The following is a blog post written by our valued partner, Wolf Star Technologies. 

Push Me, Pull You: Trailer Hitch Load

Dr. Dolittle summed up the issue very succinctly. Vehicles pulling trailers are undergoing very dynamic, difficult to classify loading. This blog article will demonstrate an easy, robust way to determine loads on anything – including a trailer hitch.

The approach for understanding the loading on the trailer hitch will be utilizing Wolf Star Technologies’ True-Load software. The process starts out with an FEA model of the structure, in this case, a trailer hitch. The hitch was modeled in Abaqus/CAE software. Once the hitch was modeled, it was meshed in Abaqus/CAE.  A unit load is simply a force (or moment) imposed on the structure, with a unit magnitude. The magnitude value is simple (say 100 lbs) and consistent for all unit loads.

For this Assembly FEM, three unit loads were applied to the center of the hitch ball (100 LBF FX, FY, FZ). The hitch was restrained at the pin center (UX=UY=UZ=0) and the top edge of the hitch tube was restrained in the UY=0 (vertical direction), which represents the hitch interface to the vehicle

The Abaqus solver created three sets of results for these three unit loads. The results from the FEA solution created a full field strain response on the structure. These strain responses on the structure will be utilized by the True-Load software to place strain gauges and develop a relationship between strain response and the unit load cases.

The Abaqus ODB file with all of the results were imported directly into The True-Load software. User input to True-Load was to select the candidate strain gauge locations. Typical candidate locations include nominal locations of the assembly but do not include part-to-part interfaces, stress concentrations such as holes, or inaccessible areas to physically lay strain gauges

In this case, none of the elements on the hitch tube could be used for gauge placement, therefore those elements were not included as candidates. The entire hitch tube is recessed into the mating feature on the vehicle.  The remaining area available for gauges was the hitch bracket and the side plates. However, on the hitch bracket, the large feature of the trailer ball takes up a large amount of area on the top and bottom of the hitch bracket. The side plates are welded to the hitch bracket, and thus it is desired to stay away from the stress concentrations of the weld. The True-Load technique requires that the candidate strain gauges are in nominal areas on the structure to assure a robust relationship between applied load and strain response. Shown here are the candidate elements used for True-Load (silver highlight).

Once the candidate elements were selected, True-Load automatically located the optimal strain gauges for performing load reconstruction. Additional user input is the number of strain gauges. Mathematically, the procedure would work with three gauges for three unit loads. However, the general rule is to use twice as many gauges as unit load cases.  In this example, we will be using six gauges for the three unit load cases.

The figure here shows where the strain gauges were placed. Two gauges were placed on the two side panels and two gauges were placed at the rear of the hitch bracket. All gauges used were uniaxial strain gauges and coincidently, all gauges were oriented parallel to the vertical (Y) axis. Once the gauges were modeled with True-Load, a set of engineering prints for each of the gauges were created, and then the physical gauges were installed.

The Data Acquisition (DAQ) system used for this was a DTS Slice micro. The compact DAQ was set up for 6 ¼ bridge strain gauge channels with a sampling rate of 1 kHz. The sampling duration was operator controlled so that multiple events could be captured. These included approaching/leaving a stopped position, backing up maneuvers, city driving, and highway driving including hills, bridges, and on/off ramps.

Once the data for all events was collected, the True-Load/Post-Test software processed the strain data using the correlation matrix extracted from the FEA modeling during the True-Load/Pre-Test gauge placement exercise. This resulted in a calculation of the operating loads on the hitch. Once these loads were known, the True-Load/Post-Test software automatically calculated the simulated strain response at all of the gauges and created cross plots of measured strain (horizontal axis) and simulated strain (vertical axis). The images here show the strain correlation plots for nine events.  These plots are showing all gauges for all time. There was some error present in the plots, due to gauge placement error which can be common and deserves consideration.

The loads were then stored in data files and together with the True-QSE software from Wolf Star Technologies, the loads were fully plotted and interrogated as well as the rest of the structure. As shown in the plot below, the force on the hitch (orange) during an extended period was largely acting upward. Whereas the fore/aft load (green) on the hitch was largely pulling rearward except at the instances when breaking/stopping occurred.

Now that a time history of loading was calculated, any response anywhere on the hitch can be calculated. These loads can even be used to produce operating deflection shapes (ODS).

This project shows the power of using True-Load to understand all of your product loadings. From just a few strain gauges and FEA model, True-Load can turn any part into a load transducer. True-Load produces strain correlation plots to give you confidence in the loads that were calculated. Once you have strain correlated loading, you have the power to design without additional trips to the proving grounds. Bringing this back to Dr. Dolittle, with Wolf Star Technologies’ True-Load you can get your parts to talk to you!