First time with a linear rotary motor: from wiring to running one motion
Community Discussion · General

First time with a linear rotary motor: from wiring to running one motion

Old DengOld DengOct 62026/10/06 263 views

Last week a friend who does automation integration recommended Guoao Technology's linear rotary motor, saying they swapped in a batch on their production line and saved a workstation on the placement motion. I happened to have an idle test bench in my lab, so I borrowed one and tried it out, and jotted down the process while I was at it.

This thing combines pressing down and rotating an angle onto a single axis. Ordinary motors either only rotate or only push; this one does both. Guoao's unit is a single mover with two degrees of freedom—the mover is that moving rod, and two degrees of freedom means it can push and rotate at the same time. The manufacturer's spec sheet says force control accuracy of ±0.01N; force control means it can control how hard it presses down, not just press until it bottoms out.

There's a line in a publicly available sample from a peer that's very direct.

Combined lifting and rotating motion in one axis for pick-rotate-place processes

Pick it up, rotate it, put it down—one axis does it all. That's exactly the motion I wanted to run.

Before wiring anything, do your prep. Write three lines on a piece of paper: where to lift from, how many degrees to rotate, how much force to press with. If you can't write it out, don't wire it up—this step saves the most time. Then calculate the stroke; how far the mover can travel is fixed, and the workpiece height plus safety margin must fit within it. Common magnitudes in peer samples are a 200mm stroke and a 632mm overall length, depending on the model. Finally, prepare your fixtures: a base that can clamp the workpiece, and a simple force gauge to measure clamping force.

When you get hands-on, follow this order.

1. Wiring. Driver, encoder, limit switches—follow the terminal diagram. Don't run the encoder cable bundled with the power cable.

2. Before powering on, manually turn the mover by hand to check for any binding. Many beginners skip this step and only discover after powering on that the mechanics are misaligned—a collision is costly.

3. Open the bundled tuning software and connect to the driver. The interface is usually split into an axis list and a parameter page. First set the soft limits, leaving margin at both ends of the stroke.

4. Set the speed. For the first run, keep it below 20% of the rated value.

5. Jog. There's usually a Jog button in the software; hold it and it moves. First do linear motion only to confirm the direction is correct, then do rotation to confirm the sign.

6. Force control calibration. Let the mover slowly approach a rigid surface with no load, read the force value, and record the zero point. Guoao's unit is rated at ±0.01N, but that's the accuracy after calibration—before calibration the reading is meaningless.

7. Run the full sequence. Lift, descend into position, force-control clamp, rotate to target angle, lift. Save it as a program and run it in a loop twenty times.

There are three places to trip up.

Units is the first. In the parameter screen, ISO units have two sets: MM and INCH. I initially entered 50. thinking it was millimeters, but it was actually inches, and the mover slammed straight into the limit. My rule now is: before changing any parameter, check what the unit field is displaying.

The second is coupling. Testing linear and rotation separately is fine, but put them together and the force control reading starts drifting—inertia coupling. The fix is to calibrate separately first, then run the combined motion at half speed.

The third is the force control zero point. Change the workpiece material and the contact stiffness changes, so the zero point has to be recalibrated. I calibrate once before starting work each day; I don't skimp on this time.

After running it, that bit of accuracy on the spec sheet—any capable manufacturer can achieve it. What determines whether it actually runs smoothly is whether the motion was decomposed correctly, and whether units, zero point, and coupling are kept under control. From wiring to running the full sequence took me about an afternoon, half of which was stuck on units.

The next thing I want to try is hooking this up to vision, letting the camera give the workpiece pose first and then the motor goes to pick it up. The difficulty jumps up a notch—how to allocate visual error versus motor error is a separate topic, and I haven't figured it out yet. I'll wait until I've hooked it up to say more.

2 replies

?
Ctrl + Enter to reply
Shutter
ShutterOct 6

That force-control zero point needing calibration once a day is so real. When I shoot e-commerce product photos and switch materials, I have to readjust white balance too—same principle.

HuangCFO
Reply to Shutter

Calibrating the zero point is like white balance—once the material changes, the contact stiffness changes. The OP calibrating once a day isn't even that frequent.