Welcome! Log In Create A New Profile

Advanced

weird corexy drifting

Posted by kerfun 
Forum List Message List New Topic
kerfun
weird corexy drifting
August 05, 2017 06:07PM
Hi, I have been busy building myself a new 3d printer with extruded v-slot and a 400mm^3 build area! For my kinematics I deiced to drive it with a corexy belt system but I ran into a problem; when I tell the printer to move just one axis 10mm or so the other axis moves too. Here is a video of someone else's corexy machine with the same problem: [www.youtube.com]. I have switch the motors out, made sure the steppers drivers are set at the same voltage and the steps per mm and microstep settings are all the same, but still I get the same result. Please help!
Reply Quote
the_digital_dentist
Re: weird corexy drifting
August 05, 2017 07:16PM
Have you told the controller that the machine is a coreXY type?

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Reply Quote
kerfun
Re: weird corexy drifting
August 05, 2017 07:22PM
Yes, I have set the kinematic settings to corexy in the firmware. I also made sure the tension on both belts are the same, and I got new identical stepper drivers but I still get the same results.
Its almost like one motor is rotating more than the other. Is there a way to adjust the steps per mm on a single motor in the corexy setup on Marlin Firmware?

Edited 3 time(s). Last edit at 08/05/2017 07:47PM by kerfun.
Reply Quote
the_digital_dentist
Re: weird corexy drifting
August 05, 2017 09:24PM
That is not the problem.

Are you using modular drivers? Are the ustepping jumpers set the same?

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Reply Quote
kerfun
Re: weird corexy drifting
August 05, 2017 09:46PM
I am using a Ramps 1.4 board so yes I have modular drivers. As for my drivers they are identical A4988s from ebay: [www.ebay.com]. I also went through the procure below to make sure the voltages where exactly the same on my x and y stepper drivers.
Reply Quote
tgmorris99
Re: weird corexy drifting
August 05, 2017 10:33PM
As noted in the comments on YouTube, that user had the jumpers set incorrectly so it's worth double checking to make sure they're set properly.
Reply Quote
kerfun
Re: weird corexy drifting
August 05, 2017 10:41PM
Just checked all of the jumpers with a continuity test on my multi-meter and verified that all the jumpers are working and set correctly.

Edited 1 time(s). Last edit at 08/05/2017 10:41PM by kerfun.
Reply Quote
kerfun
Re: weird corexy drifting
August 05, 2017 10:57PM
I just came up with this quick fix; In planner.cpp in the latest version of Marlin I found the kinematic equation for corexy and added a scaler that lets me scale a specific motor.
  #if CORE_IS_XY
    block->steps[A_AXIS] = labs(da + db);
    block->steps[B_AXIS] = labs(da - db);
    block->steps[Z_AXIS] = labs(dc);
  #elif CORE_IS_XZ
Changed to this:
  #define A_SCALE 1
  #define B_SCALE 1.2
  #if CORE_IS_XY
    block->steps[A_AXIS] = labs((da + db)*A_SCALE);
    block->steps[B_AXIS] = labs((da - db)*B_SCALE);
    block->steps[Z_AXIS] = labs(dc);
  #elif CORE_IS_XZ
So after uploading the changed firmware the error was almost gone.smiling smiley With some more fine tuning I can eliminate the error all together, but maybe this little hack could have some negative consequences by throwing some other value off?
Reply Quote
the_digital_dentist
Re: weird corexy drifting
August 06, 2017 12:21AM
You shouldn't have to hack sourcecode to fix this. There's something else wrong.

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Reply Quote
kerfun
Re: weird corexy drifting
August 06, 2017 12:31AM
I agree that I shouldn't have to hack the source code to fix this but I can't think of any other irregularities that would cause this issue.

Edited 1 time(s). Last edit at 08/06/2017 12:32AM by kerfun.
Reply Quote
o_lampe
Re: weird corexy drifting
August 06, 2017 02:16AM
Just a crazy idea, but did you count the teeth of your pulleys? 20% off could mean, 16/20 tooth pulley combination.
Reply Quote
xyze
Re: weird corexy drifting
August 06, 2017 02:39AM
Interesting that you had to set your B_SCALE hack value to 1.2 to get "almost" linear travel. That rules out microstepping as the culprit since the value would have to be a multiple of 2.

It kinda sounds like you are using a 16 tooth pulley on 1 motor and a 20 tooth on the other. 20 / 16 = 1.25. That, or there's something really messed up in your belt path.
Reply Quote
lars.arvidson
Re: weird corexy drifting
August 06, 2017 03:20AM
Quote
o_lampe
Just a crazy idea, but did you count the teeth of your pulleys? 20% off could mean, 16/20 tooth pulley combination.

Might be easier to just measure the diameter of the pulleys! smiling smiley
Reply Quote
kerfun
Re: weird corexy drifting
August 06, 2017 10:41AM
That was it! My b motor had a 16 toothed pulley on it and my a motor had a 20 toothed pulley on it. Thanks for all your help guys!

Edited 1 time(s). Last edit at 08/06/2017 10:42AM by kerfun.
Reply Quote
the_digital_dentist
Re: weird corexy drifting
August 06, 2017 10:43AM
Well that's a new one!

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Reply Quote
o_lampe
Re: weird corexy drifting
August 07, 2017 02:23AM
Quote
lars.arvidson
Quote
o_lampe
Just a crazy idea, but did you count the teeth of your pulleys? 20% off could mean, 16/20 tooth pulley combination.

Might be easier to just measure the diameter of the pulleys! smiling smiley

Right, even eyeballing would have told the truth.
But I remember one Prusa i3 vendor, who used 17 teeth pulleys. It took ages to find out why his steps/mm were off.
Reply Quote
Newer Topic Older Topic
Sorry, only registered users may post in this forum.

Click here to login