Huge Voltage spike when BMS cuts off

[Fuss]

I’m doing a few experiments at the moment. 30v spike for my 24v bank.
Let’s not discuss that BMS cutoff should not happen, the thread is about voltage spikes with lifepo4.
Do you know what is the best protection to suppress spikes with lifepo4 installations.?
I did not test yet large inductive loads like electric motors, they might produce something similar.
It would be good for everyone on the forum to have a protection solution as these spikes I suspect are coming with inverter overload shutdown, shore power overload disconnect etc.

It’s the sort of fault that’s hard to measure and possibly damages sensitive devices nice and slowly so you don’t notice until…..

Any experts on this as it would be nice to solve.

[river]

30V is not a huge spike. It is barely over charging voltage, and any equipment designed for a 24V system should already be protected from much higher voltages anyway.


If you don't trust your equipment and want to design better protection, the term to search for is automotive load dump, and the answer is probably a TVS.

[Fuss]

Yes, but I don’t have sophisticated test equipment so it probably higher than 30v if connected to an oscilloscope for example or a fluke with spike detection.

[wholybee]

Describe the system and conditions when the BMS cut off and the spike occurred. Was this during charging or during discharge? Several companies, including Balmar and Mastervolt make surge protection devices. Although they protect against much higher spikes, not 30V.

But exactly what you do to fix it would depend on what is causing it, and we need more information for that.

[Fuss]

Originally Posted by wholybee

Describe the system and conditions when the BMS cut off and the spike occurred. Was this during charging or during discharge? Several companies, including Balmar and Mastervolt make surge protection devices. Although they protect against much higher spikes, not 30V.

But exactly what you do to fix it would depend on what is causing it, and we need more information for that.

i’m experimenting, i set the charge voltage high on purpose to see what happens in a high voltage cutoff. i might later try overloading the inverter.

[Fuss]

i would like to add some device to absorb the spikes of course, it’s not that easy is it… that’s why i posted

[tanglewood]

Originally Posted by wholybee

Describe the system and conditions when the BMS cut off and the spike occurred. Was this during charging or during discharge? Several companies, including Balmar and Mastervolt make surge protection devices. Although they protect against much higher spikes, not 30V.

But exactly what you do to fix it would depend on what is causing it, and we need more information for that.

I agree. What’s charging and at what current and voltage? What loads are connected? Are you manually turning off the BMS, or is it triggering the disconnect? We really need to know your test conditions. When the disconnect happens, are only the batteries removed from the circuit and the chargers and loads are left connected?

[Fuss]

i made the bms cell voltage disconnect lower..3.4 not 3.6, then set the charger to around 20a 240v current limited so it was charging at around 45a then let the bms disconnect then i waited 10 mins and did it again.
the disconnect disconnects completely the battery

[Fuss]

My last post was written from my phone. I am not an expert on this and there is always a problem of damaged electronics which is most likely spikes...the Airmar Dst810 depth being one of the more popular examples of this at the moment.
What i think I am seeing at the moment is that the size of the spike is related to the capacity of the battery. Smaller capacity, bigger spike.
An example of starter spikes that I experienced in the past.... my original instruments were powered directly from the starter battery and were unreliable. Once I put them behind a dc dc converter, they never broke again.
Is there a DC spike suppressor that someone recommends?

[wholybee]

The size of the spike _should_ be related to the charging current, and the inductance within the charger. So the worst case would be charging with something like an alternator, and having the cutoff happen when the battery isn't full, and is thus accepting all the current the alternator can throw at it.

I would suggest a good solution is to use a Victron Orion DCDC to provide consistent voltage to your electronics, and a Balmar APD to protect against large surges that would damage the Orion.

[Maine Sail]

If you're charging with a quality switch mode charger they won't spike. If you're charging with an antique ferroresonant charger or transformer based inverter-charger or alt they can spike. A Balmar APM or Sterling APD can clamp a spike...

[Fuss]

Its an old heavy Skylla 80a so later I will test with later my Multiplus which is switch mode, but not for a few weeks.
The test changed a bit once I noticed that the size of the battery seemed to influence the size of the spike.

However with my unsophisticated Skylla test sofar, this is what I am noticing...
20ah lead acid 24v battery = larger spike
70ah lead acid 24v battery = smaller spike
older 40ah Lifepo4 with only 20ah capacity = smaller spike than the 20ah newish lead acid.
later I want to test a new 280ah Lifepo4 against the 70ah lead acid.
I have a fluke that measures inrush but not with me at the moment so these are not exactly professional tests more or less generalisations.

[team karst]

I would use a fluke scopemeter to see transient activity. Built in battery provides equip isolation

[markxengineerin]

My hamilton ferris alternator (and I would guess likely all other modern alternators made in the last 20+ years, not positive about that) already have diode(s) in them to protect against voltage spikes.

Additionally, most (but definitely not all) marine electronics have input protections and are ok with worse voltage spikes than you might think. Here is an example I accidentally recorded with my victron solar charge controller, happened when the lithium BMS disconnected due to over-charge from the alternator:


Everything was fine with the exception of a blown fuse, a couple breakers needing to be re-set, and a SCAD tank monitor brain being fried. Apparently the tank monitor was the only component on board (of many, many varying in age from the 90's to present) which did not have good input protection.

You can purchase extra components for surge protection, but based on my experience, I think it's $ better spent elsewhere.

[Dockhead]

Originally Posted by Maine Sail

If you're charging with a quality switch mode charger they won't spike. If you're charging with an antique ferroresonant charger or transformer based inverter-charger or alt they can spike. A Balmar APM or Sterling APD can clamp a spike...

Great to see you here, Rod.


What about Victron chargers? They are transformer-based, not switch mode.