What's best device for charge limit control? BMS or Device?

just thinking about controlling the charge level of a lifepo4 battery

does it matter whether the BMS is primary device to limit the battery charger by constantly tripping with the charge or low voltage limits

vs the charge controller or inverter/charger hitting its limits?

best to have a charger with custom settings.

configure charger to stop charging at 3.525 volts per lifepo4 cell.

That means four cells = 3.525 volts per cell * 4 cells = 14.1 volts total

BMS can’t achieve fine control of charge, just close the door and lock it.

hope this helps. limit at charger.

I'm going to say it depends.

Most here on the forum believe that the BMS should be the controller of last resort. I have no problem with this statement at all.

Others here say let the BMS controll more because it knows on a cell by cell basis how the battery is doing. I have no problem with this either.

To me what it comes down to is how matched is your battery bank. If you have a battery bank that is well matched and well balanced you probably do not need your BMS to cut off charging. If your cells are old or mismatched then most likely you will need your BMS to cut off charging when any one cell gets too high. If you leave it to the Charge controller you will most likely have to set the max charge voltage so low that you lose capacity.

I have battery banks in both situations and in both situations I have never had a problem with either he SCC or the BMS not doing their thing properly.

Sorry this isnt a clear answer but maybe it will give some insight to make your own decision based on your circumstances..

I try to have redundancy .... especially for over charge and low voltage disconnect. I mix whether the primary is the BMS or inverter ... or charger.
Budget issues are the primary drivers of this choice for me.

Thanks -- all makes sense --

in my setup i am trying to operate (set) the operating range of the batteries to be roughly 20% to 80%

my Inverter charger tops out at 14.2v and can't more finely adjust
my solar cc tops out at 14.4v
and based on my testing, i think i want the charging to top out around 14.0 to 14.1 (3.5 to 3.525) to achieve about 80%

so , i was thinking to have the BMS try to keep in the range...

...unless there was a reason not to?

All a BMS can do is shut off the charge when a cell goes over a set voltage. This is not the best way to charge. LFP cells as well as Lead acid and other Li Ion types all benefit from a absorption charge where the voltage is held at your maximum charge voltage for some time as the current tapers off. This dropping current will also give more time for the balancer to do it's job and get all of the cells to top balance as the charge comes to an end. A normal BMS just can't to that. Now if the cells are too far out of balance, and one cell does hit maximum voltage, then yes, the BMS needs to step in to protect that cell from being over charged and causing damage.

The discharge side is basically the same thing. Have your less important loads shut off with some power left, have the important loads able to run it a bit lower if needed. If the battery starts getting too low, shut it off. The BMS is there to shut off the loads in case a single cell goes too low while the loads are still seeing a safe overall pack voltage.

If you are not trying to get the maximum capacity, it is okay to not do an absorption cycle, but then you should have the balancer active at the lower voltage to keep them all close. Most balancers only work near the top voltage during charging.

If there is a consensus ... or vote -- for decently matched cells
to get as close to an 20/80% charge window

i current have set at devices...

Bulk / absorb / Float
14.2 / 14.2 / 13.8 - Inverter/Charger Setting -- this one i cannot change lower than 14.2
14.2 / 14.2 / 13.8 Solar Charge Controller - using the MT50, i think i can change this to 14.1

should i just go with these settings, should be about 90%? to 17% or so?

---------------

If you have an accurate SOC indication .... just monitor that to see what your voltages are at the SOC you are targeting.

GXMnow said:

LFP cells as well as Lead acid and other Li Ion types all benefit from a absorption charge where the voltage is held at your maximum charge voltage for some time as the current tapers off.

Click to expand...

If the charge voltage and the battery voltage are equal then no current will flow.
For the big blue prismatic cells that are popular here the tail current is specified at .05c which works out to 14 amps.

During the constant current phase the charger controls the current flow by adjusting the charge voltage relative to the battery.

The constant voltage phase begins when the charger no longer needs to adjust the charge voltage dow to maintain the configured current flow. During this phase the battery voltage converges with the charge voltage asymptotically and current flow decreases correspondingly.

To summarize, the charger controls the current flow during the constant current phase and the battery determines the current flow during the constant voltage phase.

Yes, but if the BMS just turns off at the absorb voltage, then there will be no constant voltage absorb charge at all.

GXMnow said:

Yes, but if the BMS just turns off at the absorb voltage, then there will be no constant voltage absorb charge at all.

Click to expand...

I'm not sure how necessary the absorb charge is with lifepo4 cells. From many tests I have not seen any difference in capacity.

If a person has plenty of capacity and is looking to maximize pack longevity, I think just cutting off the charge is a good thing ..... If needing to maximize charge floating it may be best.
There are many different requirements .... it's just not a fixed solution.

No mention of dual port BMS, so if the BMS goes into over voltage disconnect, better hope there's enough solar generation instantly available (there won't be, the charger has a reaction time) to carry the load otherwise everything falls in a heap.

Using the BMS to trip off on over current is equally terrible if the trip limit has any relation to the cell capacity. Running cells up to above their maximum charge rate is not a good idea for longevity.

If the charger itself can't be relied upon to do the job properly, either buy a better one or arrange things such that the BMS turns the charger off either agreeably or hard with a relay etc.

My single port BMS can shut off the load yet still allow charge. As well ass shut off charge and allow load. I think it has reverse diodes when it has to open one direction or the other.

If the BMS is sending a signal to tell a charger to stop, that would be ok. If you are only charging to 80%, then I agree, CV absorption is not required. You can set to a little higher voltage to get there. But I have seen manuals for charge controllers that say to never have solar panels connected without a battery. If a BMS disconnected, that is what it would do. But I still do not see any advantage of having the BMS stop the charge over a charge controller doing it. If the cells are enough out of balance that the charge controller is stopping the charge because a single cell hit your target SOC voltage, then the rest of the cells are left short. If the charge controller held the pack at the desired SoC voltage, and let the balancer do it's job for a while, then all the cells would reach the target SoC.

I Agree with gnubie. A good charger or charge controller should be the brains behind the charging. If your charging system can't achieve the desired results, then you have a choice. You can try to band-aid it with BMS settings, or buy a better charger.

It sounds like you have a dual port BMS internally but it only exposes a single terminal. Variety is the spice of life, I guess.

GXMnow said:

Yes, but if the BMS just turns off at the absorb voltage, then there will be no constant voltage absorb charge at all.

Click to expand...

I'm confused by the "Yes, but..." part.

Lets consider a single eve 280AH cell.
Suppose my cc/cv charger's charge voltage is 3.65000 volts and configured current is 56 amps and my cell voltage is greater than or equal to 3.65000 volts.
No current will flow.
The cell will not receive any coulombs(charge) if no current flows.

Said another way.
The only way to get that cell to 3.65000 volts is to charge it at greater than 3.65000 volts.

On top of that, the charge should be terminated not when the cell voltage reaches 3.65000 volts but when the current flow decreases to 14 amps(0.05c).

If I've got it wrong, I want to get a better understanding.

If you put current across the battery targetting 14.4V and some cells are at 3.5V while others are at 3.643V it's entirely probable that the high cell will hit the cut off even though the whole string is still below 14.4. We are wandering into the weeds here as this is what the BMS is supposed to do. I think the intent is to have matched and balanced cells, apply excessive charging target voltage and let the whole thing do a Wile E Coyote splat into a wall. That's not what the BMS is for, it's the stopper at the end of the rail, not the limit of travel switch.

gnubie said:

If you put current across the battery targetting 14.4V and some cells are at 3.5V while others are at 3.643V it's entirely probable that the high cell will hit the cut off even though the whole string is still below 14.4.

Click to expand...

Agree.

gnubie said:

We are wandering into the weeds here as this is what the BMS is supposed to do.

Click to expand...

If you mean we are missing the point I disagree.

Wandering into the weeds, ie off the path, not going where the trail leads in that the BMS is there to protect against faults and used in the way intended does just that. It's not intended to be a daily method of control of a deficient system, which is what is being proposed.

I think you actually agree with what I've said.

gnubie said:

Wandering into the weeds, ie off the path, not going where the trail leads in that the BMS is there to protect against faults and used in the way intended does just that. It's not intended to be a daily method of control of a deficient system, which is what is being proposed.

I think you actually agree with what I've said.

Click to expand...

I'm responding to @GMX because I want to understand the "absorb" concept.
I think this concept is right to the heart of the matter.