Battery Leads and Resistance

[DampRobot]

I was having a friendly argument with a friend on the team today, and was hoping that I could get the CD perspective on our topic of discussion.

Is it best to keep the leads between a battery, the 120 amp breqker, and the PDB as short as physically possible, or does length not reall effect robot performance? We had to change from about 4" to about 2.5' leads on the 120 amp breaker today, and were wondering if that extra wiring would make a significant difference in performance.

We measured resistance of .06 ohms across both our old and new (longer) leads for the breaker. This rather surprised me (considering resistance should be proportional to wire length), and would indicate that wire length doesn't matter in this case. On the other hand, I've personally talked to teams like 254 and 971 who keep their breaker leads super short to reduce the resistance in the wires to the battery.

To pose the question simply: do long breaker leads effect robot performance?

[AllenGregoryIV]

Here is a good thread on the topic.

http://www.chiefdelphi.com/forums/sh...d.php?t=110210

In short, it should matter, shorter is better.

[SoftwareBug2.0]

Originally Posted by DampRobot

We measured resistance of .06 ohms across both our old and new (longer) leads for the breaker. This rather surprised me (considering resistance should be proportional to wire length), and would indicate that wire length doesn't matter in this case.

You are correct that the resistence should be proportional to the wire length. What are you using to measure resistence? A standard multimeter is probably not sufficient. To measure small resistences like that, try a milliohm meter.

As for whether it makes a noticable difference on a robot, I can't say that I've tried it.

[EricVanWyk]

You are reading close to ten times as much resistance as expected, there is something affecting your measurement.

Are you using a meter capable of measuring sub-milliohm? If it has two wires coming out of it, it is not. If it has four wires, it might.

Are you measuring just the wiring, or are you including other components such as the breaker or the connector?

[IndySam]

Just as important, copper is heavy. You should keep all your wire runs as short as possible.

In helping teams make weight at competitions, shortening wire runs is one of my favorite tricks to find those last few ounces.

[Al Skierkiewicz]

A rule of thumb I have promoted for many years is the wire foot. Simply stated a foot of #10 wire will drop 0.1 volts at 100 amps. Since #6 is twice the size of #10, the same current would drop the same voltage in two feet of #6. This is a linear relationship so you can increase the current decrease the the wire size and still come up with a rough idea of the voltage drop. Remember you must consider both sides of the wire, red and black. Since the #6 wiring carries all of the robot current, it is possible with a four CIM drive to easily pass 400 amps through this wiring. For your change you added 5 feet of wiring so the expected drop would be 0.4 volts for every 2 feet, or about 1 volt for a 400 amp load current. The following lengths are equivalent for WF calculations.
2' #6
1' #10
8" #12
6" #14

[DonRotolo]

OK, to summarize:

1. A 'normal' multimeter is not sensitive enough to measure such low resistances. There are such meters, and one can also calculate resistances using current flow and voltage drop (use 10 A for example and measure the voltage across the wire, you can detect quite small resistances that way).

2. Yes, it is a good goal to keep the highest-current-carrying wires as physically short as possible, there are measurable effects.

2a. And, copper is heavy, another reason for keeping it short.

One thing not mentioned: The connections between the wire and the other components (e.g., breaker, battery, PDB) also need to have low resistance: 0.1 Ohm there is way too high. The higher the current, the better the connections need to be.

0.01 Ohm at 100 A is 1 volt (which you lose to heat), but also this is 100 Watts, which gets real hot real fast.

[DampRobot]

Thanks for all the answers! To those that asked about our resistance measuring method, we were just using a standard cheapo multimeter. We measured both wires from their crimped ends. The measurement struck me as too high at the time because at a 150 amp draw, you would see a 9 volt drop in the battery voltage.

So, we might attribute 1 volt of the drop when were at max acceleration to the length of the wires? Ie, it would be dropping our voltage by about 10%? This seems like a significant number, especially because power is proportional to the square of voltage.

EDIT: We noticed very significant heat in the wires when we changed the batteries during some long practice sessions. So, essentially you're saying that we'd be depriving our drivetrain of hundreds of watts at full throttle, and just turning it into heat?

[sanddrag]

We use 4 gauge.

[DonRotolo]

Originally Posted by DampRobot

So, essentially you're saying that we'd be depriving our drivetrain of hundreds of watts at full throttle, and just turning it into heat?

Yep.

[Andy A.]

It's worth noting that poor crimps and loose terminals are probably a much greater source of resistance in the battery cables then the wire itself. Remember to periodically check that the stud terminals on the breaker and PDB are tight; they can and do loosen up over time. Also check whatever terminals you have put on the ends of the six gauge. A lot of teams have problems getting a reliable crimp on these much larger wires. If you go the crimp route, make sure you have an adequately sized tool and whoever does it has the strength to use it correctly.

Also, the Anderson connectors are well made but sometimes the contacts get a little charred or the crimp comes loose. Take a look at the contacts and give the wires a wiggle to verify they are clean and tight.

[Joe Ross]

Originally Posted by Andy A.

Also, the Anderson connectors are well made but sometimes the contacts get a little charred or the crimp comes loose. Take a look at the contacts and give the wires a wiggle to verify they are clean and tight.

Here's what might happen if the Anderson connector is not fully engaged: http://www.chiefdelphi.com/forums/sh...threadid=18079 (Further explanation: http://www.chiefdelphi.com/forums/sh...0&postcount=20)

[Al Skierkiewicz]

Boy Joe, you took a long ride in the way back machine. In reality what we no know about these connectors is that one failure seems to override all others. Teams who use alligator clips (those supplied on the battery charger) to connect to the battery for charging, will scratch the contact surface. This causes there to be high spots on the surface of the contact that severely limit the area through which current is passed from battery to robot. The reduced area translates to higher series resistance. Ohm's Law tells us that if you raise the resistance, while keeping the current the same, increasing power is dissipated in the area of the resistance. When you try to pass several hundred amps through that pinpoint of contact, significant heat is generated. That heat can be excessive enough to melt the housing, fuse the contact(s), heat the wire attached to the terminal and it limits the amount of current available to drive the robot. There are other failures to be sure, but contact damage is the most prevalent.

[Mr V]

Originally Posted by Andy A.

It's worth noting that poor crimps and loose terminals are probably a much greater source of resistance in the battery cables then the wire itself. Remember to periodically check that the stud terminals on the breaker and PDB are tight; they can and do loosen up over time. Also check whatever terminals you have put on the ends of the six gauge. A lot of teams have problems getting a reliable crimp on these much larger wires. If you go the crimp route, make sure you have an adequately sized tool and whoever does it has the strength to use it correctly.

Also, the Anderson connectors are well made but sometimes the contacts get a little charred or the crimp comes loose. Take a look at the contacts and give the wires a wiggle to verify they are clean and tight.

You are correct adding another 5' of wire only adds another .002 ohms of resistance when using 6a wire. In other words not enough to have a noticeable affect on performance.

[cgmv123]

.002 ohms (.004 ohms accounting for both + and -) can definitely impact performance. At 100 amps, you lose .4 volts. At 532 amps (4 CIMs at stall), you lose 2.12 volts. That's just accounting for the resistance of the wire, there are other components that cause voltage drop.