Rostock head starts lifting dramatically during wide print.

[rampeh]

I've had my Rostock Max for about a year now and have finally got around to buying and mounting an Onyx heated bed for larger prints. I've had no problems printing within the build space of the old Phebe printed bed before, but it seems like there is a single spot further out from the centre on the Onyx bed which causes my printer no end of grief.

I was attempting to print Highcooley's bed levelling aid (http://www.thingiverse.com/thing:50505/) to check if my old calibration is valid for the new bed (it's not great, but workable, probably ~0.3mm out at max). When it reaches a certain point (about 130mm from the centre, near the Z tower), the nozzle starts to ascend from the heated bed rapidly. It comes back down eventually when finishing up around 180+ degrees around the circular arc, but I haven't left it long enough to see what happens after (scared of crashing into the bed). You can see it rise in the video below (sorry about lack of focus):

[youtube]http://www.youtube.com/watch?v=STMxSFOsKDU[/youtube]

The G-Code seems to have each travel still at Z=0.3 or whatever the initial layer height is set to, and it's repeatable whenever printing this same part. I've tried printing the bed leveller with a gap for a fan, and rotating this 60 degrees to avoid the trouble area, and this printed fine.

I've replicated the effect slightly with manual jogs but not to such dramatic effect. I've looked at the belts and there is not any apparent wear / stripping, as far as I can tell the sprockets on the steppers are attached well too.

Does anyone have any other ideas of what could be causing this issue? I can't think of anything that would cause such a dramatic error. When it happens all 3 towers are still moving although obviously I cannot tell which one is moving incorrectly to cause the drift upwards.

For what it's worth I'm using the old kit, with 15 tooth sprockets and 8 step microstepping. I'm using the latest Repetier from Git Hub (as of two weeks ago, first thing I tried was updating this), as far as I can tell it's correctly configured. I have installed tricklaser carbon fibre arms and am using an E3D hotend.

Any help would be much appreciated

[jmpreuss]

One possibility is that your delta radius is way off. The new repetier has horizontal radius in EEPROM. If you put your old print radius value there you will get serious print arcing.
This is the math that I think is correct.
Horizontal radius = delta radius = (printer radius - end effector horizontal offset - carriage horizontal offset)

[rampeh]

Thanks for the reply!

I didn't even notice it had a horizontal offset in the new firmware! However, it's set to 130.35, exactly the same as DELTA_RADIUS in the firmware; Must have been picked up from the config when I first flashed the new firmware.

It also strikes me as strange that it only happens in that single spot, and doesn't appear to happen in any similar points at the same radius from the centre. It leads me to believe there is trouble with a single tower. If i hadn't inspected the belts i'd say it was belt slippage, but this doesnt seem to be it.

Are the stresses on the belts any higher than usual when the carriage is nearest the bottom of its travel? The Y axis (roughly opposite the trouble spot) is only 10~ cm from the bottom of the tower when it reaches the trouble spot.

[mhackney]

That should be easy to track down. I don't think that is a software config issue! I suspect either a slipping pulley, slipping belt OR a skipped stepper steps. I also think the Y axis tower is the contributor since it is opposite the trouble spot. Check the pulley and skipped steps if you've already checked the belt.

[rampeh]

Okay! Progress!

I had a look at the steppers and they weren't obviously missing steps, but there is a quick travel from the G28 Position at the start of the print, which got me thinking about the print speed. I have my travels running at 500mm/s, which may be a bit over eager! I tried printing the same shape at 75% feedrate using the slider in Repetier, and bob's your uncle, no more ascending print head! Now, I'm not sure whether it's the acceleration or the speed which is causing the missed steps, but I'm assuming since it's such a far travel it's the first time the machine has actually attempted to hit 500mm/sec.

I've tried heavily increasing the trimpot current values in the firmware, but this doesnt seem to stop the missed steps at speed.

I have a few questions though,

1. Is the maximum speed of the print head able to be set in the firmware or EEPROM? There's a Max feedrate [mm/s], does this control travels too?
2. What is the maximum travel speed achievable by the Rostock MAX?
3. Is there a way I can set the maximum travel speed that the Repetier Host program outputs in gcode when adjusting using the feedrate slider, so even if i want to modify the print speed the travels stay the same?
4. Is there any way to tell if a stepper has missed steps during a print?

Thanks in advance!

[Eaglezsoar]

If you don't have a fan blowing on the Rambo driver area that could contribute to the missing steps. The fan helps to keep those chips cool and that can help to stop missed steps.

[rampeh]

Good idea, I've got no cooling on the Rambo at all.

Would it contribute to missed steps even if the drivers haven't had any time to heat up? I still have the problem if i turn the machine on cold and run the print sequence.

[Eaglezsoar]

Good idea, I've got no cooling on the Rambo at all.

Would it contribute to missed steps even if the drivers haven't had any time to heat up? I still have the problem if i turn the machine on cold and run the print sequence.

Cold probably not contributing but overall I think you will see an improvement with a fan down there.

[rampeh]

I am now running slower travels and the problem seems to have disappeared. I suppose the motor skips steps when it can't reach the desired rate and causes the behaviour seen.

Most of the time it doesn't happen since the travel distance is not linear enough to get to the maximum travel speed (usually it spends most of its time accelerating).

[Eaglezsoar]

I am now running slower travels and the problem seems to have disappeared. I suppose the motor skips steps when it can't reach the desired rate and causes the behaviour seen.

Most of the time it doesn't happen since the travel distance is not linear enough to get to the maximum travel speed (usually it spends most of its time accelerating).

Good job in troubleshooting that one. A lesson learned by us all. Thanks for letting us know what was causing the problem.

[mhackney]

Your original video was printing pretty slow. You should be able to go MUCH faster than that. If you stopped the problem by slowing down, my money is on a slipping pulley - the one attached to the stepper and most likely the one attached to the Y stepper given the video. If you were skipping stepper steps to get that wide of a swing you would hear the "tat, tat, tat". Did you check the pulleys?

[rampeh]

Ah I didn't see this post mhackney.

I had a look at the pulleys, and didn't see any obvious slipping but i'm not entirely sure how i'd tell if it was slipping a couple notches out of a hundred.

I have since had to replace the belts on my Y and Z axis due to a bed crash, and I've tightened them up now.

It wasn't the top speed that I've reduced, but the maximum acceleration allowed during travels. As far as I can tell, printing quickly should still work, but I am having too many extrusion problems lately to try pushing my print speed too quickly.