HV Battery Headway 38120HP LiFePo4 70S pack

You still missed the elephant in the room, this thread is about replacing the original NiMh modules with LFP cells and the traction battery recharges via regen, not mains trickle charging. Do you just turn the regen off while the little balance boards try to even out the voltage across the cells by burning off some from the high voltage cells?

As far as fast charging and balancing the cells at less than 80% SOC ... seems you have never had any hands on experience with fast charging ...... the Ionic 5 is charging at 160 amps into 50Ah cells, voltage run away can occur real quick at an over 3CA charge rate, yet settles quickly once the charging stops. By capacity shifting through the whole operation, the high current rate can continue for a much longer period before the current reduces rather than stopping completely. Once the charging has stopped, the capacity that was shifted to cells that could absorb it better, now have a higher voltage, so that capacity is shifted back to the cells it was taken from so the voltage again balances ..... that can't happen if the capacity is simply drained off can it?

T1 Terry

 

Something I've mentioned a million times on other things that have to do with this...
Lot's of people have done LiFePO4 conversions without expensive and fancy active balancing and so far I have not heard of a single failure.

Also, Tesla, Nissan, Orion BMS, Lithiumate and many others all use passive balancing without problems. I'm not saying there isn't a benefit to active balancing. But it can hardly be defined as "necessary" even for 30C NiMH to LFP replacements. There are still many DIY'ers who don't even have a balancer and just take and manually balance their cells every month or so, again without any major failures. Others just throw on a zener diode and a resistor on each cell and call it a day and again don't have problems.

Your fears of cells overheating or resistors blowing up are completely unfounded. As an EV, hybrid, retro computer and amateur radio enthusiast I board level repair and design my own electronics and work with high currents and voltages all the time. Have you ever had to bleed off 30,000V to work on a PCB? Have you ever installed a resistor that can handle up to 40 amps at hundreds of volts? Have you ever designed a PWM circuit that can put out 50kW?

 

No, but I have been installing LFP off grid battery/inverter/solar set up for around 11 yrs so I do know about cell balancing when using LFP cells to build a battery. The more cells in series, the greater the headache at the end of charging and the last bit of capacity .... roughly 10% to 20% at either end.
In this thread we are talking about 8Ah Headway cells, leave 40% still unused (20% at each end of the cell capacity) and there is only 8 X 40% =3.2, 8 - 3.2 = 4.8Ah usable, a serious drop from the 6.5Ah usable with the original NiMh cell/module arrangement.

That puts the whole idea of following the EV manufacturers battery management system of leaving 20% at each end of the usable capacity out the window .....

This whole discussion is based around top and bottom balancing the cells to get as close to the 8Ah capacity as possible, if only using 4.8Ah out of the 8Ah available, why bother even thinking about changing from the original set up that can actually handle being fully charged?

T1 Terry

 

Top and bottom balancing both on the same battery is a bad idea for EV's. Your scenario sounds like a system that has a cell with some 20% less capacity than the rest. When the other cells have lost 80% of their capacity then to get the other 20% out of them that one cell that has reached its 0% now needs to be bypassed somehow so it no longer discharges. Then yes, you can use the rest of the other cells. But then when you charge back up it will hit 100% when the other cells are at only 80%. So again you'll need to balance again.

That type of system might make sense for an off grid system that may use inferior cells but you still want to get every last drop of energy from them. In an EV or hybrid that's a huge disaster waiting to happen. Generally you never want to top and bottom balance the cells. You also never want cells that are or end up 20% from each other.

In an EV or hybrid you ideally want all your cells to be exactly the same. Ideally balancing shouldn't even be necessary. But nothing is perfect, so yes, there will eventually be differences in cells. Some will have a slightly higher self-discharge rate than other cells and will lead to an imbalance. But never in the 20% range. Not in an EV or hybrid. The Nissan Leaf will register 100% SOC before it even starts balancing. Of course 100% means 99% to 100%, so balancing happens in the last 1% of charging, not the last 20% of charging.

I've never heard of EV or Hybrid manufacturers that also bottom balance to squeeze every last drop out of their batteries. I've heard of the idea thrown around to improve range on worn batteries, but no, they don't do that. When the first cell reaches 0% (or whatever level the BMS counts as zero) the whole battery is at 0% as far as the BMS is concerned. With top balancing only in an EV, that means the whole battery is considered at end of life once the first cell reaches about 70% of its original capacity. Fortunately it is very rare that one cell reaches that long before the others. Normally most of them are going to be very close to that capacity when the first cell is at end of life.

I hate to sound mean, but again your concerns are completely unfounded and boarder line paranoic. Here's what happens in every EV and Hybrid out there with a lithium type battery. The battery discharges, normally to the point the driver decides to charge it back up. If he doesn't and runs out of range, the BMS calls the battery empty at the first cell that reaches it's 0% level. As the battery charges it reaches nearly 100% before balancing is even necessary. When the battery is balanced the current is bled of via a resistor. I repeat, all EV and Hybrid manufacturers are using resistors to balance their batteries. That's how it works, end of story, no need to fear your EV or Hybrid will only have 60% of useable electric range. No need to fear that your it has resistors that will melt down and stop working. No need to fear that your battery is going to overheat due to the balancing function.

 

Thought I'd wait a while so it didn't turn into one of those "where's the popcorn" type of back and forth discussion :lol:
Fortunately for us, we do get EV battery packs that cell top lossie balance boards have killed a few cells so the whole pack gets replaced .... we can often rescue a number of cells from a pack that will no longer power the car or scooter they were in. The cell top boards get so hot they cause the electronics that turn the resistor on/off to fail, so the resistor continues to discharge the cell 24/7. Eventually the cell gets that low that under high load the cell voltage gets dragged past 0v .... a reverse current is dragged through the cell ..... this is virtually instant death to a lithium cell of any chemistry type.
The BMS can't signal the load to be cut because the cell top board no longer functions properly to signal the BMS the cell is at a lower than acceptable voltage, so the BMS lets the load carry on regardless until the pack voltage drops to the cut off point.

The Nissan Leaf will register 100% SOC before it even starts balancing. Of course 100% means 99% to 100%, so balancing happens in the last 1% of charging, not the last 20% of charging.

You do realise the Nissan Leaf battery is Nickel Manganese Cobalt chemistry, to actually charge to even close to 100% SOC requires constant current till the cell reaches 4.2v, then constant voltage until the current drops to around 5% of the Ah rating
https://www.itechate.com/uploadfiles/2019/01/20190128145701571.pdf
so the claim that balancing happens at 99% SOC via resistor loading to balance the cell voltages is not even close to being correct ......

T1 Terry

 

You didn't understand me. I said what the BMS considers 100%. Of course it's not 100% of what the battery can actually charge to. There are zero EV manufacturers that charge their batteries to 100% because that would cut their charge cycles to a mere fraction of their current longevity. What is your point? EV's don't charge to 4.2V per cell so there's no point in talking about what happens when you charge an EV battery to 4.2V per cell.

So you say you get batteries from cars that have killed their cells due to top balancing via a resistor. Could you tell me more about that because I haven't ever heard of such a thing. If you're right, then nobody who cares about their money should be buying an EV right now. Teslas, Leafs, Bolts, etc. all use resistive balancing. Where are all the failed batteries due to balancing?

But I can tell you one thing. A BMS that drags the voltage down below 0V has nothing to do with the type of balancing circuit.

 

Then what is this about?

Sorry if I misinterpreted what you meant to say ...... I wonder if I'm the only one who is still reading this thread that made the same error ......

And here lies the whole problem, very little analysis is employed in discovering why the battery pack needs replacing, yet these battery packs can generally be recovered and end up with a second life as a house battery.
Just as a market developed for reconditioned Prius traction batteries, a market will spring up offering reconditioned EV battery packs in the near future.

A much deeper understanding of the recharging requirements of lithium batteries is required to bring what appears to be a cell that has lost capacity, back to full capacity.
EV manufacturers have no interest in actually making their battery packs last indefinitely ..... where is the after sales value in that? The added cost of designing/building a BMS system that would achieve such a result would also weigh heavily on the decision to build a BMS that is just good enough to do the job. By holding a percentage of the batteries true capacity at the top end and bottom end of the capacity curve, a proper BMS requirement is avoided .... so despite your previous assurance that "no need to fear your EV or Hybrid will only have 60% of useable electric range"
then in the nest post write
"Of course it's not 100% of what the battery can actually charge to".

Are you actually saying the cells are not discharged to 0% SOC but rather what the vehicle BMS is programmed to see as 0% SOC? Seems this contradicts what you posted and I quoted above .....doesn't it?

T1 Terry

 

So now you're getting into conspiracies. Aparantely you know more than the EV and lithium ion cell manufacturers.

Yet here's a contradiction that you just made. You just complained that EV manufacturers aren't giving us 100% of the battery capacity because I made the huge, terrible mistake of stating that a modern BMS on an EV won't give you 100% ot the battery capacity. You also said they don't care about longevity. Yet if you look at the specs for any lithium ion cell you'd see that charging up to that 100% instead of 95 to 98% cuts cycle expenctancy from the thousands down to the hundreds.

Who are we supposed to believe? You or what the cell manufacturers publish?

 

Seriously, this is just becoming silly. Any links to LFP, LYP or in fact any lithium chemistry cell manufacturers that state that in their technical data?
Winston batteries actually provide the instructions on how to fully charge their cells and give a 5,000 to 7,000 cycle life prediction. Our own Winston cell battery installs have all returned 100% of the advertised capacity when load tested using Winston's charging and 0.5CA (C2) load at each 3 yr service ... and we aren't talking one or two systems, we are talking hundreds, more than 5 of these systems are now over 10yrs in 24/7 operation and still returned the 100% capacity under load. Our BMS is designed to bring the battery pack up to 100% SOC when ever the solar or DC to DC or mains power are available for long enough to fully charge them.

T1 Terry

 

Look at ANY 18650 lithium ion cell that's fully charged at 4.2V and look at it's cycle data. You will always see a 30% capacity loss (with 70% remaining, aka, end of life) somewhere between 250 and 500 cycles when charged from completely empty (usually about 2.5V) to a completely full 4.2V every cycle.

Example: Battery Space Panasonic NCR18650B

The fact that lithium ion batteries will last several times as much if not charged to 100% has been extensively documented and well known. I don't even know why I'm responding to this. Here's a cell that goes from 250 cycles to 1,500 or more cycles when using only 60% of it's total capacity: Samsung INR18650 25R

Agreed, that LiFePO4 batteries, such as the Winstons, do have better longevity, which is a good reason to get the standard Model 3 if you were shopping for an EV. But they are also technically fully charged before even reaching 4V. Many consider them fully charged at 3.6V. If they were charged to 4.2V like other lithium cells their life expecatancy would be about the same.

Ironically for you, as a result of their charging characteristics LiFePO4 cells have a much bigger margin of error when it comes to cell balancing since keeping the cells under 4V and completely full at 3.6V is much easier than keeping a Lithium Ion cell at exactly 4.2V, no more and no less. Yet resistive balancing works just plain and fine with li-ion cells in many different cars.

This is also why LiFePO4 works ok as a NiMH replacement in a Toyota hybrid since it is not necessary to hit an exact charging voltage.

 

Interesting interpretation of what was said in the document. This part interests me:

7.10 Cycle life With standard charge and maximum continuous discharge. Capacity after 250cycles, Capacity ≥ 1,500mAh (60% of the nominal capacity at 25℃)

The definition of standard charging in this document
7.0. Characteristics 7.1 Standard charge This "Standard charge" means charging the cell CCCV with charge current 0.5CmA (1,250mA), constant voltage 4.2V and 125mA cut-off in CV mode at 25℃ for capacity.

The definition of maximum continuous discharge from the same document
3.7 Max. continuous discharge (Continuous) 20A(at 25℃), 60% at 250 cycle

Yet the nominal discharge is:
7.3 Nominal discharge capacity The standard discharge capacity is the initial discharge capacity of the cell, which is measured with discharge current of 500mA(0.2C) with 2.5V cut-off at 25℃ within 1hour after the standard charge. Nominal discharge capacity ≥ 2,500mAh Which complying to the minimum capacity of IEC61960 standard

Basically what they are saying in this document is the discharge rate has a serious affect on the cycle capacity life, not the voltage it is charged to or the voltage the discharge stops.

So, if you plan to build your EV using these cells, you will need 200 cells in parallel @ nom. 3.6v if you want to discharge at 100 amps if you wish to retain the cell capacity for more than 250 cycles ..... you would need 57 on these groups of 100 cells in parallel to build the 204v nom. Prius traction battery, that's 11400 cells altogether .... maybe not the ideal choice of cell for a traction battery replacement .... and this is what this whole thread is all about, cells that are suitable for use when building a traction battery replacement .....

T1 Terry

 

I sympithize with you. Sorry to hear about those problems. Hopefully it's something simple.

 

You can do it!!!

 

Ahhh.... the dreaded dried out module failure, shows a good voltage until loaded. A 240v filament bulb across the battery is enough load, so 2 x 120v in series, with a filament bulb it doesn't matter if it is AC or DC voltage.

T1 Terry