• Spring 2024 meetup in Calgary - date Saturday, April 20/2024. discussion Please RSVP Here to confirm and get your invitation and the location details. RSVP NOW so organizers can plan to get sufficient food etc. It's Tomorrow Saturday! you can still RSVP until I stop checking my phone tomorrow More info and agenda
  • We are having email/registration problems again. Diagnosis is underway. New users sorry if you are having trouble getting registered. We are exploring different options to get registered. Contact the forum via another member or on facebook if you're stuck. Update -> we think it is fixed. Let us know if not.
  • Spring meet up in Ontario, April 6/2024. NEW LOCATION See Post #31 Discussion AND THE NEW LOCATION

3D printed timing pulley upgrade

kevin.decelles

Jack of all trades -- Master of none
Premium Member
The CNC mill (converted manual mill) I acquired from @Janger (who previously acquired it from @Alexander) used HTD 8M timing belts on a variety of timing belt pulleys (some from cars I believe, some home-made). All the pulleys were mounted using set-screws on 'flats' on the lead-screw or stepper-motor shafts. While this works, it is prone to slippage due to the torque/weight of the tables. The set screws simply back-off and you have a pulley skip (which scores the shaft) which leads to a CNC crash. The X axis had a belt tensioner, the y-axis did not.

My goal was to mount the pulleys more securely, ideally using the 3mm key on the leadscrew, and using two-flats on the D-shaft on the stepper. I decided to start with the Y-Axis as it was the one I was having issues with, and the one without the tensioner. After doing some research on HTD 8m timing belt pulleys, I was very surprised as the cost. Not cheap, and not easy to locate a pulley with known number of teeth etc. The other issue was trying to find one with the right bore. I gave up on purchasing a pulley, and went next down the lost-PLA casting route. I had some decent results, but ultimately lost-PLA is a timely and costly process (propane/electricity/time)-- with a product that didn't have the precision/quality I wanted.

I took a further step back by reviewing the pros/cons of the Z-Axis 3D pulleys that @Janger had built. The design is logical, but the weak point was once again the set-screw contact to the shaft. With any undue torque/resistance, the 3D part cracks at the set screw hole. After staring at this for many hours, the answer I believe is to increase the contact area from a single point (set-screw) to as much surface as possible. I decided to built a steel 'collar' that fit on the shaft, and use a series of key-ways to mount a PETG or ABS 3D printed pulley.

For the Y-axis I needed 2 collars, one for the 5/8 shaft of the y-axis lead-screen on the mill, and the other 14mm for the stepper. The lead-screw collar also needed to include a recess for a thrust bearing , an internal keyway in addition to the 4 external key slots. The stepper collar was more simple, a 14mm bore w/the 4 key-way slots.

I don't have pictures of the lead-screw collar before I mounted, but here are pics of the stepper side plus pics of the final product (with belt tensionor)

IMG_4701.JPG IMG_4702.JPG IMG_4704.JPG

Picture of the pulley (black PETG w/ 1/4" keyways
IMG_4706.JPG
IMG_4705.JPG

Picture of stepper collar/pulley mounted w/key-ways full

IMG_4707.JPG

Picture of the other side and the finished product

IMG_4708.JPG IMG_4709.JPG IMG_4710.JPG


So for results? Awesome! I was skeptical that the PETG would grip the belt as well as it did. I figured it would slip or stutter. The 3D models I found are perfect matches for the belt profile, and the grip is great. Before I did this mod, the best I could get out of the Y-axis reliably was about 10 in/min jog speed. With this mod I'm up to 16.2 in/min. This mill has the usual table, but also has a sub table that allows the top table to be turned. This adds weight. Add to that the Kurt DX-6 vise at ~75lbs and there is considerable mass.

overall, very pleased so far, and I have a great consumable part that can be replace at will. The 4 key-way idea came to me as I was wondering how I would best machine a collar to fit the existing pulleys. By using the 4 key-ways, the pressure/torque is applied across 30-40mm of 1/4 key x 4. I know the PETG can withstand pressure, as it is tough to crush in the 1-ton arbor press. In this configuration I think it should last quite a while.

The belt tensioner was another piece of steel from a John-deere discer I salvaged. Great stell to work with. The collars were made from 2" diameter 12L14 which I very nice to machine and will be good for this application (low impact/hardness required.)

The other plus with the shaft collar is that I can experiment with different size pullies, simply keep the internal sizing the same. The original setup used a 15T pulley on the stepper, I couldn't find an STL for 15T so went up to 18T. I also then went up a similar ratio on the lead-screw to keep the ratio similar to before.

As for lost-PLA, I may go back to it, but first I'm going to convert the other two axis to a collar system.

Summary of machining techniques used:

Lathe: Turning, boring, drilling, facing
Mill (w/dividing head) slot cutting
Shaper Internal 3mm keyway on the lead-screw side.
Vertical Bandsaw: cutting of key-stock
Arbor press: pressing on bearings for tensioner

Supplies:

2 pieces 12L14 75mm in length
set screws (x4)
3mm key (x1)
1/4 key stock (x8 pieces)
Replacement thrust bearing
1/2 Bolt (tensioner)
5/16 bolts (tensioner)
HTD 8m timing belt (www.vbeltguys.com -- great site, cheap belts) 568mm x 12mm
PETG (I use/like the product from Spool3d.com)
3D printer - stock Creality Ender Pro 3. I bought this a month back and it is printing these pulleys stock out of the box.
 

Attachments

  • IMG_4703.JPG
    IMG_4703.JPG
    85.1 KB · Views: 0

Alexander

Ultra Member
Administrator
The car parts are from a balance shaft off a 2002 mitsubishi galant. A Volkswagen jetta also has the same pitch on its timing belt but it is wider. I did make some of the pulleys on a dividing head. But my only dividing head at the time was also home made. 1F533F5B-6070-47B3-B94C-BC290F892A42.jpeg
 

kevin.decelles

Jack of all trades -- Master of none
Premium Member
That explains the continental belt -- the cross-reference guide listed Mitsubishi Galant. I like the HTD 8m profile -- heavier than typical (5M/3M) but good for that slab of steel.

Hopefully the PETG holds up -- no reason it shouldn't.
 

Janger

(John)
Administrator
Vendor
Awesome upgrade Kevin. Really like what you have done. What happens now when you crash the axis or table into something immovable? Does the motor just stall? e.g. on the Z axis if I had the spindle locked then running the z axis motor would break a pulley at the set screw point like you explained. I've done this on the haas (oops :eek:) and the control errors out and stops the machine. After I broke the drill bit.
 

Alexander

Ultra Member
Administrator
My milltronics overtraveled last winter and it broke the timing belt. That was cheap thankfully.
 

kevin.decelles

Jack of all trades -- Master of none
Premium Member
Awesome upgrade Kevin. Really like what you have done. What happens now when you crash the axis or table into something immovable? Does the motor just stall? e.g. on the Z axis if I had the spindle locked then running the z axis motor would break a pulley at the set screw point like you explained. I've done this on the haas (oops :eek:) and the control errors out and stops the machine. After I broke the drill bit.

Well, thats a tough one, not like I want to test it out............ We'll have to wait and see, a crash is inevitable
 
Top