SOLVED: Tuning stepper motor with Gecko 201 driver

mtngun wrote:

Andypugh, I'll try setting the steplength and space to 20,000 as you suggest but ...... I'm not sure that it is meaningful. In one place the documentation says those settings are in nanoseconds and in another place it says it is in base periods. So a steplen = 1 would be one base period (the base period is 67,000 on my 8.04 box). If that is true then setting the steplen timing to 20,000 in StepConf would have no effect. This seems to be the way it works because no matter what you plug in for step timing in StepConf, the HAL steplen is always 1

A step length of 1 and a step space of 0 makes sense in the context of a "doublestep" config. It means "turn the pin on this cycle, and "reset" will turn it off again in reset-time nS. This only makes sense if the step length is a lot less than the base period, and if the step-length + step space is a bit shorter than the base period.
I am wondering if your base period simply isn't long enough to fit in the step-space that is needed (5uS is a pretty long step space, though a lot less than your base period.

If there is documentation there which says that the timings are in base periods, then it needs to be found and altered. The definitive description of stepgen is here:
www.linuxcnc.org/docs/html/man/man9/stepgen.9.html

Incidentally, the choice of software _can't_ have any effect on the sensitivity of the hardware to VFD interference. Once the registers are set to put the pins high or low then everything downstream is identical between LinuxCNC and TurboCNC. (Assuming the hardware is exactly the same, which I am no longer sure about)

mtngun wrote:

The StepConf wizard asks for timing in nanoseconds, yet no matter what I input for "Step Time," (typically 500 nanoseconds for Geckos) the generated HAL file "steplen" is only "1". So if the manual is right about nanoseconds, then the Wizard is doing it wrong. They both can't be right.

Stepconf typically creates a "doublestep" config where the step-length is actually set by the parport reset-time (which, you will see, in in nS)
The "1" of step and "0" for space are simply to ensure that stepgen doesn't try to re-zero the pin on the next iteration.

If you were to set you step + space to a time longer than the base thread period I would expect to see it switch methods.