Voltage sag followed by regen

[harrisonpatm]

In my particular build, voltage sag is kept to a minimum, but if I want to accelerate hard and full throttle for some reason, I'll see as much as a 10% drop. It doesn't trigger LVC on the controller or BMS, just wasted energy and not great for battery life. I don't like it, I'll plan my next build better.

Then what's fun for me, since I have automatic regen setup, as soon as I let go of the throttle, regen starts and voltage immediately jumps back to practically resting state. Which is neat.

The teaching is that voltage sag is "okay" as long as the battery is returning to normal voltage after the high load disengages. This is an indicator of a healthy battery that's got a lot of life left. Does having regen setup the way I do, assisting the voltage rebound, contribute to its lifespan? Not expecting a hard and fast answer, I was just curious if there are some thoughts on the subject. It really is fun for me to see voltage sag immediately disappear the second I decelerate, and I wanted to see other people's experience on it.

 

[amberwolf]

If the charge rate of the regen is high enough to be close to or more than the typical charge rate for the cells used (multiplied by the number in parallel), then may contribute to pack degradation over time (probalby not by much given the short periods it happens at).

Whether the recharge after sag helps lifespan or not, I don't know--not sure it's been tested, and would require a pair of identical systems used the same way except pne with and one without the regen to find out. (be that a pair of bikes actually used on the road or a pair of lab setups with the equivalent loading and environmental conditions).

 

[harrisonpatm]

If the charge rate of the regen is high enough to be close to or more than the typical charge rate for the cells used (multiplied by the number in parallel), then may contribute to pack degradation over time

It's not, I turned it down to a below-safe level because longevity is more important to me than performance in this case.

 

[neptronix]

Unfortunately the battery voltage jumping up significantly is another form of voltage sag and the energy is wasted coming into the battery and also you can also produce a good amount of waste heat that way too.

Charge rates are typically 1/4th or less of the rated discharge BTW.

Turning down the amps is wise for the sake of the life of the battery.

 

[harrisonpatm]

Unfortunately the battery voltage jumping up significantly is another form of voltage sag and the energy is wasted coming into the battery and also you can also produce a good amount of waste heat that way too.

Can you explain this further? Or perhaps quantify how much is wasted? Presumably if it was all wasted, regen would never be worth it.

Charge rates are typically 1/4th or less of the rated discharge BTW.

Turning down the amps is wise for the sake of the life of the battery.

Like I mentioned, it's already turned down to a safe level.

 

[amberwolf]

Can you explain this further? Or perhaps quantify how much is wasted? Presumably if it was all wasted, regen would never be worth it.

Voltage sag during discharge from a cell is caused by cell resistance, so current flow thru the cells causes a reverse voltage inside them that "cancels out" part of the voltage they generate. This creates waste heat within the resistance.

Voltage rise with current flow back into the cell (charging or regen) is caused by the same thing in reverse, if the rise ceases and then drops back to a lower stable voltage once the current flow back into the cell stops.

So both of them waste power as heat. The higher the current, and/or the higher the cell resistance (and the interconnects, etc), the greater the wasted power.

Voltage rise that *stays* at that voltage isn't the same, and is just recharged capacity.

 

[harrisonpatm]

Voltage rise with current flow back into the cell (charging or regen) is caused by the same thing in reverse, if the rise ceases and then drops back to a lower stable voltage once the current flow back into the cell stops.

But voltage will rise and return to resting state (in a healthy battery) even without charging or regen, once the load decreases or stops entirely.

Because if I throttle back, not allow regen to kick in yet, just lower the load, voltage still rises back up pretty quickly. Cuz this is a relatively new and healthy battery.

Roughly, then, what percent of the return to resting voltage is due to the battery's natural behavior, and what percent is due to regen? 80/20, perhaps? Depends on the battery's health?

 

[amberwolf]

But voltage will rise and return to resting state (in a healthy battery) even without charging or regen, once the load decreases or stops entirely.

Yes, that is because voltage sag is only occuring during load (current out of the battery).

Voltage rise (from regen) is only occuring during regen (current into the battery).

What you are seeing is first sag, then simultaneous regen and recovery from sag, then recovery from regen. If you were to log this as a chart, it would show you a curve going down then up then down again.

Without the regen it would just be down then up.

Exactly what the curve looks like and where it ends up past the curve depends on the specifics of the system and events.

Roughly, then, what percent of the return to resting voltage is due to the battery's natural behavior, and what percent is due to regen? 80/20, perhaps? Depends on the battery's health?

You can test that, by disabling regen then doing a ride (or a bench test using an equivalent load) while monitoring voltages under the same conditions and state of charge, comparing the exact amount of rise / fall and final stable point in each case, and retesting the same thing with the regen enabled (or bench test with equivalent load and charging sources).

Then you can compare the results to find the amount of recovery caused by each thing.

Probably tough to do this with a ride; hard to exactly duplicate the loading/etc in two runs. A bench setup would be easier but may require automated control of the loading/etc to identically generate pairs of runs to compare accurately.

 

[harrisonpatm]

ugh, sounds like homework!

 

[Adrian_]

Regen on it's own decreases the lifespan of a battery so I doubt speeding up voltage sag recovery decreases it but a lot.

I would watch out when going down a hill though. For example if you do a full throttle pull down a hill or just before you get to the top and start regen down before voltage sag recovery. You could face some issues if regen causes the battery voltage to instantly jump by 6v or 8v in total, above it's resting voltage. You stated you have 10% voltage sag but didn't state what voltage system you use so I'm guessing 48v in this scenario.

 

[harrisonpatm]

You stated you have 10% voltage sag but didn't state what voltage system you use so I'm guessing 48v in this scenario.

72v nominal, or 24s LiFePO4. I never see sag lower than 72V, and even 73-74v is only occasional. So from 79V actual resting, hard pull all the way down to 72V, still only 9% at the most. Like I saId, could be better, but its my first build and this is a learning process.

 

[neptronix]

Unfortunately the battery voltage jumping up significantly is another form of voltage sag and the energy is wasted coming into the battery and also you can also produce a good amount of waste heat that way too.

Can you explain this further? Or perhaps quantify how much is wasted? Presumably if it was all wasted, regen would never be worth it.

Yes, the amount of voltage lost is the percentage of difference between voltage at rest ( 79v ) and voltage at load ( 72v )
79v dipping to 72v = an 8% loss of the battery's energy at that level of load.

That loss turns into heat in the cell. Too much heat can shorten a cell's life significantly, and as cells age or have cycles put on them, their internal resistance increases and the sag ( and therefore heat waste ) gets worse as time goes on.

Same factor when you're regeneratively braking. That voltage differential is an energy waste when charging also. Resistance wastes energy in both directions.