7A

ILL SID: 06142 IL
Beginning of record displayed.

ILL Pending 19990611 Record 2 of 2
CAN YOU SUPPLY ? MS @NO @COND @FUTUREDATE
:ILL: 1606680 :Borrower: TMA :ReqDate: 19990610 :NeedBefore: 19990710
:Status: PENDING 19990610 :RecDate: :RenewalReq:
:OCLC: 37690639 :Source: OCLCILL :DueDate: :NewDueDate:
:Lender: *NO@ 11
:CALLNO:
:TITLE: Preliminary report disaster site investigation of manufactured
homes / I
@ :IMPRINT: Herndon, Va. : National Conference of States on Building Codes and
Standards, 1989. 1

:VERIFIED: OCLC
:PATRON: a martin DEPT: hist STATUS:grad
:SHIP TO: ILL/University of Memphis Lib./Campus Box 526500/Memphis, TN
38152-6500 1
:BILL TO: same FEIN # 62-0648618/WE MUST HAVE YOU FEIN# ON INVOICES.
:SHIP VIA: LIBR RATE :MAXCOST: $20ifm :COPYRT COMPLIANCE: CCG I
:E-MAIL: INTERNET: [email protected] I
:BORROWING NOTES: TRS PARTICIPANT, TENN-SHARE,SOLINE, ASERL,ARIEL
141.225.193.14 1
:LENDING CHARGES: SHIPPED:-- SHIP INSURANCE:

ALTERNATE VERTICAL ANCHORING METHOD

0
1:3
C3 0-
1:3 F-1 1:1

Opbonal Optional
RestainlrV Restainingl
Plate Winder

WITH VERTICAL
ANCHORS

Vertical
Anchor

12"x12"x0f16-
steel plate
driven next
to anchor 12- dWrWw x 12-
shaft on deM concrete cylkx*
Inside I pour amund anchor shaft

STEEL PLATE CONCRETE CYLINDER
RESTRAINT RESTRAINT
3

I
I
I
I
I
I
I
Appendix D - Texas Tornado Photos
I
I
I
I
I
I
I
I
I
I
I
I . 21

PHOTOGRAPH INDE

Photographs of Manufactured Homes Following
May 16, 1989 Tornados in Cleyburne, Texas

Photo #1 High winds shifted home off of foundation.

Photo #2 Typical improper attachment of anchor strap to I-beam. Clip
is attached to bottom of beam. Proper installation would
show strap wrapped around beam with clip attached top of beam.

Photo #3 Home flipped on its side.

Photo #4 - Typical tie-down straps located on leeward side of
manufactured home. Due to improper attachment, as described
in Photo #2 above, clips fell off of I-beam while anchors
stayed in place.

Photo #5 - Home flipped up-side-down. Note anchors hanging from the
side of home.

Photo #6 - Close-up of anchor and strap on flipped home from Photo #5.

Photo #7 - Three demolished homes. The white one near post was airborne
and destroyed the other two. All three units are at least 20
ft. from their original sites. White unit is approx. 100 ft.
from its original site. There was no evidence of tie-down
anchors at the original site of the white house.

Photo #8 - Demolished homes. Note undamaged neighboring units.

Photo

Home shifted off foundation

Photo #2

FP w ww,

Improper attachment of strap to I-beam (see circled area)

Photo

Home flipped on side

Photo #4

-7

Hooks fell off I-beam due to improper attachment

Photo #5

Si@ -t'

Z@U_@

-7-

Home flipped upside down

Photo #6

Al- F4.

T"I

4@-

Anchor pulled from soil, Hook fell off I@beam

I
I
I
I
I
I
I
Appendix E - Hurricane Hugo Photos
I
I
I
I
I
I
I
I
I -
I
I
1 22

PHOTOGRAPH INDEX

PHOTOGRAPHS OF MANUFACTURED HOME SITE INVESTIGATION
HURRICANE "HUGO". SOUTH CAROLINA
SEPTEMBER 22, 1989

Photo #1 - Demolished home (Note: chassis and floor intact)

Photo #2 - Major damage from overturning onto side

Photo #3 - Metal roof partially removed

Photo #4 - Roof loss and side wall failure

Photo #5 - Home flipped on its side (Note: floor and chassis intact)

Photo #6 - Demolished home

Photo #7 - Metal roof loss and leeward wall collapsed

Photo #8 - Metal roof loss and windward wall damaged

Photo #9 - Demolished home blown off piers

Photo #10 - Demolished home

Photo #11 - Demolished home

Photo #12 - Home rolled over the next mobile home

Photo #13 - Demolished home

Photo #14 - End wall siding pulled out at marriage line

Photo #15 - Portion of end wall overhang siding pulled out

Photo #16 - End wall overhang siding and corner siding pulled out

Photo #17 - Home shifted from foundation

Photo #18 - Home shifted from foundation

Photo #19 - Tie-down anchors pulled out of ground

Photo #20 - Tie-down strap fastened to outriggers and home shifted

Photo #21 - Home shifted off the blocks (Note: strap clip at bottom of
I-beam)

Photo #22 - Leeward tie-down strap

HURRICANE "HUGO", SOUTH CAROLINA (Continued)

Photo #23 - Tie-down anchors pulled out of flooded ground

Photo #24 - Tie-down anchor failure

Photo #25 - Foundation blocks and anchors after hurricane relocated the
home 200 ft. away

Photo #26 - Roof truss to sidewall strap failure

U
CANADA MAINE
Yr.

M CA. NEW
I - BUffalo YORK
PUNSYLVANIA ONN
lan#V Iltuburgh .-THE PATH OF
HURRICANE HUGO

KY.
Noon
VIRGINIA,
Frids Hurrk" kugo made j-,
-TEkN. WWih at Charleston,
NORTH
71c-ko CAROLINA S.C..'wb @ sustained
.-Y
OUT winds of 135'rnph
H
OUTH
CAROYNA moving northwest at 22
4 mph! The storm viss-
Colu la
downgraded to a
GEORGIA tropical, itorm Ati It
crossed kft Noft
@@rday.

QRIDA

Photo

via

Demolished home (Note: chassis and floor intact)

Photo #2

Major damage from overturning onto side

Photo #3

Metal roof partially removed

Photo #4

Roof loss and side wall failure

AvA
Photo @'5

J@
v

JEW'

Home flipped on its side (Note: floor and chassis intact)

V71
4". Photo#6

04
4@

r
'Awl

Demolished home

Photo 47

,44

Metal roof loss and leeward wall collapsed

Photo #?8

SNOW,
J
4M. 1
Ii a--

Metal roof loss and windward wall damaged

Photo #9

Demolished home blown off piers

Photo #10

-7

Demolished home

Photo

42ki 4 40

All

411,

Demolished home

Photo #12

Home rolled over the next mobile home

Photo ?g?13

tj
Y@

IIJALA.

A Al

Demolished home

Photo #14

.-ION

-:7
-, i@Az-

End wall siding pulled out at marriage line

Photo #15

Portion of end wall overhang siding pulled out

Photo #16

Owl

End wall overhang siding and corner siding pulled out

Photo #17

1 3

Home shifted from foundation

Photo #18

Home shifted from foundation

Photo #19

Tie-down anchors pulled out of ground

Photo #20

Tie-down strap fastened to outriggers and home shifted

Photo #21

Home shifted off the blocks (Note: strap clip at bottom of I-beam)

Op"

Ma
-:-;7 Photo #22,

Leeward tie-down strap

Photo #23

Tie-down anchors pulled out of flooded ground

Photo #24

F mrowxi@ J4 -

MOOED"

Tie-down anchor failure

Photo #25

*OP-1

ilk

Foundation blocks and anchors after hurricane relocated the home 200 ft. away

Photo #26
I,,, WO
C-

Roof truss to sidewall strap failure

I
I
I
I
I
I
I
I Appendix F - Hurricane Jerry Photos
I
I
I.
I
I
I
I
I

PHOTOGRAPH INDEX

PHOTOGRAPHS OF MANUFACTURED HOME SITE INVESTIGATION
HURRICANE "JERRY", GALVESTON, TEXAS
OCTOBER 15, 1989

Photo #1 - Major damage to home

Photo #2 - Major damage to home

Photo #3 - Roof and siding loss

Photo #4 - Roof loss

Photo #5 - Improper anchoring

Photo #6 - Improper blocking

Photo #7 - Roof damage

Photo #8 - Roof damage

Photo #9 - Eave damage

Photo #10 - Siding and eave damage

Photo #11 - Siding damage

Photo #12 - Siding damage

THE PATH OF HURRICANE JERRY

w v

IbL

IZZ,

It's

Photo #1

Major damage to home

Photo #2

Major damage to home

Photo #3

Roof and siding loss

Photo #4
7M
Od W@@`
- --- R@mw.,,71 W
71im- V-- A-

Roof loss

Photo #5

Improper anchoring

Photo

7 -

Three demolished homes.
The white one (near post) was airborne and destroyed the other two. Photo

A - Mai

W-7
S.- 'to
Al OL
r-7 vpt-T

a ft%v

Lwow

Demolished homes (note undamaged neighboring homes)

Photo #6

Improper blocking

Photo #7

Roof damage

Photo #8

ir:

IL-

Roof-damage

Photo #9

1
14-1
A If

Eave damage

Photo #10
77N@

Siding and eave damage

Photo #11

--Ww

Siding damage

Photo #12

M. v grew"

..........

Siding damage

I
I
I
I
I
I
I
Appendix G - California Earthquake Photos
I
I
I
I
I
I
I
I
I
I
I
1 24

PHOTOGRAPH INDEX

PHOTOGRAPHS OF MANUFACTURED HOME SITE INVESTIGATION
EARTHQUAKE, CALIFORNIA
OCTOBER 17, 1989

Photo #1 - Homes completely shifted off piers, supported by wheels
Photo #2 - Homes completely shifted off piers, supported by wheels

Photo #3 - Homes partially shifted off piers

Photo #4 - Homes partially shifted off piers

Photo #5 - Home destroyed by fire

Photo #6 - Damaged skirting and electrical service pedestal

Photo #7 - Home damaged by piers

Photo #8 - Home damaged by piers

Photo #9 - Ground anchor of shifted home

Photo #10 - Typical pier with small bearing surface

Photo #11 - Earthquake jacks

Photo #12 - Earthquake jacks

CALIFORNIA EARTHQUAKE

n Franc im
AP

4MM*AgW

s
Andreas Fau' ft
cu I

Ar
AL

picenwr:10 miles
NE of Santa Cruz,
ntore
071 on Richter s%Aale
sa

Photo #1

Homes completely shifted off piers, supported by wheels

Photo #2
'q

Homes completely shifted off piers, supported by wheels

Photo #3

VIP"

Homes partially shifted off piers

Photo #4

48a

aim&,

Homes partially shifted off piers

JOE

Photo #5

a, EWf

i6,
77.

Homes destroyed by fire

Photo

Damaged skirting and electrical service pedestal

Photo #7

Home damaged by piers

Photo #8

Home damaged by piers

Photo #9

Ground anchor of shifted home

Photo #10

Typical pier with small bearing surface

Photo

--INC
v-4aim

Earthquake jacks

_10
Photo #12
41

Earthquake jacks

I
I
I
I
I
I -
I Appendix H - National Profile of Tie-down Regulations
I
I
I
I
I
I
I
I
I
I
I
25
11

NATIONAL PROFILE OF STATE REGULATIONS
REGARDING INSTALLATION OF MANUFACTURED HOMES

State Program Criteria States (percentage of homes
potentially affected)

1. Scope of regulation and iffectiveness
of enforcement programs.
a. States having little or no in- (21 States) AK, CO, DE. GA, (43%)
stallation regula 'tions and no ID, IL, LA. MT. NE, NH, NO.
enforcement programs. OH. OK. PA, SC, SO, UT, VT,
W , WI, W
b. States having some installation (11 States) CT, IN. KA, KY. (22%)
regulations and no enforcement MD. ME, NO, NJ, NY. RI, VA
programs.
c. States having significant instal- (9 states) AL, AR, FL, MA, (18%)
lation regulations but ineffective MI, MN, NC, OR. WA
enforcement programs.
d. States have significant instal- (8 states) AZ. CA, 1A, MS, (16%)
lation regulations and also NV, NM, TN. TX
effective enforcement programs.

2. Monitoring by state of the local
government for compliance with state
laws.

a. States that monitor performance of (4 states) AZ, CA, 1A, OR 8%)
local inspection agencies and
provide training to local inspectors.
b. States that do not monitor perform- (11 states) AL. AR. FL, MA, (22%)
ance of local governments. MN, MO. NC, OR. RI, VA, WA
3. Bonding of dealers and installers.
a. States that require bonding of V states) FL. MS, NC, NM, (14%)
dealers. TN, TX, WA
b. States that require bonding of (5 states) NM, NC, TN, TX, (10%)
installers separate from dealers. WA
4. States that require licensing of (13 states) AZ, CA, MS, MI, (27%)
installers. ME, MN, NV, NM, NC, R1, TN,
TX. WA

NATIONAL PROFILE OF STATE REGULATIONS
REGARDING INSTALLATION OF MANUFACTURED HOMES

State Program Criteria States (percentage of homes
potentially affected)

S. Installation Standards

a. States that require compliance with (24 states) AL, AZ. AR, CA, (49%)
home manufacturers installation FL, IA, IN, KY. MD. MA, MI.
instructions. MN, MO, KV, ME, MS, NC, NY,
OR, TN, RI, TX. VA, KA
b. States having blocking and tie- (22 states) AL, *AZ. AR. CA, (45%)
down standards FL, IA. IN. KY, KS. MI, MN,
NV, NJ, KM, NC, OR, MO, RI,
TN. TX, VA. KA
(* No tie-down standards)
6. States that require systems component (17 states) AL. AZ. AR. CA, (35%)
manufacturers to seek approval from FL. IA, KS. MI, MA, MN, MS.
the state NO. NV. NM, OR, TN, TX
7. States that have state inspection or (6 states) CA, IA, NM, OR, (12%)
permit fees. (This does not include TN, TX
any requirement of additional fees
established by the local government.)

1DEFINITIONS OF SCOPE OF REGULATION AND EFFECTIVENESS
OF ENFORCEMENT PROGRAM

a. The states may not have any state-wide laws or programs such as the
following: state licensing or bonding of installers; blocking and
tie-down standards; routine state inspection of installation of homes;
reporting mechanism for installation; components listings. The local
government in these states may or may not have related laws.

I
I
I
I
I i
I
I
Appendix I - NCSBCS Tie-down Recoffnendations
I
I
I
I
I
I
I
I
I
I
I
1 26

Recommendati on s /Steps to Improve Installation of Manufactured Homes

The chart below lists recommended steps that HUD, the states,
home manufacturers, manufacturer associations, ground anchor or
other component manufacturers, insurance companies. and state
associations of manufactured home dealers and manufacturers can
or should take to improve instillation of manufactured homes.

SUGGESTED LIST TO FOLLOW TO HELP IMPROVE --INSTALUnON OF KMUFACTURED HOMES

* Develop training manuals, video training modules for
installation of homes.

9 Collect existing technical data about components used in
installation: provide a central source to all state and
local offices.

* Hold series of workshops with SAA, manufacturer's
associations, dealer's associations. local. county
officials. HUD area offices on installation.
* Place emphasis on the optional responsibilities of the
SAAs, such as dealer lot and installation inspection by
making them mandatory.

* Develop brochures and small handbook for homeowners
which contain Information about key installation issues
and coordinate distribution of such material through
manufacturers, dealers, and homeowner associations.

e Consider developing Federal standards on installation
hardware, such as ground anchors, buckles, piers, etc.;
or work with building codes or standards' organizations
such as ANSI and ASTM to develop such standards.

SUGGESTED LIST TO FOLLOW TO HELP IMPROVE INSTALLATION OF MANUFACTURED HOMES

HUD (continued)

* Notify lenders of problems (Title I program).

e Enhance review of DAPIA approved set-up manuals.

e Require additional data collection and evaluations on
the following:

- On-site investigations in the remaining 40 states not
covered by the study.

- On-site investigations in all 50 states of
manufactured homes on permanent,foundations.

- State programs including detailed comparison of
current state standards and regulations.
- County and city programs.
- Listing programs of installation hardware and
verification of their performance by testing.
- Homeowner manuals in complianc'e with CFR 3283.
- Cost/benefit evaluation of improved instal I ati on
procedures.

- Cost/benefit evaluation of state/local regulation of
installations.

- Reduced durability and consumer sati sfaction by
improper installation (relationship of consumer
complaints with improper installation).

SUGGESIED LIST TO FOLLOW TO HELP IMPROVE INSTALLATION QF- MANUFACTURED HOMES

STATES

* State task force should work in development of MODEL
PROGRAM for states (including laws. regulations, and
standards).

e Work toward developing uniform installation laws and
effective inspection programs
e Team up and develop training programs for state and
county inspectors.

e Establish standards for components used in installation.
e If states do not have necessary laws, coordinate meeting
with the state manufactured housing associations and HUD
to develop support for laws.
0 Place more emphasis on the *optional SAA
responsibilities," such as dealer lot monitoring and
installation inspections.

HOME KANU CTURERS

e Improve the set-up manual and homeowner manual (the
design and clarity of details).
e Provide technical training to the service personnel
either independently or in coordination with other
manufacturers.

e Coordinate the training of dealers and installers with
other manufacturers.

e If HUD or states offer training programs, participate in
those programs.
e Identify on each home by Oflag" or other means wh,ere
piers and anchors are required.

e Promote consumer awareness.

9IGGESTED LIST-TO FOLLOW TO HELP IMPROVE INSTALLATION OF MANUFACTURED HOMES

MANUFACTURER ASSOCIATIONS

* Coordinate the suggestions to HUD, the states, and home
manufacturers.

GROUND ANCHOR or OTHER COMPONENI MAMUFAMRERS

* Include installation instructions with their hardware.

* Provide training to installers.

* Conduct more testing and research.'

INSURANCE COMPANIES

# Offer better rates for homes that are installed properly
and are certified for proper installation by reputable
inspection agencies.

STATE ASSOCIATIONS of KMUEACTIUREQ HOMES

e Coordinate with states the bonding, licensing, and
training programs for dealers and installers.

EMBCS-

11
10111am
@.,- -3 6668 14107 0534 -