pic: Battery connector.....

[Nirvash]

Originally Posted by cglrcng

I did see that exact type of thing happen on our robot last year except not quite that far as they shut it off at the breaker. When the drive team came back from an early AM qualifying round, the student w/ the battery in his hands complained that I gave them a dead battery..."No way, no how" I said...as I checked it w/ a multi-meter myself I said just before handing it to him before the round.

I immed. began to inspect the battery Anderson clip Battery half (since it actually burned my hand during the semi-disgusted hand off) - and I saw melting of the clip & cables, and it was still very hot to the touch, so I took it apart and looked at the ends inside the clip...They were a bright orange copper color instead of alum. color.

I abruptly said...You do not have a battery problem folks..."You have a frame grounding problem," and pulled out the multi-meter from atop the battery cart and we began looking for the problem immediately. Sure enough, a frame grounding problem was immediately noted by a simple OHM check. That frame was grounded.

The hunt began....Sure enough, it took a while, but we found that (2 of the students had removed and reinstalled the CRIO the night before just before pit closing), and there was a nice short bright & shiny chromed Phillips Head screw near the bottom on the outside of the CRIO that was touching the lower chassis frame rail, that wasn't doing so the day before, (the CRIO's all come w/ that very unnecessary grounding screw), but, the CRIO is not supposed to be grounded to the frame in that manner at all. Same issue also cropped up w/ a grounded camera a few years back at champs that took a while to find.

Once that grounding issue to frame was cleared up...I have never seen that Anderson connector problem again. A simple ohm (00.00 meter read), check from battery to frame should tell you that you may just have the same exact problem I personally think.

I do know if they had not shut that robot power main breaker off very quickly, it could have been as bad as your pic or much, much worse. It did melt the outside insulation of the cables together right above the Anderson connector for about 2 inches (that was correctable as they pulled right apart and were still insulated properly, but those cable ends were junk), and melted slightly the connector itself...It was quite hot! That was very apparent.

Post back if you find it was a frame grounding issue please.

I don't see how a Crio grounding issue could contribute that much to the failure of the main battery connector, the wires connecting it to the PDB would take serious damage if they passed that much current.
Also if a grounded frame was passing current, you have the even bigger issue of a positive line shorting to the frame as well.

[BBray_T1296]

Originally Posted by Nirvash

Also if a grounded frame was passing current, you have the even bigger issue of a positive line shorting to the frame as well.

The same amount of current leaves (red) the battery as enters (black). Either could have melted, the red just gave first

[Nirvash]

Originally Posted by BBray_T1296

The same amount of current leaves (red) the battery as enters (black). Either could have melted, the red just gave first

What I mean is, if your frame is 'grounded', it has the same potential of the negative terminal on the battery. No current should flow across it unless something with a positive potential is also shorted to the frame.

[Michael Hill]

Originally Posted by Nirvash

What I mean is, if your frame is 'grounded', it has the same potential of the negative terminal on the battery. No current should flow across it unless something with a positive potential is also shorted to the frame.

Agreed. Just because your chassis is grounded doesn't mean you're going to have battery melting problems. Heck, the chassis of most cars and airplanes are grounded and those batteries don't melt....well not usually (I'm looking at you lithium-ion...)

[apalrd]

I do not know if there were grounding issues with either robot, but neither robot acted erratically at any time (one of them was competing in an off-season).

We know that one of them was drawing significant drivetrain current because the frame re-bent. Our 2011 robot crate was dropped while loading on it's way to the championship and the frame bent, but we hammered it straight at CMP. It re-bent at the Kettering Kickoff off-season, and melted 2 battery connectors in the same day. The replacement connector was taken from a battery as we did not have spares (they had never failed before).

The other robot was a practice robot. We geared it a bit too high, and practiced with it very hard. Eventually we added large KOP fans (the metal kind that were in the kit for a few years) to the drive motors to help them cool down (the CIM motor wires were getting very hot). The front roller was also very lossy so we threw more motors at it to get it to a reasonable speed (we had a 'joe johnson style' cross feed roller using bevel gears). The robot in general was a power hog without any additional issues. The main breaker should have tripped if we were using too much current, it did not.

During the 2012 comp season, we tripped the main breaker 3 times, then started cooling the main breaker using a cold bag of washers (set in a cooler of dry ice between matches), and replaced the battery cable to prevent failures at MSC.


The point is there are robot use cases that can melt the battery connector BEFORE the main breaker trips. This is a SAFETY ISSUE. Even if we are drawing too much current, the whole point of the main breaker is to open before any major safety issues happen.

[cglrcng]

A simple check w/ a multi-meter will tell the tale. (But I know it happened, and our Ohmmeter did not lie...There was resistance present). Removed that screw, no resistance present. No more frame grounding issue

All I can tell you is; Removing the very unnecessary grounding screw on the blunt flat end of the CRIO cured up the frame ground issue immediately. The CRIO is powered by a regulated 24V power supply from the PD Board and am told by many since the incident (we have not before or since melted any of our Anderson connectors 2.5 minutes or hours), should never touch the frame (or ground to the chassis). That screw is chromed and unpainted. (And now completely gone from all of our Team CRIO's).

Go to this link; Scroll down to the Pic below "Attach To Robot" (where they instruct teams to attach your completed Control system boards to the robot).

http://wpilib.screenstepslive.com/s/...m#!prettyPhoto

Now, make that pic larger using the symbol lower right, look at the flat end of the CRIO in the left part of the pic. You will see the grounding screw I am talking about on the right hand side. (And that one is actually very close to that frame itself though not actually touching it in the pic, but could very easily later). Ours was caused by a hasty reinstall by 2 very tired students is all.

http://wpilib.screenstepslive.com/s/...dware-overview

"The PD provides a regulated 24V power supply to power the cRIO, a special, regulated 12V power supply for powering the robot radio and a 5V power supply for powering an Axis camera."

Are you absolutely sure it could not possibly have caused that? We know why that screw grounded to frame, and we know it cured up our issue, that bot still sits in our shop and still runs today. And did the rest of that day.

[cglrcng]

Nirvash,

Ever taken a No Good CRIO apart that has been used on a competition bot before? One that has been drilled on for a season or 2, etc.....Alum. flakes are everywhere on the surface of the board. I didn't actually check that screw to see if there was any positive current present w/ the battery connected...The battery was already out and in hand when we received the bot back at the pit.

I was only saying check it first. Thousands of hours of runtime, and only a few failures, points away from the Anderson Connectors in my mind is all.

The most costly failures I have seen so far have been caused by a wayward staple (from the back of a bumper), down in a sidecar crossing about 10 pins, and some magic smoking motors turned to actual flame. But 2 major items that will frame ground are that CRIO and the rear grounding lug if used for mounting on the older cameras. Though Jags & CAN I am not at all happy with since it cost us big 2 years ago repeatedly.

I'd bet Texas Instruments could tell us why that screw is even there.