VersaPlanetary Failures

[rahilm]

Hi all,

Looking for some advice if anyone has seen failures like this before or has insight into the root cause, so we can prevent it from happening again. I've attached some pictures, but basically what we're seeing is the pins the planet gears rotate on begin to push out, eventually rubbing against the casing and causing large amounts of friction and seizing of the gearbox.

Pictures here:
http://imgur.com/a/hG4J0

Thanks in advance.

[Bochek]

We've run into that exact problem before. I can't blame Vex Pro in our situation because we where using the gearboxes far beyond their specified output torque limitations. (bag motor at 400:1 reduction)

We managed to solve the problem by firing up the tig welder and welding the pins in from the back side.

- Bochek

[Peyton Yeung]

We had this issue last year and we just pressed the pins back in. It was on our spare winch so it saw little use.

[s_forbes]

Originally Posted by rahilm

Hi all,

Looking for some advice if anyone has seen failures like this before or has insight into the root cause, so we can prevent it from happening again. I've attached some pictures, but basically what we're seeing is the pins the planet gears rotate on begin to push out, eventually rubbing against the casing and causing large amounts of friction and seizing of the gearbox.

Pictures here:
http://imgur.com/a/hG4J0

Thanks in advance.

Do you have a description or picture of the mechanism that these failed in?

We haven't run into this issue in any of our Versaplanetaries, but it might be because we use them in light applications for fear of overloading them.

[rahilm]

Originally Posted by s_forbes

Do you have a description or picture of the mechanism that these failed in?

We haven't run into this issue in any of our Versaplanetaries, but it might be because we use them in light applications for fear of overloading them.

Whoops, looks like I accidentally deleted that info from my original post. I don't have a picture at the moment, but it's for a flywheel style shooter. We're using a single stage 3:1 drivenby a 775Pro. This feeds into a 1:1.2 gearbox, for an overall reduction of 2:5:1 driving a 4" wheel. We see about 10A of current draw at 7k RPM at the flywheel, which is right in line with what we'd expect from the spec sheet, suggesting we don't have any sources of unexpected load.

It seems like it may be a good idea to press the pins back in and add some red Loctite. No access to a welder, unfortunately, so that's not going to be a possible solution for us.

[Andy A.]

Originally Posted by rahilm

we see about 10A of current draw at 7k RPM at the flywheel, which is right in line with what we'd expect from the spec sheet, suggesting we don't have any sources of unexpected load.

Presumably there is a large load spike when a ball is fed into the shooter. The motor/gearbox/flywheel go from very high speed/low load to (still) very high speed/high load as they start to accelerate the ball up to speed. That is probably causing those pins to work loose. Dynamic loads are tough!

This sort of wear is probably going to crop up for a lot of teams. You might consider some sort of cushioned coupling between the transmission and flywheel. The type I'm thinking of are usually called 'lovejoy couplers', and consist of two metal hubs with dogs that attach to your shafts, and a rubbery bit that fits between the dogs. The rubber will soak up some shock and allow for a bit of shaft misalignment.

Take a look at 'lovejoy couplings' on mcmaster, and see if you can find one that can fit between the fly wheel and gearbox somewhere. Make sure to pay attention to maximum RPM rating, though it appears most of the aluminum parts will work just fine.

[tuyauxtu]

Yup, pins moved out and chewed into the VP plastic encoder housing. We pushed them back in; so far so good.

[Foster]

This is a known problem with VersaPlanetary and shock loads (thread from 2014)

[JamesCH95]

Did you grease the transmissions prior to assembly and use? Those gear-sets look far too dry...

[MaGiC_PiKaChU]

Originally Posted by tuyauxtu

Yup, pins moved out and chewed into the VP plastic encoder housing. We pushed them back in; so far so good.

push them back and use a punch with a hammer so the pins expand a bit... Worked for us when we used VPs on our drive trains in the past years

[Tom Line]

I agree with the comment that there seems to be a lack of grease in that gearbox. However, that would not fix the problem with the pins - we have seen it as well. Like yours, ours did not come all the way out. Our solution was to use some permatex red loctite on the underside of the plate and push the pins back in.

[RyanShoff]

It has been suggested to tack weld the pins in a TIG torch. Then file back smooth. The darker colored carrier plates become grease impregnated, so you end up burning that out.

We've done this to swerve steering gearboxes that are constantly reversing direction.

[Joe Johnson]

We initially used a VP for our shooter wheels with a 3:1 single stage and a Vex 775pro input. I wasn't there but the team reports that the gearbox didn't seem happy running at those high speeds and we had higher current draw and lower speeds than we wanted, so we switched to a CIM-ile and live seems to be much much better. Note we also did a few other things that helped the shooter draw less current (mainly taped over the spokes of the wheels so that we weren't losing as much energy moving air around) but that is our experience.

YMMV.

Dr. Joe J.

[rahilm]

Thanks for all the suggestions, everyone. Much appreciated.

The gearboxes could definitely use some more grease, so we'll make sure that's taken care of when putting these back together.

It looks like the plan of action will be: clean out housings, apply red loctite and push pins back in, reassemble gearbox, stress test during today's practice session. We don't have access to a welder, but if we continue to see this, we'll have to try to find one.

[JamesTerm]

Originally Posted by Andy A.

The motor/gearbox/flywheel go from very high speed/low load to (still) very high speed/high load as they start to accelerate the ball up to speed.

Couldn't this type of problem be avoided in software via PID/Encoder where it maintains the same speed? (Assuming of course the shooter is optimized mechanically to allow desired steady rate to be lower voltage say 65%).