Anchoring Systems

As part of a larger effort to improve the performance of foundation systems, the Systems Building Research Alliance coordinated the development of guidelines for anchor system design and installation. The project was, in part, an outgrowth of work, initiated by the Manufactured Housing Institute Technical Activities Committee (TAC) Alternative Task Force with a similar goal. The results of the combined SBRA and TAC Committee-directed effort are described in the report.

The goal of the project was the development of an easy-to-use instruction card that will take the guesswork out of designing the anchor installation. The chart was the culmination of a three-step effort as follows:

• Field testing of anchors in soils typical of manufactured housing sites. The rigorous field-testing program gauged the strength of various anchor and stabilizer plate combinations. Representatives of Froehling & Robertson, an independent, nationally recognized geotechnical engineering firm, monitored the tests. F&R also took soil sample, conducted laboratory analysis of the samples and analyzed the test data..
  
  
• The raw data provided by F&R were reviewed by Edward Salsbury, PE and converted into design specifications for various combinations of design features including, home widths, pier heights, I-beam spacings, wind zones and anchor/stabilizer plate pairings.
  
  
• The engineering analysis results were then condensed into the tabular form displayed on the anchor chart. Instructions for using the chart and basic steps for proper ground installation are printed on the back of the chart

Field-testing

The field-testing employed a test caravan that spent a week on the road stopping at sites across the Southeast chosen for their relatively poor soils. Most of the tests took place in sand or sand and silt soils, which typically do not hold anchors as well as soils with higher clay content. The poor soils were chosen to represent a condition that might occur at a manufactured home site, assuring that the guidelines developed from the test results could be used anywhere in the nation and in the absence of detailed soil measurements would provide a margin of safety.

Figure 3: Ground Anchor Installation

At each site, the testing crew installed various anchor and stabilizer plate combinations and with the help of a winch and dynamometer, measured their holding power. In all, one hundred and sixty-two anchors - including almost all of the anchors in common use - were tested to determine their safe load resistance, or pull-out value, in various soils. Products tested ranged in size from 30" anchors with 12" stabilizer plates, to 60" anchors with 17" stabilizer plates. The criteria for anchor failure were three inches horizontal movement or two inches of vertical movement of the anchor head, not anchor pullout.

A second set of field tests were conducted to fill in some of the data gaps identified subsequent to the first set of tests. These included testing 30" and 36" anchors with 17" stabilizer plates. The second group of tests was conducted in Baxley, GA and Edgefield, SC.

Some of the preliminary findings were rather surprising. For example, the results suggest that a commonly used measure of the holding capacity of soil - the torque test - is not a particularly helpful tool in designing the anchor system. The torque testing results, or the measure of the force it takes to put an anchor in the ground, correlated poorly with the force required to pull the anchor out of the ground. The original plan was to produce guidelines for installing anchors in soils of both known and unknown holding capacity. The testing results, however, cast doubt on the ability of the installer to accurately measure the holding capacity of the soil with the tools typically available at the building site. Among the other preliminary conclusions from the testing were that every installation should include pretensioning of the anchor strap and stabilizer plates should be used for all types of anchors.

The testing also shed light on the actual working load of the anchors, nominally assumed to be 3,150 lbs. The tests yielded the following actual working loads for the various anchor length/stabilizer plate/wind zone combinations

Table 1 Actual Working Loads (lbs.) for Anchor and Stabilizer Plates Tested

Engineering analysis

The field test results were translated into anchor spacing recommendations by applying standard engineering procedures. The spacing estimates were developed for combinations of wind zone, home width and type (single and double section), I-beam spacing and pier height. Values were derived for common anchor sizes (30", 36", 48" and 60") with either 12" or 17" stabilizer plates. Representative values from these tables were used to construct the final chart.