Trying to set belt tension on MK3->MK3S upgrade, 3.90-3421 firmware
 

Trying to set belt tension on MK3->MK3S upgrade, 3.90-3421 firmware  

  RSS
krb
 krb
(@krb)
Trusted Member

I set my X and Y belt tension a couple years ago when I built my MK3 from kit. I didn't have any problems that bothered me, so I left them alone. About six months ago, I finally did the MK3->MK3S upgrade, so had the X-Axis apart and back together. I set the X belt tension by feel. IIRC, it ended up one tooth (in the X carriage) tighter than before, but I could be wrong. My X Motor assembly is the style with the top screw.

Over the past few months, I've had the impression (not based on anything) that the X belt is too tight. Right away, it sounded different that it was before the upgrade, but prints came out ok (ok to me, I do almost all functional parts and I've never put in the effort to really fine tune accuracy).

So last night, for some reason I decided I should think about belt tension again, did some reading and found the various guides talking about belt status numbers. I had never looked at these before, but the advice seems to be to aim for 255-270. I got my Y belt tightened up. It is now ~263, previously it was around ~290.

The X belt is confusing me. Before I adjusted anything,  the X belt was at 241-243. I tried several rounds of loosening, eventually I removed the top tension screw entirely, then loosened the X motor screws so it rotates as far right as it wants, then barely tighten the screws back so the motor doesn't move much while running, I can get the 3.9.0 Firmware belt test to show me ~255/256. But visually the belt seems much to0 loose. It vibrates during the belt test and I can pinch the top and bottom together with hardly any force.

I know the belt status numbers are just a proxy for belt tension, from what I can tell, it's really a measure of current or resistance the motor encounters during the test. So..I'm wondering if I have another problem on the X-Axis that I introduced during the MK3S upgrade. I did grease the X-axis rods last night and it doesn't seem to change the results.  I'm posting here to get ideas on what to check first. I'm thinking I'll end up pulling the X carriage apart, but before I do, I would like advice on what to check for. I guess it could be the X idler pulley, but I think is really hard to get to, so I'd like to know how to determine if that is the problem.

Any advice is appreciated!

Posted : 04/06/2020 4:27 pm
MaschinenCode
(@maschinencode)
Active Member

Same behaviour here. Assembled the MK3S some month ago and set the belt tension according to the values from the printer self test. After testing some PLA filamenents from BASF and several fucked up prints with v.3.9., due to layer shifts in X direction, I decided to have again a look at the belt tension. Ran the new belt test and got completely different values, whereas the belt tension feels nearly the same for X and Y, when doing the finger test:

X -> 234

Y -> 270

This was the point where I decided to fine tune the belt tension, as the new belt test is really quick. But as already mentioned by @krb, I am not able to reach a value below 252 for the X axis, even if the belt is completely loose and vibrating during the belt test. Re-checked the mechanics for the X-axis, applied some fresh grease but nothing changed the value.

Any help is here appreciated as well!

Posted : 10/06/2020 7:00 am
chocki
(@chocki)
Prominent Member

The numbers come down as the current increases, so as you tighten the belts, the numbers will drop.

I have mine at 198 and 214 but this is with 0.9 Deg motors, any slacker, which will bring the numbers up toward 220 and I get a loose pulley message, so I gradually increased tension, ran a belt test until the loose pulley message went.

If you connect Pronterface whilst running the belt test, you can see more detail about the measured values.

vintagepc stated that his values are around the 220 value, but again he is running 0.9 deg motors.

I think the old values of 270 to 290 are now incorrect.

Try tightening until the loose pulley message disappears, then adjust to just before it comes on and see how the belts feel then.

I haven't had time to further refine my tension with print testing yet, but my first two prints seem to have come out without any issues, but they were really quick prints to make a tool jig so not printed for quality (Printed at 150% speed and temperature bumped up and cooling reduced PETG).

Normal people believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet....
Posted : 10/06/2020 11:10 am
MaschinenCode
(@maschinencode)
Active Member

@chocki thanks for the help, but mechanics of the printer have been double checked already. Went back to v.3.8.1 again for testing and what should I say, belt tension ist alright:

X -> 269

Y -> 272

No more layer shifting so far and the print quality improved a lot compared to v.3.9.

 

Posted : 10/06/2020 1:08 pm
krb
 krb
(@krb)
Trusted Member
Posted by: @maschinencode

@chocki thanks for the help, but mechanics of the printer have been double checked already. Went back to v.3.8.1 again for testing and what should I say, belt tension ist alright:

X -> 269

Y -> 272

No more layer shifting so far and the print quality improved a lot compared to v.3.9.

Ok, so if you did the downgrade without touching the belts, this points to the 3.9.0 firmware adjusting the numbers lower for the same tension. I've got my X as loose as I'm comfortable (a little too loose I think, but not slipping) and it's around 261 on 3.9.0

 

Posted : 10/06/2020 4:15 pm
MaschinenCode
(@maschinencode)
Active Member
Posted by: @krb
Posted by: @maschinencode

@chocki thanks for the help, but mechanics of the printer have been double checked already. Went back to v.3.8.1 again for testing and what should I say, belt tension ist alright:

X -> 269

Y -> 272

No more layer shifting so far and the print quality improved a lot compared to v.3.9.

Ok, so if you did the downgrade without touching the belts, this points to the 3.9.0 firmware adjusting the numbers lower for the same tension. I've got my X as loose as I'm comfortable (a little too loose I think, but not slipping) and it's around 261 on 3.9.0

 

Changed nothing! Just flashed v3.8. ...

Did some test prints with exactly the same file and exactly the same settings:

v.3.9

 

v.3.8.

For my printer 3.8. definitely delivers the better results.

Posted : 10/06/2020 4:26 pm
Ondrowicz
(@ondrowicz)
Eminent Member

I am not really terribly experienced (less than one year with MK3S) and did no upgrade, but as I understand, the belt tension is not really the tension of the belts but rather the resistance of the system as it appears to the step motors. So I would also check carefully if the extruder on the X axis can be moved freely with no excessive resistance and with no spots with apparently tougher movement (by hand, with the step motor off). Aside of proper lubrication of the bearings, it is important not to have the screws at the back cover of the extruder (the cover holding the bearings) too tight. They should be tightened just enough to prevent any excessive movement of the extruder.

This post was modified 3 months ago by Ondrowicz
Posted : 10/06/2020 11:05 pm
MaschinenCode
(@maschinencode)
Active Member
Posted by: @ondrowicz

I am not really terribly experienced (less than one year with MK3S) and did no upgrade, but as I understand, the belt tension is not really the tension of the belts but rather the resistance of the system as it appears to the step motors. So I would also check carefully if the extruder on the X axis can be moved freely with no excessive resistance and with no spots with apparently tougher movement (by hand, with the step motor off). Aside of proper lubrication of the bearings, it is important not to have the screws at the back cover of the extruder (the cover holding the bearings) too tight. They should be tightened just enough to prevent any excessive movement of the extruder.

How the system detects the belt tension is well known and as already said, the mechanics of the printer have been double checked. Everything moves nice and evenly, nearly no force is needed when the X-axis belt is detached from the stepper to move the extruder. All screws are tightened with with a torque wrench to the recommendation of 6Nm, so too tight screws can be excluded from the list of culprits.

Did another successful overnight print with the "old" firmware and will for sure keep this setup until I find a potential test scenario to rule out all firmware related print issues.

Posted : 11/06/2020 9:21 am
Share:

Please Login or Register