So I am doing research for a side project, and I am working with high current from SLA DC batteries. I found out the maximum Amp capacity of 6 AWG wire is only 75 A. Source (https://en.wikipedia.org/wiki/American_wire_gauge, among others). I thought this was strange, as FRC robots consistently draw well over that current for extended periods. Can anyone explain this?
The duration of an FRC match isn’t really considered “extended periods” when you consider normal use for electrical wiring. Household wiring, for example, can carry current for hours and days on end without any relative change in temperature. Industrial machinery may run for months without taking a break.
In reality on our robots, we don’t normally pull 75A, with the exception of pushing matches between robots. Just driving across the field won’t do it, climbing won’t do it, shooting fuel won’t do it. When you get into those pushing matches, they’re mostly limited to a few seconds - anything over that and one team or the other is going to disengage and try to get around them. The wires can handle that, no problem.
I’d question this assertion in modern FRC, to be honest. Climbing can certainly pull around that much, depending on how many motors you have on your robot. Shooters can pull a surprising amount, too, again depending on how many motors you have on the shooter. 6CIM drives can pull that much easily when accelerating/decelerating.
That said, matches last only around two minutes.
Suggest you look at the National Electric Code article 310. This details the amperage rating of conductors and the derating methods. The current rating depends on the type of insulation used.
These amprage ratings are for continuous use. A wire can carry more current for a short period, however it will start to heat up and have increased resistance.
For a majority of teams… no. Climbing for most teams is done with a single CIM, mini-CIM, or 775-Pro. That’s not going to pull more than 40A, or you have a problem… and if you’re pushing close to that limit, you still have a problem. Plus, when you’re climbing none of your other motors are spinning!
For shooters… again, a vast majority of the one’s I’ve seen have employed a single motor. So no, not going to pull 75A while shooting. Shooting is also a stationary thing, so none of your other motors are working at the time.
As for 6-CIM drives… most team’s do go that far. Across the ~200 robots I’ve seen this season so far, I can count on one hand the number that used 6 motors on the drive train.
There are good rules of thumb that apply to most teams, and there are the exceptional cases that need to focus more on their power demands. Generally speaking, I focus on those that apply to a majority of teams.
There are no 100% efficient wires - every wire will dissipate some electrical power through heat loss. The only difference between normal wires and the heating coils in toaster ovens is by tweaking the current and resistance. Here’s a simple explanation:
Power loss through heating is P=I2R, where P = power loss, I = current, and R = resistance.
Here’s a resistance chart in Ohm’s per meter for copper wire:
http://www.daycounter.com/Calculators/AWG.phtml
If you plug in some numbers, you can see that any wire can carry any current - your only question is whether or not the wire can dissipate the power without melting. So you can send 1000 Amps through a 22 AWG wire, you’ll just make it glow red and then melt pretty quick from the immense amount of power being dissipated. Alternately, you can send microamps through a massive 6 AWG cable forever because this results in a miniscule power loss that the 6 AWG cable easily dissipates into the air around it.
The reason for this comes from the ability to dissipate the heat power - if the cable can dissipate heat faster than it builds up, you can run that indefinitely. If the cable dissipates heat slower than it builds up, you can only run that amount intermittently at best (you’ll need cool down periods, and can’t let the cable get hot enough to melt itself or the insulation), or never at worst (cable nearly instantly melts).
Applications like FRC robots fall under the latter - they only run intermittently due to the capacity of the battery and limited duration of matches - this allows wires a cool down period where they can dissipate the pent up heat.
The electric code is mostly to keep buildings from burning down. Much of the content is from buildings burning down. :] UL 508 is more applicable to machine and panels. It largely parallels and is cross referenced to the UL electrical code.
Anyway how much current you can run through a wire depends on acceptable voltage drop, the temperature rise you can accept. Temperature rise depends on current and how fast you can dissipate heat. Which is why multiple conductors in conduit might get derated. Or you can go by code where all these factors have been considered.
This is a good year to look at the DS logs. Since the game is mostly zipping up and down the field most of the game has a heavy current draw from the drive train motors. You can easily judge if your wire size is adequate by grabbing a wire after a match. If it is less than hot then it is adequate. The down side to this method is if it is inadequate, you might get a nasty burn. :ahh: Your UL safety guy will tell you better to use an IR gun or camera instead.
I know this isn’t your main point but you are forgetting about super conductors. 
However, highly impractical in most applications.
hmmm interesting… we must be an outlier… (I am sure there are many more like us…) 6 CIM drive base, two 775 pro motors for climb, and two 775 pro motors for shooter…
Not to mention 2 775 pros for ball loader and 2 for the gear floor pick up mechanism.
:ahh: That poor power distribution board! jk
I have actually melted a trailer connector by running a cim at full power for 3 hours on and off. It was a good lesson in the tolerance of our wires.
Actually there is a much better explanation.
The NEC ratings and the one linked above are determined in part by the voltage drop at the rated current for a typical length used in house design or electrical service entrance and/or the temperature rise in the insulation for continuous duty. The table above is listed for wires inside equipment. The three temperature ratings are the temperatures at which the insulation will not melt.
The ratings we use are derived from the NEC tables for open frame (not used inside equipment) DC wiring for non-continuous currents using. Wire size should be derated if bundled or run in conduit.
Bottom line: its all about the heat.
This is most likely indicative of a connector that was not sufficient for the current being carried or a connector where the contacts are damaged and do not have the sufficient mating force to have a low resistance connection.
If the current being carried by the wires is greater than what they are capable of, the PVC insulation typically used in FRC applications will get soft and droopy and possibly wet looking. At higher currents, the insulation gets wrinkly. At even higher currents, the insulation drips off and can catch fire.
“Extended periods” should be > 90% duty cycle for hours, days or weeks on end.
To summarize the considerations for wire sizing:
- Wire size versus fusing
- Wire size versus length (voltage drop for the run)
- Wire size versus insulator temperature rating / degradation
- Wire size versus temperature
For FRC robots -
-
The circuit breakers are spec’d so that the wires will not fuse.
-
Wires on FRC robots are short, and the components are fairly tolerant about voltage drops (main concern is robotrio, motors don’t care)
-
FRC robot life span is short, on the order of days or weeks of total run time. Voltage that the insulators are protecting against is pretty low (no arcflash and very little shock hazard)
-
Burns can be a concern. Wires should be spec’d so that they shouldn’t be hot to the touch after several minutes of run time.
EDIT:
Wiring Codes exist to simplify decisions. If you follow wiring code, there isn’t much to worry about. Part of engineering is doing the calculation yourself and coming to the conclusion about when whatever code costs too much (the code is “over-engineered”).
I’m not sure why you say motors don’t care about voltage drops. If you design a mechanism with a motor running at 12V but in reality the motor only gets 11V max because of resistance in the wires, you will lose 16% of your max power. That (combined with a low FoS) could potentially wreck a mechanism design.
Given a lower voltage, your motor will still operate, just not as powerfully. The roboRio or radio, on the other hand, can reboot if the voltage drops too low, which really sucks.
Much of the time on a robot, you can ignore the voltage drop from the wires. They simply aren’t long enough to give you a really significant drop. There are, of course, exceptions to that, and when you run into those situations there are actions you can take to decrease that drop.
There’s no blackout condition for motors (except when the roborio shuts them off).
FRC motors will produce torque even as low as 2V.
http://motors.vex.com/vexpro-motors/cim-motor
A mechanism might not be as tolerant as motors, but unlike other COTS parts you get to design the mechanism.
Of course the voltage drop matters more to the control system, but I wouldn’t say it doesn’t matter at all to the motors. When you design your mechanisms, you should consider the fact that your motors will often not get the full 12V because of wire resistance. Sure the motor might be able to move with 11V instead of 12, but if the mechanism is only designed to work with 12V it might not function properly. A good engineer will weigh the pros and cons of different wire sizes vs. dealing with the voltage losses. Simply saying that voltage losses don’t matter is bad engineering IMO.
Put my post in context of this thread, “Why would the GDC permit 6AWG wire for an application of 100’s of amps when code only rates it for 75? Why wouldn’t they mandate larger wires?”
In my opinion, the voltage drop was not the concern. Voltage drop is a concern for electrical code though.