Take the "Bane" out of your Banebot P60s - Solution to Banebot P60 Weaknesses

[Ether]

Originally Posted by fox46

We are using the 326:1 P60s, to articulate our arm, being driven by an RS395

Here's another way to take the Bane out, if you have room to do it:

Instead of 326:1, use a 64:1 with 5:1 external gearing (or sprockets or pulleys or winch etc). This will reduce the torque load on the tiny P60 by a factor of 5.

[fox46]

We actually already have an external 5.4:1 reduction. The P60 drives a 10 tooth sprocket which turns a 54 tooth sprocket connected to the fulcrum of the arm. If we experience any further problems though, plan "C" is to put a stud in the side of the arm to support an intermediate 10t and 30t sprocket to provide a second external reduction and then machine the ring gears down to a 3 stages and use them to power the intermediate reduction.

[Ether]

pseudo-code for rate-limiting the command to the motors:

if (newCmd>oldCmd+rateLimit) newCmd = oldCmd+rateLimit;

else

if (newCmd<oldCmd-rateLimit) newCmd = oldCmd-rateLimit;

oldCmd = newCmd;

[Chris is me]

Originally Posted by fox46

Our arm is biased with gas shocks to the point that you can release it from full height with the drive chains removed and it will "float" back down to rest position and we still were seeing signs of failure.

This isn't balanced. A balanced arm should remain stationary at maximum extension (i.e. horizontal)

[fox46]

Correct-

biased with gas shocks

I never said it was balanced. The shocks only relieve some of the load, they do not remove it.

Ether: We may be thinking the same thing but I am just having a little bit of trouble visualizing what this pseudo code will do. Are newCmd and oldCmd encoder values or motor commands? (bear in mind I have very limited programming knowlege). I was thinking some sort of function like this:

accelerationscaling = |originvalue - destinationvalue| + (0.1 * max speed)
decelerationscaling = |destinationvalue - originvalue| + (0.1 * max speed)

"Where originvalue and destinationvalue are the position of the arm as determined by the encoders. This tells the function how close it is to the target position (affects deceleration of the mechanism) and how far away from its origin position (affects acceleration). The addition of the 10% factor is to eliminate the asymptote in the speed of the arm near its target and origin positions. Otherwise it would be infinitely slow starting and stopping."

if (accelerationscaling > decelerationscaling) scalingfactor = decelerationscaling

else

if (accelerationscaling < decelerationscaling) scalingfactor = accelerationscaling

"This basically makes the function look at the decelerationscaling factor when it is time to start throttling down the motor as it approaches its destination. Or it makes it look at accelerationfactor when it is time to throttle the motor up as it moves away from the original position."

Motor speed value = Max speed value in direction of target * scalingfactor

"This part of the function adjusts the speed of the motor relative to how close it is getting to its destination position or how far away it is from its original position."

What I am trying to accomplish would be to start the motors off slowly, throttle them up to full speed as the arm gets moving, and then as it approaches its destination point, to throttle them back down. Basically a soft start and soft stop function. Is this the same sort of thing your pseudocode does?

[Ether]

Originally Posted by fox46

What I am trying to accomplish would be to start the motors off slowly, throttle them up to full speed as the arm gets moving, and then as it approaches its destination point, to throttle them back down.

Sounds like you want a closed-loop controller with a limit on the max allowed rate-of-change of the output.

[Jonathan Norris]

I see what you are trying to accomplish here but at this point I would not recommend any team trying to drive a long arm off a banebots gear box. The gear boxes simply can't handle the shock loads of an arm, even when dampened by a gas spring. You've definitely solved some of the issues with the planetaries, but there are other issues I would be concerned with:

- with that high of a planetary reduction, the loads on last couple stages are pretty significant, to the point where I don't think those mild steel planetary gears will last very long.
- I would also be very concerned with stage plates in the last couple stages, the pins that the planetary gears rotate on are only press fit into the mild steel and have been known to come loose and destroy the gearboxes.

I would highly recommend teams looking into using the Fisher Price Gearboxes included in the kit, they make for a robust arm gearbox. They are a pain to mount, but the Fisher Price is still a beast of a motor and should handle the loads well as long as the arm is dampened.

If that's not an option I would recommend teams looking into cobbling together a AM gearboxes with a cimulator and sprocket reductions to get the desired reduction on a bane bots motor.

[AdamHeard]

Originally Posted by Kevin Sevcik

Adam,

You're probably right that this will shift the failure point of the transmission elsewhere. If it's the ring gear, there's a steel version of that. I think you're wrong that the issue is stalling the motor, though. If you ever actually manage to stall one of these things, you're going the seriously damage the gearbox very quickly. We're experiencing a failure mode very similar to fox46's in that the gearbox gradually degrades over time. I think this is a combination of the weaknesses he's helpfully pointed out and some unique load conditions. We've never stalled our 256:1 RS-775 transmission, and yet we've deformed the carrier plate twice. I'm certain this is due simply to inertial loading from our arm as we lower the arm and stop it. The stop is invariably sharp and sudden thanks to the huge gear ratio, and this creates an instantaneous load on the gearbox that exceeds the capacity of the carrier plate. I expect fox46 is actually experiencing a similar problem.

His proposed solutions will definitely strengthen the gearbox and slow the degradation process, but I think they should be coupled with programming changes designed to limit the rate of change of voltage so as to lessen the sudden shocks when motion is stopped.

You're not sitting there stalling it, but when you first get the motor going, or rapidly change direction, the motor is stalling momentarily and putting out much more torque. As you already know, this probably isn't near full stall torque of the motor at 12V, but multiplied by the 256:1 it's still something rough.

Such high reduction p60s really shouldn't be used for arms, even with proper counterbalancing and software limits they are just waiting to tear themselves apart.

We've used our 4:1 p60s in our arm rough and hard, with aggressive cyclic direction changes and even full stall on a few occasions for a few seconds without damage.

Banebots isn't at fault here, they designed a gearbox that is mechanically nice, and very, very affordable. For people who want higher reduction but are aware the gearbox could damage itself from higher loads or cyclic loads, they offer a higher reduction. User error is not the manufacturer's fault.

[billbo911]

Originally Posted by Ether

Sounds like you want a proportional controller with rate limit on the output.

A limitation of the amount of change that can occur between loops is really all that is needed, similar to what Ether suggested. It will make calibrating a PID a bit difficult, but not impossible.

If you are building a proportional only controller, then adjusting the gain for a slower approach to the target will be simple. Then, use the torque limiting function, like Ether posted, to smoothly start the motion. Combined, these two should get you close to what you want.

[fox46]

Not quite actually, it is the 326:1 gearbox that drives the 10 tooth sprocket- we are indeed using the 326:1 P60. From the motor to to the fulcrum of the arm, overall we have a 1630:1 reduction.

I am very confident that these modifications will last. As it is, we have quite alot of practice with our robot- I would say it has had more run time than it will in the regional. We have NOT failed one of these transmissions yet. We only began noticing some backlash in the shaft and one of the ring gears turning back and forth against its pins. I have total confidence in the arm once these modifications have been added. Even now I would run matches as it sits in its crate. In my team's case, this mod is insurance against what we have been seeing from other teams. Trust me- our P60s have a very easy life compared to some of the other applications I have seen.

My analogy and opinion of the situation is this: Banebots has built a car. It bears resemblence to a Geo Metro- Cheap, compact, practical, simple etc etc.. But they are also marketing it to be robust and powerful. What they have in fact done is built a Geo Metro but swapped the engine to that of a V12 Lamborghini motor. This looks like an awesome little car to drive! BUT on their website they say "Do not depress the accelerator any more than 1/4" or you will ruin your car" What do you think would happen if such a car was to be sold to the public? Exactly what is happening with the P60s. Now who would you say is at fault here? The drivers for not keeping their right foot in check, or the manufacturer/designers? Its a ludicrous concept, and thats what I think of the high ratio P60s.

Banebots should stop producing P60s in any higher than ~60:1. SHOULD teams decide they want a P60 with a higher ratio, they can buy the planetary stages seperately and stack them together with longer bolts. If the thing fails, it is the team's fault for modifying the stock P60 with extra ratios but BB should have never marketed their high ratio gearboxes under the pretense that it is an appropriate reduction method for a speed 550 motor.

[artdutra04]

Originally Posted by fox46

My analogy and opinion of the situation is this: Banebots has built a car. It bears resemblence to a Geo Metro- Cheap, compact, practical, simple etc etc.. But they are also marketing it to be robust and powerful. What they have in fact done is built a Geo Metro but swapped the engine to that of a V12 Lamborghini motor. This looks like an awesome little car to drive! BUT on their website they say "Do not depress the accelerator any more than 1/4" or you will ruin your car" What do you think would happen if such a car was to be sold to the public? Exactly what is happening with the P60s. Now who would you say is at fault here? The drivers for not keeping their right foot in check, or the manufacturer/designers? Its a ludicrous concept, and thats what I think of the high ratio P60s.

Banebots should stop producing P60s in any higher than ~60:1. SHOULD teams decide they want a P60 with a higher ratio, they can buy the planetary stages seperately and stack them together with longer bolts. If the thing fails, it is the team's fault for modifying the stock P60 with extra ratios but BB should have never marketed their high ratio gearboxes under the pretense that it is an appropriate reduction method for a speed 550 motor.

Blaming Banebots because a team ignored the manufacturer's product use guidelines and the gearbox subsequently imploded is inane. It's like blaming McDonald's for selling hot coffee because you burnt your crotch while trying to drink their coffee while driving on the highway.

[fox46]

What if they sell that cup of coffee to you at a drive through? They are endorsing driving with said cup of coffee. If the coffee is indeed hot enough to cause second degree burns and they endorse driving with it - by all means they can get sued - even despite the "Caution Contents Hot" on the side of the cup. - That isnt the best analogy. We all know that many establishments have been quite successfully sued over this.

Yes they have the disclaimer not to use them with any more than 35lb-ft but obviously this spec is overrated. When using the RS395 motors on the 326:1 gearbox, at stall the combination produces ~28lb-ft of torque which is less than the rated loading of the gearbox. I don't know how you can claim a team is overloading a 326:1 with an RS395.

In our team's case, we have a 5.4:1 reduction after the P60 on our arm. At it's maximum extension in the horizontal plane, it reaches about 5 feet. Given this spec, the arm should have a capacity of 37.8 lbs. Since the whole arm itself weighs 15 lbs the P60 should reliably tolerate upwards of 20lbs of force at the end of the arm. This doesn't even consider the gas shock we have to assist it. I don't know but I doubt any of the drivers have been practicing scoring steel belted radials.

Yes the impulse loading on the arm will indeed approach this limit of 35 lb-ft but if BB was prudent they would have allowed at least a factor of safety of 2. Alas, declaring that these P60s are only good to ~17 lb-ft would definitely deter many teams from buying them. I know we would have gone with something else if this was the case. Of course this would severly impact their sales to the FIRST community so it seems they either chose to favor their bottom line rather than the reliability of their product or they have over estimated the strength of their product.

[fox46]

Ether, here are the hand drawings that the guys have been working from. Not all the dimensions are on them but anything that is missing is pretty easy to figure out. Hope this helps.

[thefro526]

Originally Posted by fox46

Ether, here are the hand drawings that the guys have been working from. Not all the dimensions are on them but anything that is missing is pretty easy to figure out. Hope this helps.

Thanks for posting these up. I showed your pictures to on of our Sponsors yesterday and it looks like we'll be able to get these done in time for our next event. I'll update with pictures and results when we're done.

For anyone else who was looking in to doing this, Stressproof Steel is sold as Alloy 1144 and should be pretty easy to find at your local steel supplier.

[fox46]

When is your next event? Be sure to post how they work out for you and if you experience any other failures.