Rhino track failier?

[Brian C]

Originally Posted by aJASONt_angles

Hey All,

My name is Jason and I'm an AndyMark Intern this year and mentor on Team 4272. I've been doing testing on the Rhino Track drive pulleys and their durability.

We put on all new pulleys and bearings on our test mule after discovering the previous ones were broken. We tightened the track belt on one side as much as we were comfortable with and left the other side a bit looser. The pulleys were assembled with 3 washers under each screw to mimic the thread engagement of the new mold.

Here are the tests we performed in the order we performed them. In between tests we drove the track drive around on carpet and concrete. The test mule was weighted down to 120 lbs.

30 Rockwall crossings, full speed on concrete.
30 Rockwall crossings with fully tightened front bolts, full speed on concrete.
Dropped from 2.5 feet onto concrete.
10 Rockwall crossings, full speed on concrete.
Hammer to the side of front bolt 3 times each side.
Dropped from 2.5 feet onto concrete twice.
Ran into metal bar full speed five times.
5 axial hits with weighted cart.
5 front end drops from 2.5 feet.
Crashing into various things in the warehouse.

Throughout the testing we stopped between tests to look at the pulleys. The pulleys showed no signs of fracture externally on either side of the track drive.

Here is a video of our axial hit with the weighted cart.

More testing is being done today and I will update those of you on Chief Delphi as soon as I have more results.

- Jason

Jason,

Thanks for the updated information. From your description it sounds as if you folks are driving your test mule much like our team drivers.

Of course we encourage them to "Drive it like you stole it" and that a robot can always be repaired BUT a match cannot be replayed.

From the looks of some of the frame parts in the video you guys have taken the "crashing into various things" portion to heart!

[Brandon Holley]

For Andy and co. -

PC/ABS may be a good alternative option to try as well. Shrink should be very similar to PC that is already being utilized so you can more than likely drop it right into your current tool. It will decrease the notch sensitivity of the material relative to impact failures (ie: making the part less sensitive to the microscopic flaws in the molded material).

Just an idea for the pile!

-Brando

[Kevin Sevcik]

Originally Posted by dradel

With all the various responses my brain has turned to jelly !! So are we saying new pulleys are on tap, or are we saying put washers between the screw heads and the pulleys? Sorry I am so lost here but has been a stressful build season for us this far with snow days and other issues

Washers are a test of a potential modification of the mold that would result in less thread engagement of the screws. It's purely to determine if pulleys from the new mold would require longer screws.

[dradel]

Ok now I'm caught up. Like I said been a bit stressful with the snow

[Michael Corsetto]

Just a thought.

If the tight fit between the bearing and the plastic hub is causing issues, could teams take a reamer to their existing plastic hubs and change a tight fit to a slip fit? Doesn't solve all the impact issues, but could reduce some of the stress in the hub.

Kudos to the AM crew for being on top of this. FRC teams are lucky to have so many dedicated suppliers!

-Mike

[Why_A_Username?]

Originally Posted by Michael Corsetto

Just a thought.

If the tight fit between the bearing and the plastic hub is causing issues, could teams take a reamer to their existing plastic hubs and change a tight fit to a slip fit? Doesn't solve all the impact issues, but could reduce some of the stress in the hub.

Kudos to the AM crew for being on top of this. FRC teams are lucky to have so many dedicated suppliers!

-Mike

You probably could ream the hole out, but I wouldn't risk causing damage to any presently undamaged parts. The reamer could put unnecessary stress onto the hole and possibly fracture it. Depending on 3D printer access I would print a new part with a stronger material and a slightly adjusted hole diameter. Just my advice.

[Kevin Sevcik]

Originally Posted by Why_A_Username?

You probably could ream the hole out, but I wouldn't risk causing damage to any presently undamaged parts. The reamer could put unnecessary stress onto the hole and possibly fracture it. Depending on 3D printer access I would print a new part with a stronger material and a slightly adjusted hole diameter. Just my advice.

You're not 3d printing a stronger part out of anything short of sintered metal. So that's not really an option.

[team222badbrad]

Originally Posted by Kevin Sevcik

You're not 3d printing a stronger part out of anything short of sintered metal. So that's not really an option.

It's possible... :-)

https://markforged.com/

[Kevin Sevcik]

Originally Posted by team222badbrad

It's possible... :-)

https://markforged.com/

I'm clearly not keeping up with the times if we're up to printing FRP. Still, their claim is the material is stronger than aluminum PER kg. A better strength-to-weight ratio is often a good thing, but if your limiting factor is volume, not weight, then it's not the important metric. In this case, the pulley geometry is already defined, so the important metric is strength per unit volume, which I suspect aluminum or sintered metal would better at.

[Chris is me]

Originally Posted by Why_A_Username?

You probably could ream the hole out, but I wouldn't risk causing damage to any presently undamaged parts. The reamer could put unnecessary stress onto the hole and possibly fracture it. Depending on 3D printer access I would print a new part with a stronger material and a slightly adjusted hole diameter. Just my advice.

No offense, but do you know what reaming is?

Reaming is an operation using a special tool used to produce a hole diameter of a very accurate size. Reaming is done by removing a few thousandths of an inch of material from a hole using a spinning cutter of a precise size. Very little stress is imparted onto the part to do this; certainly less stress than drilling a hole, and it's a reasonable suggestion here.

3D printed parts are more or less always substantially weaker than injection molded parts of the same geometry and material. There are very few material choices for 3D printers that are potentially stronger than injection molded polycarbonate, and it is reasonable to assume none of these materials / printers are within the reach of FRC teams.

And here's the kicker - in order to get accurate bore diameters using 3D printed parts, you would probably want to team the hole afterwards anyway.

[Andy Baker]

Originally Posted by Michael Corsetto

Just a thought.

If the tight fit between the bearing and the plastic hub is causing issues, could teams take a reamer to their existing plastic hubs and change a tight fit to a slip fit? Doesn't solve all the impact issues, but could reduce some of the stress in the hub.

Kudos to the AM crew for being on top of this. FRC teams are lucky to have so many dedicated suppliers!

-Mike

This could be reamed out, of course, but I really don't think that a too tight bore is the problem. We can press bearings into the current pulleys we have (existing design, Rev3) with a strong push of a thumb. It's not a tight press fit.

Here is an update with our current status:

1. So far, 5 teams have contacted us regarding cracks or fractures on their pulleys. We are sending emails to those 5 teams today. Oddly enough, all of these orders were shipped out on Jan. 10th or 11th.

2. We are continuing with an improved mold for our Rev4 pulley. It has all of the improvements listed above, AND we were able to change the interior (non-bearing) bore to 0.754. This bore would be able to accept a bushing if the user would like to install one. We don't think it will be needed with the other improvements.

3. We will send out improved pulleys to the 5 teams who have sent us input regarding cracks and fractures.

4. As you can see from Jason's post above, we are quite confused about how we can't break these pulleys like we broke our original test mule. We are still testing. The current testing shows that our existing design is very robust for climbing over things and for surviving drops on concrete floor from a considerable height (2 feet).

5. If more teams are seeing cracks and fractures, please send a note to sales@andymark.com, telling us which pulleys are breaking (1, 2, 3, or 4), your Rhino Track Drive order number, and with pictures of the breaks. We will replace these pulleys with pulleys from the new mold.

6. We are hesitant to send out new pulleys to all customers due to the positive results we are seeing in #4 above. If customers want to contact us to get replacement pulleys, please do so by sending an email to sales@andymark.com. You will need to tell us your AndyMark order number (this saves us time) and tell us the number of pulleys requested.

Andy B.

[ChuckDickerson]

Andy,

Do you have an estimate of when the new improved pulleys will be available and ready for shipment?

-Chuck

[BoilerMentor]

Originally Posted by Andy Baker

1. So far, 5 teams have contacted us regarding cracks or fractures on their pulleys. We are sending emails to those 5 teams today. Oddly enough, all of these orders were shipped out on Jan. 10th or 11th.

Andy B.

Andy,

Our pulleys would have gone out in that range of dates. We have yet to observe a failure. We've been checking them at fairly regular intervals.

I believe the reason we have yet to see a failure is related to our dead shaft design. We use a .375" OD tube between drive plates with a 1/4" bolt though the center, so the axle acts as a standoff. If spacers are fabricated properly it makes it impossible to put a clamping load on the inner race of the bearings in the pulley by tightening the bolt.

That's my two cents. I'll report back if we do see a failure.

This situation is an example of the world class customer service we've come to expect from Andymark.

Thank you.

[Jimmy Nichols]

Originally Posted by Kevin Sevcik

I direct you to the Brecoflex page on their spliced belts. They say only about half the tension members carry the load through the spliced area.

I haven't looked but the belts could be extruded.

[Why_A_Username?]

Originally Posted by Chris is me

No offense, but do you know what reaming is?

Reaming is an operation using a special tool used to produce a hole diameter of a very accurate size. Reaming is done by removing a few thousandths of an inch of material from a hole using a spinning cutter of a precise size. Very little stress is imparted onto the part to do this; certainly less stress than drilling a hole, and it's a reasonable suggestion here.

3D printed parts are more or less always substantially weaker than injection molded parts of the same geometry and material. There are very few material choices for 3D printers that are potentially stronger than injection molded polycarbonate, and it is reasonable to assume none of these materials / printers are within the reach of FRC teams.

And here's the kicker - in order to get accurate bore diameters using 3D printed parts, you would probably want to team the hole afterwards anyway.

I know what reaming is thank you but I appreciate your concern, but injection molds can be structurally compromised from even small modifications. My team has modified the CADD file of the part, combining both halves into a solid part and filling in the empty space around the bearing socket. I will gladly share this part if it works after our testing.

-Jacob