501 needs your help – P60k 256:1 Banebots Transmission

[JGecko146]

Team 501, the PowerKnights of Manchester, NH, are in desperate need of a Banebots P60k 256:1 transmission. We had one and it suffered catastrophic failure. We’ve bagged and tagged and are desperately looking for the transmission so we can replace the old one at the BAE Regional. Banebots is backordered past the date of the BAE. Does your team have one we could purchase from you or do you know of any team that does? We will gladly pay for the transmission and expedited shipping. Any help you can provide would be greatly appreciated.

Thanks,
Jen Hosker
Team 501 Mentor

[Swampdude]

FYI: I sent you a private message

[sanddrag]

On a side note, would you be willing to post the details and pictures of the failure?

[fox46]

Lets take some bets here- its the output shaft/stage rounding off or the ring gear torquing free?

I absolutely hate these things..... We are trying to re-engineer ours to keep them from failing. I don't understand the mentality of marketing and selling a gearbox that can't handle the power produced by the motor is it sold for.

[Teammax]

I really have nothing to say that would help your situation. But, it amazes me that Banebots is still so far behind on filling orders. This really needs to be looked at before next season

[MrForbes]

I'm not amazed by them being that far behind.

I'm also not amazed that they would sell a gearbox that can't take the abuse the motor it's designed for can dish out.

I'm also not amazed that teams design robots that use a gearbox that can't take the abuse the motor it's designed for can dish out, in such a way that it's likely to fail.

This is all normal stuff.

[thefro526]

Can you define the "Catastrophic Failure"?

We've had three gearboxes fail, and all but one of them was easily fixed with a New Carrier Plate and Output shaft, both of which can be purchased from Banebots, and they should ship within three days, so if you selected overnight shipping you could have them by next Monday or Tuesday.

The gearbox that we haven't fixed yet failed a little bit more spectacularly than the other two. The Ring Gear/Case started to spin which ruined the roll pins/spring pins and the holes they sit in. I believe Swampdude posted a fix for this somewhere that involves drilling new spring pin holes and milling the ring gear slightly.

[fox46]

It's not that teams have done a poor engineering job or are overrating their components. BB clearly states that the gearboxes are not designed to take more than 35 lb-ft of torque. The mechanism on our robot is neutrally biased by a gas shock- you can lift the arm with one finger however, the gearboxes are just not robust. They work for a bit and then parts start wearing, the aluminum galls and things come apart.

Their gearboxes, especially those 128:1 and greater produce a TON of torque. The only think keeping the ring gear from spinning free are four 3/32" roll pins pressed into aluminum, loaded in bending against two small flats! its like trying to tighten a 12 sided aluminum bolt with a crecent wrench. This could have easily been avoided by keying the faces between the end plates and ring gear.

The output shaft although 1/2", transmits torque through a 3/8" diameter shoulder with two flats machined on it. The stress concentration is so high at the edges of these flats that it distorts the final reduction carrier that it fits into and rounds it out. I don't understand why they wouldnt machine the final reduction carrier and shaft as one and avoid this problem all together.

[MrForbes]

Originally Posted by fox46

The mechanism on our robot is neutrally biased by a gas shock- you can lift the arm with one finger

How much inertia does the arm have? how much load is on the transmission when you change arm travel direction?

Before using a specific gearbox, should the team take one apart and see if the design is sufficient for what they're going to do with it? especially if they read that the torque rating of the gearbox is less than what it can generate with the motor they're using?

It's a tough situation, and a good way to show how things work in real life, unfortunately.

[fox46]

How much inertia does the arm have? how much load is on the transmission when you change arm travel direction?

This is exactly why we are having issues but that comes back to the marketing of a gearbox for a motor which it can't handle. The load on the motor is almost nothing but the speed at which they start and stop is what is wearing them out. We have the programmers working on a ramp function / soft start function for them which should help.

This is still a highschool competition and these machines are being built by highschool students, not engineers.... or at least thats how it should be. When we got our gearboxes, they were given to a couple of students who were told to disassemble these, figure out how they work, then reassemble them with grease. Of course they aren't going to be able to identifty the weaknesses in a design- they are just beginning to learn this stuff. I just would have preferred they learn about it in a different way in this case.

[AdamHeard]

Originally Posted by fox46

This is exactly why we are having issues but that comes back to the marketing of a gearbox for a motor which it can't handle. The load on the motor is almost nothing but the speed at which they start and stop is what is wearing them out. We have the programmers working on a ramp function / soft start function for them which should help.

This is still a highschool competition and these machines are being built by highschool students, not engineers.... or at least thats how it should be. When we got our gearboxes, they were given to a couple of students who were told to disassemble these, figure out how they work, then reassemble them with grease. Of course they aren't going to be able to identifty the weaknesses in a design- they are just beginning to learn this stuff. I just would have preferred they learn about it in a different way in this case.

Banebots posts maximum torque ratings for the gearbox on the page they sell them on, and based on those ratings the 256:1 is a poor choice for anything that will see stall loads.

I would stick to the 64:1 or less (4:1 for us), and make/get a different style gearbox for the comp bot, even if you take measures to alleviate the load, the potential for failure with the 256:1 at full motor stall is always there, and could totally wreck your season.

[MrForbes]

Originally Posted by fox46

This is still a highschool competition and these machines are being built by highschool students, not engineers.... or at least thats how it should be.

That's debatable. If you have an engineering mentor who understands how to design this stuff, that mentor should be helping the students figure out how to make a workable design--that includes making sure the parts are strong enough for the application.

I realize that many (most?) teams don't have a mechanical engineer helping them, the result is that we kind of need to have indestructible parts available. That's a tall order.

[sanddrag]

For those of you with failing BaneBots planetaries, you may want to look into Dewalt Drill gearboxes, and/or a hypocycloidal lug nut torque multiplier. Ours has been holding out so far, but it hasn't seen a lot of heavy use. We have the 256:1 with a steel ring gear. In our design, it should see no more than about 15 lb-ft, and it is well supported on the end. I hope it lives, but I'm prepared to deal with it if it doesn't.

[Isaac501]

Originally Posted by fox46

Lets take some bets here- its the output shaft/stage rounding off or the ring gear torquing free?

I absolutely hate these things..... We are trying to re-engineer ours to keep them from failing. I don't understand the mentality of marketing and selling a gearbox that can't handle the power produced by the motor is it sold for.

First both of the output bearings failed; We had to re-work the output mount with new bearings. Then the shaft failed (rounding out) and then the new shaft we machined failed (rounding out).

The gearbox can handle the power and torque the motor puts out.

I get the sneaking feeling we've exceeded the 35ft-lb rating by .. quite a bit on backdrive.

Then again, I don't know.. I'm a semiconductor engineer. What's this torque stuff?

We have the facilities to harden the steel.. I am wondering if it might be a good idea or if it will just put the point of failure further down the line.

[fox46]

The problem with hardening the steel is that it will get brittle and you will risk snapping the shafts before rounding them out. My plan so far consists of remaking the end plate with larger 1/2" bearings spaced further apart to tolerate more side load (we havent had any bearings fail yet though) and then machining a 1/2" ouput shaft with the carrier on the end as one intergral piece. Pop some hardened dowels into the new shaft/carrier and replace the final carrier and shaft as one. The ring gear I plan to run an end mill accross it and machine the new end plate with a face key so that when assembled with high grade 6-32 socket head cap screws it will key together. I toyed with the idea of TIGing everything together- shaft to carrier, ring gear to face plate, but I am doubtful that I can pull it off without distorting the ring gear and having it bind.