Reversing direction causes motors to jerk

Despontene wrote: I'm using the Avid power supply with their motors, so I assume that's not the problem. I have measured the voltage at the Geckos between 46-48 depending on load. This also happens when I only have a single motor hooked up.

OK just to finish up on this, what is the total amps the power supply can output?
What is the amps required for each drive and what is the total?
The total of the drives can be about 40% higher than the power supply.

A couple more thoughts to eliminate hardware/build issues.
Are the pinions slipping on the motor shaft under inertia?
Is it possible to lock the spring loaded pinions (eg bridge with a fixed link)?
Is the Z axis drive still using a spring loaded drive?

Total output amps: 20.8
Amps per drive: 7
Drives: 4
Total drive amps: 28

The Z axis motor is directly connected to a screw. It's not spring loaded and doesn't use a rack and pinion like the other axes.

Just in case the Gecko drive does something funny if direction changes without a step pulse
you might try setting the P term to 200 (this will prevent any dithering of direction at near 0 velocity)

So the Gecko current setting is set to 7 amps? via the switches (eg all on)?
See what PCW's suggestion yields.
My experience with gecko drives is limited to blowing one up before replacing it with some Chinese ones.
AS long as nothing is slipping on the motor shaft, then it does seem to point to an issue with the drives.

How are you changing direction? With Gcode? If not, try doing it that way so the trajectory planner can kick in and deal with the acellaration paramters in your ini file.

rodw: Yep, all switches on.

I set P to 200 and made sure to move with G Code, no dice on either one.

Check the wiring to the drives, something is fishy there.

I checked the connectivity of the motor wires. I think the pins are paired correctly. two of the connected pins are wired to A/~A and the other two are connected to B/~B.

I've checked/redone the wiring from the drives to the break out board. I think that's also correct.

I also redid the wiring from the drives to the power supply.

Is there anything else I can do to check the wiring?

I would try to scope the signals of enable, dir and step with.

linuxcnc.org/docs/2.7/html/hal/tutorial....ec:tutorial-halscope

Mike

Thanks Mike, this looks great. Exactly what I was looking for. I'll give it a shot and report back.

Update: I checked HAL scope and all of the wiring again with no luck. The folks at Gecko thinks this could be related to a firmware issue they had. They're going to send me a new set of drivers to try.