Overkill BMS and Handling Current Surges

I am currently building 2 lifepo4 batteries with 2 overkill 8S 100A BMS's. My inverter pulls a large current surge to charge its caps inside and when you connect it to say a car battery, this current is evident by a pretty large spark. My concern is that when the BMS turns off the battery because its fully discharged, then turns it on later when its charged up enough, OR maybe the charge controller turns off the load, however its wired doesn't matter, will the BMS be able to handle that inrush current? Unfortunately there is no way for me to measure it currently (hah, get it?)

It's recommended to pre-charge the inverter capacitors using a resistor before connecting a battery to it. Lots of information on this if you search the forum.

Gazoo said:

It's recommended to pre-charge the inverter capacitors using a resistor before connecting a battery to it. Lots of information on this if you search the forum.

Click to expand...

@Cookieman101 is concerned about the case where the BMS disconnects the batteries and then reconnects them with or without intervention.
Here is a scenario.
Your battery disconnects just after the sun goes down and doesn't reconnect until the next afternoon when the solar charge controller raises the battery voltage to the bms release trigger value.
The caps could be discharged in the interim.

smoothJoey said:

@Cookieman101 is concerned about the case where the BMS disconnects the batteries and then reconnects them with or without intervention.
Here is a scenario.
Your battery disconnects just after the sun goes down and doesn't reconnect until the next afternoon when the solar charge controller raises the battery voltage to the bms release trigger value.
The caps could be discharged in the interim.

Click to expand...

I have a 1500 watt@24 volt inverter and will use it with one Overkill BMS and expect the BMS to survive this use case many times.
If the user is continuously hitting the low voltage cutoff in a solar application they probably need to increase their storage or decrease their consumption.

I don't know the answer.. I thought there was some discussion about this and FilterGuy had a solution or was headed in that direction. That's why I suggested searching. Good question though.

Cookieman101 said:

I am currently building 2 lifepo4 batteries with 2 overkill 8S 100A BMS's. My inverter pulls a large current surge to charge its caps inside and when you connect it to say a car battery, this current is evident by a pretty large spark. My concern is that when the BMS turns off the battery because its fully discharged, then turns it on later when its charged up enough, OR maybe the charge controller turns off the load, however its wired doesn't matter, will the BMS be able to handle that inrush current? Unfortunately there is no way for me to measure it currently (hah, get it?)

Click to expand...

So the thread is here...lots of reading to do:


Maybe @FilterGuy could chime in here with thoughts.

I have not been able to verify it, but there are a couple indications that the BMS that Overkill sells has a pre-charge circuit built in.
* There appears to be an extra FET with a resistor in series to it. This would be a good way to build a pre-charge.
* A few people have reported an error that indicates pre-charge failed. (I saw the error once myself)

So.... there is a good chance that the BMS you have will work OK when the BMS turns on.

Note: This circuit will *not* help when physically connecting the battery to the inverter. The circuit can only help when the BMS is turning on.

If anyone can add info about the pre-charge capability, please let us know.

Well that's interesting. I wonder if this applies to the 8S BMS as well? Guess it's best to contact @OverkillSolar to confirm.

It would be nice if OverkillSolar released a spec/feature sheet on his BMS's that described things such as this, but if all else fails ill contact him.
I guess another question to bring up would be this: Do the caps on an any given inverter discharge within the time period of say one night? i just watched a Will Prowse video about precharging inverter caps, and he said that some inverters caps discharge over weeks / months. Food for thought. Also i read the other thread on this topic and i do like the delay switch idea, although i would need a high current switching device, be it electro-mechanical or solid state, supporting a current of 100A in operation.

Cookieman101 said:

Do the caps on an any given inverter discharge within the time period of say one night?

Click to expand...

If the inverter is left on, they will discharge in minutes....

Right i should've been more specific. if the inverter was turned off by some sort of switch that follows the bms shutoff, then how long would the caps stay charged. I guess it would be down to a test for my specific inverter.

I recently purchased an Overkill BMS and it has some user settings for surge including a range of times for length of surge.

I beleive the surge settings are for a short surge of current due to the load. I don't think it is related to pre-charge.

Cookieman101 said:

Right i should've been more specific. if the inverter was turned off by some sort of switch that follows the bms shutoff, then how long would the caps stay charged. I guess it would be down to a test for my specific inverter.

Click to expand...

If the inverter is on and in use when the BMS shuts off, it will very rapidly discharge the capacitors till they hit the low voltage disconnect level and it shuts off the inverter function, after that it will be powering itself for a little while from the energy in the capacitors. When the voltage gets low enough the inverter will completely shut down with some very small charge left in the capacitors. After that, the capacitors will continue to discharge through the inverter circuitry even thought the inverter logic is no longer functioning.

I have a battery disconnect and when I switch it off, it is similar to the BMS shutting off. My experience is that even if there is no load on the inverter the low voltage disconnect is nearly instant. After that is only about a second or so before the capacitors are drained enough that the inverter completely shuts off.

My working assumption is that if the BMS shuts off while the inverter is on, the inverter capacitors are completely drained. This assumption may be a little worse than reality, but it is probably closer to reality than assuming they are charged.

If the inverter is shut off and then the BMS shuts off, the capacitors start out completely charged. They will slowly discharge due to internal leakage but presumably there is no leakage through the rest of the inverter circuitry. I do not know how long they will go before they are discharged..... it probably depends primarily on the characteristics of the Capacitors used. I have never tried to measure it..... I just assume they will be discharged.

If the BMS turns off due to the cells being too low, then it probably will not turn back on till something is charging them.... in this case the charge device will have to charge the inverter capacitors to get the system voltage up, so there would not be a surge issue when the BMS turns back on.

However if the BMS turned off for other reasons (like low temp or high temp), the turn-on event might see the full surge of charging the capacitors.

FilterGuy said:

I have not been able to verify it, but there are a couple indications that the BMS that Overkill sells has a pre-charge circuit built in.
* There appears to be an extra FET with a resistor in series to it. This would be a good way to build a pre-charge.
* A few people have reported an error that indicates pre-charge failed. (I saw the error once myself)

So.... there is a good chance that the BMS you have will work OK when the BMS turns on.

Click to expand...

One of the BMS options is "LoadDetect" -- which (in the sparse documentation I could find) tries to see if there's still a short-circuit before turning the BMS back on again:
Load detect: after short protect, if BMS detect the load, then do not release, only the load removed then release protect status. After short, if load not remove then BMS cut off, if load removed then BMS turn on again.

That function might be related to the extra fet.

And my favorite quote from that doc:
if the Inrush current is too high, will trigger over current prortect, you can set the value high, to avoid Inrush current, normal use default value. if the board can work well, do not care, if triggered by Inrush current, you can set more high.


I suppose, if the board works well, one can set more high.

The BMS is only there if some unforeseen failure at the individual cell level requires disconnect. I switch the load off the inverter before a LVD shuts discharge down. LVD shuts off discharge before the BMS disconnects. I have to reset a latching relay before restarting the inverter. It is a really very quick "surge" to fill capacitors.

What I have observed: (speaking about a 12v system)

I test every BMS with a 4 cell battery and a 3500w pure sine inverter, so I've had the chance to test this from every angle.

1. If the BMS if active but not connected to the inverter, and the inverter caps are discharged, then connecting the wiring to the inverter produces a large spark, and the BMS usually remains active, no faults. Sometime this triggers a "short circuit protect" fault. This starts a 32 second delay, after which the BMS will try to reconnect. If it is still wired to the inverter there will be another inrush surge, and it may or may not trigger another fault.

2. If the BMS is inactive and is wired to the inverter, then when the BMS becomes active (either by plugging in the balance connector, or enabling discharge via the app), it will usually trigger the "short circuit protect" fault. It will try again in 32 seconds.

Sometimes the BMS will be able to charge the caps and come online without a fault, sometimes not. Sometimes it succeeds after a few attempts.

There are a few factors:
Is the inverter switched on? If so it will discharge itself between attempts and the BMS may never catch up.
The size of the inverter matters, bigger inverter, bigger caps, bigger surge.

The BMS does not have a soft start circuit of any kind.

My suggestion, if you get down to the discharge cutoff, is to switch off the inverter until the system is charged up a bit and the BMS re enables discharging.

oh, and if I switch off the inverter before disconnecting, the caps stay charged for days. No inrush next time.

OverkillSolar said:

What I have observed: (speaking about a 12v system)

I test every BMS with a 4 cell battery and a 3500w pure sine inverter, so I've had the chance to test this from every angle.

1. If the BMS if active but not connected to the inverter, and the inverter caps are discharged, then connecting the wiring to the inverter produces a large spark, and the BMS usually remains active, no faults. Sometime this triggers a "short circuit protect" fault. This starts a 32 second delay, after which the BMS will try to reconnect. If it is still wired to the inverter there will be another inrush surge, and it may or may not trigger another fault.

2. If the BMS is inactive and is wired to the inverter, then when the BMS becomes active (either by plugging in the balance connector, or enabling discharge via the app), it will usually trigger the "short circuit protect" fault. It will try again in 32 seconds.

Sometimes the BMS will be able to charge the caps and come online without a fault, sometimes not. Sometimes it succeeds after a few attempts.

There are a few factors:
Is the inverter switched on? If so it will discharge itself between attempts and the BMS may never catch up.
The size of the inverter matters, bigger inverter, bigger caps, bigger surge.

The BMS does not have a soft start circuit of any kind.

My suggestion, if you get down to the discharge cutoff, is to switch off the inverter until the system is charged up a bit and the BMS re enables discharging.

oh, and if I switch off the inverter before disconnecting, the caps stay charged for days. No inrush next time.

Click to expand...

Thank you for a very nice document on the BMS - much appreciated.

Cookieman101 said:

Right i should've been more specific. if the inverter was turned off by some sort of switch that follows the bms shutoff, then how long would the caps stay charged. I guess it would be down to a test for my specific inverter.

Click to expand...

Mine takes a couple days before the caps discharge enough to cause a spark when I reconnect.

I know this has been hashed out many times but I do my best to remember using this when reconnecting my inverter in case of a battery disconnect. Not automatic or built in but works very well
Two of these were $5. And they’re shiny. ?