[arvidj]

There are a number of threads that are using timing belts and pulleys in the implementation but little information about why the particular belt and pulleys were selected. Looking at belts and pulleys there seems to be a number of characteristics:

• Pitch - 2mm, 3mm, 5mm, 8mm, ... all the imperial pitches
• Tooth Profiles - tangential, round, various flavors of 'improved round', etc.
• Width - multiple choices
• Pulley size - the number of engaged teeth vs. the higher belt speed that is associated with "a larger pulley means more engaged teeth".
• Other items that my inexperience knows nothing about

I am wondering what process others have used to determine what belt and pulleys they would use.

What ever process you used --- be it "its what I had on hand" thru "this is what someone else used" and on to "put criteria X, Y, Z into spreadsheet ABC and got answer LMN" --- I and possibly others would like to know how the decision was made.

If a 'reference implementation' would help with the responses then: "A large RF45 table ... 1000mm by 260mm travel, 5 pitch lead screws, ~800 oz/in servos for X and Y and 4 pitch lead screw with a ~1100 oz/in servo for Z."

Any thoughts or real world experience would be appreciated.

Thanks,
Arvid

As a side note, please lets not get sidetracked with a "just use steppers" discussion.

[SCzEngrgGroup]

Go to sdp-si.com, and download their design belt drive guide, or order the free hardcopy version. You'll have more information than you'll ever need about the strengths, and weaknesses, of all the major timing belt families. Which type is "best" depends very much on the specific application - power, speed, backlash, cost, etc.. The perfect for one application may well be a poor choice for another application.

Regards,
Ray L.

[arvidj]

Ray,

Thanks for the quick response. I have visited their site and the term "overwhelmed" applies.

Based on the inputs I provided it seems like a 5mm pitch belt would need to be over ther maximim listed width of 15mm.

A narrower 8mm pitch belt might be appropriate but I'm not able to find software and documentation on their site related to the 8mm pitch belts.

But both of these seem to be "bigger than I expected them to be". During the calculaitons I provide known spefications like motor torque, RPM, ratios, etc. But then there is a "service factor" that is essentually a 'fudge factor' that they suggest should be between 1.5 and 2.0. As can be imagined this simply increase the values calculated values by 150 or 200%, which you then to look up the belts on the chart. I am wondering if my selection of 2.0 as the "service factor" may have been a little off the mark for a non-produciton level machine.

I'm still wondering about real world experiences anyone might want to share.

Arvid

[691175002]

In general the new curvilinear belt profiles (GT2, GT3) are universally superior to the trapezoidal profiles, although they cost more.

You can check the maximum power ratings for belts via their calculators, most of the time the 2mm and 3mm pitches are more than enough.

In most cases the rpm/horsepower ratings for belts are very conservative, and in a servo application where very little of the time will be spent at max speed/power you could probably use a much smaller belt if you so desired.

The only real pitfall is avoiding profiles with backlash. HTD along with most of the trapezoidal profiles (except AT) may have backlash.

[arvidj]

In general the new curvilinear belt profiles (GT2, GT3) are universally superior to the trapezoidal profiles, although they cost more.

You can check the maximum power ratings for belts via their calculators, most of the time the 2mm and 3mm pitches are more than enough.

In most cases the rpm/horsepower ratings for belts are very conservative, and in a servo application where very little of the time will be spent at max speed/power you could probably use a much smaller belt if you so desired.

The only real pitfall is avoiding profiles with backlash. HTD along with most of the trapezoidal profiles (except AT) may have backlash.

Thanks for the suggestions. You are correct that my calculations were based on absolute worst case conditions with the maximum if-lightning-strikes horsepower values.

And thanks for the low backlash tooth profile suggestion. Does pitch have any effect on backlash? It would seem that for a given diameter pulley a smaller pitch would have more teeth engaged which would seem to be a plus for reducing backlash, but that is just an uninformed guess.