Banebot 56mm gearbox - double D - RELOADED

[ZZII 527]

Originally Posted by Arkorobotics

Funny thing is that it does run smoother and no signs of binding, yet (hopefully never). How much torque does it really take to backlash it and break it?

Yep, ours ran much more smoothly too after some tweaking. I was just saying that I think the alignment and spacing issues are separate from the double-D issues. Solving one won't fix the other.

An earlier estimate from Dr. Joe was that 350 in-lbf, in repeated, high-cylce loading was a safe estimate for the amount of torque the plate could see before a problem occurs.

[Arkorobotics]

Originally Posted by ZZII 527

Yep, ours ran much more smoothly too after some tweaking. I was just saying that I think the alignment and spacing issues are separate from the double-D issues. Solving one won't fix the other.

An earlier estimate from Dr. Joe was that 350 in-lbf, in repeated, high-cylce loading was a safe estimate for the amount of torque the plate could see before a problem occurs.

40N*M? thats a lot for our drive system. Our drive systems stall torque is 38N*M.

[Gdeaver]

I know the best solution to the carrier plate is a new properly heat treated replacement, but could brazing the shaft and carrier together overcome the problem? This might help if the spacing issue agravates the carrier wipe out.
A good 16% silver brazing rod and the proper flux should penetrate well. What woud taking the carrier and shaft up to 1300 - 1400 F do to the hardness? Water quench after?

[dtengineering]

While the brazing may well reduce impact loading by reducing the amount of space for "play" in the Double D, and could serve as an emergency fix, I would suggest holding off on doing it for the moment, if you can.

I would be concerned about getting proper alignment of the carrier plate while brazing... yes, it can be done, but if it is done wrong you are deeper in the hole.

I share your concern about hardening... or softening... of the shaft. You don't want to weaken the shaft, yet at the same time you don't want to make it brittle. While I can't say for sure whether the brazing is likely to affect the shaft, or whether quenching would re-harden it, I agree with you that this is an unknown of no small concern.

I would be concerned that since the braze material is softer than the steel of the carrier plate that it will deform under static load, leaving you no further ahead than you were before brazing.

If a replacement carrier plate made of hardened steel becomes available it will likely be machined to fit the double-d of the shaft... which is now brazed into the old carrier plate.

I am hoping (rather anxiously now...) that the next team update from FIRST will address an official response to this problem, and that the official response will include replacement final stage carrier plates that will be available to all teams. (I am basing this on Dr. Joe's comments that everyone at FIRST and Banebots is working hard to address this issue and the high respect I have for the FIRST organization... I have no inside knowledge on which to base this.)

So if you've got a spare transmission on which to test this fix, I'd say sure... go for it... and let us know how it works. It just might work out okay, after all. But if you can still test drive your robot, despite increasing backlash, and have no spares, then I would suggest perhaps waiting a day or two before trying this fix.

The optimal solution does seem to be in getting some proper carrier plates manufactured and distributed to teams so that the gearboxes can do the job that they were specified to do. It shouldn't be that hard, nor that expensive, particularly if it is a co-ordinated response for all teams.

Jason

[MrForbes]

Perhaps you didn't see one of the earlier threads on this subject, where welding the two pieces together was discussed, or where making the shaft one piece was discussed. The shaft is 12mm at the DD joint and the bearings, and 1/2" (12.7mm) at the other end. So, you need to braze it with the shaft installed in the bearings and the end mounting block. I would hesitate to do that....the heat would probably be hard on the bearings! plus you can't disassemble it without reheating and melting the braze.

[Joe Johnson]

If the carrier material is the proper hardness, I don't think that brazing will significantly improve the performance of the joint. The problem is not the impact from the slop in the joint but the impact of the robot's momentum vs the rotational inertia of the armature of the motor.

This is why grippy tires play such an important roll in making the joint open up. Without grippy tires, the force from the robot momentum is limited by tires slippage.

As to the solution, I don't know all the final details, but everything I have heard so far seems to be reasonable. I am sure that some folks will still complain, but I think that most level headed folks will be okay with the plan.

Joe J.

[ChrisH]

Originally Posted by dtengineering

I am hoping (rather anxiously now...) that the next team update from FIRST will address an official response to this problem, and that the official response will include replacement final stage carrier plates that will be available to all teams. (I am basing this on Dr. Joe's comments that everyone at FIRST and Banebots is working hard to address this issue and the high respect I have for the FIRST organization... I have no inside knowledge on which to base this.)
Jason

I was at FIRST HQ over the weekend on a different matter and spoke with the person who is responsible for the testing being done there on this issue. They were very frustrated at the time it was taking to get the new "improved" parts delivered for what will be a much more rigorous round of testing. At that time, Sunday afternoon, the parts were somewhere in transit.

Assuming the parts were delivered yesterday, as expected, today would be earliest we could reasonably expect results. Depending on the test program, which was not elaborated on to me, it might be several days before results are ready for general release. One thing they do NOT want to do is release a fix that then does not work. That would be worse than the situation now. So they are taking the time to do it right.

Another thing that needs to be understood is the small crew FIRST has working on FRC. I was introducted to about five of them. That was the whole FRC staff. These people all have duties to perform that are related to the Regionals and are therefore time-critical. Somebody has to plan the loading of all the boxes for the trucks, make sure all the equipment gets loaded in the boxes, make sure the boxes are properly loaded into the truck, and make sure the truck leaves on time, among other things. These tasks were planned and people and other resources were allocated well in advance.

The current test program is an un-planned task. It is a very important task and FIRST is very concerned that it be accomplished successfully, but it is still something that they have to divert some of their limited resources to cover. This is also their busiest time of year and people, even robot people, need to sleep sometime. So in spite of everybody's best effort it may take some time to get this resolved.

The good news is that the failure mode is slow and progressive and the fix can be installed at a Regional if needed. That gives at least an extra two weeks to deliver a fix. (Obviously the parts would be considered COTS items and could therefore be purchased if needed during the interim between ship and competition)

So keep on building your robots, handle them gingerly during testing and watch out for the start of any failure behavior. Make sure your trannys are accessable and easily removable. If you do these things you should be OK in the long run.

[Dick Linn]

If I had to do something and had no machining capability, I'd be tempted to heat and quench the existing carrier in a very cold brine solution, one containing about a pound of salt per gallon of water. I might even add some Jet Dry or other type of surfactant (or some "water wetter") to ensure a very hard quench. (People used to use a lye solution, but that's a bit dangerous.) The idea is to raise the boiling point and reduce the tendency for the water to form vapor bubbles at the interface between the quenchant and the steel, thus ensuring maximum heat transfer rate.

It is my understanding from research that steels with as little as .20 to .25 percent carbon can still be hardened to as much as 36-37 Rc by this method. PLEASE DO NOT TRY THIS WITH STEELS CONTAINING MORE THAN .40 PERCENT CARBON. THEY WILL SHATTER.

Some people swear by the "superquench" formula containing some dishwasher soap. I found this to be of interest: http://paaba.net/WeekendMet.htm

[billbo911]

Team Update #9 is out. In it the Carrier plate is addressed. Currently BaneBots is in the process of manufacturing new parts. The method of distribution is being worked on.

Excerpt from Team Update #9
"...Replacement plates are currently being manufactured by BaneBots for all transmissions distributed to the FIRST community. These plates will be available and distributed to teams by BaneBots at no additional charge. We are currently working on a distribution method to insure that teams receive them promptly...."

[WaterFreak]

Originally Posted by billbo911

Team Update #9 is out. In it the Carrier plate is addressed. Currently BaneBots is in the process of manufacturing new parts. The method of distribution is being worked on.

Excerpt from Team Update #9
"...Replacement plates are currently being manufactured by BaneBots for all transmissions distributed to the FIRST community. These plates will be available and distributed to teams by BaneBots at no additional charge. We are currently working on a distribution method to insure that teams receive them promptly...."

YEAH, good news indeed :-)

[Kevin Sevcik]

Originally Posted by Dick Linn

If I had to do something and had no machining capability, I'd be tempted to heat and quench the existing carrier in a very cold brine solution, one containing about a pound of salt per gallon of water. I might even add some Jet Dry or other type of surfactant (or some "water wetter") to ensure a very hard quench. (People used to use a lye solution, but that's a bit dangerous.) The idea is to raise the boiling point and reduce the tendency for the water to form vapor bubbles at the interface between the quenchant and the steel, thus ensuring maximum heat transfer rate.

It is my understanding from research that steels with as little as .20 to .25 percent carbon can still be hardened to as much as 36-37 Rc by this method. PLEASE DO NOT TRY THIS WITH STEELS CONTAINING MORE THAN .40 PERCENT CARBON. THEY WILL SHATTER.

Some people swear by the "superquench" formula containing some dishwasher soap. I found this to be of interest: http://paaba.net/WeekendMet.htm

Just for completeness, we got the results back from a friend with a material lab. His considered opinion is that it is either 1018 or 1020 steel. So it's virtually impossible to harden without carburizing, likely even with your superquench method.

[Doug G]

Just the news we're hoping for!!! Because they still haven't figured out distribution method, I figure teams will be making this fix at their first comp. I mean for teams to receive them by the end of next week, they'd have to ship out next Monday. Can they even make a couple thousand plates in less than a week?

[Kevin Sevcik]

Originally Posted by Doug G

Just the news we're hoping for!!! Because they still haven't figured out distribution method, I figure teams will be making this fix at their first comp. I mean for teams to receive them by the end of next week, they'd have to ship out next Monday. Can they even make a couple thousand plates in less than a week?

Check your rule book. The FIX-IT WINDOW rules allow for teams to buy/manufacture spare parts and replacement parts. Since stronger carrier plates are obviously replacement parts, the ship deadline is obviously not as important. Especially considering this is an official FIRST replacement part. At worst, I can see FIRST distributing these at the regionals. I feel safe in saying that any team could replace both carrier plates in under an hour with a little help.

[Ricky Q.]

Originally Posted by Kevin Sevcik

Check your rule book. The FIX-IT WINDOW rules allow for teams to buy/manufacture spare parts and replacement parts. Since stronger carrier plates are obviously replacement parts, the ship deadline is obviously not as important. Especially considering this is an official FIRST replacement part. At worst, I can see FIRST distributing these at the regionals. I feel safe in saying that any team could replace both carrier plates in under an hour with a little help.

Still only gives teams until the Friday after ship, 2/23.

The timing of these “FIX-IT WINDOWS” is at the discretion of the team, but all work must be completed by 5:00pm on the Friday following the ROBOT shipment deadline.

[Billfred]

Originally Posted by Ricky Q.

Still only gives teams until the Friday after ship, 2/23.

Keep reading.

<R22> During the “FIX-IT WINDOW” following each Regional Competition weekend: During this
period, all teams (not just those teams attending a Regional Competition) may utilize one or
two 5-hour FIX-IT-WINDOWS to manufacture SPARE, REPLACEMENT and UPGRADE
PARTS and develop software for their ROBOT at their home facility (not at the competition
site). The timing of these “FIX-IT WINDOWS” is at the discretion of the team, but all work
must be completed between the opening of the Competition (at 8:30 am on the Thursday
of the Competition weekend)and 8:30 am on the Thursday following the Competition
weekend. At the conclusion of a regional competition event, teams may take a limited
amount of broken or malfunctioning COMPONENTS or MECHANISMS back to their home
facility to make SPARE or REPLACEMENT PARTS. The purpose of this rule is to allow
teams to make critical repairs to existing parts to enable them to compete in following
events. The intent of this rule is not to have teams take their entire ROBOT back home and
make large-scale revisions or upgrades to the ROBOT. Teams may manufacture and/or
repair all the parts they want, but the amount of parts they can bring to the competition
event is limited (as specified in Rule <R31>).

That said, a competition-day switch wouldn't faze me in the least. I have empirical evidence that shows even a marketing major can tear a pair of these down and rebuild them in well under an hour with one allen key.