To:     Mr. Chris Stephens, Assistant Fire Marshal
          Manufactured Housing Division

From: John Taylor - Founder - TAISMHO
           Deborah Chapman - Chair - NFMHO

Subject:  The proposed MHRA test protocol for soil anchors

HUD, the States, and the industry have been testing, retesting, and retesting, and retesting helix soil anchors since 1973. When the results do not provide the desired results, they write new test protocols, and new test criteria, and test again, and when this does not work, they do it all over again. What is significant, is they still have not, been able to come up with test data, which supports the past, current, and continued use of the helix soil anchoring system, and the MHRA proposal actually points out that the past and currently accepted industry standard, for anchor performance is invalid.

From at least 1982, until now, there has never been creditable evidence, that this equipment meets the federal standards. It is argued by the industry, that there is no proof that these devices do not work. It can be argued these devices do work, since they will provide for some resistance to windstorms. The problem lies in; they will not provide the necessary resistance required, to assure the safety of the occupants of the manufactured home, to the level, which is being claimed. Their argument has no merit given the test data, and is a deleterious one, for the occupants of manufactured housing. The proof is in the test data, and the data is supported by the high percentage of manufactured homes destroyed, as compared to the damage suffered by conventional housing, caused by severe weather events. It seems rather deleterious to continue chasing a test protocol for helix soil anchors, which will support the continued use of these devices. Since it is apparent that HUD and the industry are not going to acknowledge, in the name of affordability, the inherent weaknesses in this type of tie down system, we will address the MHRA Anchor spacing chart, by addressing the problems with helix soil anchor tie down system.

Clarification of Terms

(kip) = 1000 pounds
(COV) = Coefficient of Variable - (COV x 100 for percentage of probability of failure)

Note:   Table and Figure numbers are not the same as in the reports from which they were taken

It is concluded from the MHRA report "Field Testing of Ground Anchors", that the testing of helix soil anchors did not include testing of these devices without stabilizer plates, which prevents comparison to anchors loaded without stabilizer plates. This is of particular concern given the large coefficient of variation noted by Marshall (1994), for anchors when stabilizer plates were used. Since the MHRA report is void of this information, we will rely on previous test data. Second, the ultimate load, under inclined loads was excluded from the testing.

Coefficient of Variation

Marshall (1994)
It was not possible to install the anchors to full depth in every case and those anchors have been excluded from the summary of results listed in Table 1. Also excluded from Table 1 are certain test results classified as "data outliers" that ranged from a low of 35 lbf to a high of 3,400 lbf at a horizontal displacement of 4-in. The effect of including these outliers is to almost double the COV. Mean resistances and the corresponding standard deviations are shown in Figure 1, A-B, along with the ANSI A225.1 requirements.

Marshall (1994)
It is seen from Anchors without stabilizer plates developed only about 21% of the required resistance the horizontal displacement limit of 4 in., and the use of stabilizer plates resulted in only a marginal improvement (26% of the required resistance). An additional problem with stabilizer plates is the large variation in resistance (COV = 0.41 at the limiting horizontal displacement) that must be accounted for when assigning an allowable working load or resistance factor. Even at maximum horizontal displacement, of the order of 10 to 20 inches, the anchors developed only about 80 percent of the required minimum-working load.

For Both figures 1-A & 1-B, and b1 = 45 degrees  b2 = 105 degrees (anchor installed at 15 degrees of vertical) (Person et al. 1991)

A larger than acceptable coefficient of variation was again apparent in the MHRA report (2000), although there does not appear to be any consideration given, to its significance in determining the in-service working load of the anchors. It is also noteworthy that in the MHRA report, that when the coefficients of variations were calculated, COV values (0.21) were almost half of those that were calculated (0.41) from the Pearson report (1991). This could be explained in part, by the reduction from 4-in to 3-in for the limiting horizontal displacement (4-in, phase 1-Pearson, 3-in MHRA), and the soil at the selected test sites.

At the shallow depth that stabilizer plates engage the soil, they develop relatively low mean resistance and are associated with an apparent large coefficient of variation. Given the available data, it would appear that stabilizer plates cannot be considered, an effective means for limiting the horizontal displacement, which is necessary to ensure pier stability, and prevent damage to utility connections, such as gas lines.

Double 4-in Helix anchors vs. Single 6-in Helix Anchors

There is no reference made in the MHRA report as to whether or not the anchors tested were single helix (6-in.), double helix (4-in.), or a combination thereof. From the photographs provided in the MHRA report it appears, only single helix anchors were tested. This again raises a serious question. Yokel, et al. (1982) carried out an extensive test program for soil helix anchors in which it was noted that there was a significant difference in the performance between the two types of helix soil anchors. Pearson (1991) also noted a significant difference in the performance between the two types of anchors. The final spacing chart from MHRA does not specify single or double helix.

Yokel et al. (1982)
There was a considerable difference in stiffness between anchors ST 91 (4-in, double helix) and ST89, namely the 6-in. single helix anchor had smaller lateral displacement than the 4-in double helix anchor. This difference was consistently observed in all the anchor tests and was not anticipated, since it was thought that the long slender shaft of the 6-in single helix anchor would provide less resistance to lateral displacement.

Pearson et al. (1991)
It will be noted that Fig. 2 contains four sets of results for the 30-in. anchor with two 4-in. diameter auger disks. The anchor was installed at approximately 15 degrees with the vertical axis and pulled initially at 45 degrees with the vertical axis. No stabilizing plate or concrete collar was used at the surface.

In our opinion, the last to loading curves are not representative load-deflection curves for this soil anchor. For curves labeled "3rd" & "4th" loading, the loading rod bound against the horizontal member of the frame and, therefore the measured load was not applied to the anchor. This problem was rectified by modifying the load frame. Several other anchors were pulled using the same procedure. The results were very similar to those reported in Figure 2. It was clear from the above description and results in Fig 2, i.e. large deflections at relatively low loads, that these test results were totally unexpected.

It would appear that Pearson (1991) had access to Yokel's report (1982) and therefore should not have been surprised by this phenomenon. Yokel's testing revealed that in general, the pullout capacity of vertically installed, axially loaded, double helix anchors, was approximately half the capacities of the 6-in. single helix anchors, which were similarly loaded. Other double helix anchors, installed, and loaded at various angles had similar results. Yokel drew the following conclusion as to why this occurred.

Yokel et al. (1982)
Several factors combined to produce the difference in performance:     1. The embedded depth of the helix of the 6-in single helix anchor is deeper (3.7 ft vs. 2.6 ft):    2. The 6-in single helix plate has a larger area (2.25 times the area) and   3. The 4-in double helix anchor has two helixes. Many authors claim that anchor capacity is proportional to the area of the anchor plate.

Corrosion

It was revealed by both Yokel et al. (1982) and Pearson et al. (1991) that the corrosion resistive coatings for anchors are badly damaged, and in some instances, completely stripped during the installation of the anchor.

Yokel et al. (1982)
After installation, painted anchors do not seem to have effective corrosion protection because of paint stripping.

Pearson et al. (1991)
All anchor shafts yielded during the installation phase as exhibited by the flaking of paint.

Marshall (1994)
Protective coatings currently used on helix-soil anchors are badly damaged or totally removed during anchor installation.

There is sufficient evidence that painted anchors have little, if any chance of ensuring effective corrosion control.

It is not sufficient to argue that the devices are provided with corrosion protective coatings, when it is known that this protection is rendered ineffective, by the installation of the device. It is also not sufficient to argue that the anchors will provide windstorm protection upon installation, while there is a high probability that the anchors will succumb to the effects of corrosion, which will result in significant reductions in load capacity, long before the in-service life of the home has been reached. Yokel et al. (1982) and Pearson et al. (1991) both noted that anchors should be galvanized.

Pearson et al. (1991)
Anchors should be galvanized

Yokel et al. (1982)
Galvanized anchors would probably perform adequately

Some galvanized anchors are currently available on the market. However, the paint-finished anchors are not prohibited from use by HUD, State jurisdictions, or by any of the Manufacturers, whose installation manuals we have reviewed. This makes it highly unlikely that the installer will select the more expensive galvanized anchors. The obvious benefit towards assuring the homeowner's safety, during the in-service life of their home, is well worth the relatively low, increased cost to the homeowner.

Installation

Pearson et al. (1991) noted that they had difficulty installing (fully embedding) the 48-in anchors. It was also apparent from the information provided in the report by Froehling et al. (1999)(MHRA Report), that only anchors that were fully embedded, were part of the testing. This is significant since it is common to observe, on completed installations, that approximately half of the anchors for a home, are not fully embedded. Table 2, on the following page, shows the correlation of anchor depth to anchor capacity for coaxially loaded anchors.

Yokel et al. (1982)

It should be understood that anchors, which are not fully embedded would be subject to larger horizontal displacements, and significantly reduced inclined load values, as well as the reduced vertically loaded values for vertically installed anchors noted in Table 2.

Pearson et al. (1991)
Almost immediately, difficulties were encountered with installing the anchors. During the installation process, the four-foot (48-in) anchors typically stalled after being driven two to three feet into the sand.

In all cases, the only obstacle encountered was the frictional resistance of the sand acting on the auger disks. This condition was verified in several instances by digging up the anchor and examining the subsoil for rock, boulders, or other obstacles in the vicinity of the auger. No obstacles were found. Because of these difficulties, less than half of the ninety test program anchors in the field were installed to their full-intended embedded length.

Pearson et al. (1991)
The majority of 48-in anchors were unable to be driven to their full depth at the test site. Because of the type of soil, it was expected the soil be the most favorable for installing these types of anchors. This is very significant as anchors, which were not completely embedded carry very low load levels and exhibited large deformations at these low load levels.

Given the industry's unwillingness to abandon soil anchors for competitively priced, alternative systems, it is not likely that this detrimental installation problem will be resolved. This excluded factor makes it impossible to assure, even with the proposed reduced loads, noted in the MHRA proposal, that helix soil anchors can be considered an effective means of manufactured home anchorage.

Marshall (1994)
Data from Vann, et al. (1978) indicates that the actual breaking strength of installed strapping may be only 80% of the strength indicated by ASTM D3953 due to flexing involved with strap installation, imperfect wrapping of the strapping around tensioning devices, and stress concentrations caused by buckles and similar hardware used to secure strapping.

Pearson et al. (1991)
Two of the five steel strapping products evaluated failed to meet the minimum breaking strength of 4750 lbs required by the Federal Specification QQ-S-781G. In addition, two of the other strapping products did not meet the minimum thickness requirements of that specification

Results of laboratory tension tests on cold-rolled steel strapping (Pearson et al. 1991)

Marshall (1994) drew from the data collected by Pearson et al. (1991), and Vann et al. (1978) and provided the following analysis.

Marshall (1994)
Considering the test data presented herein, it appears that the anchor system requirements of the MHCSS, and ANSI A225.1 are based almost entirely on the nominal breaking strength of 31.75 x 0.89 (1 1/4 x 0.035 in.) cold rolled strapping with a 1/3 reduction for allowable working load. There appears to have been no recognition of the variability of ultimate strength from product-to-product or to the adverse effects of strap installation on ultimate strength. While the variability of ultimate strength for a given product is small (COV typically less than 0.02), the variation between products and the detrimental effects of installation need to be accounted for. Unfortunately, test data on the strength of installed strapping are insufficient to quantify the variability, but the limited data that are available suggest a reduction factor of 0.80 and a corresponding in-service ultimate strength no greater than 16.90 kN (3,800 lbf).

Due to a lack of any additional data, or at least available data, there seems to have been little, if any interest in discovering whether or not, there is a significant problem in the quality of steel strapping, from manufacturer-to-manufacturer and, what the effects of installation and connecting hardware, have on the ultimate breaking strength of cold rolled steel strapping. However, Pearson (1991) did recognize this as a serious problem and, recommended that this issue be addressed.

Pearson et al. (1991)
Enhanced follow-up testing and quality control procedures should be imposed on steel strapping materials to assure conformance to Federal Specification QQ-S-781G.

The available data, supports a much lower ultimate load for cold rolled strapping (3800 lbf), than that required by the MHCSS (4725 lbf). It would be prudent to address this issue before moving on with a protocol, that excludes this variable, along with the ultimate load capacity for helix soil anchors, which is intended to be used for justifying the use of this equipment, for anchoring manufactured housing.

Cones of influence

Marshall (1994), made it clear that soil anchors should not be used in wind zone II and III due to the cones of influence, and they should only be used in wind zone I if they were pre-loaded (pre-loading should not to be confused with pre-tensioning). The effects of the cones of influence on the ultimate capacity, will not likely be addressed until a severe weather event, forces it to be addressed, or another few years passes, which seems to be the cycle for separately re-addressing, in some modified way, one of the many individual weaknesses inherent, with the helix-soil-anchor tie down system.

Marshall (1995)
The factored loads for a basic wind speed of 100 mph resulted in a maximum anchor spacing of 7.3 ft. At higher wind speeds, the anchor spacing becomes so small that the cones of influence begin to overlap significantly. Therefore, even with preloading, the traditional shallow anchor/tie/pier system is limited in application to basic wind speeds less than about 44.7 m/s (100 mph).

Froehling et al. (1999)(MHRA Report) noted that the engineering analysis suggested anchor spacing values that were less than the anchor length.

Froehling et al. (1999)
The MHRA committee overseeing the project recognized that at such close distances between anchors there would be overlapping "cones of influence" where the same soil is being counted on to hold more than one anchor. The committee acknowledged the importance of the "cone of influence" to the performance of the anchoring system is not well understood and is an area for subsequent research.

Froehling et al. (1999), were concerned enough about the cones of influence, that when the engineering analysis suggested, in some instances, very close anchor spacing would be required; they discussed this problem with the committee overseeing the project. There appears to be an unwillingness to recognize the effect, that the cones of influence have on the load capacity, of helix soil anchors, since this would certainly eliminate helix soil anchors, as the sole means of windstorm protection, in wind zones II & III.

Yokel et al. (1982)
It has been determined in previous investigations [1] that there are two types of failure mechanisms for soil anchors: the so-called "shallow" anchors fail by pulling with them a body of soil (cylinder, truncated cone or other) which extends to the ground surface; the so called "deep" anchors fail without causing substantial disturbance to the ground surface, since the failure (slip) surface surrounding the anchor does not extend to the ground surface.

As discussed earlier, the problem with the so-called "deep" anchors is the inability to consistently install the anchors, to their full-intended depth and, this problem is not going away, with this type of anchoring system.

Varying Soil Conditions

Yokel et al. (1982)
Figure 4 compares the load-displacement characteristics of vertical anchors installed on the three test sites and pulled at a 40º angle. Note that the initial anchor stiffness on the sandy site was less than that on the silt and clay sites. However, the stiffness on the sandy site increased rapidly with increasing loads and the peak resistance was reached at a smaller displacement than that in the silt and clay site. It is noteworthy that while on the sandy and silty site the load capacity of the vertical anchors pulled at an angle tended to be higher than that of the axially-pulled vertical anchors, a similar increase in load capacity did not occur on the clay site. This can be explained by the fact that the compressive forces exerted in this loading mode on part of the soil mass surrounding the anchor substantially increased the shear strength of the sands and silts, which increases with increasing confining pressures, but not that of the clays, which entirely depend on cohesion and thus tends to be independent of confining pressures.

Vertically installed 6-in single helix anchors pulled at a 40º angle on the sand, silt, and clay sites (Yokel et al. 1982)

It is noteworthy, that the description of the soils in the MHRA publication (2000) at the chosen sites emphasized SAND and silt, with a low to non-existent plasticity index, which indicates granular soils. It is pointed out by Yokel et al. (1982) that these sites would be more favorable for anchors installed vertically, under inclined loads, for achieving higher load capacities at lower horizontal displacements, even without stabilizer plates, as is shown in Figure 4. The Baxley, GA site did indicate some clay content, although it was not significant. The table of results in the MHRA Report (2000), for the 48-in anchors, with 17-in. stabilizer plates, at the Baxley, GA site shows a high degree of variation between the three test subject anchors, ranging from 1450-to-3900 lbf.

Soil Types

Yokel et al. (1982)
A great number of soil types are encountered in nature, and oversimplification cannot be avoided if an attempt is made to condense these into a few typical cases. There is evidence [1] that there is a fundamental difference between anchor performance in granular and cohesive soils, since these soils have different strength, drainage, and strain hardening characteristics.

Prediction of Anchor Load Capacities on the Basis of Tests on Similar Sites

Yokel et al. (1982)> The coefficient of variation of anchor strength on the sites ranged from 0.18 to 0.40. There probably would be more variation if anchor test results from one site are used to predict anchor strength at another site.

Yokel et al. (1982)
Miscellaneous hypothesis have been developed which correlate the load capacity of anchors to the shear strength of soils, which in turn can be measured by various in-situ and laboratory tests. However, it was concluded on the basis of available data that the correlation between calculated and measured anchor-load capacities, particularly in granular soils, tends to be poor [1]. In part, our inability to make reliable predictions of anchor-load capacity on the basis of the shear strength of soils is attributable to our inability to make reliable measurements of the in-situ shear strength, particularly that of granular soils. This measurement is even more severe at shallow depths, where soils are subjected to many disturbances, such as freezing and thawing, changing moisture content and the effect of root systems and organic matter.

The testing of helix soil anchors has consistently shown, a much lower nominal load capacity for soil anchors, than the current 3150 pounds, and the MHRA study supports this truth. While it is enlightening to see this admission, without full industry recognition that the currently stated nominal working load is invalid, this latest attempt is not likely to yield any significant changes, in the current industry anchoring practices.

There is no recognition of the ultimate load capacities of soil anchors, which is especially critical for wind zone II & III. While the intent is to limit, the horizontal displacement associated with this equipment, by lowering the nominal working load of helix soil anchors, the previous testing, and analysis of data, makes it doubtful that this equipment will perform as intended when most needed, even at the lower capacities.

Every few years, the previously used test equipment, which was developed and/or used, is abandoned, along with many of the test results obtained with this equipment. New test equipment or some modified version of previously used equipment is developed, and new test criteria and/or protocol are developed, which exclude, or modify some key element of the previous criteria and/or protocol, which revealed the weaknesses of helix soil anchoring system. The result has been to create confusion, as to the effectiveness of these devices and, to create the appearance that these devices will perform when needed.

The first noted attempt, which we can find, to create the appearance that these devices would perform as intended, occurred when the ANSI committee, removed from the A225.1 standard, the horizontal displacement failure criterion. Of course, when you turn off the test, the device being tested will pass. It was noted in the 1991 by Pearson, that the criterion proposed in the ASTM standard, was ignored in all subsequent testing after it became apparent, that the testing of helix soil anchors, under the proposed criteria, had no hope of achieving the requirements of the proposed ASTM standard.

Marshall 1994
It can be seen in Figure 5 that none of the load/installation configurations meet the ANSI A225.1 limitation on horizontal displacement at the prescribed working load of 3,150 lbf. In fact, those anchors installed in silty soil with b2= 135 degrees developed only about 14 percent of the required working load at a horizontal displacement of 4-in. Compare this with values of 21 to 26 percent obtained by Pearson et al. (1991) for loose to medium dense sand. Given the ABYSMAL performance of conventional soil anchors under the action of inclined loads, it is unfortunate that the committee responsible for ANSI A225.1 chose to DELETE the displacement limitations all together rather than address the problem directly. Furthermore, a simple and workable solution to the problem was proposed more than 10 years ago (Yokel et al. 1982), a fact that makes the current situation all the more inexcusable.

Pearson et al. (1982)
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, to test anchors that were fully embedded and record load and deflection until one of the following conditions existed: a. the applied load began to decrease with increase in deflection or b. the horizontal deflection exceeds approximately 18-in. The test was stopped if one of these two conditions was exceeded.

The significance of the problem with fully embedding anchors is being ignored by HUD, and HUD is allowing for horizontal deflections, that far exceed what is necessary to assure pier stability, and prevent damage to gas line connections, and other utility connections. (See Note 1, after the references for additional commentary.

Results of field pull-out tests on soil anchors installed and loaded at various angles (Yokel et al. 1982) Note: Shaded areas of the bars denote loads at limiting horizontal displacement; non-shaded areas of the bars denote ultimate load capacity.

In the final analysis, while the MHRA effort is commendable, it will do nothing to resolve the problems associated with this type of anchoring system. Until the industry and/or HUD accept the reality that is presented by the currently available test data, this system will continue to be used, and consumers will continue to be mislead into believing, that their homes meet the federal standards and, therefore are safe, an assumption that could have deadly results, for the family occupying the home.

Before more manufactured homes are installed with inadequate anchoring systems, immediate action should be taken to discontinue the use of soil auger anchor systems and either use permanent concrete foundations or other anchoring systems that have proven, under actual pull tests, to meet federal standards. There are price competitive alternatives for helix soil anchors such as, the self-seating swivel (steel cable versus threaded rod) and, the Vector Dynamics system, a relatively new innovation for manufactured home tie down, both of which appear to provide superior performance to helix soil anchors, in all soil conditions.

[1] Kovacs, W. D. and Yokel, F. Y., Soil and Rock Anchors for Mobil Homes, A State-of-the-Art Report, NBS Building Science Series 107, National Bureau of Standards, Washington, D.C., October, 1979.

[2] Yokel, F.Y., Yancey, C.W.C., and Mullen, C.L. (1981). "A study of Reaction Forces on Mobile Home Foundations Caused by Wind and Flood Loads". NBS Building Science Series 132, National Bureau of Standards, Washington D.C., 74pp.

[3] Yokel, F.Y., Chung, R.M., Rankin, F.A., and Yancey, C.W.C (1982). "Load-Displacement Characteristics of Shallow Soil Anchors". NBS Building Science Series 142, National Bureau of Standards, Washington D.C., 147pp.

[4] Pearson, J.E., Meinheit, D.F. and Longinow, A. (1991). "Testing of Soil Anchors and Strapping". Report No. 901798, prepared for the U.S. Dept. of Housing and Urban Development, WJE Associates, Inc., Northbrook, IL, 67pp.

[5] Marshall, R.D., (1993). "Wind Load Provisions of the manufactured Home Construction and Safety Standards - A review and Recommendations for Improvement." NISTIR 5189, National Institute of Standards and Technology, Gaithersburg, MD, 90pp.

[6] Marshall, R.D., (1994). "Manufactured Homes - Probability of Failure and the Need for Better Windstorm Protection Through Improved Anchoring Systems." NISTIR 5370, National Institute of Standards and Technology, Gaithersburg, MD, 42pp.

[7] Marshall, R.D., and Yokel, F.Y. (1995). "Recommended Performance-Based Criteria for the Design of Manufactured Home Foundation Systems to Resist Wind and Seismic Loads." NISTIR 5664, National Institute of Standards and Technology, Gaithersburg, MD, 60pp.

[8] Froehling and Robertson (1999) "Field Testing of Ground Anchors", Froehling & Robertson, Greenville SC, 7pp, Note: This report became the MHI/MHRA document, titled - "Guidelines for Anchor System Design" (January 2000).

Note 1: The removal of the horizontal displacement limitations from the ANSI A225.1 standard, is the type of activity that occurs when, affordability is placed ahead of safety. This is exampled again in Texas, where the solution in 1999, in the name of affordability, when difficult soils are encountered, is to allow rock anchors to be installed instead of soil anchors, or some other alternative anchoring system. The federal regulations state that all equipment, components, etc. must be approved for their intended use, and rock anchors are approved only, for use in solid rock. The Texas standard does call for doubling the number of rock anchors, but 2 x 0 = 0. In addition, rock anchor testing was performed by an independent testing laboratory, in 1998, and was witnessed/observed by representatives of the TDHCA, Manufactured Housing Division, and representatives of the Texas Manufactured Housing Industry. This testing found that this was not an acceptable option, since rock anchors, used in these difficult soils, produced very low resistances to inclined loads, even when stabilizer plates were used.

Mayer et al. (1998)
In summary, the cross-drive anchors used in this test program did not generate the pull-out loads required in the Department's regulations in these difficult soils because of excessive lateral displacement and subsequent rod slippage (ignoring the failure of the components of the anchor).

Note: the Department being referred to is the TDHCA, Manufactured Housing Division, and the Department's regulations shall be equal to the Federal Standards, as per the Texas Manufactured Housing Act.

Mayer et al. (1998)
This test program demonstrates NO improvement in rock anchor performance in these soils with the addition of a soil plate.