During an off season event we mysteriously stopped working due to an electrical problem. The robot would lose power and reboot, we would get communications back and everything appeared to be fine until we tried to run any motor, which would cause us to lose power again.
We pulled every breaker but one and tried to see if it was a short in any of the individual circuits, but every single one of them still caused the robot to reboot. It was concluded that the problem must be the power distribution board so we ended up borrowing one from a local team after the event. We plugged this new one in but got the same results.
We then pulled all the electronics out of the robot and tested them on a table. When doing this we decided to take off the connections to the main breaker to make sure they looked fine.
When investigating the main breaker we discovered that one of the screw on connectors we were using looked rusted/corroded and had melted the electrical tape covering it. We replaced this one connector and it appears the problem is fixed. We then switched back to our own PD board and it still worked.
What do you think caused the connector to become rusted on the screw and melt the electrical tape covering it? We were using a compressed air can held upside down can to cool the breaker so that may have caused the problem. Could the connection being messed up like this cause the behavior that we experienced in the robot?
**Sorry for my phones bad quality photos
Aloha,
First off, from the looks of it the tape issues could be due to the heating up of the wires during heavy load draw from the battery. Not enough to trip the breaker but might be close. If the connection was loose it will contribute to it. Not knowing more about the robot makes it hard to know for sure
Second, the adapter you are using is more than likely steel, The wires from the connector are copper. We see this a lot here in Hawaii. This is called dissimilar metal corrosion or known as galvanic corrosion.
http://en.wikipedia.org/wiki/Galvanic_corrosion
Was the robot stored with a battery left connected?
Check the wire and make sure to clean / cut off the corroded end. And toss that adapter…
Hope this helps
Max,
This is a typical failure for this type of connector. The screw is designed for solid wire or a stranded wire that fully fills the opening or is covered in a ferrule. When you tighten the screw on stranded wire, the individual strands “ride up” the screw threads until only a few are actually be contacted at the end of the screw. In this way you are trying to force the several hundred amps through essentially a #16 or smaller equivalent wire. The result is heat. Even if you get many strands tight initially, some of the strands will work out and that will loosen the bite on the remaining strands. Again, heat is generated. What appears to be rust may actually be the plating on the parts detaching under the heat stress. That coupled with any chemical reaction with the tape and adhesive will lead to the pitted, textured look of the connector. Screw terminals that actuate a clamp to hold the wire do not suffer this failure. The screw pushes against the back of the clamp body which in turn applies the force to the stranded wire.
Are there any markings on the connector? I will bet there is something like 8-2 or 10-4. That is the range of wire that can be used with this connector. Many electricians read that as #8 solid or #2 stranded. When we had to use this type of clamp on the old brass distribution blocks, we folded over the #6 so that it would completely fill the void before we tightened the screw.
Contributing to the problems Al brought up, it’s a slot head screw, and hardly anyone knows that they need to hold the connector housing tightly (like with pliers, a wrench, or in a vise) when tightening the set screw. The result is that the screw was probably not as tight as it should be, which will result in a bad connection, with resulting heat cooking things and causing more problems.
I think the answer is to use crimp connectors, if you can fit them in, and have the proper crimping tool. Or make sure who ever is working on it understands how tight the set screw needs to be.
yes, we’ve been bit by this issue over the years…
There is a 2-14 on the connector.
So I am assuming that the connection was good enough to carry a little bit of current, enough to power the control system and nothing else, then when we tried to draw any real amount of current for motors it wasn’t able to handle it through this bad connection, which would cause the loss of power to the cRIO?
From now on its crimp connectors only!
Thanks for the help.
Screw clamp terminals can work, if you do things properly. Crimps are better.
As a former electrician, this looks like a grounding lug. With the vibrations of running the robot, it probably is not a good choice for a connection. It would work loose after a while. The dissimilar metals is also a factor in the failure you see. If you use this type, get a copper one and tighten with a torque screwdriver.
Max,
The 2-14 suggests this connector is rated to be used with #2 to #14 wire. Again, that is for solid conductors in most cases. Since the connection is loose, there is significant resistance across the connector. As you start to draw current through the connection, there will be a voltage drop across the connector. If we assume that the connection is about 0.01 ohm. Then for every 100 amps, there will be a 1 volt drop. (V=I*R) That resistance is also real so it will dissipate power (P=I^2 * R). So for every 100 amps, you will see 100 watts, pretty much the same heat as a 100 watt light bulb. Considering that a CIM draws 131 amps at stall (start) or that the battery can deliver 600 amps fully charged, you could be talking several hundred watts being developed in that small connector. I am surprised the main breaker didn’t trip just from the heat conducted by the connector.
Another practical note going forward. The 2015 PDP has a cover that protects the terminals. If a connector is too wide or too tall, it will not fit. The connector show is too tall (we tested similar connectors). The connectors sold by AndyMark and Cross The Road Electronics fit.
Is tinning the wire an option in this scenario?
Steve,
I really recommend that teams do not tin wires for several reasons. In the case of the WAGO connectors, the manufacturer has designed the terminals for stranded specifically or wire terminated with their ferrules which are square. For other screw type terminals, the tinning prevents the screw from biting in a uniform way and if the wire were to turn, the connection would be loose. In the new hardware, tinning would again prevent a good contact and tight retention in the push type contacts. I only tin wires that are intended to be soldered into terminals meant for solder.
