Homing not smooth
- andypugh
-
- Offline
- Moderator
-
Less
More
- Posts: 23178
- Thank you received: 4862
08 Nov 2019 12:40 #149911
by andypugh
That would tend to suggest that you are moving close to the theoretical max step rate.
Looking at the numbers:
MAX_VELOCITY = 27.3430828642
MAX_ACCELERATION = 500.0
STEPGEN_MAXACCEL = 625.0
SCALE = 944.867009968
Max velocity of 27... x Scale of 944... = 25kHz
So the absolute maximum base thread period is 40nS.
And this is possibly the root of the issue you are seeing. At maximum speed the machine has a choice of pulsing every 1 period (25kHz) or every other period (12kHz) so the speed that the motor is being driven at varies by a factor of 2 several times a second.
You might want to experiment with a lower microstep setting to give a bit more overhead.
(I would suggest moving to an external step generator such as Pico or Mesa, but would feel guilty if that didn't affect the problem)
Replied by andypugh on topic Homing not smooth
Thanks - I did make the change and received an immediate Joint error on the zero axis (which happens to be X at present). The error persisted each time I tried to initiate Home All so I reverted to the previous figure.
That would tend to suggest that you are moving close to the theoretical max step rate.
Looking at the numbers:
MAX_VELOCITY = 27.3430828642
MAX_ACCELERATION = 500.0
STEPGEN_MAXACCEL = 625.0
SCALE = 944.867009968
Max velocity of 27... x Scale of 944... = 25kHz
So the absolute maximum base thread period is 40nS.
And this is possibly the root of the issue you are seeing. At maximum speed the machine has a choice of pulsing every 1 period (25kHz) or every other period (12kHz) so the speed that the motor is being driven at varies by a factor of 2 several times a second.
You might want to experiment with a lower microstep setting to give a bit more overhead.
(I would suggest moving to an external step generator such as Pico or Mesa, but would feel guilty if that didn't affect the problem)
Please Log in or Create an account to join the conversation.
- ytiuqibu
- Offline
- Senior Member
-
Less
More
- Posts: 75
- Thank you received: 3
22 Nov 2019 13:58 #150922
by ytiuqibu
Replied by ytiuqibu on topic Homing not smooth
Thanks Andy,
I will return to investigating this issue during next week. If I remember correctly the suggested microstepping figure for our Gecko 203v drivers is different from the actual figure set up in the StepConfig process. During a revision of our build we replaced three Chinese drivers that did not seem to work adequately with the Geckos but probably did not alter the StepConfig settings to those recommended for the Geckos. I will check soon.
I will return to investigating this issue during next week. If I remember correctly the suggested microstepping figure for our Gecko 203v drivers is different from the actual figure set up in the StepConfig process. During a revision of our build we replaced three Chinese drivers that did not seem to work adequately with the Geckos but probably did not alter the StepConfig settings to those recommended for the Geckos. I will check soon.
Please Log in or Create an account to join the conversation.
- ytiuqibu
- Offline
- Senior Member
-
Less
More
- Posts: 75
- Thank you received: 3
30 Nov 2019 10:17 #151560
by ytiuqibu
Replied by ytiuqibu on topic Homing not smooth
Thanks,
Having made a couple of changes I obviously need to gain a better understanding of the parameters involved.
I changed the Driver type in StepConfig from Other to Gecko 203v and the driver timing settings drop considerably but seems to make no difference to the behaviour.
I halved the driver microstepping which caused the distances of movement on each axis to double (which seems incorrect).
I will have another look on Monday when we finish a few jobs on the machine that we need to do first - we just have to do zeroing manually which is undesirable but manageable.
I simply don't understand how the movement towards homing switches is so intermittent when returning to designated zero point is smooth and at a reasonable rate.
Having made a couple of changes I obviously need to gain a better understanding of the parameters involved.
I changed the Driver type in StepConfig from Other to Gecko 203v and the driver timing settings drop considerably but seems to make no difference to the behaviour.
I halved the driver microstepping which caused the distances of movement on each axis to double (which seems incorrect).
I will have another look on Monday when we finish a few jobs on the machine that we need to do first - we just have to do zeroing manually which is undesirable but manageable.
I simply don't understand how the movement towards homing switches is so intermittent when returning to designated zero point is smooth and at a reasonable rate.
Please Log in or Create an account to join the conversation.
- andypugh
-
- Offline
- Moderator
-
Less
More
- Posts: 23178
- Thank you received: 4862
30 Nov 2019 13:32 #151564
by andypugh
Are there any encoders involved?
Replied by andypugh on topic Homing not smooth
No, that is as-expected. Halving from 16 microsteps per step to 8 microtosteps means that each step pulse makes the motor move twice as far.I halved the driver microstepping which caused the distances of movement on each axis to double (which seems incorrect).
No, it is very odd indeed. And seems to be unique to your machine / config.I simply don't understand how the movement towards homing switches is so intermittent when returning to designated zero point is smooth and at a reasonable rate.
Are there any encoders involved?
Please Log in or Create an account to join the conversation.
- ytiuqibu
- Offline
- Senior Member
-
Less
More
- Posts: 75
- Thank you received: 3
26 Jan 2020 11:22 #155787
by ytiuqibu
Replied by ytiuqibu on topic Homing not smooth
The jerky and slow motion has been solved by adding 0.1 microfarad capacitors to the input pin circuits near their connection to the breakout board.
This was done on advice from the designer of the breakout board and, although we are still experiencing some joint issues, is a huge step forward as it is allowing us to at least do homing.
This was done on advice from the designer of the breakout board and, although we are still experiencing some joint issues, is a huge step forward as it is allowing us to at least do homing.
Please Log in or Create an account to join the conversation.
Time to create page: 0.074 seconds