drive train wearing out?
 

We're having some trouble with our drive train and launcher and haven't been able to determine the problem.

We were able to get an entire week of driver training in at the end of the build season, the first time ever for us. We pushed the robot hard (drove almost constantly, swapping batteries every 10 minutes or so then throwing them on the charger; only stopping when the CIMs got too hot) all week and never had a major issue and it was working great. However, at the Wisconsin Regional, we ended up with some major issues.

1) drive train didn't turn as well as it did in practice. We couldn't spin around in circles in high gear like we had been able to.
2) The robot didn't seem as fast sometimes. (posibly related to #1)
3) Sometimes the robot wouldn't move when I moved the joystick.
4) launcher (using 2 CIMs) worked well but couldn't throw it as high or as far as in practice.

We have kitbot 6wd with center wheel lowered. All IFI high traction wheels. Each side is powered by 1 CIM through an AM supershifter and a 1:1 sprocket ratio. The launcher is powered by 2 CIMs through a toughbox and an 11t chained to a 45t sprocket.

We have a list possible problems and solutions, but no one has been able to pinpoint anything specific.

Help is greatly appreciated.

Regards,
Andrew

Re: drive train wearing out?
 

Hmm, That seems very strange. Over time a motor will gradually wear out but, it should take longer than that. Perhaps greasing your transmissions will help with some of your power loss. Other than that I'm stumped. Possibly changing the motors could be an option if you have the time.

Re: drive train wearing out?
 

I have two immediate thoughts.

First, did you do your drive practice on carpet? Robot handling is very dependent on the surface it's on. Many teams find that a robot which turns okay on tile or concrete won't turn at all on carpet.

Second, the general description you give, with separate systems seeming to be affected, points to a common problem. How does your battery wiring look? It sounds like you have a poor connection somewhere in the high current path. Double-check all the contact points: the main breaker, the distribution block, the breaker panels, etc. Feel for hot spots. Look at your battery voltage while you are running the motors.

Re: drive train wearing out?
 

Sounds like Alan has it pegged....carpet makes a big difference in high gear turning ability. Also go thru the electrical supply system completely, the screws that hold the wires into the distribution block and the Maxi Fuse panel need to be TIGHT, and have enough bare wire (correct strip length) to be fully inserted. This bit us at the AZ regional, the robot worked fine at home, but had all kinds of weird problems like you describe, ending with it sitting still on the field during our last elimination match (which we lost, of course). Inspection at the next regional revealed a loose and melted wire at the fuse block, replacing it and tightening all the connections properly resulted in no odd behavior during that regional.

Re: drive train wearing out?
 

We plan on trying that.


Yes. We had an entire practice field (a cheaper version of the real thing). During that week, we wore our treads down past the rivets, some had even fallen out. We never replaced the treads at the drive's corners, figuring it wouldn't help our problem.

I forgot to mention that the robot's on its way to Atlanta so we can't do this for a while. Also, our forks used to pick up the ball were always controllable. I'm not the electrical guy, but I think this would rule out many poor connections. Anyhow, Its on the list to check.

Also, The intermittent loss of drive response seemed to only happen when the robot wasn't moving, and it would only last a second or 2.

Re: drive train wearing out?
 

I think it is important to note a few other things -

First off, the launcher and drive motors are on separate fuse panels (drive motors on 40A maxi block, launcher on 30A). I second that the problem is not an electrical connection problem, though we are going to check to make sure when we get to Atlanta.

Second, about the turning - I would say that we encountered a 75-99% loss of turning ability at the Wisconsin regional. At the practice field, we would give the robot a 100% turn command and it would turn extremely fast. In Wisconsin, we gave it the same command and it would either turn much slower or just give a bit of a jerk, then stop.

Finally, it seemed like the robot would drive better (more like it was at the practice field) when in low gear, but it just runs so slowly that we cannot waste that much time driving around the field.

I am personally leaning toward our CIM motors going bad because our lifter, which is using the Taigene motor, seems to still be functioning perfectly. Has anyone else ever run into CIMs going bad after quite a bit (I would guess at least 15-20 hours) of running their robot?

Re: drive train wearing out?
 

An electrical problem could be farther up the line. Check the distribution block.

I don't think you can kill a CIM in a mere 15-20 hours. 330 has been running their practice robot for I don't know how long and hasn't killed a CIM yet that I know about. Wheels, yes, motors, no.

One other thing--did you change the code at all?

Re: drive train wearing out?
 

One of our joysticks was wearing out and not sending solid(ish, well as solid as these joysticks can) signals. Double check your Joysticks for your drive issues.

-Jim

Re: drive train wearing out?
 

We're getting a full command sent to our victors (solid LEDs)

Re: drive train wearing out?
 

I'm really leaning towards a sneaky electrical problem in the high current path as well. As noted, it's really rather difficult to kill CIMs, especially to kill them as much as you claim to have done. Having a robot act strange under high current draws is pretty darn easy, though. The fact that your other motors are working fine doesn't really change anything, as they don't sound like they're drawing very much current at all.

I'd really, really recommend checking you connections, as it's really the only thing those two systems have in common. I can't imagine your launcher motor was running for the exact same time your drive motors were. I'd bet it was running 25% of the time or less, so it should have deteriorated much more slowly.

If you really can't be bothered to check all the connections first, you can do a quick check with a pair of multimeters. Measure the voltage across your battery and across the V+/V- terminals of your launcher's speed controller while the launcher is loading. I'd recommend doing the drive wheels as well, but you need some way to safely load them while the robot isn't moving. The difference you see is entirely dependent on how much wiring you have and how good your connections are, but I don't think it should be more than 1.5V. Maybe 2V. I'm sure the other experts in this thread with have good suggestions for values, but you're dropping 2 or more volts on the way to your speed controller, you definitely have a problem in the power path.

Re: drive train wearing out?
 

If you really can't be bothered to check all the connections...you probably should not have asked for help????....

Seriously, just because you don't think this is the problem, until you actually check the connections (really check them), you won't know if they are OK or not. You have described symptoms which are commonly caused by poor power distribution connections, so it would be time well spent seeing if that is the cause, even if you have never experienced this situation before--others have, and several folks here zeroed in on this as the most likely culprit.

Re: drive train wearing out?
 

We had a similar problem, robot would work fine but sometimes it would stop then jerk once in a while. found out that our battery connections were bad. between the AM cables and the actual battery terminals, make sure the bolts are very tight and wrapped in a lot of electrical tape.

Re: drive train wearing out?
 

Sorry if I seemed ignorant of the power distribution. Like I said, I'm not even on the electrical team. It's definitely something we'll look at.

It sounds like its an electrical problem which hopefully wont be too difficult to fix. If I'm understanding this right, a lose connection would increase resistance in series with the motors and dropping the voltage available to them. Also, it might lose contact and cause a momentary loss of power to the motors. Is this correct?

Thanks to everyone who has responded so promptly!

Re: drive train wearing out?
 

Yes, that's correct. The result is bizarre behavior from the robot.

Re: drive train wearing out?
 

You can easily kill a CIM in 30 minutes if it gets too hot.

Lot's of practice without proper cooling time will greatly shorten the life of your motors.

But I would agree that it's more than likely a connection problem.

Re: drive train wearing out?
 

Our robot's odometer reads a little under 75,000ft on it's drivetrian now (about 14.2mi :ahh: ) and the chains have started to gum up from carpet fuzz (among other things).

Our (currently tired) robot is due for an overhaul that we'll be doing thursday morning in Atlanta. It'll be back up to it's old midwest-regional-speed self first thing Thursday.

But yeah, you are not alone in your robot's symptoms gurellia53, they aren't much different from the ones ours developed.

Good luck finding your problem,

-q

Re: drive train wearing out?
 

Well if it helps at all, we were seeing temperatures of around 150 degrees farenheit using a non-contact thermometer after a decent amount of use. This was the temperature of the casing of the motor, so it could have been hotter in the center.

Thanks so much for all of the replies so far. We will definitely be putting a lot of effort into the robot on the practice day in Atlanta. I'll be sure to check everything you guys have recommended. The only reason I am thinking that it is not a loose connection is that we have noticed a significant difference in the drive motors, a smaller difference in the launcher motors, and no difference in the lifter. As some have said though, the lifter gets very little use.

Re: drive train wearing out?
 

W are experiencing similar issues too. The Electrical paths is one of the items we will address at Atlanta before anything else.

I want to add that the Ground returns are just as important to check along with the power side.

All of those return molecules have to go somewhere. Think of a Dam.

If the water leaving the dam is restrained, it backs up behind it. If the dam is wide open and the water to the dam is a trickle, the same amount of water is still leaving the dam. The trick is figuring out what scenario is happening. Sometime heated wires can pinpoint where the problem is occurring.

At the Florida Regional, we must have redone connections on the battery terminals at least 5-6 times. Those screws would work loose and then you would have a resistance build up or intermittent connection. That surprised me the most. If you think a connection is tight enough, give it another 1/4 turn for safe measure.

The yellow jumper may also be an issue on the distribution block. Move your wires around so the maxi-block is fed by the same block as the power coming in without the yellow jumper being a factor.

All of our other wire connections are soldered to minimize resistance, so the focus will be from the maxi-block back to the battery.

Re: drive train wearing out?
 

I think the drivetrain is way under powered. We have 2 CIM's per side and a AM Shifter and a ~3:1 sprocket ratio, and we have a hard time turning sometimes with this setup.

You have one CIM and a 1:1 sprocket ratio. It is geared for speed and not for turning. I think like "Alan Anderson" said the surface you were on at your school made you able to turn and now you are on a surface with more traction and can't turn.

Re: drive train wearing out?
 

I'm with Kevin on this one... we built a six wheel drive kitbot with the centre wheels lowered and 6" IFI traction wheels on each corner for the Aim High game a couple years ago. Even with two CIMs per side and a slight reduction in the sprockets the robot had difficultly turning on carpet. It may be that the carpet you tested on was lower friction than the carpet at the tournaments. (Edit: or that your robot was lighter at the time or had a different weight distribution.)

If you can try slapping a pair of omni wheels on the heavier end of your robot, I am willing to bet that all your turning issues will suddenly disappear. Yeah, your traction will drop a bit, but given the nature of this year's game and the gear ratios that you have that should have little effect on your machine's performance.

Good luck with it!

Jason

Re: drive train wearing out?
 

Andrew,
Come and find me early in Atlanta. I will be one of the Lead Inspectors so I should be easy to find. There is a lot of evidence that points to an electrical problem but the drive motors sound like they have been stressed. 150 degrees is likely 30-50 degrees cooler than the armature temp. These motors do start to short windings which limits output power with extreme temperatures. Be ready to replace in Atlanta. If they are shorted, they will be putting an increased load on the battery that was not there during practice. However, if all problem motors are driven from the 40 amp block, that is the first place to look.

Re: drive train wearing out?
 

Well, the carpet may have been slightly different between the practice field and competition field, but there was such a small difference that it would not have as drastic of an effect as we are seeing. It performed very well on the practice field and now with the same robot and an almost equal field. it seems to us that our drive and launch systems are far less effective. We are thinking that it is either a mechanical or electrical problem somewhere between the battery and the transmissions.

Al, thanks so much for your offer for help. We will be sure to find you in Atlanta!

Re: drive train wearing out?
 

This is from my RC Car racing days, But could you have thrown the Solder?

No doubt that the motors got hot from doing turns in high gear. I've seen it when motors lose all power and efficiency from throwin the solder.

Re: drive train wearing out?
 

Try changing the chain's on your DT. The chain on our comp. bot is really tight, and the bot is a little slower as well , while our practice bot's chain is fairly loose... we are going to change out the chains thursday for the first match and see how we do...

-Keaton

Re: drive train wearing out?
 

Hi guys,

I have been working with DC motors for a very long time. The ones I work
with are called starter motors on very big diesels. These have lifetimes
like all DC motors. The big problem on small motors that aren't heat sinked
is breakdown of the varnish that is on the armature windings.
A bigger problem is the commutator surface. Normally it is an even surface of copper
that the brushes contact. Over time this surface gets hit with constant
current spikes. If brush wear is even then the current is distibuted evenly.
If hot spots occur then the surface starts getting burned. The more
burning the more nonconduction of the surface and the motor starts losing
performance. Take an old DC (or AC) brushed motor apart and look at the
commutator. I have been amazed at some of these still working despite their
appearance. If your motor doesn't turn and some one moves the drive
wheel ( do this with an object not your hand) and all of a sudden the motor
starts moving then you may have commutator brush problems. When you
move the motor it goes to a different spot on the commutator that isn't
burned then it starts to spin.

HOT Cims also can affect brush springs. As they get softer they don't exert
as much force on the commutator. Less force less contact less current. The
dust from the brushes also gets in the brush retainers thus they don't slide.

The problem with CIMs is like all DC motors they will spin like crazy with
no load and sound just fine. You must load the motor to test it. A little
Dyno would be perfect.

I advocate what many of you already said. Run your bot for a test run
on the carpeted practice area. Then feel all the connections. Any heat
is lost voltage. Its that simple. A simple thing to do is carefully place
a voltmeter in the Robot where you can easily see it. Make leads that
attach from the battery lead to the input side of your motor controller.
When the robot is running it should not drop more than a few tenths of
a volt. If it is you are losing voltage thats a bad thing. If you are then
just start probing from the battery to the next thing and so on until you
find out who is dropping all the voltage. A loose connection can get so
hot it will burn :ahh: your fingers so be careful.

If you find you are delivering full voltage to the controller then you need to
see what voltage is being applied to your motor at full throttle. It should
be at least 95% of the applied voltage. (Your motor control uses parallel
MosFets that act as switches that turn on and off very fast, the motor
armature is an inductor so this integrates an RMS current through the
winding and produces Electro motive Force to spin the motor ) As you
approach full signal the mosfets turn on for such a long time verus off
that it will appear as the MosFets are fully on. (they still have Rds on though)

I put a lot of buzz words in here so you guys can hit Wikipedia to learn
more.

The summary is Check your system for voltage drops, (heat or voltmeter)
You need to have the robot under load to do this so
DO IT SAFELY, have a mentor around to observe
If you have full voltage then clean out your drive line
of carpet fuzz, check bearings (for string,hair), clean chains
If you still seem to not have the power it may be prudent
to change out the CIMS. If you have two maybe you can
change one at a time to see if only one is bad.

Sorry for the long post but knowledge is Power :] (ha-ha)

Re: drive train wearing out?
 

We had that SAME EXACT problem with our drive train. We used a direct front wheel drive 1 CIM per side configuration with traction wheels. (We had omni's in the back) Our issue was we would test on our hard linoleum surface and be fine. We also tested at the play date. We were fine (but we only ran one match per trial). We got to Chesapeake and nothing worked. If you use a Gyro, disable the gyro code and any other sensors that would affect your drivetrain. These could be hindering your turning ability because they attempt to control the spin and causing you to be unable to turn. When the motors are cool, I bet you have decent performance. As motors heat up, the become less efficient causing more power to be drawn. Did you ever notice a "stuttering"? This would be caused by circuit breakers popping and resetting. If possible, I would strongly consider changing your gear ratio or between matches putting ice packs on all of your motors. We did this with great success at Philly and I think it can work for your team too!

Re: drive train wearing out?
 

Thank you everyone!

I think we'll plan on changing the motors (we have to open the transmissions anyways) and check for loose wires. We also have some sort of a thermometer on our multimeters (how awesome is that :] ) so we'll probe around for hot spots. Also, not mentioned in this thread, some people on our team think that we may have bad victors. And since we have 2 empty victors on our robot, we'll switch the drive motor victors.

Al, thanks for your offer! I'll definitely look for you in Atlanta.

-Andrew

Re: drive train wearing out?
 

I haven't seen anyone throw this out yet, but you should also check your Victors. A few years ago, our robot had a weird issue of driving very weird in forward, but perfectly in reverse. When I took the fan off the victor, I saw that one of the transistors was blown, causing less power to go to the motor. With the stress of one motor driving the robot, the Victors take more abuse, which may cause them to fail.

To check this, just pop off the fan, get a small flash light, and look at each transistor and see if there are any pieces missing from them. Also check for melted plastic where the transistor enters the case. This is also a sign of something wrong.

Re: drive train wearing out?
 

A much easier way is to use your nose. When Victors pop the FETs they give off an very unique odor. Please remember that in my experience, about 90% of all Victor failures are due to metal dust and shavings.

Re: drive train wearing out?
 

Another tip to remember: nothing is EVER maintenance-free. Check for loose bolts and for pieces of the bot that may have shifted or come undone. This may be causing a not-so-obvious mechanical interference that causes an inefficiency.