Poor man's discharger, preliminary report

[DimensionDude]

Hi all,

Got the replacements for the multimeters that I destroyed (been in industrial maintenance for 20 years, still can't believe I made such a newbie mistake). So, I'm doing a charge, discharge, charge cycle.

The use of the dual-lamp work light as a discharge load is not my original idea, it came from the website for the really nice grid charger.

The work light has been modified by wiring the lamps in series to survive the high voltage coming from the IMA pack. I considered wiring it as a dual purpose light (discharge load or work light) but thought better of it (see above reference to newbie mistakes). Much safer to have it as just a discharge load.

The photos are from very early in the discharge cycle. The first charge went to 172.2 vdc and stayed there for 2 hours so I considered it done. Total charge time was approximately 20 hours at 260 ma.

The discharge started at 1.8 amps and slowly dropped, ending at 1.3 amps when the voltage reached 132 vdc and I stopped the discharge. For the moment, it's going to be completely manual but I'm working on a few ideas to automate it. The discharge took approximately 2.5 hours, maybe closer to 3. I logged the voltage but not the current during discharge, a more powerful laptop is on the way.

The battery was allowed to rest for a couple of hours with the fan running and then the charge cycle was started.

Feel free to ask any questions, I know I haven't explained it thoroughly here.

Kent

[retepsnikrep]

What software is running on the laptop and what exactly is it measuring/logging?

[aisbell]

How do you detect when the cell with the lowest charge level is about to reverse which could damage it?

[DimensionDude]

Originally Posted by retepsnikrep

What software is running on the laptop and what exactly is it measuring/logging?

The software is (I think) DMMView and it came with the Mastech M9803R multimeter. In this particular setup I'm logging only the pack voltage but I'm looking into logging voltage and charge/discharge current.

I'm not thrilled with DMMView as it saves to its own proprietary file type. I'm going to experiment with QtDMM and QtDMM2.

Originally Posted by aisbell

How do you detect when the cell with the lowest charge level is about to reverse which could damage it?

The short answer is, I don't. It's a calculated risk but I'm hoping to reduce that risk by being conservative with the cutoff voltage. Also, I'm going to review the discharge voltage graph for any sudden drops before I perform any more discharges.

I'm always on the lookout for useful software and may try to exercise my own rusty programming talent.

Kent

[FrugalSteward]

I have a few questions before I do.
1. Is a sudden voltage drop the key as to when to stop the discharge?
2. What voltage does anyone else think it is safe to go down to?
3. Using the same 600W (2-300W) work light with lamps in series, is there anything else I should consider?

[DimensionDude]

Originally Posted by FrugalSteward

I have a few questions before I do.
1. Is a sudden voltage drop the key as to when to stop the discharge?
2. What voltage does anyone else think it is safe to go down to?
3. Using the same 600W (2-300W) work light with lamps in series, is there anything else I should consider?

Before you try this, be advised that there is a chance of damaging the battery pack due to cell reversal.

I did this only because my pack is on its last legs so I didn't really have much to lose.


But, to answer your questions:

1. A sudden increase in the rate of voltage drop is the real key, but it's a bit easier to see a quick drop in voltage (such as a sudden 0.5v drop).

2. I chose 132v (1.1v per cell) as a conservative cutoff value owing to the unknown condition of the pack. I'm not sure what the absolute minimum cell voltage is, 0.9v comes to mind.

3. Consider that sitting and watching a voltmeter can be more than a bit tedious but it's easier than pulling the pack and cycling the sticks. Granted, the individual stick cycling is safer (for the pack) and much more informative, but it takes so long to complete, and the car is not a joy to drive without assist.


One important thing, a discharge may be performed ONLY upon a freshly grid charged battery so that all the cells start from the same point.

[Mike Dabrowski 2000]

It is safe to go down to the point where the first cell drops out. That can happen between 148 to 120V. some cell drop out graphs are here:
http://99mpg.com/projectcars/underst...ilverpacktest/

Don't blink

[3-Wheeler]

Originally Posted by Mike Dabrowski 2000

It is safe to go down to the point where the first cell drops out. That can happen between 148 to 120V. some cell drop out graphs are here:
http://99mpg.com/projectcars/underst...ilverpacktest/

Don't blink

Mike,

To add to what you said above....

Since replacing a weak stick, with another one purchased "in the wild", my pack currently stops Assisting at 157 volts. Yes, this is not a misprint, as I saw this value just two nights ago when the BCM stopped the Assist.

This voltage is solely based on a stick that I put in the car, and it self-discharges at a very high rate. I was foolish enough not the check this before installing into the pack.

In summary, with a "normal" pack, I would say that Mike's comment is correct, however with my current pack, the cutoff voltage is much higher.

Jim.

[eq1]

Does the car detect cell drop out, and does it then disable assist?