Rated torque on pulleys?? Why is there not a problem here?

discussion of design changes / improvements / suggestions
northbear
Posts: 254
Joined: Mon Apr 09, 2012 9:21 pm

Rated torque on pulleys?? Why is there not a problem here?

Post by northbear » Wed Jul 25, 2012 5:54 pm

Looking at using larger motors then the standard design and after reading comment by Improbable Construct (http://www.shapeoko.com/forum/viewtopic ... =539#p4441) regarding his motor and driver and that the belt skips before the motor stalls got me to thinking.... What is the rated torque for these pulleys. And the result I come up with is concerning to me, so hopefully I am missing something! According to a couple of charts from Sdp-SI and York shown and linked below the maximum torque these pulleys can handle is 19.2 oz*in :o Well below the standard motor torque of 68 oz*in!!

My concern is if I go with large motors the pulley will fail! Any thoughts on this?

http://www.sdp-si.com/D810/PDFS/Technic ... 00T061.pdf --- page T-79
111.PNG
Page T-79
http://www.sdp-si.com/D810/PDFS/Technical-Section/8100T061.pdf
111.PNG (113.2 KiB) Viewed 10204 times
http://www.york-ind.com/print_cat/engineering.pdf ---- Pg E23
222.PNG
Page E23
http://www.york-ind.com/print_cat/engineering.pdf
222.PNG (40.96 KiB) Viewed 10204 times
My buildlog is here

Improbable Construct
Posts: 997
Joined: Tue Apr 10, 2012 3:21 am
Location: Fairhope, AL
Contact:

Re: Rated torque on pulleys?? Why is there not a problem he

Post by Improbable Construct » Wed Jul 25, 2012 6:19 pm

Wow that is some interesting information about the torque ratings of the pulleys. I just double checked and the belts do indeed skip on the teeth of the pulley if I run the gantry into the side plates. It does not appear to do any damage to the pulley or the teeth on the belt. I am running my alternate belt configuration. http://www.shapeoko.com/forum/viewtopic.php?f=11&t=227 I have the belts pulled tight enough that you can pluck them like a guitar string. This configuration does lower the contact patch of the belt on the pulley, but I have not had it skip teeth when cutting, only if I crash into the end stops. There are other belt and pulley options that I have seen on 3D printers. Some of them have bigger teeth, but I don't know what the torque ratings are. Also there are questions abut backlash from the tooth engagement area.
Shapeoko #Classified some of the bolts may be original parts.
Shapeoko 1 # ???? Stainless plates, still in the box.
Shapeoko 2 # 3926 not stock
Shapeoko 3 # 0003
Store:
http://ImprobableConstruct.com
Twitter:
https://twitter.com/ImprblConstruct

Digitalmagic
Posts: 123
Joined: Sat Jun 23, 2012 8:47 am

Re: Rated torque on pulleys?? Why is there not a problem he

Post by Digitalmagic » Wed Jul 25, 2012 7:21 pm

The values shown in the SDP/SI table for MXL are surprisingly low.
But after studying this, I really prefer GT2 pulley/belt instead: better accuracy, better teeth design against backslash.

If you read the Table 34 on the PDF you linked, you find:
GT2 pulley for 3mm pitch / 6mm belt width / 18 grooves. We get 1.84 Nm, which converts to 261ozf.in !

MXL belts can yield good results, but GT2 is definitely a step ahead.
Last edited by Digitalmagic on Wed Jul 25, 2012 8:22 pm, edited 2 times in total.
s/o #650 - Wish my digits are ... magic!

northbear
Posts: 254
Joined: Mon Apr 09, 2012 9:21 pm

Re: Rated torque on pulleys?? Why is there not a problem he

Post by northbear » Wed Jul 25, 2012 7:43 pm

Improbable Construct wrote:I just double checked and the belts do indeed skip on the teeth of the pulley if I run the gantry into the side plates....... I am running my alternate belt configuration. .......
Looking at that belt configuration I estimate that you have the belt wrapped about 40 deg around the drive pulley, which would give you about 2 teeth in contact with the belt. ((40deg/360deg)*18teeth = 2 teeth). The recommended tooth engagement for timing pulleys is at least 6 teeth -- in this case a minimum of 120 deg of belt wrap. My guess is this is the reason you would skip teeth before faulting the motor.
My buildlog is here

Improbable Construct
Posts: 997
Joined: Tue Apr 10, 2012 3:21 am
Location: Fairhope, AL
Contact:

Re: Rated torque on pulleys?? Why is there not a problem he

Post by Improbable Construct » Wed Jul 25, 2012 9:58 pm

northbear wrote:
I estimate that you have the belt wrapped about 40 deg around the drive pulley
Good guess! I just measured it and you are correct! I hope that the new motor plates that Edward made up will help with my low tooth engagement. The new plates move the idlers a bit lower and closer to the motor than I have them and also allow the motor to move up some. I think this will give me more belt wrap. I will say though that the only time the belts skip is if I crash into the frame. I have not had them skip under normal cutting.
Shapeoko #Classified some of the bolts may be original parts.
Shapeoko 1 # ???? Stainless plates, still in the box.
Shapeoko 2 # 3926 not stock
Shapeoko 3 # 0003
Store:
http://ImprobableConstruct.com
Twitter:
https://twitter.com/ImprblConstruct

northbear
Posts: 254
Joined: Mon Apr 09, 2012 9:21 pm

Re: Rated torque on pulleys?? Why is there not a problem he

Post by northbear » Thu Jul 26, 2012 3:33 am

Digitalmagic wrote:The values shown in the SDP/SI table for MXL are surprisingly low.
But after studying this, I really prefer GT2 pulley/belt instead: better accuracy, better teeth design against backslash.

If you read the Table 34 on the PDF you linked, you find:
GT2 pulley for 3mm pitch / 6mm belt width / 18 grooves. We get 1.84 Nm, which converts to 261ozf.in !

MXL belts can yield good results, but GT2 is definitely a step ahead.
The concern I would have with this pulley is the diameter is larger then the 18 tooth MXL and so you would get less applied torque and less resolution. Looks like you can go down to a 17 tooth GT2 3mm pitch pulley with a 1/4" bore (https://sdp-si.com/eStore/PartDetail.as ... roupID=218)
So the circumference would be 17teeth * 3mm/tooth = 51mm = 2.01" with compared with 1.44" (18 teeth x.08in/tooth = 1.44") for the MXL. This gives about 188 steps/in with a 400 step/rev motor (94 steps/in for a 200 step/rev motors that I am looking at :shock: ) vs the standard set-up with 278 steps/in

Now I guess you have to ask does this matter? We are talking .01" resolution with the 200step/rev motor and the 17tooth GT2 pulley vs .0036" for the standard set-up. This is 2.77 times worse, but I guess it is all about trade-offs. Do you want higher resolution or more torque capability? And there is nothing saying that the MXL pulley won't work, it is just not "rated" for the torque.

Also because of the larger pulley diameter you would get about 71% less applied torque to the system which means you would need about a 96 in*oz motor just to keep the same applied torque with the GT2 pulley!

To me, the choice is not that simple!
My buildlog is here

Digitalmagic
Posts: 123
Joined: Sat Jun 23, 2012 8:47 am

Re: Rated torque on pulleys?? Why is there not a problem he

Post by Digitalmagic » Thu Jul 26, 2012 8:44 am

Comparing MXL and GT2 obtainable linear resolution value needs to take in account accuracy and backslash properties of MXL vs GT2. If your "extra fine" MXL resolution value go below its accuracy/backslash characteristics, positioning is not following your equation.
MXL has a 0.02" positioning error, so you can't target 0.01" for MXL.

About the pulley/belt width worsening torque, you make a mistake.
Moment (rotational) to force (linear) transmission is independent of the transmission geometry.
The aim of choosing a belt technology, and its dimensional characteristics, is of course to convey this force, but first and foremost, it is accuracy, repeatability and anti-backslash properties.
a wider belt reduces the belt sectional traction and reduces the pulley/belt contact pression by increasing the surface.
So, I plan to use GT2 6mm width belt, but curious to experiment 9mm.

Finally, if you target a given accuracy, like your 0.01", in a non loaded, open-loop, theoretical estimation, the real outcome will be based on the weakest component on your system, and the bit against material to be cut will generate a counter force fighting against your pulley/belt system.
Here, dynamic factors are critical, not static and theoretical ones.

1 - Positioning (GT2 versus MXL)
GT2-Belt-2.JPG
GT2-Belt-2.JPG (53.03 KiB) Viewed 10113 times
2 - Belt tooth profiles & Belt working tension:
http://www.sdp-si.com/D265/PDF/D265T015.pdf
2mm & 3mm pitch GT2 allows more belt tension (fighting better against bit/material counter-force).
s/o #650 - Wish my digits are ... magic!

northbear
Posts: 254
Joined: Mon Apr 09, 2012 9:21 pm

Re: Rated torque on pulleys?? Why is there not a problem he

Post by northbear » Thu Jul 26, 2012 1:52 pm

Digitalmagic wrote:Comparing MXL and GT2 obtainable linear resolution value needs to take in account accuracy and backslash properties of MXL vs GT2. If your "extra fine" MXL resolution value go below its accuracy/backslash characteristics, positioning is not following your equation.
MXL has a 0.02" positioning error, so you can't target 0.01" for MXL.
[/i]
You are absolutely correct! I neglected the accuracy in the belt and as you pointed out the GT2 has a better positioning error of a bit over .001 vs .002 for the MXL. Just for fun, I put a static backlash picture from Misumi below

http://us.misumi-ec.com/pdf/fa/PDFViewe ... &page=1096
888.PNG
888.PNG (36.49 KiB) Viewed 10094 times
Digitalmagic wrote: About the pulley/belt width worsening torque, you make a mistake.
Moment (rotational) to force (linear) transmission is independent of the transmission geometry.
[/i]
I probably wasn't clear here! I did not mean to say the torque the was less or changed with with belt width. In fact I kept saying "applied torque" when I should have said "force". Torque = Force * distance. The distance in this case is the pitch radius of the pulley. Since Force = Torque/Distance with a larger distance (larger pulley radius) at the same torque the force will be less. Thus you will have less force to push the cutter through your material. Is this more clear?
Digitalmagic wrote: Finally, if you target a given accuracy, like your 0.01", in a non loaded, open-loop, theoretical estimation, the real outcome will be based on the weakest component on your system, and the bit against material to be cut will generate a counter force fighting against your pulley/belt system.
Here, dynamic factors are critical, not static and theoretical ones.
[/i]
Yes! This! Like you say the .01" is a theoretical resolution and not real world! All a bunch of jibber-jabber until put into practice!
My buildlog is here

Digitalmagic
Posts: 123
Joined: Sat Jun 23, 2012 8:47 am

Re: Rated torque on pulleys?? Why is there not a problem he

Post by Digitalmagic » Thu Jul 26, 2012 4:00 pm

Based on a given torque for a given stepper, and if you target the GT2 accuracy/strengh, you can:
- choose a less groovy ( :D ) GT2 pulley, to transfer a better force,
- use a 3mm pitch, with a wider belt, like 9mm (if assembly compatible).
To get a good compromise between force transfer and pulley to belt contact surface.

But at a certain point, to reach a good "bit against material" strength, you would need a torquey stepper anyway.
To increase linear resolution, you can also choose stepper finer than 1.8°/step(200/rev) or 0.9°/step (400/rev)
Of course, these 2 refinements at a cost, for a boosted shapeoko!
s/o #650 - Wish my digits are ... magic!

northbear
Posts: 254
Joined: Mon Apr 09, 2012 9:21 pm

Re: Rated torque on pulleys?? Why is there not a problem he

Post by northbear » Fri Jul 27, 2012 10:02 pm

Looks like the makerslide idler is slightly below the "recommended" pulled diameter. (.689 vs .75). Shouldn't make a difference.

http://www.sdp-si.com/D265/PDF/D265T056.pdf
999.PNG
999.PNG (61.2 KiB) Viewed 10027 times
My buildlog is here

Post Reply