Gen II Prius Individual Battery Module Replacement

Glad to hear I'm unknowingly following in other's footsteps...

Steve

 

Yep, I have rebuilt about a hundred batteries. And I advertise on Craigslist. I offer mobile installation and a one year warranty for $700. Anyone need a battery in Minnesota? My dischargers are antiques from before lithium RC batteries. Thanks to S Keith for finding a modern charger capable of high discharge rates. I will eventually wear these out.

Five dischargers ready to go. No need to compress the modules for discharge only.



Current stock of tested modules.


This is an Electrophoresis power supply. It will do constant current or voltage. Limits are 500V and 2.5A.



This is my latest test. The pack is all low capacity high resistance modules that are not fit for use in a customers car. The only good quality these modules have is low self discharge. They have been grid charged at 333mA for 56 hours starting at a 70% SOC. So I pumped over 16 Amp hours through 1 to 2.5 Ahr capacity modules. I used a spare duct and battery fan for cooling. Discharge test to soon follow.


Questions?

Brad

 

Great pics of your setup, thanks.

Does that EC570 charger allow you to adjust both voltage and current or just current? (looks like just either or)
In other words is there any way to set it to charge at 350V and 350 mA?

 

I can build a grid charger new for about $40. It consists of 3X constant current LED Power supplies in series. They work great. No adjustments, but they hold current and increase volts as the pack allows. That's about the same $$ I threw away on the barely-functional bio-rad PS.

 

strawbrad,
Great pics!
I have the same EC570 (plus some others)
The racks for sorting are just what I think most rebuilders need. Easily matches like modules.

I see the nice set of pouch cells you have as well. Perhaps from an Enginer system?
Let me know if you want more of those, I occasionally get calls from up in MN with used/non-functioning Enginer systems.

 

Smart idea on using the constant current LED power supplies, I'd love to see one of these built if you ever get around to it.

 

It's not worth taking a picture. It can be as easy as 3 of them strapped together with their output wired in series and their input in parallel with an AC power cord.

 

The Honda Insight folks thought of this quite a few years ago.
Easy.
Wire in series.
Add a 12v to run a fan.
Maybe even a diode and fuse for safety.
Good to go.

I use a similar setup for charging an electric bike of mine.
Not fancy, or fast (at 350ma) but it does the job and is cheap.

 

Hey, Thanks guys! I am happy to share. That is how we all move forward.

It is one or the other, not both at the same time. I use just the constant current.

It is looking good so far. 2000 to 2500 mAhr are now putting out 5000 to 5500 mAhr. These "junk" modules are destine to end up in my car.
After I discharge to 5.4 volts the modules just sit in that low state of charge in my racks. They will bonce back to 7.3 + volts. After a week or so it is easy to weed out any that drop voltage.

Are you using diodes in the lines? The electrophoresis power supplies do not like a battery pushing back at them.

Yep, Enginer. I also have a Plug In Supply and Hymotion kit. I think all they are good for is motorcycle or bicycle batteries.

Something like this?

That's a 12 to 18 bulb LED driver. Cost about $3.
Cheap and easy for the DIY.


Gotta go to work now. It's been fun.

Brad

 

Strawbrad,
Yep. need some diodes in there. What ones are you using?

 

I tried a 1N4005 (600V/1A), and it seemed to "prefer" that as resetting it was easier, but the only time I can get any kind of function from it is if my set voltage is pretty much at my current voltage. It will then hold voltage and adjust current. It appears that it doesn't hold constant current reliably.

Good fortune on those modules. That's quite a recovery.

 

400V/1A straight from my local Radio Shack. I found that one diode was enough for a Gen II pack but I needed a diode in each line for the higher voltage Gen I packs.

I tried a smaller power supply before the 570. It just did not work.

Here are the results of my last test. One module had the discharger overheat and terminate the discharge. For one other module I used a 1 amp discharge rate to see if the results would be vastly different. That module went from 2971 to 5490 mAhr. Just one module, the 11th from the left did not respond well to the long slow over charge. The lower numbers and dates are from my first tests. These modules were all rejects from multiple packs. The packs where charged at 500 mA until a delta peak was reached from 241 down to 230 volts. Modules rested at least overnight and then discharged at a 20 amp rate to 5.4 volts. After the discharge all modules will bounce back to 7.3+ volts. Some will then slowly self discharge to 6.5 to 7.0 volts and are recycled. All of these modules held at over 7.2 volts for weeks even when left at a very low state of charge. The modules were all charged to about 70% SOC and then sat for weeks again. Then the pack was charged at .333 amps for 56 hours for 18.6 Ahrs pumped into a almost full low capacity pack. Cooling air blowing in from the bottom kept the pack at ambient temperature. Swelling was minimal. The modules were allowed to rest for three to four days before being discharged at 20 amps to 5.4 volts.

 

The Bio-rad 1000/500 is 5-1000V and up to 500mA with a 250W max, so a 240V set @ 500mA is well under the power limit. I really think I just got a lemon, but it definitely reset and attempted to recover quicker with the diode in place. It auto-adjusts to either constant voltage or constant current depending on the pack voltage, but it fails to maintain constant current when that is the operating mode. It often reads over the 500mA max even when the set point is well under that. I really think that it just works 1 out of 50 attempts...

Those are some pretty awesome recoveries based solely on overcharge. In the IMA world there has been some evidence that extended grid charging (36+ hours until peak voltage out to a couple decimal places is reached) can provide significant recovery. I am building a pile of marginal IMA sticks upon which I can conduct my own testing some day.

 

I know most are talking about overcharging at low current rates for long periods of time but what about deep discharging?

Would bringing the cells down to very low SOC also help to bring back capacity in the modules?

how big is the risk of cell reversal if you are using very low wattage loads?

 

Hybrid Automotive has recommended progressive deep discharges of both IMA and Prius packs for some time now. They show a chart of bulbs vs. volts to take the current lower as the voltage progresses lower, and they recommend 3 progressively deeper discharges.

 

Yes, I knew about the HA recommendations but was wondering why others were not using the same techniques when dealing with capacity loss on battery modules (sounds like time is the issue?). I'm looking to use this on my Highlander but I have to be alot more careful since it's much harder to find good modules for rebuilds if I damage something by mistake. I was hoping to pick up one of their grid charger / deep discharge setups but wanted to see what others were suggesting for best practices too.

 

Understood. I think a lot of it has to do with the higher failure rate of the IMA packs that necessitated an alternative to whole pack replacement. A lot more work in restoration and capacity recovery techniques has been done.

Does the MiniVCI work on the Highlander? If so, I would pursue that first to identify suspect modules.

Steve

 

Great insights, that puts my concerns to rest given the results you've seen. Yes, mini VCI works great on the highlander too and all my resistances look fine now. At this point I've already gone through the pack and replaced any bad modules so the deep discharging would be just to improve MPG so you can imagine why I wouldn't want to damage a perfectly functioning pack (wife uses it as a daily driver). I'm certainly willing to take the time to do this right but I want to make sure what I do won't undo what I've already accomplished with rebuilding my pack. I'm very curious to see how mine will respond to the grid charging / deep discharging process too.

 

I firmly believe that grid charging and discharging is a very effective tool to maintain balance and improve life. It is VERY important for the IMA packs, but less so for the prismatic types as they are less prone to imbalance.

Short cycling reduces capacity due to voltage depression, and grid charge/discharge is typically very effective at restoring capacity lost due to voltage depression.

Steve

 

How low would you suggest to take the voltage when doing a discharge w/ 60W bulb?

I assume the goal is to never allow it to hit 0V or will that not damage the modules?