Stairstepping- Belt problem, missing steps, something else?

[npolanosky]

Hey all,

I've had my rostock for about a month, and it started out pretty awesome. Lately though, I've had problems with it turning my prints into staircases. My belts are tight, stepper power is set to recommended values, and nothing is binding.
It seems to happen randomly, and it hasn't always done this- I have some 8 days of print time onit, and it's been doing it in the last 20 hours of printing, with an occasional successful print in there amid failures.

I'd really appreciated it if anyone has any suggestions. Thanks!

-Nick

[Tincho85]

What motors do you have? Check the stickers.
If they are not Wantai but the label says Kysan you need to change the the motor current.

Load up your firmware, configuration.h, line 701

change:

#define MOTOR_CURRENT {175,175,175,200,0} // Values 0-255 (RAMBO 135 = ~0.75A, 185 = ~1A)

to:

#define MOTOR_CURRENT {155,155,155,165,0} // Values 0-255 (RAMBO 135 = ~0.75A, 185 = ~1A KYSAN MOTOR CURRENT SETTINGS )


http://forum.seemecnc.com/viewtopic.php ... 24&start=0

[npolanosky]

They're neither- They're all black motors instead of black and silver, and are labeled "Automation Technology Inc", not kysan or wantai.

Edit: They're these ones: http://www.automationtechnologiesinc.co ... 3d-printer

[KAS]

Those came stock with the printer?

I'd check the pulley on each stepper to make sure it's not slipping. On that link they appear to not have a flat section for the set screw, hopefully it's that simple of a fix.

[npolanosky]

Yeah, they came with mine. They do have a flat and I used loctite, but its worth a check anyway.

[Polygonhell]

One other thing to check is the inside of the belts about one third of the way up from the base on the inside, just flip the belt over and verify it isn't worn out.
In general they should last a LONG time, but if the head hit the part or the bed it can cause the pulleys to eat the belt, I destroyed my first belt at about 10 days, but I have the original metal pulleys, and the plastic ones are mulch less of an issue.

[npolanosky]

The belts and pulleys all check out OK. Could this be caused by belts being too tight? They seem about the same, or just a little bit tighter than the ones on my Prusa i3 and my Overlord Pro.

[Eric]

You should be able to tell which axis is acting up by the direction the staircase is going.

[npolanosky]

I've completed a few prints with no problems. It seems to happen semi-randomly. I'll update if I have it happen (I'm sure I will).

[teoman]

Check that the end effector moves freely while the printer is not powered.

I had similar problems and it turned out to be the universal joints seizing up. (Mine were aluminium.)

[npolanosky]

I'll check it out in the morning. Just confirmed it definitely still does it, it's moving left along the X-axis.

[npolanosky]

Update: My girlfriend called SeemeCNC while I was at work and explained the issue to them. They recommended checking tightness of eccentric spacers on the cheapskates. Basically, if you can hold the bearings/wheels still with your fingers while you move the carriage up and down, it's too loose. She tightened them (A few were loose-ish, especially on the rear tower) and we have done a few small to medium prints this afternoon with no issues. I'm not getting my hopes up too much, but initially it looks good.

[geneb]

When the cheapskate is properly adjusted, it should fall just slightly slower than "free fall" when you drop it from the top as shown in the example video I did a while ago.

g.

[npolanosky]

Cheapskates have been adjusted and re-adjusted. Still did it in the middle of a 6hr print today. I think it definitely must be something binding or stepper current being incorrect; the motors run very warm to hot (Poor man's thermometer says ballpark 160-170F) and I have no idea what the proper current value is.

If anyone has any idea what I should set stepper current to for Automation Technology Inc motors it would be much appreciated. These motors did come with my kit, purchased on 5/24/15, but obviously they are not mentioned in the manual.

Thanks.

[geneb]

If anyone has any idea what I should set stepper current to for Automation Technology Inc motors it would be much appreciated. These motors did come with my kit, purchased on 5/24/15, but obviously they are not mentioned in the manual.

Then I recommend you read page 194. You might want to pay special attention to the little bit in bright yellow.

Just sayin'.



g.

[npolanosky]

Aha! Didn't see that.

Nevertheless, I adjusted my stepper current and still no joy. Stairstepping like mad.

I'm wondering if I'm looking in the wrong place here. I actually saw it do it for once today and it looked like the hotend got caught on the print and missed steps that way. Could that be the horizontal radius being off or maybe overextrusion?

I'm going to see if I can rig up a dial indicator and see if flat actually=parabola, and if so fine-tune the horizontal radius.

[npolanosky]

Dial indicator attached. With center as zero and moving 100mm from center, I'm reading 10 thousandths high in the +y axis, 10 thousandths low in the -y axis, 2 thousandths high on the -X axis, 10 thousandths low on the +X axis, and near the towers it gets wonky. I'm going to go back through the leveling process since that's definitely part of it, and then dial in the horizontal radius, and then I'll see if there is still any abnormality.

Precise measurements are definitely the way to go.

[npolanosky]

I wasn't off too much, but now it's level within 1/1000 at each tower and in the center. If I bisect the towers along the same circle it's a slightly different story. within 1/1000 between Z and Y towers, but about 5/1000 high between z/x and x/y.

What gives? Is it some weird geometrical issue or am I being too picky at this point?

I'm going to do a test print, but I'm all but convinced that this wasn't the issue. Even 5/1000 over the ENTIRE build plate should not have been the end of the world. The stock arms have slop 4 times that.

[npolanosky]

Final update...I guess I'll call it resolved. By some miracle, it seems like my bed leveling, assembling and disassembling the hotend, recalibrating everything, and everything else I tried has finally fixed my issue. I'm sad to say I'm not sure exactly what it was, but so far I've got another 24hrs of print time on it or so and its doing well.

[ZakRabbit]

Final update...I guess I'll call it resolved. By some miracle, it seems like my bed leveling, assembling and disassembling the hotend, recalibrating everything, and everything else I tried has finally fixed my issue. I'm sad to say I'm not sure exactly what it was, but so far I've got another 24hrs of print time on it or so and its doing well.

The important part is that it's up and running again. Thank you for reporting back.