Double X-Axis gantry problems

More
27 Jul 2012 11:04 #22514 by Miselph
Hello all,

My Father and I have recently built a 3 axis router. We have two motors on the X, one on the Y and one on the Z. All four motors are steppers and I am using a MESA 5i25 + 7i76 combo.
We have managed to get all the motors to run independently using an XYZA configuration with the second X on the A axis.
My question is this: How do I get the X-axis to run together and home to two switches to square the gantry?
I have searched like crazy and everyone says "Look at gantrykins" so I did but to be honest I am still quite new to this and it doesn't quite make sense. The best I have managed is to get the two X motors to run together very slowly but if I increase the jog speed then I get joint follow errors.

Please help, I am tearing my hair out here!
I have attached my (much hacked at) configuration file


Thanks

Chris

File Attachment:

File Name: Realistic_V1.zip
File Size:9 KB
Attachments:

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 12:22 #22517 by andypugh
Miselph wrote:

I have attached my (much hacked at) configuration file


The stepgens on the paired motors _must_ have the same maxvel and maxacc as each other. (these numbers are in your HAL file, you might want to create a tag in the INI for them, and make both stepgens share the same INI tag.

One non-obvious thing about gantrykins is that after homing you need to press $ to switch to XYZ mde.

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 13:51 #22519 by Miselph
Thanks for the quick reply.
I have now ensured that the acceleration and max velocity settings are consistant but I am still getting joint follow errors.

I can move each motor independently in Joint mode at any jog velocity but in world (XYZ) mode i get errors if I jog faster than about 13mm/min or try and jog repeatedly.
At the moment I have got AXIS set so that it doesn't require a homing before allowing movement as we have only just (as I type) finished fitting the limit switches.

Any ideas?

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 14:01 #22520 by andypugh
Miselph wrote:

Any ideas?


In the case of a step machine, following-error normally just means that software has not been able to make pulses fast enough.

One thing I see is that your MIN_FERROR is only 1.6 steps. I would be tempted to make it (and FERROR) somewhat larger.

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 14:31 #22521 by Miselph
OK, I'll try that and get back to you. Is there a way of calculating what FERROR and MIN_FERROR should be? Also, how are those values used by the program?

Sorry if these questions seem daft.

Thank you for your patience.

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 14:52 #22526 by andypugh
Miselph wrote:

OK, I'll try that and get back to you. Is there a way of calculating what FERROR and MIN_FERROR should be? Also, how are those values used by the program?.


The f-error is the allowed max difference between the position-cmd pin and the position-fb pin for that axis at max speed.
the min-f-error is the allowed difference at low speed (or zero speed).

In the case of a stepper system with no actual feedback it is generally used to keep track of the step generator and its rate. (for example, with software step generation you can tell if the computer simply isn't fast enough to make pulses at the required rate).

In your case it is quite possible that thread-jitter might make one servo period see a step too many, and the next see a step too few, giving a difference of two steps, or 0.0125mm.

To be honest, I would probably set them to 1 and 0.1 to spot gross errors but avoid nuiscance tripping.

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 15:52 #22532 by PCW
Not sure if its related to your problem, but if your gantry is scaled in mm, the accel and velocity limits seem much too low
(stepgen maxaccel 2.5 mm/s^2 seems snail like)

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 18:57 - 27 Jul 2012 19:00 #22537 by Miselph
Thanks, that got rid of some of the problems and it is now homing correctly.

However, for some reason when jogging in world mode the acceleration information is completely ignored and it tries to jump to full speed instantaneously. I am still getting joint follow errors when trying to move at rates over about 900mm/min. I am at a loss as to why it is doing this.

I have attached the new configuration files.
[edit] I did have an acceleration rate of 50mm/s^2 but I turned it down to check the problem and may not have turned it back up in the attached file [/edit]


File Attachment:

File Name: Realistic.zip
File Size:7 KB
Attachments:
Last edit: 27 Jul 2012 19:00 by Miselph. Reason: Calrification

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 19:04 - 27 Jul 2012 19:16 #22538 by PCW
When you say you are getting following errors, is this in only in world mode or all the time?
I also see that you have ferror and min_ferror values swapped which may cause some trouble (min ferror is expected to be the smaller of the two)

Also, have you run a latency test on this machine? large latency can cause apparent ferrors
Last edit: 27 Jul 2012 19:16 by PCW.

Please Log in or Create an account to join the conversation.

More
27 Jul 2012 19:21 #22539 by Miselph
It is only ignoring the acceleration and giving me follow errors in world mode. In joint mode it works fine and I can jog up to 1200mm/min which is my max.

Please Log in or Create an account to join the conversation.

Time to create page: 0.137 seconds
Powered by Kunena Forum