Limit current to each battery string in a system with multiple parallel batteries.

[uzevla]

Lets say this is my starting system:
https://d31ezp3r8jwmks.cloudfront.net/e63qawtqq2yyqmd6x0i38axlms1a
A connector box w/ multiple MPPTs, batteries in parallel, 48V....
Plan to expand this system with even more MPPTs and battery strings (e.g. plan to have about 35kW of solar, 65kWh of batteries). The current coming from the solar will be too high for each battery string. So, if any imbalance happens, and the current becomes to high for a string, it's BMS will trigger. Likely then other BMSs will trigger.

I would like to have some kind of current limiter on each battery string so that the BMS doesn't trigger due the imbalance. Is there some standard practice when it comes to handling this ? Tried reading on parallel BMSs but didn't get too far. Maybe some kind of current source (PWM?) in each battery string, that limits the current to the batteries spec.

Or maybe nothing is needed; in normal operation, w/ batteries having similar spec, and with BMS in each string, the current imbalance will be too small and will not cause any overload. E.g. 500A into 4 strings, each handing 200A, should allow for quite some imbalance. Thus, I shouldn't expect BMS to trigger on overload unless there is a really a faulty condition, which is ok.

 

[Supervstech]

Solar does not push amps.
Charge controllers PULL amps.
As much as the batteries need, within limits of the chargers.

35kW of solar is a LOT.
Are they facing same direction?
On a good day, thats over 190kWh of production.
Wow.

 

[Supervstech]

How many controllers will you have on this array?
With nearly 800A of potential solar, i would love to see the setup!

Pics?

 

[uzevla]

@Supervstech I am still working on the overal design, trying to understand all the details. Hoping to have it built later this year. Professional electrical engineer will work on this if I get stuck (have electronics background, career in programming), but i do want to understand the system as much as possible, in any case.

Lets make assumption it'll be like in the picture with possibly 3x solar, 2x batteries, to make it a bit smaller for discussion. All the PVs are facing the same directions.

Lets say I have 6 85A chargers and plan to allow up too 200A into each battery string (4 of them). You wrote "As much as the batteries need, within limits of the chargers." . But I am concerned with the opposite situation (hope I got your comment right) - chargers can provide more than what eeach battery string can handle, thus I need to limit the current in each string. Don't want one string to overload due to any imbalance and keep triggering BMS

Related to 800A comment - should I maybe bump up the voltage to around 100V ?
Thanks a lot!

 

[Supervstech]

The controllers dont push amps, they simply provide what the batteries draw.

 

[PreppenWolf]

I like the parking lot analogy for how batteries charge.

Your battery is a giant parking garage, like at an airport.

Electrons are cars.

Chargers are where the cars come from. The charger has an infinite number of cars (you can think of it like an even bigger parking garage that is trying to kick everyone out.

If your battery is empty and the charge controller turns on, this is like cars going from a full garage to the new garage. They need to find a parking spot. At first thr parking spots are immediately available and electron cars very quickly pick spots in thr battery garage. But as the battery garage has less and less spots available, thr cars start to drive around in circles looking for spots, slowing how many cars can leave the charger station and enter the battery.

Thus is viewed on a multimeter as current flow in rush very high, and then tapering off as it's harder and harder to find a parking spot. Until the battery garage is full and there is no where else to park. At that point there are still cars in the charge lot that want to move, but there is no where for them to go.

The current is limited solely by how fast parking spots can be found. Not how many cars the charger wants to kick out.

 

[Supervstech]

I like the parking lot analogy for how batteries charge.

Your battery is a giant parking garage, like at an airport.

Electrons are cars.

Chargers are where the cars come from. The charger has an infinite number of cars (you can think of it like an even bigger parking garage that is trying to kick everyone out.

If your battery is empty and the charge controller turns on, this is like cars going from a full garage to the new garage. They need to find a parking spot. At first thr parking spots are immediately available and electron cars very quickly pick spots in thr battery garage. But as the battery garage has less and less spots available, thr cars start to drive around in circles looking for spots, slowing how many cars can leave the charger station and enter the battery.

Thus is viewed on a multimeter as current flow in rush very high, and then tapering off as it's harder and harder to find a parking spot. Until the battery garage is full and there is no where else to park. At that point there are still cars in the charge lot that want to move, but there is no where for them to go.

The current is limited solely by how fast parking spots can be found. Not how many cars the charger wants to kick out.

Yeah, and once the lot is near capacity, the attendant closes most of the entrances, and starts directing cars where to park, slowing the exchange.

 

[SoggyShoes]

I like the parking lot analogy for how batteries charge.

Your battery is a giant parking garage, like at an airport.

Electrons are cars.

Chargers are where the cars come from. The charger has an infinite number of cars (you can think of it like an even bigger parking garage that is trying to kick everyone out.

If your battery is empty and the charge controller turns on, this is like cars going from a full garage to the new garage. They need to find a parking spot. At first thr parking spots are immediately available and electron cars very quickly pick spots in thr battery garage. But as the battery garage has less and less spots available, thr cars start to drive around in circles looking for spots, slowing how many cars can leave the charger station and enter the battery.

Thus is viewed on a multimeter as current flow in rush very high, and then tapering off as it's harder and harder to find a parking spot. Until the battery garage is full and there is no where else to park. At that point there are still cars in the charge lot that want to move, but there is no where for them to go.

The current is limited solely by how fast parking spots can be found. Not how many cars the charger wants to kick out.

@PreppenWolf that is such a cool analogy! Thanks.
To me, the real masters are not the ones who who understand the complexities, but the ones who understand the complexities so well that they can put it in terms that the rest of us can have an idea of what's going on.

 

[Supervstech]

@PreppenWolf that is such a cool analogy! Thanks.
To me, the real masters are not the ones who who understand the complexities, but the ones who understand the complexities so well that they can put it in terms that the rest of us can have an idea of what's going on.

Totally agree.
A genius that cannot communicate is only impressing themselves.

 

[再生可能 エネルギー]

@Supervstech I am still working on the overal design, trying to understand all the details. Hoping to have it built later this year.
Professional electrical engineer will work on this if I get stuck (have electronics background, career in programming),
but i do want to understand the system as much as possible, in any case.

Just to add to the fray, I would recommend reviewing some of the 'Off-Grid Garage' videos (here is a playlists summary)

The whole 'Off-Grid Garage' Solar System uses three batteries racks, so you migh find some clues regarding your quest.
I would like to add that using various types of BMS and reviewing their settings, was a particular subject analysed by the author.

As a good start, you can look at: How I've built the Off-Grid Garage. Whole system installation and design explained.

 

[Andrewr05]

I like the parking lot analogy for how batteries charge.

Your battery is a giant parking garage, like at an airport.

Electrons are cars.

Chargers are where the cars come from. The charger has an infinite number of cars (you can think of it like an even bigger parking garage that is trying to kick everyone out.

If your battery is empty and the charge controller turns on, this is like cars going from a full garage to the new garage. They need to find a parking spot. At first thr parking spots are immediately available and electron cars very quickly pick spots in thr battery garage. But as the battery garage has less and less spots available, thr cars start to drive around in circles looking for spots, slowing how many cars can leave the charger station and enter the battery.

Thus is viewed on a multimeter as current flow in rush very high, and then tapering off as it's harder and harder to find a parking spot. Until the battery garage is full and there is no where else to park. At that point there are still cars in the charge lot that want to move, but there is no where for them to go.

The current is limited solely by how fast parking spots can be found. Not how many cars the charger wants to kick out.

Huh never thought of it like that, but that's a great laymans explanation.

Sort of like how I describe to people the difference of computer storage versus computer memory.

Imagine sitting at a desk.
Your storage drives (hard drive or solid state drive) are the drawers of your desk.
and the files are in the "file folders" of those accessible drawers.
The more drawers you have the more storage you have and the things you access more often are stored at the front of your drawers.

Your computers memory "RAM" is the physical size of your desks usable area, the literal desk top.
The more surface area the top of your desk has then the more "room" you have to do your paperwork, projects etc i.e. the more things you can work on at once without having to take things in and out of your drawers each time you need them.

 

[SupraSPL]

It is possible for a BMS overcurrent to be triggered when one battery inevitably fills up first and tapers off, the remaining battery eventually could take too much current as it is topping off. The solution to that is to have a large enough battery system that excess current cannot overwhelm any of the BMSs. As long as that is the case, they will balance safely on their own every time they are topping up.

I would describe the panels as electron pumps. They create the voltage/pressure. The MPPT is like a valve, limiting both the voltage/pressure and the current/flow rate into the batteries. Therefore it is possible to overcharge the batteries if the MPPT voltage or current were set too high or if the battery were too small.

 

[uzevla]

@Supervstech @PreppenWolf
I do agree the charge controller doesn't push the voltage and can simply provide what the battery draws.
BUT, an mostly empty battery will draw too much and it will overheat - it's resistance is simply too low.

That's why a safe charging speed spec exists for all kinds of batteries, as a charger shouldn't give a battery all it can draw. For Li-Ion it's about 1CA. In my case (304AH battery), it's 300A, and 800A will overheat the battery. Thats why BMS overcurrent/overcharge protection exists. Battery simply must not get all it can draw.

To make an analogy:
When 300 cars have to fill a garage with 300 spots they can do it at 10mph or 100mph.
If doing at 100mph they will fill the garage very quickly, which is great but all those cars would crash when taking corners, or can burn the tyres causing the fire hazard in the garage.

Lets say I have 4 battery strings, 3 have the same resistance, 4th one is half that value. All batteries are almost empty, can draw a lot and I have 800A avaiable from the chargers This means first 3 batteries will draw 160A, the last one will draw 320A. Thats above the max safe current.

The referenced video is great as it comes somewhat close to my system but 140A changes things a lot.
My example with one battery having half the resistance results in these currents: 35A, 35A, 70A. No BMS gets triggered. Note that I have no idea how big the imbalance can be - just showing it's a less of a problem for lower currents. If some normal imbalanse (due to all batteries being slightly different) is never more than e.g. 20%, I am just fine even with 800A.

 

[AntronX]

I would like to have some kind of current limiter on each battery string so that the BMS doesn't trigger due the imbalance

Batteries will self-balance and share current about equally. They have plenty of internal resistance for that. I would not worry about it. Try it and see for yourself.

 

[timselectric]

I do agree the charge controller doesn't push the voltage and do simply provide what the battery draws.

It does push the voltage.
But the battery must draw the amps.

 

[uzevla]

It is possible for a BMS overcurrent to be triggered when one battery inevitably fills up first and tapers off, the remaining battery eventually could take too much current as it is topping off. The solution to that is to have a large enough battery system that excess current cannot overwhelm any of the BMSs. As long as that is the case, they will balance safely on their own every time they are topping up.

I would describe the panels as electron pumps. They create the voltage/pressure. The MPPT is like a valve, limiting both the voltage/pressure and the current/flow rate into the batteries. Therefore it is possible to overcharge the batteries if the MPPT voltage or current were set too high or if the battery were too small.

Just replied with a very long text, before seeing this one. Thanks a lot. I am just not clear if my battery system is "large enough"? For this I need to know the what usually happens during the imbalance and how big it usually gets. I don't know any of that, so theoretically I just came up with the current limiter, to solve the problem "by design" and not "by experience"

 

[uzevla]

Batteries will self-balance and share current about equally. They have plenty of internal resistance for that. I would not worry about it. Try it and see for yourself.

Thanks, will go with that assumption. I just went with some theory, assuming that resistance might get really low.

 

[SupraSPL]

I could see that being a concern with the numbers you described, 35 kW of solar and only 65 kWh of battery.

To give some perspective, my system is 4.2kW with 40kWh of battery. That results in a maximum charge rate of about 0.2C (I've seen 1500W going into a 7.7kWh battery) under these rare conditions: no load on the system from the inverters, full sun and one last battery is in the process of topping off.

In my case I have 100 amp BMSs on each 25.6V battery, maximum each BMS can handle is maybe 2800W. So despite having 5 batteries in parallel, made from different cell manufacturers with slightly different capacities, that charge with different voltage curves, that's never a problem with a panel/battery ratio of that size.

 

[teal95]

Your math is a bit off, it's ~2 hours charge time 1/2 c.

 

[SupraSPL]

Corrected thnx