‘Normal’ LiFePO4 post-charge settling?

[fafrd]

Once you hit the current which the resistor will bypass (V=I*R) across that cell, the cell will stop charging. The current passing through the resistor will continue to charge the other cells. Eventually you will have all but one cell bypassed, and that cell will be charging at the resistor bypass current. Takes a little fiddling, but as long as your BMS is working, there isn't any risk of cell damage.

I’m intrigued but not yet understanding.

I’ve got an 8S battery, a 10A Charger, small parallel Port sense wires that can manage 1-2A but no more, a small 1.5A 5.2V Charger, and a 2.5ohm 50W power resistor.

Are you talking about a resistor between the charger and 8S battery leads or if not where?

 

[fafrd]

Once you hit the current which the resistor will bypass (V=I*R) across that cell, the cell will stop charging. The current passing through the resistor will continue to charge the other cells. Eventually you will have all but one cell bypassed, and that cell will be charging at the resistor bypass current. Takes a little fiddling, but as long as your BMS is working, there isn't any risk of cell damage.

Read more carefully and I think you are talking about just a single resistor between 10A charger and 24V + terminal.

My BMS has rated balance current of 76mA, so if I put my 2.5ohm resistor in series with my charger when it is in CV mode and assuming it is supplying no more than 76mA, that means I’ll be dropping 190mV over the resistor or dropping my CV voltage from 28.75 to 28.66V.

Cells being balanced down by the BMS should not charge any higher while cells not being balanced down should continue to charge up to 3.5825V - is that what you are talking about?

 

[Ampster]

1/ I’ve run two discharge cycles and identified 2 ‘weakest’ cells which hit 2.5V while the other cells appear to be at ~10% SOC.

I expect that there could be 10% difference in capacity with the cells I purchased. I am not concerned about that because I saved much more than that on the purchase. My strategy is to get them all to the same voltage at the top and not discharge anywhere near the knee of those weakest cells.

2/ At full charge (28.75V / 3.6V avg), if these two cellls are fully charged when the charger cuts out, I’ll be getting as much capacity from the battery as is possible (while if they are low by ~10% or more, I’ll be sacrificing available capacity).

3/ Once I’ve got the cells balanced in a way to maximize capacity (meaning weakest / fastest cells are fully-charged with the others), the battery will either maintain that balance or will lose balance over time / cycles. If balance is getting lost, I believe that means either I have a bad cell or a bad BMS which is why I’m trying to nail down stability right now.

@nebster gave me this analogy. Think of your cells like popsicle sticks, each with a different length. You can't change their capacity. You can decide which end to make them even, either at the top or at the bottom. The weakest cell will limit the pack capacity. The only solution is next time you buy cells spend twice as much and get perfectly matched cells. We use the term balanced very loosely. Unless we have perfectly matched cells they will never be balanced to the same capacity. We can choose to have them reach the same voltage at the top or at the bottom but never both.

 

[fafrd]

I’ve just been free-thinking about what would happen if I connected my charger through my 2.5ohm resistor.

Battery is at 27.6V now, so 1.15V below charger CC voltage of 28.75V.

At that voltage, the charger will enter CC mode and start pushing 10A.

10A would drop 25V over the resistor so that dog’s not hunting and the charger will switch to CV mode (or may not ever go into CC mode to start with).

In CV mode, there will be 1.15V across the 2.5ohm resistor, meaning there will be current of 460mA across the resistor charging the cells.

The cells will start charging and battery voltage will increase.

When first cell hits 3.55 V, balance current of 76mA should kick in for that cell.

Let’s assume all other cells are at 3.5V when that happens and battery is at 28.05V.

So voltage across resistor has dropped to 0.75V and current has dropped to 300mA.

High cell continues charging but at only 224mA while other cells charge at 300mA.

Let’s now assume all cells but my low cell are at 3.55V, high cell is at 3.65V and low cell is at 3.45V.

Battery is at 28.4V, voltage across resistor has dropped to 350mV, CV charge current has dropped to 140mA, 140mA is charging the low cell while all other cells are only charging at 68mA.

Other cells increase towards 3.65V, the high cell charges towards 3.7V and the battery voltage reaches 28.56V (average of 3.57V per cell).

At that point voltage across resistor has dropped to 190mV, charge Current has dropped to 76mA, the highest cell is not charging further while all other cells are either not charging as well or continuing to charge towards the voltage of the high cell (depending on how the BMS has been designed) and the low cell is charging at 76mA.

Let’s assume the low cell reaches 3.5V, all other cells are at 3.575V and the high cell is at 3.7V. Battery is now at 28.65V, voltage across resistor has dropped to 100mV and charge current has dropped to 40mA.

High cell is discharging at 32mA, other cells are either charging at 40mA or also discharging at 32mA, and low cell is charging at 40mA.

This will continue until the low cell reaches 3.595V at which point battery will be above 28.75V and charger will switch to float mode and all other cells will be drained down to the voltage of the lowest cell.

Am I understanding this correctly? If so, it seems like a brilliant way to balance - aside from it taking longer to charge and wasting I^2C of energy in the resistor, is there any other downside?

 

[fafrd]

I expect that there could be 10% difference in capacity with the cells I purchased. I am not concerned about that because I saved much more than that on the purchase. My strategy is to get them all to the same voltage at the top and not discharge anywhere near the knee of those weakest cells.

@nebster gave me this analogy. Think of your cells like popsicle sticks, each with a different length. You can't change their capacity. You can decide which end to make them even, either at the top or at the bottom. The weakest cell will limit the pack capacity. The only solution is next time you buy cells spend twice as much and get perfectly matched cells. We use the term balanced very loosely. Unless we have perfectly matched cells they will never be balanced to the same capacity. We can choose to have them reach the same voltage at the top or at the bottom but never both.

Exactly how I’m thinking about it. I’ve been struggling to get the shortest popsicle stick to be reaching 100% first or along with the others (exacerbated by the fact that I’m not yet sure the BMS is functioning properly).

 

[Ampster]

I’m not yet sure the BMS is functioning properly).

Do you have any way of monitoring the BMS? Leds, Bluetooth? Have you tested the HVD or the LVD at a pack level or at the cell level?

 

[Luthj]

I am talking about clipping an appropriate resistor across the cell itself. Directly to the terminals. No different than any passive balancer works. By adjusting your charge current limit to the bypass resistors current (3.5V=I*R), you effectively can stop the high cells from charging while the low cells catch up.

 

[fafrd]

Do you have any way of monitoring the BMS? Leds, Bluetooth? Have you tested the HVD or the LVD at a pack level or at the cell level?

The BMS has no Bluetooth - my only way OV monitoring is to measure the voltage of the sense wires through my parallel port. That port was set up for a BattGO monitor, but the parallel channels are so inaccurate (+/-5mV) that the BattGO is next to useless (perhaps usable to know when the weakest cell has hit 2.5V, at best).

Since I use the same multimeter to measure the sense wires, my precision is +/-0.5mV, even if accuracy is probably less.

I have exercised both high-voltage cut off at 3.75V as well as low voltage cut-off at 2.1V, so that part of the BMS is working fine (and seems reasonably accurate).

I’ve not found a way to prove balancing is working correctly but if I understand this ‘charge through a resistor’ idea correctly, that should allow me to prove that functionality.

 

[fafrd]

I am talking about clipping an appropriate resistor across the cell itself. Directly to the terminals. No different than any passive balancer works. By adjusting your charge current limit to the bypass resistors current (3.5V=I*R), you effectively can stop the high cells from charging while the low cells catch up.

OK, now I understand - you mean discharging individual cells as I have been while charging.

Problem I have with that is I’ve got no control over charge current, plus it is a lot of screwing around for me to connect multiple discharge resistors in parallel.

Inspired by your post, I walked through the logic of charging the entire battery through a resistor - do you think that will function as I outlined? Have you ever heard of anyone else using a charge resistor to top-balance with a BMS?

 

[Luthj]

Inspired by your post, I walked through the logic of charging the entire battery through a resistor - do you think that will function as I outlined? Have you ever heard of anyone else using a charge resistor to top-balance with a BMS?

You mean using a resistor to limit charge current? Sure, I do something similar with a long wire run to limit my alternator charge current at low SOC.

Actually picking the right resistor watt/ohm is a bit tricky If you know the current and desired terminal votlage (they must correspond for the SOC you are targeting), then you can pick one. For example you want terminal voltage of 27.6V, and your charger is outputting 28V. So you need a 0.4V votlage drop. Now you need to know what current the battery will accept at 27.6V. If you know that is say 10A (at current SOC), then you just work backwards to find the correct R, and then calculate the wattage needed. In this example 0.4V= 10A * R R= 40mohm. Power is 4W. In this example the easiest resistor would be ~6ft of 18AWG wire.

 

[fafrd]

You mean using a resistor to limit charge current? Sure, I do something similar with a long wire run to limit my alternator charge current at low SOC.

Actually picking the right resistor watt/ohm is a bit tricky If you know the current and desired terminal votlage (they must correspond for the SOC you are targeting), then you can pick one. For example you want terminal voltage of 27.6V, and your charger is outputting 28V. So you need a 0.4V votlage drop. Now you need to know what current the battery will accept at 27.6V. If you know that is say 10A (at current SOC), then you just work backwards to find the correct R, and then calculate the wattage needed. In this example 0.4V= 10A * R R= 40mohm. Power is 4W. In this example the easiest resistor would be ~6ft of 18AWG wire.

I think I mean something a bit different, more like only using a resistor once a charger goes into CV mode to balance the cells using the BMS.

I don’t know exactly how my charger decides to go into CV mode, but once the fan turns off, I’m pretty certain it’s in CV mode and battery voltage is usually around 28.5 to 28.6V when that happens and current is obviously well below rated maximum current of 10A.

If I put a 2.5ohm resistor in series with the charger at that point, it’ll limit the charge voltage to 0.2V/2.5ohms = 80mA until the first cell passed 2.55V when balance current should kick in for that cell.

That cell will keep charging at ~8mA while the others continue to charge at 80mA capacity until the battery voltage increases to 28.57V at which point the charge current will have decreased to 72mA and the high cell will stop charging (since balance current being pulled out equal charge current going in) while the other cells keep charging at 72mA.

This process should continue all cells are at 3.57125V or every cell has reached an equilibrium where balance current out = charge current in.

I don’t know what cut-off current my charger uses to go from CV mode to Float so it may go to float before balancing is achieved, but other than that, does it sound like an approach worth trying?

 

[John Frum]

I don’t know exactly how my charger decides to go into CV mode, but once the fan turns off, I’m pretty certain it’s in CV mode and battery voltage is usually around 28.5 to 28.6V when that happens and current is obviously well below rated maximum current of 10A.

Assuming a relatively sane cc/cv + float charger....
When the current draw drops below the configured limit your charger is in cv mode.
When the voltage drops its in float mode.
Didn't I already explain this?

 

[Ampster]

I don’t know exactly how my charger decides to go into CV

Once again @smoothJoey beat me to the answer. When the voltage of the battery reaches the voltage set on the charger is when CC ends and CC begins. There actually may be some overlap but fan noise is probably a good indicator. Interestingly when I am charging my 4 ¹/² year old Tesla the compressor noise of the battery cooling process gives me a clue to how many kWs it is pulling from the charging cable. It doesn't tell me when it transitions from CC to CV mode because think it varies depending on temperature.

 

[fafrd]

Assuming a relatively sane cc/cv + float charger....
When the current draw drops below the configured limit your charger is in cv mode.
When the voltage drops its in float mode.
Didn't I already explain this?

If you did and I missed it, my apologies.

I understand about CC to CV transition being triggered by current dropping below some threshold (without knowing what that threshold is).

But from what I’ve seen, voltage continues to increase at that point so I assumed voltage rising above some threshold or current dripping below some threshold was the trigger to go into float.

My charger is 28.6V, but seems to be similar to one of these: https://m.aliexpress.com/item/32870...oXLneU6uekqms2lniRRoCi1kQAvD_BwE&gclsrc=aw.ds

It states that the CC to CV transition is at 5% of the charge current, or 500mA for my 10A model, but there is no spec for how it makes the transition from CV mode to what it calls ‘trickle’.

 

[John Frum]

Once again @smoothJoey beat me to the answer. When the voltage of the battery reaches the voltage set on the charger is when CC ends and CC begins.

If the voltage of the battery >= to the voltage of the charger no current will flow.

 

[Ampster]

If the voltage of the battery >= to the voltage of the charger no current will flow.

To further clarify when the voltage of the battery reaches the voltage set on the charger the charger increases the voltage so current will flow. We don't see that voltage increase above the setting because across the battery/charger interface the voltage is at the CV setting.

 

[John Frum]

I understand about CC to CV transition being triggered by current dropping below some threshold (without knowing what that threshold is).

cc ends and cv begins when the battery draws less amperage then the charger is configured to provide.
cc/cv charging is that simple.