I got out my working notebook with it's data acquisition system, and wrote a quick labview program to monitor, display, and write to a file what is happening on each cell. I will add AH computations tomorrow.

The high current connections worked reliably through out the test, with only a slight rise in the #10 wire temperature to show for it. The discharge test at 35A, showed an initial voltage difference on cell #1 (the shorted one)of about .08V it lagged the other cells by about that amount for most of the test.
I got a long 1-3 minute discharge at @35A,beforethere was a rapid drop off of cell #1 voltage, just as the others started to accelerate their downward plunge. It was so fast I couldn't get my cameras turned on fast enough to shoot it.
So the shorted cell performed at say 80% the capacity of the others, and had a lower fully charged voltage than the others at full charge.
I can't help but think that I need a better zapper.

I took some photos of the big alligator clip and Stainless steel bolt head that was my capacitor zapper switch.The amount of plasma damage suggest that we may have lost a good deal of the energy in the contacts.
Now that I made a spotwelder that can put out 10,000A, controllable length pulses, I need to try this again.

We must remember that these cells are designed to produce 100A, so we need to hit it with many times that if we hope to vaporize the dendrites.