Response to the Richard Mendlen email;

The 1994 report (NISTIR 5370); is a good source of information as well. It was based on standards, and practices, in November 1994. We will let Frank Walters with MHI/MHRA speak for the 1994 NIST report (NISTIR 5370).

Frank Walter PE, January 10, 1995
It is a continuation of the earlier study (NISTIR 5189) by the same author, Richard D. Marshall, which was dated May 1993. HUD used that study to justify the revised wind standards for the new high wind zones.

The attached study, NISTIR 5370 (Nov, 1994), "Manufactured Homes-Probability of Failure and the Need for Better Windstorm Protection Through Improved Anchoring Systems", is particularly critical of traditional soil anchor installation practices. It concludes that major changes in design requirements are needed.

The report includes a number of recommendations for the installation of anchoring systems. It suggests that a new approach is needed to provide windstorm protection in the high wind regions.

Richard D. Marshall, 1994 (NISTIR 5664)
The traditional shallow soil anchor/tie/pier system should not be used for windstorm protection where the basic wind speed exceeds 44.7 m/s (100 mph).

This above statement from the 1994 NIST report is what Mr. Walter is referring to when he says; it suggests that a new approach is needed for windstorm protection in high wind regions. The statement by Frank Walter, PE leaves it open as to whether this means a new approach to installation, or a new type of anchoring system for manufactured homes. Marshall's statement makes it clear that it is the latter of the two.

It can be said, that the May 1993 report was primarily recommendations by NIST, based on assessments of damage, caused by Hurricane Andrew. It can be said, that the November 1994 report, which you failed to mention, was a continuation of the 1993 report. This of course does not mean that the 1993 did not address support and stabilization. Marshall, as well as any Structural Engineer that we have spoken with, and consider non-biased, DO NOT consider "support and stabilization" (foundation systems) to be separate from the structural design of a home, nor does HUD. Restated, each structural component of a unit is dependent on the integrity of all of the other structural components (3280.305). If one of these components is not capable of carrying its load, the integrity of the other structural components cannot be assured. Hence, the federal standards, and regulations require the manufacturer to provide the design for the support and stabilization (foundation system), and include this design in the design package (3282.203(b)(10)). The DAPIA then approves the manufacturer's design package as being compliant with the federal standards, or rejects the design.

This is real simple guys, a manufactured home, in wind zone II for example, which is destroyed, or structurally damaged at winds of 100 or less mph, as a result of inadequate support and stabilization, was not capable of transmitting the design loads specified in the federal standards, to the ground, without exceeding the allowable stresses and/or deflections. It is irrelevant what caused the home to exceed the stresses and deflections specified in the federal standards. The fact remains that the home exceeded the stresses and deflections, specified in the federal standards, and therefore the manufactured home did not perform as the manufacturer certified that it would, to the federal standards for the performance of manufactured housing.

The manufacturer provides the design for the support and stabilization system for the home, and certifies that their design for the support and stabilization will without failure; transmit the design loads specified by the federal standards, to the ground. The DAPIA authenticates the manufacturer's certification by inspecting and then approving the design for the manufactured home. Every time anyone allows, or requires a home to be installed in ANY manner that is inconsistent with the manufacturer's installation instructions, they let the manufacturer off the hook, for the certification that they made, and for their warranty. If it does not perform as advertised, and certified, then the home did not, and does not comply with the performance requirements of the federal standards.

The original Act, by the language contained within, prohibited the Secretary from establishing any excusatory criteria, which ended his authority, to inspect, and investigate, and order the repair of a manufactured home that did not perform to the federal standards. In other words, the Secretary's authority to enforce the federal standards extends to the product produced, so long as the product produced has been certified by the manufacturer to be compliant with the federal standards. Without the manufacturer's certification, the Secretary of Housing has no legal authority over the product produced, and/or the manufacturer who produced such product. Inversely the manufacturers of this product could not construct a product that bore the nomenclature, "manufactured home" unless they agreed to be responsible for their product's compliance with the performance requirements of the federal standards. This agreement arises each time the manufacturer certifies that their product complies with the ALL-applicable federal standards.

The original "Manufactured Home Construction and Safety Standards Act" never prohibited the manufacturer of this product from building to other codes. It merely gave the manufacturers the option, or privilege of building to one code (National Preemption), to satisfy all code jurisdictions in the United States. In exchange for this privilege, the manufacturers agreed to be responsible for the performance of their product, as designed and constructed, no matter where, or in whose possession this product was. This agreement by the manufacturers gives the Secretary of Housing authority to regulate every aspect of manufactured home construction and assembly, well beyond the back door of the factory. If the manufacturers did not like the conditions that society attached to National Preemption, they were free to certify before beginning construction on a product that they would produce, that they would build their product to some other National Building Code, and thereby, the product they are building cannot be a "manufactured home". This is where the majority get lost; National Preemption is not some legal right that society owes to the manufacturers of this product, it is a privilege that society has granted to the manufacturers of this product, in exchange for the manufacturers assurance that the product they produce, will assure quality, durability, and safety. Under the current system, the manufacturers disavow responsibility for their products FULL compliance with the federal standards. Once the product leaves the factory, the manufacturer cannot assure society that they are upholding their end of the deal, and therefore the manufacturers are not entitled to the privilege of building homes to standards that carry the weight of National Preemption.  

We will now address the only two recommendations, in the 1995 report (NISTIR 5664), "Recommended Performance Based Criteria for the Design of Manufactured Home Foundation Systems to Resist Wind and Seismic Loads".

Richard D. Marshall, 1995 (NISTIR 5664)
When conventional shallow soil anchors are used for windstorm protection, they should be installed vertically to maximum depth and pre-loaded in the direction of the service load to ensure sufficient system stiffness and withdrawal capacity.

Let us discuss "pre-loading" anchors, which Richard D. Marshall is referring to in his 1994, and 1995 NIST studies of manufactured home foundations, and anchoring systems. Comparing the "pre-loading" of anchors, a method devised by Yokel, which Marshall referenced in the 1994 and 1995 NIST reports, to the current practice of pre-loading anchors, which is described in the MHI/MHRA report, is like comparing a ten story building to a one story building, and saying they are the same height.

Yokel first noticed the advantage of pre-loading anchors, during his 1982 study of soil anchors. Yokel suggested that anchors be pre-loaded to 1.25 times their working load. To explain what Marshall and Yokel mean when they use the term "pre-load", we will describe this for those of you who do not already know.

First, you toss out the 4725-pound capacity for anchors, since it is unrealistic. The manufacturer would instead, stipulate the total working load that the anchoring system must resist, for the wind zone where the home will be installed. The manufacturer's agent (retailer) in the field would then determine the number of anchors required to achieve this performance, based on anchor capacity that can be achieved at a particular site. The spacing would have to be coupled with a minimum, and maximum anchor spacing to avoid unreasonably close, or wide anchor spacing, where dense or loose soils are encountered. The implementation of this approach would require an initial estimate of the anchor capacity (ultimate load) at an installation site, either by load test, or previous experience. The manufacturer's agent would then divide the anchor capacity for that given site by 1.5 to obtain the working load for the anchor. With this value, the manufacturer's agent would determine the number of anchors necessary to achieve the manufacturer's stipulated total design load. The manufacturer's agent would then multiply the working load by 1.25, to determine the preload value for each anchor in the anchoring system. The manufacturer's agent would then install the anchors, and pre-load them to that value. For the industry's mythical working load value of 3,150 pounds, this would call for a preload value of 3,938 pounds, which is far more than any of the manufacturer's agents can achieve by turning a monkey wrench, on a anchor head bolt. The allowable displacement is as much as is necessary to achieve the pre-load value. This would call for displacements, in many cases in excess of 4-inches, which is why anchor stabilizer plates cannot be used, when this approach to pre-loading anchors is used. The manufacturer's agent would install the anchoring system connecting hardware, after the preloading of the anchors had been accomplished.

It is the horizontal displacement of an anchor, which occurs in the direction of the load, after vertical with the ground plane where the benefits of preloading are realized. This is nothing more than simple geometry. When you install anchors at 15 degrees of vertical, away from the direction of the load, and then pre-load using the industry method, 3-4 inches in the direction of the load, you have simply returned to vertical. This leaves the majority of the necessary horizontal displacement, required to achieve pre-loading, as Yokel and Marshall meant, unaccomplished. It is doubtful if the pre-loading being practiced by this industry can be credited for any more than a minimal improvement in the performance of this type of anchoring system.

There are many benefits to using Yokel's approach, and HUD and the industry know what these benefits are. According to documents, it appears that HUD and the industry were more concerned about the physiological effects on the manufacturer's agents (retailer), of using this approach, than they were with assuring the safety of the users of this product, and the public. We have already gone beyond what is necessary to point out the differences in what the industry practices as pre-loading anchors, what HUD appears to mean when they use the term "preloading", and what Yokel and Marshall of NIST meant, when they used the term "pre-load", therefore we will not go into further detail on this subject.

Richard D. Marshall, 1995 (NISTIR 5664)
The traditional shallow soil anchor/tie/pier system should not be used for windstorm protection where the basic wind speed exceeds 44.7 m/s (100 mph).

Marshall states unequivocally that even if anchors are pre-loaded, as described above, that traditional soil anchors should not be used in high wind regions. This is because anchors, even if pre-loaded, generate pullout resistance that is lower than what is necessary to prevent unusually close anchor spacing, much the same as those being sought by the industry, making multiple anchors dependent on the same soils, to resist pull-out (cones of influence). In addition, this MHI/MHRA initiative, which appears to have HUD's support, shows that HUD is supporting an initiative to legitimize the continued use of traditional soil anchors in high wind regions, not the discontinued use of anchors in these regions. As well, this industry-sponsored initiative proposes to replace the current mix of performance and prescriptive-based criteria, with purely prescriptive criteria. This is in the opposite direction of what was recommended in the 1995, NIST report, and this is moving to strictly performance-based criteria. With the exception of a small percentage of manufactured homes, which HUD will issue Mortgage Insurance Policies for (see HUD document 007487), it appears that HUD, as well as the industry, have disregarded the recommendations in the 1995 NIST report (NISTIR 5664), as they related to windstorm protection.

Richard Mendlen, HUD
It is my understanding that such testing is being considered by MHRA.

Is it HUD's position that they will wait an undetermined amount of time for the industry being regulated, (during which time more manufactured home buyers may be placed at risk) to decide if they will perform a study, to determine if what they are doing, is safe? This reinforces our opinion that a rather deleterious relationship has developed between the regulators, and the industry being regulated.

Richard Mendlen, HUD
However, as a result of the new manufactured home legislation, we "will" be able to address these issues in the future.

Richard Mendlen, HUD In the fourth line of the fifth paragraph, please revise my message to read as follows:

"...legislation, we "may" be able to address these issues in the future."

It appears that HUD has taken the position that the newly passed legislation, may not after all, give HUD the authority to regulate the installation of manufactured homes, is this correct?

Richard Mendlen, HUD
In particular, certain statements in your report suggesting the Department was ignoring the anchor embedment issue or permitting large horizontal deflections that would affect pier stability are simply not correct.

HUD's 1991 report WJE NO. 901798
New failure criterion - the failure criteria suggested in the proposed ASTM Standard was ignored in subsequent testing. For subsequent testing, it was decided in consultation with the HUD GTR (Inserted by author: HUD GTR - Richard Mendlen), to test anchors that were fully embedded, and record load and deflection until one of the following conditions existed: a. the applied load begins to decrease with increase in deflection or b. the horizontal deflection exceeds approximately 18 in. The test stopped when one of these two conditions was met.

No other explanations for taking this course are given within the report. Given the removal of the horizontal displacement criterion from the ANSI A225.1 standard in 1987, and the use of the term "ignored", it is questionable if this was taken completely out of context. I will however accept the argument that the objective became a study to determine ultimate failure, since the report does not contain enough information to assess your current statement on this matter, and determine otherwise. As noted in Appendix A, of this same report, as well as the dozens of manufactured home installations observed by this author (1999-2000), installers pay little, if any attention to anchors, which are not fully embedded, and the percentage of anchors not fully embedded, reflect those disclosed in the report. They simply hook them up, and move on to the next one. It is apparent that some five years later, if we accept that this was addressed in 1995, that the importance of fully embedded anchors, has not been recognized where it is counts. This issue cannot be disregarded when determining or recommending load values, given that in practice, this there appears to be a total disregard for it importance. Since the MHI/MHRA study, which HUD appears to support, has also disregarded this problem, by using only load values for fully embedded anchors in determining the load values recommended by MHI/MHRA, I stand by my original statement that HUD has the large deformations and low load values associated with anchors that are not fully embedded.

Sincerely,

John Taylor
"The American Internet Society of Manufactured Home Owners"
(TAISMHO)