# Battery Power Consumption

What is the best way to measure power consumption on the robot using a 12 volt battery? Our team is worried that we will be taking up too much power during a match and we want to have enough power to last us through the entire match. Is there a calculator that we could use to do this? Should we be concerned about this?

It is definitely a factor in the way that an end-of-match battery will have a lower maximum voltage and the voltage will drop more as the match goes on. As for literally running out of juice, you’d have to really try. 240 A (DOUBLE the nominal main breaker capacity**) * (2 minutes 30 seconds) is only 10 Ah (our batteries have a nominal capacity of 18 Ah.) EDIT: This is incorrect. Refer to smarter people below me.

The main breaker can take more than 120A in spikes, but a sustained draw of 240A would pop the breaker in less than 30 seconds according to the data sheet. So even in this unrealistic scenario, you’ll only use half the capacity of the battery.

I’m not a battery expert, so please correct me if there are more nuances to this situation.

As to my first point (that voltage will decrease) you do need to think about that, like using voltage compensation in your software, or using PID and sensors to be independent of percent voltage.

So really if you’re concerned about draining the battery, instead be concerned about popping the main breaker.

It’s not super easy to calculate. Datasheets list out some info:

http://files.andymark.com/am-3062.pdf , specifically the discharge characteristic curve.

Ultimately, what you care about is whether the maximum load on your battery will cause brownout conditions (Vbat < ~6.5v) by end of match. This requires knowing both what your robot did to the battery throughout the match, as well as the age & condition of the battery, and what sorts of loads your robot can produce.

At any given moment in time, the voltage your battery supplies is a function of both its present internal parameters (OC voltage, equivalent internal resistance, and others), as well as how much current you draw from it.

For a fully charged battery, pulling a transient 100A will drop the voltage by an amount. Probably not enough to be concerning, but somewhat.

For a partially or fully discharged battery, that same transient 100A load will pull the voltage down a lot more. This is where you hit issues with brownouts.

I’m not sure if I have a good way to calculate this beforehand. The models I know of and data I’ve seen is all good in the steady state, and tends to fall short in the transient analysis. I suppose a good transient analysis would require modeling portions of the chemistry of the battery, which is well beyond scope of what I know I can do.

I do know that some battery testers can do pretty advanced cycles to discharge the battery. If you have a good idea of what your transient load profiles will look like, and a lot of money, you could get one of these. I don’t really know what sorts are the best though. I think Team 16 has had one in their pits in years past? You could potentially use a setup like this to empirically test if your load profile can be supported by the battery.

<EDIT: Removed incorrect information based on poor research about what a “CA” was.>

http://batteryuniversity.com/learn/article/bu_503_how_to_calculate_battery_runtime may be a good source of additional info, although I feel I’m not qualified to fully interpret the info which is contained there…

As nickbrickmaster mentioned though, the good news is the main breaker helps keep you out of the “dead battery” territory by putting somewhat of an upper limit on how much current you can draw steady-state.

It is impossible to accurately predict how much of the battery SOC will be used because the loads are going to vary significantly over the time of a match. Sometimes you’ll be drawing every amp that the system is able to deliver and other times a small fraction of that. The 18ah rating of the battery isn’t really that useful in calculating the time that it will last because of that varying current demand. Discharge it at say 10a for 10 min and you will have a higher final SOC than discharging it at 100a for 1 min, even though 10x10=100x1.

However with the 4cim drive of the kitbot you will not have a problem of running out of battery if you start the match with a fully charged (12.77v) battery unless you have a serious amount of other motors that are being used with a very high duty cycle.

One thing you can do to extend your useable battery life is to ignore all of the people who encourage you to upsize the battery cables and wires between the PDP and motors. The slight non-noticeable increase in performance will be offset by a potentially noticeably shorter usable battery life.

After you build it, run a full practice match as hard as you can.
Then you will have a graph of battery usage in the DS log file.

This can be done if you have an older robot that can be mocked up with the number/type of motors that you plan to use.

Here’s a good read on the topic:

https://wpilib.screenstepslive.com/s/4485/m/24166/l/289498

I don’t see how this affects your battery life. It increases your ability to push maybe 5 extra amps to your motors, which maybe increases your torque by a percent or two and maybe it’ll also increase your rate of discharge and wear your battery down faster.

Running wire larger than 6GA isn’t going to have any noticeable effect on either performance or battery life in the FRC

realm. 99% of your battery life is the battery’s condition and how many high-amp motors you use. I call BS.

The resistance of your wires to the battery is impacted not just by gauge but by the length of the wires. If the wires are long increasing the gauge can offset the resistance.

The teams that tend to oversize their wiring also tend to run 6 cim drives and lots of other high power motors that do occasionally drain their battery to the point that it’s voltage sag can cause brown outs.

So yes it does increase the rate of discharge and you run out of battery just a little bit sooner. It also allows you to unintentionally draw your battery down to just a little lower SOC which affects the number of cycles it will last.

When that does occur it will be noticeable while the fact that you shaved .01 sec off your cross field sprint will never be noticeable.

You are ignoring that the internal resistance of the wire also acts like a load resistor. I would vastly prefer to move closer to the engineering fantasy of zero resistance wiring and consume all my available power actually doing something other than contributing to global warming on the field. The manual suggested minimum wire sizes will get noticeably hot to the touch after hard matches. Upsizing your high power wiring runs will reduce this. The key fact is that they will reduce this for exactly the same current draw from the battery, delivering more of your precious power to the actual motors.

You also need to consider your power budget through the match. The endgame is often a high power draw activity. For example last year we played with different motors and gearing on our climber after the first event to try and speed it up. The fastest one was designed at the edge of current limits from the PDU, and geared for the optimal motor rpm power point. It provided a faster climb with a good battery. It provided a faster climb on what we believed represented an average match. It failed on a higher than average match. End result was we went back to the slightly slower setup that successfully climbed under all endmatch battery conditions.

However, the biggest factor to battery performance is where in its life cycle the battery is, and how well you manage the charge/discharge cycle. Best option is to buy a fancy load tester and actually manage/test your batteries. Reasonable compromise is to use good battery chargers, have a battery management system so you rotate through all your batteries, and don’t let them sit discharged in the offseason.