Case Study: Accuracy Is Key in AHP Hydraulics’s New Ball Joint Tester

The extremely precise secrets of SIDEM NV’s new tool came into being with help from Delta Motion.

March 4, 2025

Tim Gessner

AHP tests ball joints at its headquarters and at customers’ locations, for example at Belgian supplier SIDEM NV’s facilities. (AHP Hydraulics)

Belgian solutions integrator AHP Hydraulics faced a certified stumper of a challenge from client SIDEM NV, which develops and produces automotive suspension and steering units. Ball joints play a critical role in these suspension units and thus require rigorous testing to ensure that production batches can be QA-approved and sent into supply. The client had a ball joint tester from another company, but it wasn’t exact enough to meet the application’s needs.

AHP ball joint tester system. (AHP Hydraulics)

AHP’s solution uses two hydraulically controlled axes: a horizontal (y-axis) and a vertical (z-axis), both with linear actuators and electric torque axes. The hydraulic cylinders deliver position and force feedback into a programmable user-machine interface for better precision and reliability.

SIDEM NV told AHP it wanted a force tolerance of ±100 newtons over a total range of 6000 newtons, said AHP Hydraulics automation engineer Ewout Reuman. On the rotation, it specified a total tolerance of 0.5 degrees over a total range of 90 degrees while cycling at 4 Hz.

“The rotational motion and the hydraulics had to be very accurate, and there I was on Google and YouTube trying to learn about the subject, quickly realizing how much I needed to pick exactly the right control product,” Reuman said.

Identifying the unknowns

AHP’s first step was procuring the client’s current ball joint tester and disassembling it into the mechanical parts. This allowed them to examine the electronics and consider how to improve the horizontal axis (y-axis) and vertical axis (z-axis) of hydraulic circuits and components. Ultimately, AHP rebuilt the tester from bottom to top and re-engineered the electronics and software. That included scrapping the conventional PC for motion control in the electrical cabinet and replacing it with a Beckhoff PLC functioning as the master and the Delta Motion controller as a slave for the hydraulics.

Reuman found himself tasked with handling the ball joint tester’s electrical engineering and software programming. Once he understood the accuracy needed in the solution and why prior attempts had fallen short, he reached out to Delta Motion, which agreed to send a representative to Brussels to work alongside the AHP team.

AHP procured Delta’s RMC75E dual-axis motion controller with Ethernet connectivity for communication with the system’s master PLC. Reuman paired this with Delta’s MA2 magnetostrictive I/O module and the AP2 analog I/O module for measuring force feedback from load cells.

AHP ball joint tester control cabinet featuring the Delta RMC75E motion controller. The two linear axes are controlled by a hydraulic servo valve and cylinder. The complete testing machine is controlled by a PLC, with the PLC providing set points and controlling both the rotational axes as well as the Delta Motion controller via an Ethernet IP connection. Within that PLC/Delta loop, in which the Delta serves as a network slave to the PLC master, the Delta RMC controls the hydraulic accuracy, error correction, and other factors.

Diagram of the basic PLC/Delta functionality. (AHP Hydraulics)

The nature of the directional forces within the apparatus led to some vexing challenges. Imagine the central ball joint continually rotating back and forth within its housing while being pushed on both from the side and below, all while a boom allowed the ball joint to swing back and forth, and the give within that joint housing allowed the ball to be constantly off-center. With these multiple forces interacting, there was never a perfect center state; the system was always in a state of deviation and change. Asynchronous force testing was always inaccurate. That alone was something AHP’s client could accept, but the deal breaker was the system’s propensity to show force steadily growing over time. The customer demanded that asynchronous force be stable.

“I tuned my axes to specific frequencies, force amplitudes, and oil temperatures and the specific supply pressure for the pump and accumulators,” Reuman said. “But our setup is highly dynamic ─ the operator can select any combination from 0 to 6000 Newtons and 0 to 4 Hertz. Environmental factors also play a role. For example, if the power unit is located outdoors, weather conditions ranging from -30°C to +40°C will impact oil viscosity during test runs. To address this, we installed both a heater and a cooler. However, at the extremes of this temperature range, the system may struggle to maintain the setpoint consistently.”

“Going into the project,” he adds, “I wasn’t aware how much influence that would have. I would get immense overshoots and undershoots, so I was constantly retuning the system when test frequencies, amplitudes, or oil temperature changed.”

Figuring out the fix

Graph from RMCTools showing asynchronous AHP ball joint test data for the Fz (vertical push/pull) axis. Note the rise and fall in the wave peaks. (AHP Hydraulics)

The situation called for rethinking the tuning. Delta suggested that adaptive amplitude control could fix AHP’s precision and reliability. AHP could measure every sine wave cycle and compare it to the target. Then the controller adjusts the amplitude and offset for the following sine wave cycle until the target and actual forces coincide.

Reuman identified a repeating wave pattern in the force curve. Delta’s sample project provided the foundation for implementing adaptive amplitude control, and Reuman added the wave pattern solution he developed. If all axes were synchronous, then the peaks would be flat. Based on the input frequencies, Reuman used Delta’s RMCTools software to identify the wave pattern in the wave crests and troughs and arrive at a method of offset and amplitude gains to compensate for it. For example, if the highest peaks occurred with every fourth wave, then the instance of one peak wave could inform how to compensate for the next fourth peak. (In reality, Reuman ultimately compensated for every sixteenth peak.) This dynamic approach proved far better than AHP’s previous approach of applying one set of offset and amplitude gain values for every wave. With wave peaks brought back within acceptance tolerances, AHP could successfully use PID control to control force between the peaks and thus rectify that runaway force growth.

The combination of adaptive control and PID tuning was the missing key. Reuman mapped the changes he worked out in RMCTools to his own application. No additional sensors or other equipment were needed to successfully implement adaptive control. It simply dropped into place via software.

With Delta’s RMCTools software and support, Reuman had his sine wave, complete with initial tuning, running “literally in a couple of hours.”

“One really helpful feature with Delta Motion is that it can autotune,” Reuman said. “During autotuning, it measures the system behavior and creates a system response. Based on the Actual Position, Target Position, and Control Output, the Auto Tuning algorithm calculates a system response model. That’s very important when controlling hydraulics because there are so many unknown factors in the system.”

Reuman pointed to oil temperatures and the temperature fluctuations that impact system behavior. The variation in oil temperature caused large fluctuations in system response. Implementing a Gain Schedule enabled the controller to dynamically adjust control response based on the system (or in this case, environmental) conditions. The RMC Motion Controller has the ability to dynamically adjust parameters to maintain the required performance levels.

The precise results needed

It took AHP roughly one week to configure the hydraulics and another one to two weeks to fully tune and integrate the Delta RMC with the entire ball joint testing system. Like many solution integrators, AHP isn’t privy to whatever ROI metrics the client used for its ball joint tester project. However, in a sense, the ROI boils down to a 0 or 1 proposition. The automotive manufacturer can do something it simply couldn’t before. Alternatively, ROI might be estimated using hypothetical failure rates.

“You have to imagine that when a batch is coming off the production line, and they are distributing those parts all around the world, a batch with production failures — because it wasn’t well tested — is going to cause very high expenses,” Reuman said. “That’s why we picked the best components and partners for this job. There are so many factors in hydraulics that aren’t the same for electric-driven axes.”

Tim Gessner is director for Business Development at Delta Motion.