OpenBuilds Actuator Test Rig

OpenBuilds Accuracy Test Rig
Please note: This report and results shared are intended solely for information purposes. (see Disclaimer)

When deciding which type of actuator to use for your build, it would help to know how it will perform. How precise will the movements be? How much force can it produce? How fast can it go? As with all machine components, the OpenBuilds components have minor imperfections and tolerances that will affect the overall performance.
One can calculate the theoretical resolution of the actuator in steps/mm, but how does this work out in practice? How close will the actual distance be to the programmed distance? In an effort to answer these questions, a test rig was built to measure the real world performance of the OpenBuilds actuators.



The heart of the test rig is a digital machine scale that is accurate to 0.01mm. This is attached to the actuator to measure the actual distance traveled. Mounting points on the base plate and custom feet allow any of the actuator types to be attached. In order to measure the force the actuator can produce, an air cylinder is used to provide resistance. Increasing the pressure via the regulator increases the force. The electronics to run the motors consist of a CNC xPro controller along with a 24V power supply. This setup allows direct measurements of the following: accuracy, repeatability, max force, and max speed.



Test procedure

For the positioning test, a simple program was written that moves the actuator in 20mm increments from zero up to 100mm, and then back to zero. When each target is reached, the distance to be recorded. The process is repeated 6 times. These tests are done under no load, so the air cylinder was not used.

A spread sheet was used to collect the data and calculate the results. A sample is shown below.



The formula used to calculate accuracy takes the difference between each data point and the target distance, and then averages them out. The repeatability is similar, except each data point is compared to the average of all the data points instead.

The speed test was done by gradually increasing the acceleration and maximum speed values on the xPro controller until the motor stalled out. Once again, this is done under no load conditions. Measuring the maximum speed depends a lot on friction in the actuator, so these values are approximations at best. If knowing the maximum speed of your build is important to you, the test is fairly simple to do. Just make sure the distance you use is large enough to allow the motor to accelerate to the top speed. Shorter moves mean the maximum speed will never be reached.

The force test was done by gradually increasing the pressure in the air cylinder until the motor stalled. Admittedly, this is a fairly crude method for measuring force. There are a lot of factors such as the motor speed, acceleration, current limit, etc that have some effect on the force. For simplicity, these were all kept constant. The force values are approximate and will vary a fair bit from one application to the next.

Results

The table below lists the various actuator types, and the test results. Please keep in mind that these values are representative of an average build, but they are not guaranteed. Your results could be better or worse, depending on how much care is taken during assembly and calibration. The point here is that these measured results serve as a guide to help make informed decisions on which actuator type to use for your particular application. For example, if you want high speed and don't need much force, a belt drive actuator is probably the best choice.

You can check out all the actuator test results from the testing we did here Actuator Test Specs




Lead Screw Pitch Correction

During testing an interesting discovery was made with the lead screws. The theoretical resolution for a screw actuator is 200 steps per mm. But early testing showed that longer movements had larger errors, and shorter movements less error. This behavior was consistent with several different screw actuators. The cause was found to be the leadscrew pitch, which should be 2mm, but in reality is slightly more. A series of calibration tests were done, and it was found that reducing the steps/mm to 199.1 yielded the most accurate results, with consistent errors that no longer varied with the distance.

Conclusion

Hopefully these results are helpful when comparing the different types of actuators you can make with OpenBuilds. As you can see, some of them excel in slower, more precise movements which would suit a 3D printer, while others offer quicker movements which may be better suited to a CNC router. And maybe you don't know what type of performance you need for your project.

The beauty of OpenBuilds is that it is easy to build and test different actuators.
The best thing you can do is start experimenting!

DISCLAIMER
This report and results shared are intended solely for information purposes. Information is obtained from sources believed to be reliable, but is in no way guaranteed. No guarantee of any kind is implied. In no event should the content of this report be construed as an express or implied promise, guarantee or implication.