Facebook Twitter LinkedIn YouTube
toll-free: (800) 947-5598

Resources Technical Library

Fatigue Testing

The use of load cells and a data logging system are a necessity in the majority of situations where materials, parts, or assemblies are fatigue tested to destruction. This is true because an accurate record of the forces at every moment of the tests is the only way that an engineer can analyze the stresses which occurred in the moments just prior to the ultimate failure. No one can accurately predict exactly when the failure will occur, nor which part of an assembly will be the weakest link which eventually fails.

When designing a test protocol, serious thought should be given to the possibility that some of the parameters of the test will need to be changed as the result of information learned in the early test cycles. It may be that the test frequencies, force levels, location or angle of force application, or phasing of test waves will need to be changed. It is therefore prudent to start out with equipment which can accommodate an increase, a decrease, or other change without a major redesign of fixtures or a major expenditure to convert or replace high cost test equipment.

Fatigue Capacity

“Fatigue Rated” is an exact Interface specification which defines a special class of load cell design and construction.

  1. Design stress levels in the flexures are about one-half as high as in a standard Low Profile load cell.
  2. Internal high-stress points, such as sharp comers and edges, are specially polished to avoid crack propagation.
  3. Extraneous load sensitivity is specified and adjusted to a lower level than in a standard Low Profile load cell.
  4. All fatigue rated Interface Low Profile cells have a specified service life of 100 million fully reversed, full capacity loading cycles.

Not all manufacturers adhere strictly to the stringent discipline necessary to produce true fatigue rated load cells on a consistent basis. By contrast, the history of Interface Low Profile cells shows a zero return rate due to fatigue failure, for fatigue-rated cells used within ratings.

Use of Non-Fatigued-Rated Cells in Fatigue Applications

Although Interface does not recommend it, there are times when circumstances force a user to apply a large number of test cycles on a non-fatigue-rated cell. The following guidelines may assist the user in deciding how long to carry on such a test before installing the properly sized fatigue-rated cell.

  1. All Interface non-fatigue-rated load cells can be considered to have a useful life of at least one million cycles of single-mode loading (loading in only one direction), at full nameplate rating. Therefore, if used in single mode at 50% of rating, it is likely that a cell would survive at least ten million cycles.
  2. If used at 500/0 of the cell’s rating, the material used in steel and stainless steel Interface cells would be stressed below the endurance limit, the level at which the steel itself would resist failure indefinitely.


If this were the only failure mode, any Interface steel or stainless steel cell could be used as a fatigue cell, if operated below 500/0 of its rated capacity. However, other minor failure modes would take over, because the load cell would be missing the “hand crafted” steps of the manufacturing process.

  1. Aluminum alloy load cell material does not exhibit an endurance limit, the horizontal flattening of its S-N Curve, the operating curve of stress versus number of cycles to failure. Therefore, while use at 50% of rating will substantially increase the number of cycles it will withstand, no exact number of cycles can be predicted.
  2. One failure mode which has not been tested on non-fatigue models is the number of cycles to failure of the connecting cable versus the excursion the cable is subjected to on each cycle. Users should take steps to support the cable in a way which reduces the stresses for larger excursions.
  3. Since the application of dynamic forces during a test involves bi-directional motion, there is a larger risk of contact resonance due to clearances in the driving mechanism. (See the section on Contact Resonance, page 14.) Any contact resonance generates non-sinusoidal forces having a high frequency content which has a greater effect on the cycle count on the load cell and also generates peak loads which are greater than the measured average.

Fatigue Capacity With an Added Fixed Load

Many test protocols require a fixed load plus a dynamic load to be applied to a test sample simultaneously. An Interface fatigue-rated load cell is well suited to this type of application. However, there are limitations which should be applied to the loadings, to insure that the cell will be operating in its linear range and to avoid overloading the load cell or reducing its fatigue life.

The Goodman Curve shown in Figure 50 is a useful nomograph (visual calculating graph) for easily figuring combined loading limits for fatigue-rated cells. Notice that the dynamic loading limit by itself is 100% of the fatigue rating, and the static loading limit by itself is 2000/0 of the fatigue rating. In between these two end points, the limit is the diagonal straight line which connects the end points.


The Goodman Curve applies only to fatigue-rated cells. Using it to calculate combined loadings on a standard load cell could result in damage to the cell.

In the example shown on the graph the dashed line indicates a situation where we want to apply a fixed load of 70%, and we want to know how much dynamic load we can apply simultaneously. By taking a straight horizontal line at the “70% Static Load” level across to the limit line and then projecting downward from that intersection, we find that the intersection on the dynamic scale at “65% of Rated Fatigue Capacity.” This means that, on a 1000 lbf rated fatigue cell, we could apply a fixed load of 700 lbf combined with a dynamic load of 650 lbf peak in both modes.

Checking the graph, note that if we needed to apply a fixed load of 160%, we could still apply a dynamic load of 20% for a total load which varies between 140% (lower peak) and 180% (upper peak). This would mean that the cell would be operating outside the limit of the normal factory calibration on fatigue cells of 100%. If the utmost accuracy is desired, it might be advisable to have a static calibration done on the cell up to 200%, which can be done on a fatigue-rated cell by special order.