1

Topic: Belt slop / backlash

I have firmly determined that the cause of the y-axis backlash I'm experiencing is caused by very sloppy/loose belts on the y-axis. These are the long belts on the carriage, not the short one connecting the stepper. Unfortunately, it doesn't look like solidoodle provided any way to adjust them. 

The mounts are molded plastic.  I am at a complete loss, other then the hope of someone maybe designing a new carriage?  Seems unlikely. Is there any kind of external belt tensioner or something that could be added? 

I guess the backlash may just be something I have to live with.

2

Re: Belt slop / backlash

Is there any reason you couldn't print a simple tension device that slides over the belt?
Rough drawing but you get the idea.

http://i.imgur.com/q8rs3r7.png

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

3

Re: Belt slop / backlash

Something along those lines was percolating around in my head after I posted this... I was trying to think of something that would be adjustable, but your sample drawing is much more straightforward... and probably much more achievable.

I'll see what I can come up with.

Thanks!

4

Re: Belt slop / backlash

... and it turns out there are a number of designs on thingiverse... should have googled a little more.  I'm anxious to try this out! smile

5

Re: Belt slop / backlash

Ya, these things are dirt cheap and sold nearly everywhere that sells printer parts:

http://www.robotdigg.com/product/41/Tim … ner-Spring

SD4 w/ RUMBA, E3D Volcano, all bearings, glass bed

6

Re: Belt slop / backlash

Argh.  Solidoodle has the belts in a plastic "channel" and it's not possible (or at least not very easy) to get to the alternate side of the belt to put a tensioner on (so that it won't interfere with the pulleys).  I'm going to see if I can squeeze one in there, but I'm doubtful. 

I may have to dissassemble far enough to remove a pulley... didn't really want to do that... I'm worried about getting things out of alignment.

7 (edited by jagowilson 2015-02-19 02:40:03)

Re: Belt slop / backlash

There has to be a way to adjust the tension. How could they ensure equal tension otherwise? From my experience if the belts are taught enough you can't install them . . .

8

Re: Belt slop / backlash

I think the belts on the y-axis have a section (in the y-carriage) that holds two points of the belt.  You should be able to slip the belt out of the holder at one end of the holder and pull the belt so that it catches a tooth further along, making a slack part between the two sections of the belt.  I don't think you should be needing to put a tensioner in there unless you need really fine adjustment (less then one tooth difference in belt length).

9

Re: Belt slop / backlash

android78,

Yes, that is the design.... at least on the left side.  On the right, the two tabs are so close together, there really isn't room to leave any excess belt.  The bigger problem is that those clips are injection-molded and very thin...  I'm pretty sure a hair too much pressure would break them (based on my experience with a few other pieces on the press.).

What I've done for now is print a small tensioner and placed it directly next to the carriage.  I then reduced the Y-axis size in Repetier to be 190mm, to prevent it from hitting the pulley.  That seems to have worked, from visually inspecting the situation.  I'm going to try some test prints shortly.

10 (edited by jagowilson 2015-02-19 02:56:10)

Re: Belt slop / backlash

Are the belts open loop or closed loop (cut or not cut)? If they are cut, it sounds like the right side was cut too long. Just a guess.

11

Re: Belt slop / backlash

closed loop.  I can see where they adjusted the tension on the left side... they just didn't make it tight enough.

the left side is actually worse... I think they just didn't adjust it properly... the right side isn't *as* bad, but as I said, it seems like there is very little room for adjustment there.  I think they are relying on the tolerances of the molded parts... but I'm new to this, so there may be something I'm missing.

I did ask solidoodle for a procedure to adjust the tension... the case is still "open".

12

Re: Belt slop / backlash

How about providing some high quality pictures? I bet we can figure it out.

13

Re: Belt slop / backlash

Yeah, the left belt is easy to adjust the tension. I did that a couple of days ago. The right belt on mine isn't too bad, so I haven't touched it yet. I might try one of those printable tensioners though. My big problem in my X belt. It is really loose! And I haven't found a good way to adjust the tension on it yet.

14

Re: Belt slop / backlash

Fredjikrang wrote:

Yeah, the left belt is easy to adjust the tension. I did that a couple of days ago. The right belt on mine isn't too bad, so I haven't touched it yet. I might try one of those printable tensioners though. My big problem in my X belt. It is really loose! And I haven't found a good way to adjust the tension on it yet.

I wonder if taking the screws off the bottom of where the x belt passes through (I think there are 4 screws there) will reveal a similar mechanism as the y belt tensioners.  Mine is WAY too tight to the point where the x-axis makes a terrible noise and I can see the belt wearing already after only a dozen or so prints.

15

Re: Belt slop / backlash

I give up for now.... The belt tensioners make things look great... but I still have cubes that are .4 mm shorter in the Y axis.  I guess there is some other factor causing backlash.

If I print a part that has Y-direction measurements that are mid-stream (such as a Plus shape), the dimensions are accurate, but a cube or circle results in the Y-direction being .4 - .5 mm shorter than the X-axis.

half a millimeter probably doesn't matter for the stuff I'm likely to print... it just bothers me. (OCD?)

16

Re: Belt slop / backlash

Not OCD... .5mm is the difference between the filament being laid on top of the previous layer and it being totally off.

17

Re: Belt slop / backlash

There are no issues with the filament missing the layer... it just seems that the part is shorter in the Y axis.  I printed the Plus shape because I was wondering if the steps/mm were wrong... but the "arms" of the plus were exactly 10mm wide, while the overall length was still about .5mm short.

It's very repeatable... I don't have any wavyness or anything in walls.  Just not sure what I need to adjust.  when I move the pulley by hand now I don't see the sloppiness in the belts... It seems odd that tightening them didn't change the amount of backlash at all though.

18 (edited by jagowilson 2015-02-19 22:37:01)

Re: Belt slop / backlash

Timing belts are picky about tension. Too loose, and you will see backlash. Too tight, and you will also see backlash. There is a tension that is "just right", and the tensions of the left and right belt need to be equal. I believe the press drives the Y axis with a short motor belt below the back Y drive rod, and it needs to be tight. Your description indicates backlash on the Y axis, which is the most problematic axis on machines of this type.

An effective way to measure backlash is with a dial imdicator mounted directly on the X guide rods. That way you can put a number to it and eliminate some guesswork.

19

Re: Belt slop / backlash

Fredjikrang wrote:

Yeah, the left belt is easy to adjust the tension. I did that a couple of days ago.

Sorry to revive an older thread but my left belt tension (y?) is very sloppy how did you adjust the tension.  I cant see m to see anyway of doing this?

Thanks

20

Re: Belt slop / backlash

I used a printable tensioner, as mentioned above, though those tensioning springs are probably a better idea.

It helped a little.  Mostly I just decided I could live with the backlash in most situations.