Raspberry Pi3 GIPO & Single axis
- tommylight
-
- Away
- Moderator
-
Less
More
- Posts: 18563
- Thank you received: 6220
13 Mar 2020 13:03 #160010
by tommylight
Replied by tommylight on topic Raspberry Pi3 GIPO & Single axis
You should have no problems even at 100us, granted there are no excursions and real time delays.
Please Log in or Create an account to join the conversation.
- Todd Zuercher
-
- Away
- Platinum Member
-
Less
More
- Posts: 5000
- Thank you received: 1409
13 Mar 2020 17:37 #160060
by Todd Zuercher
Replied by Todd Zuercher on topic Raspberry Pi3 GIPO & Single axis
With software stepping, if you have 50us jitter your step timing will also have 50us jitter.
Please Log in or Create an account to join the conversation.
14 Mar 2020 13:22 #160169
by thadwald
Replied by thadwald on topic Raspberry Pi3 GIPO & Single axis
After having spent some time with the pi in software stepping, I feel I can now answer my own question: how much jitter is acceptable?
The answer is that it depends on the step rate, as I suspected. Any jitter that is more than 1 or 2 percent of the step period will be audible. This is not noticeable at low speeds, say up to around 50 full steps per second, probably because the actual motor position is not tightly controlled between full steps due to the nature of stepper motors. At higher speed where the chopper keeps hitting full steps with different speed demands due to the jitter, the motor makes a metallic grating sound, as if it had a bad bearing. This quickly becomes strong enough to throw the motor out of sync and stall. A 5% jitter visibly shakes the motor.
Do those Trinamic drivers help? Actually, they do, somewhat. They make those slow step rates quite usable. When configured for full step mode, a single step is interpolated to a quadrature sine wave, causing the motor to gently advance to the next full step. Starting at about 12hz, it begins using the previous step spacing to smooth the steps into a continuous motion.
The TMC2209 that I have also do something at higher speeds that is should take care of the jitter problem, but I haven’t seen it work successfully. At around 750 full steps per second, the controller should switch over to an internal step timer, thereby causing the motor to run jitter free at higher speeds. Perhaps the amount of jitter in my signal is too high for it to work? More testing is needed here.
The answer is that it depends on the step rate, as I suspected. Any jitter that is more than 1 or 2 percent of the step period will be audible. This is not noticeable at low speeds, say up to around 50 full steps per second, probably because the actual motor position is not tightly controlled between full steps due to the nature of stepper motors. At higher speed where the chopper keeps hitting full steps with different speed demands due to the jitter, the motor makes a metallic grating sound, as if it had a bad bearing. This quickly becomes strong enough to throw the motor out of sync and stall. A 5% jitter visibly shakes the motor.
Do those Trinamic drivers help? Actually, they do, somewhat. They make those slow step rates quite usable. When configured for full step mode, a single step is interpolated to a quadrature sine wave, causing the motor to gently advance to the next full step. Starting at about 12hz, it begins using the previous step spacing to smooth the steps into a continuous motion.
The TMC2209 that I have also do something at higher speeds that is should take care of the jitter problem, but I haven’t seen it work successfully. At around 750 full steps per second, the controller should switch over to an internal step timer, thereby causing the motor to run jitter free at higher speeds. Perhaps the amount of jitter in my signal is too high for it to work? More testing is needed here.
Please Log in or Create an account to join the conversation.
- twoflowers
-
- Offline
- User is blocked
-
Less
More
- Posts: 120
- Thank you received: 32
26 Mar 2020 14:07 #161566
by twoflowers
Replied by twoflowers on topic Raspberry Pi3 GIPO & Single axis
I had that metallic grating sound on my RPi3, too, and could not figure out where it came from :/
What I see on my setup: without OpenGL, the jutter is ~ 1/5th, but that eleminates axis as a GUI.
What I see on my setup: without OpenGL, the jutter is ~ 1/5th, but that eleminates axis as a GUI.
Please Log in or Create an account to join the conversation.
Time to create page: 0.073 seconds