Welcome! Log In Create A New Profile

Advanced

How do i test this

Posted by BackEMF 
How do i test this
July 27, 2014 10:48AM
I noticed something suspicious.

When the Y axis makes long moves then the X axis seems to be moving on its own. Slowly but i can see the motor axis move.
I noticed this during auto bed levelling. when the x axis supposed to be steady and the Y moves to the center then the x moves veeeery slowly too.
the center grid line is probed, and then the Y moves to the back and it happens again.. the x moves slowly again (visually the motor axis turns a few degrees as long as the y axis is moving.)

Now I wonder if this only happens during auto bed level, or also during a print!

And then I’d like to find out of this is software related or an electronic issue.

Please could someone check if this is a common thing and if it is a real issue or something that is only happening on my machine. (that would make it a electronic issue)

A shame that I don't have a video camera to show what I see...
Re: How do i test this
July 27, 2014 12:58PM
Looks like it is a electrical issue:

[forum.pololu.com]

This topic sounds similair though in my case it is less extreme. Now i think that this is also the cause of the skewed print i had with the rack & pinion.
everytime the flat side was printed in Y direction the X motor turned 1 degree on its own skewing the next layer by about 0.2 mm

It all starts to make sense now, but is this bad stepper driver design? Or just plain bad luck. I'm in it for the learning, but man do i want o get on with this and print a few weeks without issues hehehe.


And the journey continued...
Re: How do i test this
July 27, 2014 08:55PM
Quote
BackEMF
When the Y axis makes long moves then the X axis seems to be moving on its own. Slowly but i can see the motor axis move.
I noticed this during auto bed levelling. when the x axis supposed to be steady and the Y moves to the center then the x moves veeeery slowly too.

During auto bed leveling or just after it?
Re: How do i test this
July 28, 2014 11:07AM
I can see the movement during auto level, and when printing a Rack from the rack & pinion i get skewed results. So it always happens on long straight moves.

It is a crosstalk issue, though i'm a bit puzzled why i'm seem the only one with this issue. I don't see a reason why my electronics would behave different. And even after some changes to the machine the issue hasn't changed.

Still when i print circles or 100% straight lines then the results are ok. but when diagonal lines come in the mix then prints get skewed.
Many blamed the fishingline drive to be slipping but I now can assure everyone that this was not the case.

I now make a 100 x 100 square in gcode and it is clear that the x axis suffers most from this when the y axis moves.
I swapped the steppers and the issue is still on the X.

So now I'm now heading to the RAMPS guys with this issue....
Re: How do i test this
July 28, 2014 04:59PM
Quote
BackEMF
I swapped the steppers and the issue is still on the X.

I don't know how complicate it may be, but you may think about changing the firmware so that the blamed slots of the RAMPS are assigned to other two steppers (for example Z and the extruder), and then check that you still have cross-talk on the same slots. This would be to exclude any interference by the software.

Of course if you have another RAMPS around it is easier if you just change it to see if the issue persists.
Re: How do i test this
July 28, 2014 07:26PM
Hm indeed, i'll give that a go. I have another RAMPS in the pipeline so within a few weeks i can try both.
Re: How do i test this
July 29, 2014 12:18AM
Quote
BackEMF
Hm indeed, i'll give that a go. I have another RAMPS in the pipeline so within a few weeks i can try both.

Maybe try using shielded cable for the motors?? Worth a try I think.
I have been lucky so far and not needed it, but I did read somewhere that a guy had this problem with a Mach3 setup and had to use shielded cables.
Maybe you could try twisting the cables around each other too?? Keep the wiring as separated as possible from each motor set of wires so there is less chance of cross-talk.


[regpye.com.au]
"Experience is the mother of all knowledge." --Leonardo da Vinci
Re: How do i test this
July 29, 2014 11:46AM
Yeah i have been playing with that but nothing seems to change, In another thread nophead said that this could ne 'normal behaviour' of the auto bed level though that doesn't explain all of it.
Still trying to get my head around it.

I have had to many things happening to this machine already and it is now beginning to become a bit complex. What is really going on? I'm probably thinking to hard and dont see the real issue.

And i also need a way to make te Z axis more stable.

Every now and then i can make a good print, but there is no consistency. I do like to learn more, but it also would be nice to have a stable machine so that i can join the evolution process. :-)

About shielding and twisting

I now have 'twisted pairs' each motor winding twisted and with a different number of turns. And that also made no difference.
So if it is crosstalk then it happens at RAMPS. I think i need to step away for a while and come back with a fresh mind, though that wouldn't help me ahead either.

I'm starting to have ideas how to do things completely different but i'm not up to the challenge. i first need a stable printer. It's becoming a bit frustrating to see others build and print seemingly without issues. But it is what it is, and i just need to keep moving and get thisone right.
Re: How do i test this
July 29, 2014 06:16PM
Quote
BackEMF
It's becoming a bit frustrating to see others build and print seemingly without issues.

Same here. I don't know about the others, but I have been struggling a lot to make this little jewel work reasonably. And I am still very far from what I want, always dealing with clogging and printing quality issues.

But at least now I see that it can give amazing results, it just needs a great deal of dedication.
Re: How do i test this
July 30, 2014 04:55AM
It sure does. hehehe all my spare time goes into this.

And to build a printer you still kind of need to have a printer to be able to print fixes and upgrades.
I'm still wondering why the Huxley hotend didn't work for me as how i see it work for others. And yes. I had a few good pints from it.

The supposed crosstalk seems not to be crosstalk after all. The X moves are only happening during the auto level and are part of the auto level process.
So yes i have learned something.

Still this doesn't explain why my smartrap messes up a Rack & Pinion print by skewing it.
When i print round objects (like LM8UU) or something square, that all comes out good.

And yes I do have a few prints that came out perfect, so I'm not that far away from having a stable printer.
Sorry, only registered users may post in this forum.

Click here to login