Welcome to the SeeMeCNC Community

I'm still trying to figure out exactly how it happened but two of my 3 Rostock Max belts have an area on them that has been stripped. After assembly, I was having what appeared to be belt lash issues so I tightened the belts to felt to me should have been tight enough. I continued to have the symptoms of belt lash so I tightened the belts even more. I'm not sure if I went too far or if something else happened. One other strange thing is that I had some motor connector problems on the axes that ended up getting stripped and that was causing the motors to jump. An interesting thing I noticed is that the damage is in exactly the same spot on the two belts, at the motor gear when the head is centered and about 3 in above the bed.

I have the exact same issue on the X and Z belts. In fact, X only hanging by a thread at this point. I will not be printing anymore until I get some new ones.

No idea what happened. It has gotten worse in the last few days.

I'm still not confident I know the exact cause but one clue appears to be that the damaged area is near the motor pulley around the time when the nozzle would be at the bed. While I was tuning the machine I had other issues that were causing the nozzle to hit the bed. Maybe during those times the motor pulley slipped on the belt and slowly started eating it. I don't remember any violent impacts though so I'm not sure if this would explain it or not. Maybe a combination of that and having the belts too tight? I ordered new belts so hopefully I'll get them by Tue. In the meantime, I have taken the two damaged ones and pulled them off and changed them end for end. So now the damaged area is running up over the idler. They stretch too much due to the fact that you can see the bare threads and it's pretty weak but it does seem to let me keep printing this way.

There are a couple of factors that resulted in your eaten belts - I have the same issue.

First off, you may have gotten a set of drive gears that were not wire brushed after manufacture. This will leave the gear teeth very sharp.
The other issue involves what you've already found - the hot end being too close or in contact with the build surface. What happens is that instead of the stepper motor stalling, it starts chewing on the belt. It's my understanding that Delta geometry printers really don't have belt lash issues like Cartesian printers do - if you're seeing something that would normally be attributed to belt lash, chances are pretty good that one or more of the stepper motors is having Belt a La Carte for lunch. Because the early gears are sharp, it does pretty serious damage to the belt in a very short period of time.

g.

I am persuaded that my X belt was chewed by the gear during an incident where the nozzle dragged on the build plate for about an inch. I went from tape on bed to tape on glass on bed and forgot to adjust the height.

The nozzle dragged from left to right as it started the loop around a test object, so that's probably why only X was affected. My Z belt has what looks more like chafing along the edge for about 8 inches. I probably won't replace it immediately, but I'm ordering 3 new belts.

I had a few g-code issues that contributed to the mess as well. When you drive the carriage all the way to the top or to the bottom, that loud buzzing noise you hear is the stepper motors eating the belts.

g.

geneb wrote:I had a few g-code issues that contributed to the mess as well. When you drive the carriage all the way to the top or to the bottom, that loud buzzing noise you hear is the stepper motors eating the belts.

g.

Then you should turn down the digipots or further tighten the belts, ideally you want the steppers to skip rather than damage the mechanical components, steppers will skip all day without hurting themselves.
Backlash on Delta printers looks a bit different than on conventional printers, but it's still very much present, it causes a layer to "tilt" (be lower at one end) in the direction of motion, but enough will result in the familiar infill not reaching the perimeter issue most associate it with.

FWIW I just checked and on mine if I manually stop a carriage moving, the steppers skip rather than the drive gear eating the belts.

What variable needs to be changed and what value should it be set to?

tnx.

g.

The digipot settings are in configuration_adv.h, I have mine set to 160, the default is 195, but I turned mine down because the stepper drivers were getting too hot, not to prevent the belt skipping. The GT2 belts have a rather shallow tooth profile, so they also have to be very tight to prevent the pulleys from slipping teeth.

That's very intersting, I just turned my digipots down to 130 which may be too low, I'll have to play with it. Since my belts are chewed anyway, I will try stalling the carriage and seeing if the motors will skip without chewing the belt. I'm used to mid size CNC routers which will lose steps before breaking anything so I was surprised to see damage before missed steps. I also had an overheating problem which is why I turned the digipots down. I added a fan blowing on the rambo board as well so if I turn the digipots back up a little I should still be ok I think.

I also had the belt lash issue at first because I had the belts pretty loose. I was afraid that I had tightened them too much but I definitely have no belt lash symptoms now and if I can get the motors to stall before chewing the belts I'll leave them tight.

My understanding is that a properly tensioned belt should produce a bass note when plucked. Correct?

g.

geneb wrote:My understanding is that a properly tensioned belt should produce a bass note when plucked. Correct?

g.

Sure but I don't know how you'd actually verify that, I can't "pluck" my belts, because they hit the outside of the extrusions.
I'd go with as tight as you can get them within reason, if you can see them vibrate when the axis are moving, or the belt on one side of the carriage has obviously different tension than on the other they are too loose. Mine aren't as tight as on my other machines, but a lot of that is the length of the belt.
You can usually also see backlash, if you watch the carriage and write some trivial gcode to move it up and down, the change of direction should be instantaneous.

I've gotten my motors working, but can someone please explain what I look for/listen for to see if the motors are slipping? So far I've had one issue when I clicked the wrong Z arrow button in Repetier and the z carriage moved up against the limit switch and kept running. Haven't figured out why that happened. I haven't calibrated yet so could that z carriage running up against the limit switch have to do with setting the z axis in the firmware?

Richard

That's one of the things that contributed to the demise of my belts. You would think that hitting the limit switch would stop the travel no matter what.

As to the sound, it'll be a vibratory "bumping" noise.

g.

Thanks g,

Why didn't the limit switch stop the travel? Did you find telltale black dust in the niche where the toothed pulley lives? I didn't let mine go on for more than a second or two, but I was surprised it happened. John?

And, can you further explain the issue of a vacuum cleaner and static electricity? Is the RAMBo more sensitive? I've vacuumed around computers and assorted electronics many times and have never had any experience of a problem. Why am I not supposed to get the vacuum nozzle how close to what?


Richard

Craftgizmos wrote:Thanks g,

Why didn't the limit switch stop the travel? Did you find telltale black dust in the niche where the toothed pulley lives? I didn't let mine go on for more than a second or two, but I was surprised it happened. John?

And, can you further explain the issue of a vacuum cleaner and static electricity? Is the RAMBo more sensitive? I've vacuumed around computers and assorted electronics many times and have never had any experience of a problem. Why am I not supposed to get the vacuum nozzle how close to what?


Richard

It's relatively normal for repraps to ignore limit switches except while homing, it's a firmware flag.
It prevents noise on the limit switch wires prematurely ending a multi hour print.

USB connections are very susceptible to ground noise, RAMBO shouldn't be any worse than RAMPS or any other of the standard printer exlectronics, but some people do have issues, a lot depends on your wiring.

The black dust is the remnants of the belt teeth.

High velocity air moving across the mouth of the vacuum is what causes the static electricity. If the nozzle were to accidentally touch the circuit board, it could discharge the static electricity into components on the board, silently killing them. I learned this lesson the hard way after destroying a $366 386DX-40 motherboard in the late 90's.

g.

Does anyone have a source for the GT2 belts by the inch? SeeMeCNC.com is out of stock, I think I bought the last they had but it was only enough to replace 2 of my 3 belts, all of which are chewed. The only other online sources I've found only have fixed length belts like 40in which is not enough for the Rostock Max.

wonderemporium wrote:Does anyone have a source for the GT2 belts by the inch? SeeMeCNC.com is out of stock, I think I bought the last they had but it was only enough to replace 2 of my 3 belts, all of which are chewed. The only other online sources I've found only have fixed length belts like 40in which is not enough for the Rostock Max.

Try https://sdp-si.com/eStore/ they sell if by the meter because it's a metric product.

Perfect, thanks! They had 90 meters in stock. I ordered enough for two complete sets. Hopefully I can get this thing to stop eating them, but I'll be set just in case. My gears do seem to be razor sharp. I'm going to try gently filing the edges of the teeth to dull them a little.

Chaulk up another victim http://forum.seemecnc.com/viewtopic.php?f=54&t=1537

I want to know what causes this. It is not happening at a point on my belt where a crash could occur (ie at the limit switches). I have not crashed into the bed or had any other mishaps. Something is wearing it prematurely. Is it too much tension? Not enough? It will be expensive and a pain in the a$$ if I need to keep replacing belts every 3 weeks.

I just lost one of my Rostock GT2 belts. What are the advantages of the new 20 tooth plastic pulleys? I've had some issues with the black dust and I wonder if the new ones will cut down on the abrasion. Any one have thoughts?

Thanks.

plastic wont hurt the belt as much that's for sure, might be another reason they did away with the metal ones they were using also. mine have really sharp teeth on them. its just a matter of time when somebody drags their head across the platform or crashes into it. once it starts it gets worse and worse over time. I have cnc machine that use metal pulleys also with a belt but the teeth don't do this type of damage when and accident occurs.

The choice of the pulley material (metal vs. plastic) is a matter of price, desired precision, inertia, color, magnetic properties and, above all, personal preference based onexperiences. Plastic pulleys with metal inserts or metal hubs represent a good compromise.

(SDP-SI).

If the pulley has been manufactured to the specs, it will not have a significant effect on belt wear. The problem is not the pulley teeth wearing the belt, the problem is either the teeth "skipping" over the pulley grooves (which would result in a very obvious motion problem - i.e. the head would not travel in a plane - during the print. Of course, rehoming would reset the axis with the skipped teeth) OR stress put on the belt teeth that happens just below the force required to skip. I suspect this later issue is the major contributor. Skipping makes a thunk sound that is pretty hard to miss.

There are machines out there with 1000s of hours on metal and plastic pulleys (at the same or similar numbers of teeth) and belts with no signs of wear.