I will get one!

The bulb part number is OEC1157 The O might be a C. Its hard to tell.

Thanks - Jen

 

1157 is the important part. 2.25A @ 12V.

Since the bulb will flow less current at lower voltage, and you're discharging between 8.2 and 6.0V, your current will be lower but not linear like a resistor. I would suspect that you're closer to an average current of 17.1A, so you're probably looking at a high current capacity of 4750mAh. That's still fine.

Note that the relationships between current and voltage drop for bulbs is what I've observed on 120V light bulbs used on whole packs. The 12V bulbs may be sufficiently different that the relationship doesn't apply. In any case, I believe my 17.1A number is conservative, and you'll probably see more with the power analyzer I linked.

NOTE: That power analyzer will also work during the load test, so you'll also get the current and capacity consumed in the load test.

Since the shut-off feature is in question, I would take the next fully-charged stick and let it go below 6.0V. As I mentioned, the shut-off is probably in the vicinity of 5.4V. Voltage should plummet from 6.0V, and you should hit 5.4V very quickly.

The thing to watch for is a drop that suddenly stops or slows dramatically. That's an indication of a cell reversal. Short term reversals even at high current aren't going to damage anything significantly... we're talking seconds here. If you see that, terminate the discharge ASAP.

Provided the voltage is dropping fast, I would let it run to 5.2V to test the auto-shut-off feature. The amount of capacity extracted below 6.0V will be minimal and doesn't affect the capacity calculation.

I suspect it is important to start with a fully charged stick. It's probably making some cut-off decision based on starting voltage. A fully charged 6 cell battery is going to be higher than 8.1V unless it's sat around for a long time (days). If you put a stick that's been freshly discharged on it, it might run it down too low.

I've glanced at your sheet. I'll give some thoughts shortly.

Steve

 

The bulb device does shut off at 5.2V and as you rightly surmised the sticks had sat for a couple of days to give the 8.1V start reading but that is the same as the 12hr post charge reading. Should I be doing the test after topping off the battery? - Jen

 

You're capturing a lot of great data.

When you're finished, three things will matter the most:

1. Voltage drop after 50A discharge (measuring direct load and capacity with the power analyzer will be great a big bonus).
2. 20A capacity
3. Self-discharge

We haven't talked about self discharge but between it and increased IR, that's what kills these sticks... usually just a single cell in them. Your charger's IR measurement may be useful as a comparison tool. The numbers themselves are horrible, but they might have value if you notice a correlation between measured IR and 50A voltage drop. The 50A voltage drop is essentially testing IR. More drop = higher IR. The numbers you've posted are pretty good.

Self-discharge is a hairy one. The conventional method is to take a reconditioned stick of known capacity, charge it, let it sit for a week and then discharge it. The lost capacity is what self-discharged over 7 days. Unfortunately, this isn't a realistic representation of what happens in the car since the car only uses the sicks in the 20-80% capacity range, so it has limited value. A stick at 100% SoC will self-discharge more than it will at 50%.

NASA published this about LiPo cells:

http://ntrs.nasa.gov/archive/nasa/ca...0120000040.pdf (pages 11-14)

They are focused on open circuit voltage recovery following a deep discharge.

Basically, do they bounce back after discharge? The above concept seems to apply to NiMH chemistry based on my experience with Prius modules. I have yet to see how to apply it to IMA.

I suggest that after the 20A discharge, you record the stick voltage 24, 48 and 72 hours after the test. What we're looking for is consistent bounce across the sticks. They should eventually bounce back to nominal (7.2V) or higher.

 

5.2V is fine as long as the drop was fast from 6V.

I have a suggestion for the remaining 10 sticks:

First Cycle:
Set NiMH sensitivity to 20mV
Set capacity safety to 9600mAh
Set charge for 0.6A
Set discharge for 1A to 5.4V
Set cycles to charge/discharge, 1.
Initiate cycle.
The safety MAY cause the cycle to terminate once hit. If that's the case, initiate the discharge manually. Depending on model, the safety just triggers the next cycle operation.
Record capacities

Set NiMH sensitivity to 4mV (or your previous default)
Disable capacity safety
Set discharge for 0.2A to 3V
Set charge for 6A
Set cycles to discharge/charge, 1.
Initiate cycle.
Record capacities.

The cycles will be longer but likely more effective than 3-5 of your previous discharges.

You will now have a fully charged stick for the 50A load test.

 

I've never heard anyone recommend going as low as a 3V discharge. Are you sure that won't hurt the batteries? - Jen

 

When you're talking consumer grade cells, polarity reversals will damage and potentially ruin them. They have relatively high internal resistance because they're only designed to deliver very low currents and to retain their charge for longer periods of time.

Again, these are 100A cells with very low internal resistance. They are designed to flow in and out. Hybrid Automotive makes grid chargers and dischargers and they recommend discharging IMA packs to 60V and 12V for the whole pack.

I've taken entire IMA packs to < 2V. Yes, the ENTIRE pack voltage was 2V and 168 cell Prius packs taken down to 1.2V. I've taken sticks down to 0.1V. In fact, every IMA stick or Prius module I own (11 packs worth) have been taken to extremely low voltages. The key is to do it at low current for a reasonable amount of time.

I've had cells reverse for several minutes at 20A current, and yes, that will damage them in the form of lost capacity, but they can take it several times before they are ruined. These are very weak cells to begin with, and they are unsuitable for use in the car.

A member of the Insight community (who pioneered the grid charging and stick reconditioning concepts) REVERSE CHARGED a cell at 2A until it read -1.8V and charged. He did this several times. He then did multiple 30A discharge cycles and compared it to a cell that was never reversed. The performance between the cells was indistinguishable from one another.

I'm not recommending anything to you that I haven't tried directly or tried something analogous that will produce the same results. In fact, I'm recommending more conservative versions of things I have tried.

Steve

 

Okay, I will do the 3V discharge. Thanks. BTW, I've heard that the 3 sticks that have the temperature sensor wires attached to them can be charged in the pack. I am doing that right now and blowing a fan through the pack. But it makes it so that I don't get temperature measurements for sticks 12, 14, and 15. I didn't want to take the chance of damaging the wires or the clips for them. - Jen

 

I got excellent bounce back at 24hrs for the first five sticks (7.52, 7.53, 7.50, 7.8, and 7.52 Volts).

 

The 7.5 numbers are expected and good. 7.8 almost seems like it isn't discharged completely. I am in the process of discharging some Prius modules today to build a pack, and they've been about 7.75-7.81V, and they've had over 1200-2900mAh remaining at 20A load.

Is that a typo?