Gen II Prius Individual Battery Module Replacement

Squaring off with the Professor, this ought to be interesting. Who has popcorn?

 

The imbalance problem is usually not cells with lower capacity, it is cells at a lower State of Charge (SOC) than the other cells in each module. I have successfully rebalanced Prius NiMh modules and restored them to a uniform usable capacity.

JeffD

 

What are the units of State of Charge (SOC)?

 

Percentage of battery (cell in this discussion) capacity.

JeffD

 

Cells or modules with the same SOC but different capacities will take differing amounts of time to discharge at a constant current. The only way to ensure that they track each other when the pack is operated over the normal range of 40 - 80% SOC, is to input a known quantity of charge into each module. That quantity should be 80% of the capacity of the weakest module, because this will be absorbed with close to 100% efficiency by every cell.

Over charging may get every cell in a module to 100% SOC, but at what cost to longevity? The SOC of each cell or module will immediately begin to diverge as the modules discharge, because the current passing through every cell is the same but their capacities are different.

 

You keep talking about "weak" cells. The OE modules were tested when new to have the same high capacity and low effective resistance. Their cells were well matched as well and they have all experienced the same use. Some of the modules have been subject to higher temperatures (middle of the pack) and may have aged a bit more than the other modules, but we are discussing balancing a module at low current so that the heating of the fully charged cells is not damaging.

In my 2004 Prius at 195k miles, one of the modules had gotten out of balance and when the overall battery was discharged to the lower limit, one cell became fully discharged and then reversed which killed that cell. The other 27 modules initially had widely differing capacities, but after 3 charge/discharge cycles all had recovered to 6.5 amp-hour capacity as measured by the charger (this was measured at a low discharge rate so the actual capacity of all of the modules would have measured as less at a 1C discharge rate). These cells were then successfully used by 6 other Prius owners to bring their batteries back to life,

JeffD

 

I can see that because the pack operates over a restricted range of SOC (40 - 80%), the pack can tolerate modules with varying capacities. Obviously the closer the capacities, the greater the range of usable SOC.

 

That is how Toyota can rely on most batteries surviving for 10 years or 150k miles and fail well beyond the warrant period without a true BMS that would maintain cell balance in the pack.

The manual balancing process that we are discussing, if done periodically (a pain unless you get a full pack charger), would greatly extend the average life of our NiMh battery packs.

JeffD

ps: My 2016 Prius has a LiOn battery pack, but did not get to it's end of life due to a Kamikaze deer. My current 2020 AWDe has the older NiMh battery pack and I am confident that it will make it through the warranty period.

 

Your experience implies that when discharging the pack as a whole, it would be best to monitor the voltage of individual modules and stop the discharge if any one module falls below say 0.9V/cell (5.4 volt), rather than relying on just the overall pack voltage.

 

Yes, but even better would be to have a full BMS that maintained module balance when recharging the pack. The BMS allows some current to bypass fully charged units while still charging the others. The low terminal voltage of NiMh cells makes doing a BMS at the cell level impractical. Our Prii monitor pairs of modules, but I don't believe that this information is used to decide when to run the ICE to charge the battery.

I was charging/discharging individual modules and did indeed stop each discharge at 6V ( 1V per cell).

JeffD

 

I've ordered a LPC-150-500 so that I can DIY my own full pack charger. Planning to balance the pack out of the car so that I can monitor individual module voltages during discharge/charge cycles and also clean the bus-bars.

 

Just be careful. Either the battery or the LPC-150-500 can kill you if you make a mistake.

JeffD

 

300 volts is good for a 34 module battery. A battery of 28 modules may not charge well. The Prius 20, 30 charger (250v) from Hybrid Automotive won't charge the Prius c battery, it just charges to a certain voltage and does not go higher. For a battery of 28 modules, 245 volts would be ideal.

 

The LPC-150-500 is 500mA constant current supply, with an output voltage swing of 150 - 300V DC. The voltage spec is very similar to the LPC-100-350 which has been used successfully by others as the basis for a DIY grid charger, so I think that it should be fine.

 

Reading through this thread somehow led to me watching this webinar video from NuVant Systems:

At the timestamp 25:12, the rep from NuVant, Eugene Smotkin, says this:

"so what reconditioning actually does is is it breaks that barrier that you have through a series of deep charge/discharges at very low voltages at high current. If you’re high current, you can bring a module voltage very low."

I was a bit shocked (no pun intended) to hear this because after reading 129 pages of this thread, I was thoroughly convinced that high current (e.g. >1A) during deep discharges (i.e. discharging NiMH below 1.0V/cell, so 6.0V for Prius modules) was a big no-no. It's my understanding that people shy away from this for fear of causing cell reversals. However, it appears NuVant prefers high current to elicit the phase changes necessary to restore function to the nickel oxy/hydroxide electrode.

Does anyone know why Eugene appears to be saying the opposite to what most people on this thread are saying? Am I simply misunderstanding?

Thanks!

 

Why is each discharge time shorter than the previous one? Doesn't that imply that the total battery capacity is decreasing not increasing?

 

Voltage depression occurs at higher discharge currents, but it is supposed to be insignificant below 1C(6.5A). Obviously discharging at higher rates is quicker, and time is money!

When I reconditioned my battery I discharged at 0.5C to 168V (6V/module) and the voltage would quickly rebound to 190V when the load was removed. The one time that I discharged the pack to 151V (0.9V/module), it rebounded back to 170V. At higher currents the voltage can collapse rapidly, so you need to be ready. Details here...

 

That is a very good question. I will pay attention to this next time I do a maintenance cycle.
However the assumption being that the discharger is uniformly using the same voltage to discharge the batteries, which it may not be, I don’t know. It appears to be a smart-discharger, not a series of incandescent light bulbs that pull a constant draw.
Whatever the case, it seems that every time I do this maintenance cycle on this 10 y/o OEM (unmolested) battery it seems to become more stable and less prone to rapid fluctuations while driving. I also appreciate the work you’ve done to document the results of your charge cycles and will try to follow your method next time.

 

I’ve noticed this time that the Prolong smart-discharger is indeed blowing hotter air at different times and drawing differently from the battery so the amount of time it takes to discharge a traction battery is not indicative of the overall capacity but limited by the heat generated by any block of cells which the smart-discharger is monitoring and adjusting its draw accordingly. Weaker cell blocks generating more heat are likely putting the brakes on the smart-discharger which draws less power over a longer period of time to protect those cells. A faster discharge to the target voltage might indicate a more balanced battery? That’s my theory anyway… what do you think?

 

In this context you mean module/modules, not cell/cells. Just so we're all on the same page.

No, that's a bad theory. Faster discharging outside of the 40% - 80% range will just lead to issues with reversed cells thereby consigning the module that contains that cell to the trash.