Consideration about belt tension/uneven belt tension but still perfect cut

Hello everyone|

For the last 2/weeks I had the machine work perfectly, and I mean REALLY perfectly! I pretty much cut stuff on MDF which is not a friendly material to work with 'cause it tend to make the bits dull in a really short time, and we know they are not cheap.

Anyway as I said, I work with Rhino and RhinoCAM, and after a long battle with the machine setup on the last 2/3 weeks I had the most dead perfect cut ever. Now, the last monday I noticed that the left belt tension was loosened about 0.5/0.8kg. I had back setted the Y axis belt tension around 3kg (pretty much) and a little more for the X axis.

The funny fact is that the machine with this uneven belt tension has continued to cut dead perfect stuff till today since I noticed that the bit was really dull so I decided to make a complete checkup of the machine, but know I’m asking myslef what is the correct tension for the belt: The left one (about 2.5kg) or the right one (about 3.0kg)?

I would like to not touch anything else, though I think I need to re-calibrate the stepper motor. …right?

Any thought or idea about this? I know that there’s a lot of post about belt tension/tensioning but I think is a pretty uncommon thing that happened to me!

1 Like

Sadly I had to re-tight some screw and the left Y belt was really flopping on the last cut and I noticed it was slipping off of the belt clip too much.

Now I have retightened (is that a word!?) it to the same tension of the right Y belt. I really want to test it right now, but the bit is really dull so maybe I will not get a tue ‘answer’ from the machine, and is late here in Italy now.

For instance, what material do you usually cut? 2.1kg is pretty much the tension that I had (on the left belt) on my latest cut this evening.

1 Like

Mm that is interesting. So maybe the correct tension is really around 2.5kg… My theory is that the machine just need the tension to grip the belt on the pulley. More tension can cause torque problem?

I have a zip tie on each end, but on the Y belt I had to put 2 because it continued to slip off of the belt clip. Don’t know why

Consider that the original machine worked w/a single motor / pulley driving one side — adding a second motor (or a drive shaft) was an upgrade:

So you’re saying that today my machine was working with only a motor driving the Y axis? (exasperated meaning). In that case the correct tension would be around 3kg or little less more

Well, I don’t exactly know when the Y belt started to get loose, but I noticed it on monday. Before this and untill today I had perfect cut. I mean (for example) 10.1mm square on the CAD and 10.1mm square on the cut. This of course for any sort of drawing/cut I did in the last 2/3weeks. And I did a lot!!

What I know for sure is that I was cutting perfect stuff with an uneven Y axis belt tension from this monday

1 Like

Ok so that also means that I have made a really good motor calibration indeed (I guess). Hope to have the same great results tomorrow

1 Like

Mmm that look logic (to me). So is a matter of equilibrium? I mean, when I reassembled all the machine (we talked about that back if you remember) I setted both the Y belt to 3kg or less more, and about 4kg for the X as I thought that it run with only a motor. The results were perfect so I pretty much did the same this time, but I only had to tension one belt instead of 3. I need to recalibrate the motor anyway, right?

That is totally true. And using the same logic the results is that I was more than lucky LOL!

1 Like

Oh well the bright side of the thing is that I cutted all the stuff that I need from now till end of the year.

Anyway, we’ll know tomorrow if all those theory were right or not!

1 Like

Did 2 test cust today, with a dull bit (the only one I had a.t.m.) and cut were still perfect. Tollerance is between 0.05/0.07mm which is quite perfect for my needing.

The only thing I noticed today is that, as you said to me some weeks back, I have 2 of 4 motor pulley with the screw not ‘touching’ the flat side of the shaft. The X axis is totally off. But guess, thanks to the upgraded screw instead of those tiny grain is still ok. I’ll put them in the correct place once the C-Beam arrive to me (if I will be ever able to pull off the pulley from the shaft). I feel I don’t need the whole upgrade kit right now.

1 Like

[Disclaimer: I’m not a mechanical engineer. At all.]

I don’t think this is the case at all. Tension means two things that you care about: It changes the length of the timing belt, and it increases the force on the stepper motor spindle and pulleys (thus increasing friction).

Really I think what you want is for the pitch of the two belts to be precisely the same. In practice, the easiest way to approximate that is to assume that the belts are cut from the same batch and are the same age, so if their tension is the same, then they’re ending up with the same teeth-per-mm measurement. (Belts are certainly built with the same pitch, but they change over time.) You can’t fix this with steps-per-mm in the driver, since there isn’t an adjustment for each side.

But the amount of work being done by the motors is only affected by tension when the belt pushes down on the shaft. The more tension, the more friction on the motor and the supporting pulleys, and the more the motor has to work to turn a step. But these are stepper motors; you may produce slightly more heat than the other side to do a step, but if you’re not losing steps, you’re moving precisely the same number of teeth on each side. You’d notice if you were losing steps.

My hypothesis: the tension has to be significantly different before the length of the belts are different enough to matter. Both sides are making the same number of steps, which means that both sides are doing the same amount of work; different tensions can only mean a) the same number of steps (and the side that has to deal with more friction works harder and generates more heat), or b) you’re losing steps and you’re getting bad cuts.

You’re right, but my point is that the amount of that “more work” can be ignored; it means you use some tiny amount of electricity that turns into heat instead of motion. If you’re not losing steps, you don’t care. (Once you start losing steps, the entire game changes, of course.)

There’s a fairly easy experiment to do if you think one side is doing more work than the other: just point a thermometer at the motors. 75% more work requires a significant temperature difference.

Well guys I really don’t know why it countinued to work perfectly. For sure what I learned is that you need just enough tension to prevent belt slippage over the pulley, And that can be achieved only by trying to cut stuff. Ironically the machine seem to cut better hard wood than soft wood or MDF.

Once I figured how to clamp guitar body on it I’ll also try 3D stuff with my current settings.

That’s one nice! I’ve seen better stuff from ESP company, they usually build insane stuff, like this http://cdn.growassets.net/user_files/esp/photos/002/240/840/xlarge.jpg?1390580313

Anyway, I’m a guitar builder and for now X-Carve helped me building templates of my own model, which I used to build by hand. But now is time for the machine to help me cut stuff like bellycarve and armrest. All other stuff I will continue to build by hand as usual

1 Like