9mm belt upgrade, losing steps now, thoughts?

I’ve run into an interesting situation after upgrading to 9mm belts. I’ve started to lose steps on occasion and I can’t figure out why exactly. The best example would be a simple toolpath I use hundreds of times a year, a simple t-slot. I ran the toolpath after the 9mm belts in some OSB and I ended up with a oval entry. I clearly lost some steps on the Y axis on some of the most simple G-code I have. The belts are tensioned to 3.8lbs on my Y axis, I don’t see any issues visually. I made sure to check my output on the x-controller and adjusted them all to be 2.8A as required for my 270oz steppers. Any thoughts on what might be causing this?

Tight belts. Try a lower tension setting.

3.8lbs is a tight belt? What tension is everyone running? I considered that as well, but this tension is looser than my factory stepper use to run. Does the pitch on the 3gt3 belts require less tension?

Pinion/pulley not 100% secured to motor shaft?
What is the diameter of “oval entry”?

3.8lbs isnt tight per other user reports here, so I dont think it that giving you issues.

That’s a good point. I didn’t consider the pulleys may be slipping on the steppers. I made a diligent attempt to seat the set screw on the keyed flat of the shaft, but I suppose I should check to make sure it didn’t back out. I probably should put some thread locker on it.

Well, I would guess on that very simple toolpath it probably ended up losing 1-2mm of Y axis step for it to be that obvious. The entry and exit are supposed to be exactly the same location, but it ended up exiting like 1-2mm off. On a much more complicate carve it held together fine, it was about a 6hr carve. However, there was a very clear point where it lost step on one portion. Luckily it was one of the very last parts, so I was able to salvage it.

You gave me a good point to start. I’m going to remove the pulleys on my steppers and see if I can identify marring on the stepper shaft to prove it was walking. If I see it I’ll threadlock the set screws in and assume that was the problem.

What feed rate,rpm,depth per pass & bit details are we talking about here?

It’s really pedestrian work, it should never lose steps. I don’t even recall what the speeds were on that toolpath as I’ve just been running saved gcode for years. It’s a simple downward plunge, pecking, and then adaptive clearing forward slot. It isn’t aggressive at all, it should never lose steps on it. The whole toolpath probably takes 20-30 seconds. When I get home I can post the gcode.

If your not is getting dull it could create more load on the machine.
I think T slots are somewhat aggressive cuts.
Could be a material change as well.

it’s a new t-slot bit, it may have only cut around 20 or so. I’d wager it’s still extremely sharp. It may be the OSB, as that stuff is pretty dense. But, with 270oz steppers, I really can’t imagine it would struggle with that. It never did with my factory NEMA23’s. And, it also lost steps on other jobs since.

As a matter of fact, now that I remember, I think it’s lost steps on every job I’ve done since the upgrade. I was doing a very shallow vcarve as a test in some pine, and it skipped on the Y axis on a .05" depth vcarve, which is a joke. I had forgotten about about that one. The weird thing is it didn’t loose steps on much more aggressive cuts in the same job. I’ll start with checking the set screws.

I had a similar issue when i upgraded to 9mm belts, they were so much stronger than the 5 mm that the motors couldn’t move them when they were tighter.

I regularly move the machine slowly without power (some people say this is a bad idea but I’ve never blown anything up) and have a good idea about how much force it takes.

With the new belts I adjusted the tension until it felt like a similar force to move it manually. The result was that the tension in the Y Axis seemed rather slack. Since these adjustments I have not had a single missed step.

okay, that has me wondering. Do you measure your tension or do you do it by feel?

I lowered my tension to 3.2lbs but I don’t think any lower is going to do any good. Honestly, I think I’m barking up the wrong tree. I do think I may have found the problem. My dust collector crapped out a few weeks ago and I’ve been running with only a air blast to clear debris temporarily. I have a feeling dust in the vwheels may be causing my problems (it was pretty…severe). I honestly never ran the machine without dust collection so it’s something I didn’t consider.

1 Like

Just by feel initially, then use a luggage scale (gives an approximate reading, can’t remember the numbers, just checked they were similar)