BMS cable lengths - same length?

[WindWizard]

I can still see the cell voltage even when the BMS is balancing my cells, did I buy special BMS

The voltages are stored in memory. It may look like they are occurring at the same time but they are really not.

 

[MisterSandals]

Sorry, plain wrong.

I do accept the possibility of WindWizards (subsequent) explanation that balancing and voltage reading are not done at the same time. I think it would be easy to test with the right equipment (not sure if a simple DC clamp meter is good enough/fast enough).

I'm here to learn, and as an engineer i'd prefer to know definitively how it works.

Absolutely!!!!!! Otherwise my system would not work with balance leads of 6 inches to 10 feet.

WindWizard has a big system that sounds like it is working well (great, nice work!) but i am not so willing to accept that the BMS has to work as he describes for his system to work.

 

[WindWizard]

Lets think this through logically. We all agree that when current goes through a wire it produces a voltage proportional to the amount of current and resistivity of the wire. Now if you were the design engineer for a BMS would you want it measuring voltage when you were drawing current out of a balance lead? The answer of course is no because the voltage would not be accurate.

 

[MisterSandals]

Now if you were the design engineer for a BMS would you want it measuring voltage when you were drawing current out of a balance lead? The answer of course is no because the voltage would not be accurate.

Agreed 100%

But what about the design engineer for this POS capacity tester? It fails miserably because it does measure the voltage while drawing current (fatal flaw).

https://www.amazon.com/gp/product/B07HB3XK7L/

Logic and good design are not to be taken for granted.

 

[Skypower]

Thanks for explaining this. I'm building my first LiFePO4 24 V pack from 8x EVE 280Ah cells, and I'm using a JK B2A8S20P BMS. Actually I'm building two such packs to run in parallel. Now I'm trying to decide if I'm going to top charge before I assemble the packs so the cells get off to a good start, or just ignore it and let the BMS sort it out over time. The cells are supposed to be matched in IR and capacity, and also in the same SOC when they are shipped (Luyuan shop). But you'll never know what shape the cells actually arrive in.

In your earlier post you gave the tip to set the charger to 13.6V (3.4 V/cell), do a full charge then let the BMS take a few hours to balance out the cells, then repeat the following days and each day increase the charger voltage in 0.1V steps (0.025V/cell). If I top balance at 3.55V, then it would take 6 days. At that point you can set the charger back to what ever normal voltage you use. Did I get that right? Sounds logical to me.

The only scenario I would suspect it doesn't work is if the BMS needs a very long time (days/weeks) to figure out the actual capacity and IR of each cell. Then it would do a poor job initially. People who don't top balance and use active BMS'es, what experience do they have?

Please correct me if I'm wrong, but on some level I don't see the point with top balancing, unless it is super important that the pack is optimized for max capacity from day one. The BMS will eventually make it optimized, and the cells will not get damaged in the process. I suspect that it is important to allow the pack to really get up to 100% charge every time (not limit it to 90% or less) during this process, because that's where the SOC/voltage curve really takes off so the SOC is easier to read for the BMS. So first a few weeks/month of balancing, then set the charger to whatever voltage/profile you want long term.

Thoughts?

In my opinion you have one of the best bms’s on the market. With a reasonably well matched set of cells that 2amp balance should be able to do the job but it could take a while the first time charging. You may need to greatly reduce the charge current once it starts to disconnect to give the balancer the time it needs to work. The default voltage when balancing starts is usually 3.4 volts, so you need to get it there for the magic to happen or temporarily reduce that setting if you’re having problems getting there because of a high cell voltage protection (3.65). If that’s triggered,the charging will disconnect but balancing will continue as long as you are also within the balance voltage range. The app will tell you if it’s balancing if by looking at the balance current, on some JK’s the “Balance On” just means it’s enabled, not active. There will be a yellow notification letting you know about the disconnect. I try to avoid the disconnect by lowering the charge and giving it time. I sometimes manually augment the process but that’s a whole other story. Yes, hwse is on the right track.

 

[Quattrohead]

The voltages are stored in memory. It may look like they are occurring at the same time but they are really not.

I can see the voltage being pulled lower when I sell is being balanced, are you really going to make me make a video of this showing you. It's real I promise you.

 

[42OhmsPA]

I had to extend the last few on my JBD, I hope I don't have any issues. I did use the same length for every extension and soldered them.

 

[Brett V]

I've always dressed out the BMS leads. No problems.

 

[WindWizard]

I can see the voltage being pulled lower when I sell is being balanced, are you really going to make me make a video of this showing you. It's real I promise you.

No need for a video. You can see the Cell voltage pulled lower. If it balances like my Overkill then it balances a group of Cells. If the Cell voltage pops backup again when it goes to the next group then I would concede that it is using the voltage while pulling current. If it just goes down then I would be inclined to say it is measuring and storing the values and then using the stored values to determine which Cell needs to be balanced and is alternating between measuring and balancing or somehow doing a calculation based on wire resistance and then displaying the correct voltage.

My Overkill BMS uses a tiny bleed resistor so it really does not matter how different the BMS lead lengths are.

I was forced to add a 5 amp balance board. The balance board does not have bluetooth so I cannot see the voltages on the Cells.

The balance leads on my active balance board are all over the place in length. I was interested in a clean layout and was not interested at all in coiling up a bunch of surplus wire. It works great. It does take several hours to balance my Cells. The varying wire lengths have no effect on my Cell balancing.

Now lets assume that your BMS is measuring voltage at the same time it is sucking current out of the Cell. If you have a 5 foot wire difference and you are running 12 awg wire then it measures about 5.211mil ohms per meter or about .0079 ohms.

So if you are sucking out 5 amps of current from your cell then it will drop about .04 volts or 40 millivolts in the wire. You would see an immediate drop in voltage depending on your wire length and the current draw of your BMS.

So my setup works fine with the wire length all over the place. If yours is using the resistance and then somehow compensating for wire length then it would not matter either if your wire length was all over the place. One experiment that you could do but I probably would not recommend would be to add a 10 foot wire to one of your BMS leads and then see if it makes a difference. I suspect it will have no effect.

 

[WindWizard]

I've always dressed out the BMS leads. No problems.

Very nicely done. Your wires are dressed very well. The only thing I do not like is putting your BMS wires under the Busbar nuts. I prefer drilling a hole in the busbar and attaching the BMS leads there. Just a personal preference and it probably does not make a difference. My thought pattern on this is it could add some resistance in connecting the Cells together. I purchased a torque wrench and torqued all my busbar nuts all to spec.

 

[Brett V]

Very nicely done. Your wires are dressed very well. The only thing I do not like is putting your BMS wires under the Busbar nuts. I prefer drilling a hole in the busbar and attaching the BMS leads there. Just a personal preference and it probably does not make a difference. My thought pattern on this is it could add some resistance in connecting the Cells together. I purchased a torque wrench and torqued all my busbar nuts all to spec.

If you look at the second picture, the battery on the right, I drilled and tapped the bus bars. There was no discernible difference. That was my first set of cells and they came with the rigid bus bars. Subsequent cell purchase came with the semi flexible bus bars and I prefer them so I’ll be replacing everything with those on my new build. I have two shelves done and when complete, the rack will have four identical shelves.

 

[hwse]

I like to compress cells and don’t like to disassemble them once they are charged, so since I don’t use diagonal cable jumpers/busses for future builds, I now only use flexible buss bars. So the JK is the future balance from now on, preferably with time to do it. You got to be around to set the charger/system back to normal when it’s done.

Not really. Just set it to whatever charge parameters you like best and let it work. It will take a bit longer, but it will still get the job done. In the meantime, there is no problem with the cells being out of balance other than that they will not have quite as much capacity as they could.

 

[hwse]

Thanks for explaining this. I'm building my first LiFePO4 24 V pack from 8x EVE 280Ah cells, and I'm using a JK B2A8S20P BMS. Actually I'm building two such packs to run in parallel. Now I'm trying to decide if I'm going to top charge before I assemble the packs so the cells get off to a good start, or just ignore it and let the BMS sort it out over time. The cells are supposed to be matched in IR and capacity, and also in the same SOC when they are shipped (Luyuan shop). But you'll never know what shape the cells actually arrive in.

In your earlier post you gave the tip to set the charger to 13.6V (3.4 V/cell), do a full charge then let the BMS take a few hours to balance out the cells, then repeat the following days and each day increase the charger voltage in 0.1V steps (0.025V/cell). If I top balance at 3.55V, then it would take 6 days. At that point you can set the charger back to what ever normal voltage you use. Did I get that right? Sounds logical to me.

The only scenario I would suspect it doesn't work is if the BMS needs a very long time (days/weeks) to figure out the actual capacity and IR of each cell. Then it would do a poor job initially. People who don't top balance and use active BMS'es, what experience do they have?

Please correct me if I'm wrong, but on some level I don't see the point with top balancing, unless it is super important that the pack is optimized for max capacity from day one. The BMS will eventually make it optimized, and the cells will not get damaged in the process. I suspect that it is important to allow the pack to really get up to 100% charge every time (not limit it to 90% or less) during this process, because that's where the SOC/voltage curve really takes off so the SOC is easier to read for the BMS. So first a few weeks/month of balancing, then set the charger to whatever voltage/profile you want long term.

Thoughts?

In my opinion, some of your ideas are correct and some not so much. As far as how I top balance, you have it right except that after you get the pack balanced at 13.6v it will not take days, it will take minutes to get to each new high point. By the time you get all cells to 3.45v, you are at 98%-99% SOC and pretty well balanced. It is getting to 13.6v that can take a long time. The reason I choose 12.6v is because it is low enough to let the charger run but limit the charge current without tripping any of the cells for over voltage (OVP). If you have really badly out of balance cells, and the BMS triggers an OVP, lower the CV setpoint so that it will charge without tripping.

As to the second point, I see no downside with letting the JK do the work. You get your cells, build your pack and put it to work and let the BMS do the hard work. You will have lower total capacity until the cells get balanced but that is not a problem unless you are really pushing the limits in which case you should have a bigger battery bank.

I take exception with your next statement. I believe that it is very important to top balance. I just choose to let the KJ do it rather than to manually do it in parallel with a benchtop power supply. If your pack is not top balanced, you are losing capacity. I also have two 460Ah batteries in sealed 8D cases. They unfortunately have JDB BMS's which makes balancing very difficult but not impossible. I was able to reduce the imbalance on the worst of them from 300mv to 50mv which increased the tested capacity from 445Ah to 480Ah.

Your last statement about charging to 100% is incorrect. It is not necessary or even desirable to charge LFP to 100% every time. That is a carryover from AGM which are damaged by not charging to 100% and being stored at a partial SOC. LFP are very happy at partial SOC and not harmed by it. In fact, for storage you want to lower them to about 50% - 60% SOC. In my boat and motorhome, I charge conservatively with bulk/absorb at 13.8v for 30 minutes. I then float at 13.3 if I am using the batteries and at 13.2v if they are in storage. That means that they are down about 20%-40% when in storage. Before I head out on a trip, i cycle the charger which fully charges it back up to 13.8v to top it off.

 

[hwse]

The only scenario I would suspect it doesn't work is if the BMS needs a very long time (days/weeks) to figure out the actual capacity and IR of each cell. Then it would do a poor job initially. People who don't top balance and use active BMS'es, what experience do they have?

Just a point of clarification. The BMS does not figure out the actual capacity and IR of each cell. It is a safety valve that does not allow the cells to be over or under charge for voltage. It also limits the continuous and max charge and discharge amperage. It also shuts down charging and discharging due to temperature extremes.
The JK WILL find resistance of the sense wires and configure itself accordingly. To get an actual capacity, you will need to do a full discharge test with a seperate battery monitor to track the Ah discharged and a fixed load to pull the battery from max voltage of 3.65v/cell until the BMS shuts down dicharge when the weakest cell reaches 2.5v.

 

[hwse]

In my opinion you have one of the best bms’s on the market. With a reasonably well matched set of cells that 2amp balance should be able to do the job but it could take a while the first time charging. You may need to greatly reduce the charge current once it starts to disconnect to give the balancer the time it needs to work. The default voltage when balancing starts is usually 3.4 volts, so you need to get it there for the magic to happen or temporarily reduce that setting if you’re having problems getting there because of a high cell voltage protection (3.65). If that’s triggered,the charging will disconnect but balancing will continue as long as you are also within the balance voltage range. The app will tell you if it’s balancing if by looking at the balance current, on some JK’s the “Balance On” just means it’s enabled, not active. There will be a yellow notification letting you know about the disconnect. I try to avoid the disconnect by lowering the charge and giving it time. I sometimes manually augment the process but that’s a whole other story. Yes, hwse is on the right track.

Not everyone can reduce the current that their charge source sends to the battery. But for LFP almost everybody can set the absorption voltage to a custom setting. Once that is done, the charge controller will reduce the current to a level that will not send the pack above that point. I have found that most out of the box cells can be charge to 13.6v without tripping the OVP. I prefer to never require the BMS to trigger OVP. I temperarily lower the voltage to a level that will leave the charging without OVP and that lowers the current to a very low level to give the JK time to do its thing.

 

[Skypower]

Not really. Just set it to whatever charge parameters you like best and let it work. It will take a bit longer, but it will still get the job done. In the meantime, there is no problem with the cells being out of balance other than that they will not have quite as much capacity as they could.

I’m kinda in the camp of not wanting to have the FETs turning off and on anymore than they have to, especially with a higher current. I like it to be there of course if SHTF

 

[Skypower]

Not everyone can reduce the current that their charge source sends to the battery. But for LFP almost everybody can set the absorption voltage to a custom setting. Once that is done, the charge controller will reduce the current to a level that will not send the pack above that point. I have found that most out of the box cells can be charge to 12.6v without tripping the OVP. I prefer to never require the BMS to trigger OVP. I temperarily lower the voltage to a level that will leave the charging without OVP and that lowers the current to a very low level to give the JK time to do its thing.

True, that not everyone can reduce the charge current. I don’t have that problem so I didn’t consider it.

 

[hwse]

If you look at the second picture, the battery on the right, I drilled and tapped the bus bars. There was no discernible difference. That was my first set of cells and they came with the rigid bus bars. Subsequent cell purchase came with the semi flexible bus bars and I prefer them so I’ll be replacing everything with those on my new build. I have two shelves done and when complete, the rack will have four identical shelves.

I am also a big fan of the multi-ply humped flexible buss bars. My batteries are all in vehicles (sailboat and motorhome) so major vibration is a real thing. I fix my cells so that they cannot move and then connect the buss bars which can flex in all 6 degrees of restraint. That way, any thermal or vibration stresses go into the fixture and not into the terminals.

 

[hwse]

I’m kinda in the camp of not wanting to have the FETs turning off and on anymore than they have to, especially with a higher current. I like it to be there of course if SHTF

According to the log file on my JK, I have hit UVP twice. Once when doing a capacity and system test. The other time was when I forgot to check my SOC and the heater ran longer overnight than expected. I have my UVP set to 3.0v so that I can lower it to 2.6v, turn off all nonessential loads and still have navigation and running lights for a while.
The log indicates that I have hit OVP maybe 6 times. Once for testing and the rest while I was getting my alternator's voltage regulator dialed in.
That is one of the things I do not like about Battle Born batteries. They tell you to charge to 14.6v which forces their BMS to be a charge controller and trip all the time. It is part of how they balance but I think it is a really bad idea.

 

[Skypower]

You’re blind to what’s happening inside a Battle Born. Is ignorance really bliss? No app, no display = less customer complaints ? Same for Big Battery.