Belt tension ?  

Re: Belt tension ?

Acceptable belt tension levels are 200 - 280. Your belts are too loose.

Re: Belt tension ?

So your saying if I tighten my belts the numbers will drop

Re: Belt tension ?

Yes.

High # = low belt tension.

Low # = high belt tension.

I know thats counter intuitive but its the way it is.

RE: Belt tension ?

I had the same issue. Thought the higher values meant that my belts were tighter then they should be. I tightened them and my values are going down and prints should be working flawlessly now. 

RE: Belt tension ?

The belt tension display on the LCD is not a valid indicator of belt tension.  The numbers are not even meaningful for what's really going on.

My belt tension has been as high as 30 lbf, way too high, yet my LCD readings are 295 & 296.   No where near the suggested 240.  

If you get to 240 by tightening the belts, you are probably damaging the belt, the motor bearings, and warping the printer frame. Unless you have really bad bearings... that will also get you to 240.

RE: Belt tension ?

But you know that the belt status is only updated if you do the self-test?

If you want a current measurement, you have to start the self-test again.

RE: Belt tension ?

The LCD numbers are meaningless. lol.  I have watched them after many many self tests and an extremely wide range of measured belt tensions.  They are meaningless for belt tension.  I went from slack to more than 30 pounds tension (self testing in between) and the numbers moved about 2 points. Do NOT use the LCD numbers if you are trying to adjust the belts. A waste of time.

Belt Tension Math

If you want lower numbers, just tighten the bearing mounts a bit (add axis friction and drag).

RE: Belt tension ?

I have a belt frequency meter and have worked to this approximation which works out and feels about right.

With the Y axis all the way back, the distance from the belt clamp to the pulley centre is 240mm. (S)

Using a tension of 8lbs this is 35 Newtons (T)

Now using these pages from Gates  http://dpk3n3gg92jwt.cloudfront.net/gates.pt/pdf/drivedesign_lightpower.pdf

I get the belt mass constant of 5 (M)

The belt width is 6mm (W)

Then using the formula from here:  https://assets.gates.com/content/dam/gates/home/resources/resource-library/operating-manuals/sonic-tension-meter-manual-jan2018.pdf

T = 4 x M x W x S^2 x f ^2 x 10e-9

And transposing it to get f= Sqr (T/(4M x W x S^2 x 10e-9)  which is Sqr (35 / 4 x 5 x 6 x 240^2 x 10e-9)

Which = 71 Hz for the Y axis

RE: Belt tension ?

Thanks Chocki - now what is X supposed to be?  And the database needs both at 6 lbf, which is for the LDO motor radial load spec bearing safety.  lol

Oh - and did you use the freewheeling second pulley or the fixed gear to gear lengths?  It affects the static deflection method, not sure it affects the sonic method.

RE: Belt tension ?

x is multiplied by.

I have actually got my radial loading too high, it should be max 6.4 lbs (I read your previous posts then did some homework - ooops!), but that is because any lower, stall guard does not work and I get Y axis self test failures, maybe due to 0.9 Deg OMC motors.

I am going to play with the stallguard setting to get the tension lower and reliably pass self test. I am also going to weigh the belt to make sure 5 is exactly right, it's extrapolated from the link to gates belts, but there's nothing like accuracy.

We use frequency meters at work for setting up belts, hence I have access to one, but I would not like to buy one of these, it costs more than a new printer!., it is basically an electret microphone, some circuitry and an lcd display. It must be possible to build something cheaper using some good op amps, filters, automatic gain control and an esp32 with oled display. (That's the internal company price)

Most cheaper frequency meters for belts give out above 60 Hz, but these go up to hundreds of Hz

Found it: https://www.okw.com/en/Applications/Tension-meter-for-measuring-belt-pretension.htm

RE: Belt tension ?

Sorry, that link shows the meter, but it is actually the case manufacturer, this is the meter specifications: https://www.walther-flender.de/en/couplings/timing-belts/drive-belt-accessories/product/products/WF-TM%20Nano%20tension%20meter-227/

Oh and I found it a little cheaper at €486   ermmm, don't think I will be buying one myself.

RE: Belt tension ?

I used the length between two fixed points, one of them being the belt clamp under the heated bed, and the other being the freewheeling pulley where the belt touches the pulley, you could also use the belt underneath between the pulleys, this would be a slightly longer length so (S) would be larger and the resulting frequency would be lower.

At work we have a fixed length clamp of 0.5m for lift belts, you simply clamp the fixed frame to the belt, tap the piece clamped within the frame and measure the frequency.

RE: Belt tension ?

Now all we need is some sort of printable accurately repeatable device that anyone can use and calibrate it against my belts, since I have an accurate way of setting the tension (Assuming I don't set it too high like I have! 🙂 )

RE: Belt tension ?

You don't need a meter to hundreds of €. You can just download a free app to your phone... https://play.google.com/store/apps/details?id=org.intoorbit.spectrum&hl=en_US

I just measured my belts to 85Hz - both of them. And it seems to be a reasonable value.

The LCD values are X:265, Y:293. I have tried to get the Y value down, but it gets too tight with strange sounds from the motor. I believe the LCD values are crap! The manual should instead recommend a frequency range.

Maybe people could share their well working belts frequencies here?

RE: Belt tension ?

Higher "tension" numbers in the Prusa LCD mean you will be happier since they show drag and friction, not belt tension.  X and Y belts are different lengths, and should have different frequencies.  People have broken their printers trying to adjust the belts to get to that phony 240 number the manual suggests is ideal.

RE: Belt tension ?

Posted by: chocki

Now all we need is some sort of printable accurately repeatable device that anyone can use and calibrate it against my belts, since I have an accurate way of setting the tension (Assuming I don't set it too high like I have! 🙂 )

There certainly is one, although Tim dislikes it:

https://www.thingiverse.com/thing:2230598

I would love to see what your standard reading is. I would use the same number on mine.

Note: this does expect repeatability between Pentel G2 pen springs. I would expect them to be very good.

RE: Belt tension ?

I think maybe something like this, but with a weight pushing down, not a spring, last time I checked Gravity was fairly consistent around the world, yes I do know there is some variance, but not enough to affect belt tensioning.

RE: Belt tension ?

A simple balance approach.  X should lift 15/64; Y should lift 12/64.  And it is semi-accurate/repeatable and actually allows you to know what the belt tension is in lbf.  Not included is a rule to show deflection values.  But I think those are easy enough to come by.

Cool Stantia