Today we we’re testing driving over the defenses with a KOP chassis + pneumatic tires. We kept having the radio reboot when we’d get stuck. At first we thought we just had a loose connection somewhere but after more testing we figured out we we’re browning out.
We think what might be happening is that in some situations the belts get pinched. With the belts pinched they can’t skip on the gears and the 4 CIMs end up fully stalled.
Since we can engineer our way around getting the belts pinched we moved on in testing. Our next test was pushing full power into a wall. We almost insistently popped the 40a breakers on the motors. This is better than dropping power to the radio but is still a problem.
A few more pieces of info; we had rather old 40a breakers in that had been tripped many times in the past. Replacing them with new breakers seems to have helped some. We also had the robot up to around 140 lb. for testing. We had no issues when it was just the drive-train and battery.
My question is, has anyone else experienced similar issues with the pneumatic tire setup?
Are you wheels close to square geometrically? You may want to reconsider 4 traction wheel drive if this is the case. The main reasoning is the scrub force is too great and the power required to turn exceeds the limitations of the electrical system. You may want to consider a wide 4WD wheelbase or opt to go 6WD with a drop center wheel (you don’t need a full size wheel).
A fun fact is that breakers can creep mechanically over time. For instance, if you have a 40A breaker that you regularly pull close to 40A without tripping, you can actually affect the breaker’s rating and it will possibly trip at a greater current.
2 CIMs per side, 8 inch Pnuematic (AndyMark) Tires, 12.76:1 ratio
JVN’s sheet says ~12fps and ~70Amp per motor during a pushing match.
I would say you’re probably playing with fire with that current in a pushing match number.
It’s not surprising that you end up tripping breakers and possibly browning out (where the RoboRio protects itself by turning off motors).
I hate to say it but I fear this is the tip of the iceberg where the brown out protection that FIRST put into place last year doesn’t actually end up keeping the important bits alive after all when robots have less than friendly interactions (aka any year except last year and that terrible 4vs0 year).
Back to your particular problem. I recommend you look into less grippy tires or more gear ratio. 3 CIMS would help but only so with the breakers tripping (they’d still trip but at 47Amps per motor you’d have a bit more time pushing) it wouldn’t help with the brown out at all.
Keep us informed on your progress – especially radio lock up issues.
Your absolutely correct. We are overloaded. I had a typo in the CoF in my spreadsheet. Bad part is I think we’re maxed out on the ratio that will fit in the TB Mini Gearbox. Might need to swap out tires…
I hope all the teams that bought up the stock of pneumatic wheels are testing. It’s been many years since we’ve considered using such large tires and some of us might be a bit rusty with our calculations :yikes:
Maybe convert to chain and get some added ratio by using larger sprockets on the larger tires?
I love chain*. Haven’t converted to this new fangled belt stuff…
Dr. Joe J.
*chain let’s be be wrong late and recover. I can laser cut new ABS sprockets an afternoon and make up new chain lengths and I’m off an running with a new iteration. Belts make me order new pulleys, new belts, … I just don’t see the benefit for most FIRST applications.
What is the lowest voltage your seeing reported in the DS before you lost communication? The VRM should be able to maintain the radio power down to a very low voltage.
I think we had “bad” breakers. I’m not sure what exactly that means but when we replaced all the 40a with new breakers the RoboRio started doing what it is supposed to do and turned the motors off. It is now keeping the rest of the systems running. I have no idea how changing the breakers could do that. Maybe something else got moved around while changing the breakers.
RufflesRidge: I believe we we’re dropping near 5v at some points.
Joe: Chain may fix this for us. Thanks for the suggestion. Also…your cutting ABS sprockets for drive trains :ahh: As someone who loves laser cut ABS that never occurred to me. About 40% of our robot last year was ABS. Do they hold up as sprockets?
As breakers trip, they start to trip sooner and sooner.
If you know this is going to be a problem, I believe there is a way to check current draw and/or voltage. When you start getting too high (low), back off the motors yourself. Don’t rely upon the RoboRio to do it for you.
As my grandfather might say if he were around to see this competition: Tripping breakers is like writing a bad check. It’s hard the first time, but then it just gets easier and easier. (No really, a breaker will trip sooner on successive high-current loads because it’s been broken in.)
You may also want to try different air pressures in your tires or different tires entirely. 200x50 tires like AndyMark sells are a semi-standard size, so they’ll mount to the same hubs. We bought several of these and found they have less grip than the AndyMarks, but they’re also a smaller diameter than the AndyMarks. Only you can say whether that’s a good or bad thing.
I love ABS sprockets. The hold up nicely if you can follow the rules. 1 keep the chain tight (preferably by having a good chain tensioning system designed in from the start) 2 keep the sprockets and chain aligned and in one plane 3 all torque transfer sprockets have over 120 degrees of wrap.
If you break these rules and the chain jumps, ABS is going to get munched much more than aluminum but that’s the way the cookie crumbles.
For 25 chain:
I typically use 1/8 ABS sheet with bevelled edges for the teeth and make a sandwich with 1/8 sheets on the sides. ABS solvent bonding glue is your friend (and amazing).
For #35 chain:
You can use the above technique or you can make a one piece no glue design by using 1/4 ABS sheet but you need to really bevel the edges. If you have a belt sander and patience it isn’t too hard. Makes a mess but the results are nice.
A lathe would make the bevel edge trivial however, the teams I help typically don’t have access the a lathe big enough to spin the sprockets I like to use. The Epilog Laser cutters I have access to can cut a 12X24 envelope so I tend to make giant sprockets (just under 12" diameter).
Take a look at the sprockets on Scorpion (#246’s 2015 robot*) and our “Standard Three Joint Arm.” There is some serious ABS Sprocketage going on there
Never underestimate the power of simple tools mixed with resourcefulness and patience to make up for lack of fancy pants equipment.
Dr. Joe J.
*By the way, if your team doesn’t name your robots (both practice robots and competition robots as well as off season experimentation chassis), I feel sorry for you. Honestly, don’t you get tired of saying “our robot from Aerial Assist” and “that mecanum wheel chassis we experimented as part of Cool Kid’s senior project”? Of course you do… AND… You’re missing out. You could be calling your robots Scorpion & Rover & Hungry Hippo. Increases your cool factor by 50% right there.
