LIFEPO4... Small BMS for smaller loads, bypassing BMS for large loads

[ericfx1984]

So I have a question so I have a relatively large 12 volt system currently on FLA golf cart batteries

And I'm seriously considering going to LIFEPO4 but one of my biggest Hang-Ups is that I have a 3000 watt inverter.. and while I rarely go above 2,000 w it would be nice to have the capability for things like starting an air compressor or kicking a motor over

It seems like as long as you stay 200 amps or less on the BMS it's reasonably affordable but the second you step up to anything above 200 amps the price steps up pretty quickly

Now I know one option would be simply to build something in the hundred to 200 amp range and then add another battery Bank a little later

One person even suggested leaving my existing fla battery Bank in place to handle those relatively short but High amp draw loads... Well this is intriguing it seems like it would require some electrical wizardry and quite frankly I'm lazy these days

But then I read about the option of installing your high amp draw directly to Lifepo4 battery Bank and then everything else would be installed after the BMS this still allow the affordable BMS to handle charging and balancing... While still allowing significantly higher current draws

This seems like a workable option however I would think that you would still have a strong possibility of over discharging one or multiple cells especially if they weren't particularly well matched...

So I don't know

 

[DThames]

You can parallel BMSs. Super capacitors at the inverter input will help start loads. Not good if it is truely a large load, not just a start load.

 

[DJSmiley]

Biggest power user by far is an inverter.

If you're installing it directly on the cells, before the BMS, you aren't limited by the max current the BMS can provide, only limit is the cells discharge capability.

And yes, you don't have any cell monitoring. The inverter has its own low-voltage shutoff, but that isn't aware of any unbalanced cells, whereas a BMS is.

However, this can be (ratherly easily) fixed: Most inverters do have just a dumb on-off switch
Replace that switch with a simple low-current relay.

Use the existing switch to power the relay (rewire it), with the power provided by the BMS.

So: If BMS detects any issue, power is released, thus the relay is released, and inverter does shutoff.
You do have the features of the BMS, without the current passing the BMS. Much cheaper than a heavy BMS, and no issues with any surge currents, as long as the wires/cells can handle it.

Only disadvantage: the BMS isn't registering the current, thus any SOC on the BMS isn't properly displayed. But in my opinion they suck anyway, so an external SOC meter with a shunt is the way to go. (And that WILL register the current from the inverter as well)

Some other devices do have a remote on/of which you can use for the same purpose It's not only the low voltage disconnect, but also the high voltage if it's an inverter/charger combination.

Paralleling BMS is also an option, added benefits is redunancy, if 1 BMS fails, you still have power (altough the max current is limited due to 1 bms missing)

 

[Short_Shot]

You can just power a relay/contactor from the BMS so that the large load can come through that, but when the bms shuts off it cuts the contractor...

The contactor coil will pull some current on its own but likely will be insignificant compared to the load.

You'll also need to put a shunt between the battery and bms/contactor if you intend to monitor capacity. This has a downside of a tiny power drain still in place after the bms cutting off but should be insignificant.

Edit: I see everything I just said was already covered lol

 

[DLTooley]

I’d be comfortable avoiding the BMS for a short peak, but not for a typical heavy use of 2000 watt loads.

This is not my area of expertise, but it sounds like a super capacitor might be the ticket. I’ve no idea how user friendly that tech has become.

Another question - is it possible to parallel a bms on the same battery?

 

[chrisski]

Could you parallel smaller batteries each with a 200 amp BMS?

If you buy two BMS of 200 amps, that would be 400 amps. If you’d intended on a single BMS with four 200 ah cells, two BMSs with twice the number of batteries, two BMS each with four 100 ah cells will get you what you want.

I may pull up to 150 amps, and I making two batteries each with a 100 amp BMS.

 

[grizzzman]

I’d be comfortable avoiding the BMS for a short peak, but not for a typical heavy use of 2000 watt loads.

This is not my area of expertise, but it sounds like a super capacitor might be the ticket. I’ve no idea how user friendly that tech has become.

Another question - is it possible to parallel a bms on the same battery?

Answer on second question is yes. I have done this in the past. I have also provided a relay to allow an inverter to be powered directly from the battery. It worked fine. With that said. I was not comfortable with doing this so I moved on and the system amps can be raised as needed.

 

[ericfx1984]

It seems like a simple way to do this on a budget would be a set of 4- 280 ah batteries and 4, cheap 100 amp 4s BMS(s)

Then maybe add 4 More later... Big advantage here is redundancy... If one BMS should fail, the other 3 SHOULD be able to carry the load

 

[Short_Shot]

I'm not sure why this is being overcomplicated. Using the BMS to power a relay/contactor is the solution.

Its vastly cheaper than multiple BMS units, and still provides the cutoff functionality of the BMS to the higher inverter load. Its no different than the many BMS designs on the market which just use a relay instead of a mosfet for the load.

 

[ericfx1984]

It's probably being overcomplicated because I was not aware of such a thing

 

[BretS]

One of the many reasons I went with a REC active BMS... Contactor type with no current concerns.