Persistent Pause Print Failure/Belt Grinding + Layer Shift

Having a problem? Post it here and someone will be along shortly to help
Post Reply
exalt
Plasticator
Posts: 16
Joined: Sat Jan 02, 2016 2:53 pm

Persistent Pause Print Failure/Belt Grinding + Layer Shift

Post by exalt »

Hello everyone. I would like to pick everyone's brains before trying to hit up customer support, as I have already filed a few tickets for previous problems and it seems a little demanding to file another.

Using a recently assembled rostock max v2, assembled to the guidelines in the assembly manual v2, and then following the v3 manual(mine has the older melamine carriages and non-ball arms). Repetier firmware, the only variables I changed were jerk to 30 and acceleration to 1600(unlikely to be factors).

There are two separate symptoms which I have which might be related.

The first is that sometimes, when I use the "pause" function on the LCD menu, the printer tries to raise the carriages/arms but before reaching the end stops the machine starts to grind quite loudly. I suspect one of the belt or belts might be grinding against the drive gear(s) making a loud grating noise.

I tried manually pressing the end stop but the grinding didn't stop immediately.

It stopped grinding on its own after a few seconds.

At this point, any attempts to "resume" the print result in the platform staying at the same height, moving around and extruding filament some 25cm or so above the print(the Z height is completely thrown off).

This only has happened during the pause function on the LCD, not on any other Z movement(maybe the pause function movement somehow clashes with the printers current movement, or it uses a much higher speed/acceleration than other z moves?), and has happened a handful of times.

I tightened the belts' tension but it still seems to happen, albeit not all the time. Now I try not to use the pause function since it might cause print failure.

The second problem is with layer shifting. I have encountered a persistent problem where in the middle of larger prints(10+ hours) the layer would shift noticeably to the left by about 0.5mm and also slightly toward the front of the printer(facing the printer from the LCD side).

In such taller prints, sometimes the first few levels would be very straight and well aligned, and then there might be slight perturbations perpendicular to the direction of the Z-axis, going upward, increasing in frequency the higher the print goes. Then eventually there might be a part where the print head shifts to the left.

It doesn't happened with shorter prints(<2cm), and it seems the layer perturbations set in more the higher a print goes. After a while, a major leftward and frontward shift happens.

Sometimes, the print head would seemingly overcompensate, and a couple of layers later, it would bounce back to the right and to the back. This would often be severe and sharp, exceeding the original shift. This bounce back resulted in one shift of over 1.1mm to the right.

However, I have also printed a tall print which had no layer shift issues.

After the first two layer shifted prints, I tried tightening the belt tensioners and this did no seem to resolve the major layer shifting (0.4mm). While it does seem to get rid of the smaller perturbations, the larger layer shifts still happened. The reason I believe this problem and the pause problem might be related is because I had one of these pause-grind events just before attempting a the 3rd print where the layer shift happened again.

I found a couple of other threads describing layer shifts here:

http://forum.seemecnc.com/viewtopic.php ... 63&p=82222" onclick="window.open(this.href);return false;
http://forum.seemecnc.com/viewtopic.php ... 17&p=82597" onclick="window.open(this.href);return false;

If I do have to take out the stepper motors, grub nuts, belts, arms, etc for inspection, is there a good way to go about doing this without removing the towers/base, cutting wiring, etc? Right now, to my mind, there are a lot of things that could have gone wrong, but I don't know a good process to rule out possible causes. I don't want to rush headlong into disassembling and maybe cause more problems.

I have attached a couple of photos of the various layer shifting/perturbations.
Layer Shift 1a.jpg
Close up of layer shifted print number 3:
Layer Shift 2a.jpg
Layer Shift 3a.jpg
Layer Shift 4a.jpg
Layer Shift 5a.jpg
Layer Shift 6a.jpg
Thank you for your attention and help.
Last edited by exalt on Sat Feb 06, 2016 10:31 am, edited 1 time in total.
User avatar
thedoble
Printmaster!
Posts: 131
Joined: Fri Dec 25, 2015 8:32 pm
Location: Brisbane, Australia
Contact:

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by thedoble »

Not sure on the pause question - I have never had any luck 'pausing' a print - I always end up with problems, so just avoid it these days.

When it comes to layer shifting, I found all my layer shift issues came back to simply printing too fast.
What do you have your travel speed set to? What about other print speeds?

If you print too fast, it's possible for the motors to miss a step, after which it keeps printing at the new 'offset' point - resulting in a layer shift.
Smoke me a kipper, I'll be back for breakfast.
exalt
Plasticator
Posts: 16
Joined: Sat Jan 02, 2016 2:53 pm

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by exalt »

Thank you for the reply thedoble. So far I've kept the print speeds in the range of 30-45mm/s for Perimeters and Infill. Travel speed for non-printing moves is 300mm/s. I don't usually crank up the speed but one of the prints I did at around 50mm/s didn't have any layer shift, and that one had a jump distance of over 8cm. It's an interesting point. I think that if the print would develop a shift due to too much torque causing the drive gear to skip a tooth, it would more likely be attributable to the acceleration, travel acceleration, or jerk settings rather than the speed setting itself per se. I figure it's like skidding in a car, it's the rapid acceleration/deceleration that causes the wheels to skid, not the overall speed per se. I also don't know if the print would reach the full travel speed of 300mm/s during the middle of the print job, since the travel moves are contained within the perimeters(looking at the G-code, the longest moves are about 8cm long). My acceleration and jerk are lower than stock settings though.
magicmushroom666
Prints-a-lot
Posts: 32
Joined: Mon Dec 17, 2012 5:27 am

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by magicmushroom666 »

The grinding might actually be the steppers missing steps, this can sound like a belt grinding or similar. if it was actually grinding the belt there would be signs of it such as belt dust in the area of the stepper. Missing steps could be due to the wrong motor current.
Polygonhell
ULTIMATE 3D JEDI
Posts: 2430
Joined: Mon Mar 26, 2012 1:44 pm
Location: Redmond WA

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by Polygonhell »

300mm/s is very high for travel speed, though how close you get to it will depend on you acceleration and jerk settings.
I run 200mm/s travel, and I've seen a lot of people run 150, it's worth just going very conservative with speeds to see if the issue goes away, you can then creep them back up to find the point it starts.

The grinding sound is more concerning, you can destroy the belts if the gear is slipping against them, but that would only happen if something were impairing the motion of the carriage.
Check the belts are OK, when the head is homed you care about the inside of the belt about 1/3 up from the base, but check the teeth on the belt all the way up.
Run the carriages up and down by hand to dee if you feel any resistance to motion.
It could also be bad digi-pot settings, but that seems unlikely unless you changed the firmware, or a bad motor wire.
User avatar
thedoble
Printmaster!
Posts: 131
Joined: Fri Dec 25, 2015 8:32 pm
Location: Brisbane, Australia
Contact:

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by thedoble »

I would reduce your travel speed to 200mm/s or even 150. I used to run 300 and had problems with layer shift too. Your other speeds sound fine.

Regarding the 'grinding' sound - I don't think you're hearing grinding, I think you're hearing the motors move very very slowly - it can sound like a grinding noise, it sounds totally different to them moving at normal speed. If the carriages aren't touching the base or the end-stops then there's nothing to cause grinding.

You can recreate this situation by sending a g-code move command with a very low F rating, like somewhere between F200 and F50. I bet it will make a similar awful noise :)

G1 Z100 F200
Smoke me a kipper, I'll be back for breakfast.
exalt
Plasticator
Posts: 16
Joined: Sat Jan 02, 2016 2:53 pm

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by exalt »

magicmushroom666, Polygonhell, thedoble thank you for your replies.
magicmushroom666 wrote:The grinding might actually be the steppers missing steps, this can sound like a belt grinding or similar. if it was actually grinding the belt there would be signs of it such as belt dust in the area of the stepper. Missing steps could be due to the wrong motor current.
I double checked the repetier file, and can confirm I've updated the line to

"#define MOTOR_CURRENT {155,155,155,165,0}"

as per what was recommended in the assembly guide for Automation Technology motors, which are the ones I have. However, a bit confusingly, the inline comment in the repetier.ino/Configuration.h asks for 140xyz and 160 for Kysan motors instead of the 155xyz and 165 as per the assembly guide for Kysan. Should I switch to 140xyz and 160 instead?

I did find some belt dust, not a lot, just some crumbs of rubber. There were more black crumbs around the X tower, both near the stepper and the idler pulley, compared with the Y tower, which had just a few crumbs. This seems to be in line with my observation that the layer shifts always seem to toward the left and slightly frontward, indicating the X tower/motor has the most grinding going on. I have attached some photos of the dust:

X Tower Base:
X Tower Base.jpg
X Tower Idler Pulley:
X Tower Pulley.jpg
Y Tower Base:
Y Tower Base.jpg
Y Tower Pulley:
Y Tower Pulley.jpg
I did not find any traces of dust around the Z tower. I now strongly suspect that the X tower is having significant issues with belt grinding, and the slipping of teeth might be causing additional resistance for the Y arm, causing it to grind/slip too.
thedoble wrote:I would reduce your travel speed to 200mm/s or even 150. I used to run 300 and had problems with layer shift too. Your other speeds sound fine.

Regarding the 'grinding' sound - I don't think you're hearing grinding, I think you're hearing the motors move very very slowly - it can sound like a grinding noise, it sounds totally different to them moving at normal speed. If the carriages aren't touching the base or the end-stops then there's nothing to cause grinding.

You can recreate this situation by sending a g-code move command with a very low F rating, like somewhere between F200 and F50. I bet it will make a similar awful noise :)

G1 Z100 F200
I ran the G-code, it's not quite the same. The grinding I get during the pause-then-grind is more of a fast movement at first(initially, this moves the carriages very quickly), then a much louder, more grating kind of noise. It's sounds more pulsating and dull-sounding, but still quite rapid. The best description I can give is it sounds like the ratcheting sound of an automatic screwdriver when the screw is fully seated already, but more dull. Thank you for the advice on turning down the travel speed, I will try that.
Polygonhell wrote:300mm/s is very high for travel speed, though how close you get to it will depend on you acceleration and jerk settings.
I run 200mm/s travel, and I've seen a lot of people run 150, it's worth just going very conservative with speeds to see if the issue goes away, you can then creep them back up to find the point it starts.

The grinding sound is more concerning, you can destroy the belts if the gear is slipping against them, but that would only happen if something were impairing the motion of the carriage.
Check the belts are OK, when the head is homed you care about the inside of the belt about 1/3 up from the base, but check the teeth on the belt all the way up.
Run the carriages up and down by hand to dee if you feel any resistance to motion.
It could also be bad digi-pot settings, but that seems unlikely unless you changed the firmware, or a bad motor wire.
I ran my fingers along the entire printer-facing side of both the X and Y tower belts, there weren't any missing or obviously damaged teeth but there was definitely some wear: I noticed some black powder on my fingers. Moving the carriages up and down feels stiff, but probably not disproportionately so. Your point of there being an impairment to movement got me to closely inspect the idler pulleys and I believe I might have found a possible cause, the top idler pulley on the X tower was slightly off square to the vertical axis. This meant that due to uneven tension the right side of the belt was constantly rubbing against the spacer on the right. This is possibly a source of impairment to the motion. I was very careful when mounting the grub screw and drive gear, so any misalignment is either on one of the idler bearings or on the mounting of the steppers.
Skewed Idler Pulley.jpg
I will readjust the pulleys for alignment, lower travel speed and try printing again. Thank you all for your replies.
User avatar
mhackney
ULTIMATE 3D JEDI
Posts: 5412
Joined: Mon Mar 26, 2012 4:15 pm
Location: MA, USA
Contact:

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by mhackney »

Lower your accelerations too. This kicks in when the print head moves from location to location. Accelerations too high can all cause missed steps and perturb mechanical issues like binding.

Sublime Layers - my blog on Musings and Experiments in 3D Printing Technology and Art

Start Here:
A Strategy for Successful (and Great) Prints

Strategies for Resolving Print Artifacts

The Eclectic Angler
exalt
Plasticator
Posts: 16
Joined: Sat Jan 02, 2016 2:53 pm

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by exalt »

Thank you for the reply mhackney.

I will try the accelerations, although I have already lowered the acceleration from the default 1850 to 1600. Perhaps I should lower the travel acceleration, currently 3000, too, since that seems much more likely to cause the grinding.

I have attempted to square the top pulley to stop the side of the belt from rubbing against the idler bearing side spacers, but it seems impossible. Inevitably the belt pulls towards either side and sticks to it, regardless of how I try to eyeball it. It's probably not possible without adjusting the base idlers at the same time and some more precise tools than 2 screwdrivers and a wrench so it's just going to have to be software adjustments.
User avatar
mhackney
ULTIMATE 3D JEDI
Posts: 5412
Joined: Mon Mar 26, 2012 4:15 pm
Location: MA, USA
Contact:

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by mhackney »

yes, lower this travel accelerations. Travel is usually where this sort of issue presents itself.

Ah, your comment triggered another thought. Are you using Astrosyn dampers? I recently discovered on my machine that the stepper motor is not level, it was being pulled up (at the pulley end) but the belt. I simply printed a little support pillar and wedged that under each of the stepper bodies. Solved the problem completely.

Sublime Layers - my blog on Musings and Experiments in 3D Printing Technology and Art

Start Here:
A Strategy for Successful (and Great) Prints

Strategies for Resolving Print Artifacts

The Eclectic Angler
exalt
Plasticator
Posts: 16
Joined: Sat Jan 02, 2016 2:53 pm

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by exalt »

Thank you for the question mhackney, no I don't have the AstroSyn dampers. I don't want to mess with the screws holding the stepper motors in place because I already threadlocked them and the screws/motors do not seem loose to the touch. They probably aren't perfectly aligned/squared to the vertical axis but that kind of the main problem for me. I don't want to take apart everything I already built if I can avoid it. I guess if the layer shift keeps happening despite it all then I'll have no choice but to find some kind of spirit level to make sure they are vertical, break the threadlock, then tighten the screws to a perfect levelling.
User avatar
teoman
ULTIMATE 3D JEDI
Posts: 1783
Joined: Sat May 24, 2014 5:43 pm

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by teoman »

You fiddled with the motors and they are tight right?

Put a little spot of nail varnish/paint between the shaft and gear. That way you can see if there is anything loose.


When i had this problem it was my joints binding. (Aluminium joints with axels) i solved that by putting molybdeniumdisulphide based grease on them. But there have been 2 iterations since then so i doubt that is your problem.
When on mobile I am brief and may be perceived as an arsl.
exalt
Plasticator
Posts: 16
Joined: Sat Jan 02, 2016 2:53 pm

Re: Persistent Pause Print Failure/Belt Grinding + Layer Shi

Post by exalt »

Thank you for the reply teoman.

I don't have any nail varnish available but I guess a bit of threadlocker is just as good(?)

So after decreasing the travel speed to 200 and travel acceleration to 2000 the issue of huge layer shifts seems to be resolved. I reprinted no.3 and there was no obvious shift/belt grinding. I did still encounter small pertubations in the layers at the same area, but it's a huge improvement. Thank you everyone for the fine advice!

I will continue to try with taller prints to see if the problem pops up again.
Post Reply

Return to “Troubleshooting”