I forgot to document earlier that turning the controller up to 120% per Kepler's advice gave me a top end of 51+ mph, and really highlighted that I needed better brakes. Recently I freed up some vacation time to finish sorting out the Norco. The brake adaptor was one of the recent challanges holding things up, but I also needed to sort out the rShunt for my methods controller.
I bought a CycleAnalyst v2.2 CA-SA from someone here on E:S in May 2011, but never installed it. I tried to use it to calibrate the method controller's shunt, but it the results were not making sense (I later found out it is defective) so I decided to add a third CA to the mix:
I have two other shunts I had cut off from V1 CA-SA, but I could only find one of them, and the rshunt value on it rubbed off. I checked the thread for known shunt values, and found my own post documenting that several of the V1 CA-SA I bought all had an rshunt of 1.399 (lo-Mode), so I used that value, and the results appeared to be in the right ballpark.
Testing with the third CA gave me an rshunt of .2418 (Hi-Mode) for the Methods Controller. The CA-SA I was working on seemed to have an rShunt of 1.50 (lo-Mode), but sometimes it worked and sometimes the values were wildly incorrect, so I determined it was defective, probably the shunt itself. I'll buy a replacement from ebikes.ca, and re-calibrate the Norco CA again.
Now that I could measure discharge with reasonable accuracy, I could take it for an official Maiden Voyage. Since I was baselining the pack on this trip I'd be monitoring the CA for rapid voltage drops (indicating a bad cell), and figured I'd use a mental LVC of about 60v (2.5v/cell) to terminate the test. The current pack is (24) of the a123 16ah cells I got from cell_man with 1,279wh capacity, so I hoped to get 1,023wh from it with an 80% DOD.
I stayed close to home and used an aggressive duty cycle, running up and down hills, stopping at lots of intersections and accellerating hard away. I used the "street" preset with current/speed limits that I programmed into the CA-V3, as the "Unlimited" preset would pull 10kw and make it hard to track the voltage decline as the battery discharged.
The throttle became unfcomfortable so I need to re-engineer it. The brake adaptor I milled worked perfectly, the 9" Zero Moto rotor gave better grip than the 8", but I had to pump the Gatorbrakes lever a few times before each stop to build up pressure, so I guess I need to change the fluid. The chain is only long enough to use the middle front chainring, I need to make it longer to accomodate the large front chainring so I can pedal at speed. When I was done, the motor was hot but touchable, the methods controller was merely warm, and the battery packs were warm but not hot or in any way puffy.
Dispayed Amps peaked at 152.5a, and I pulled an indicated
1078.2 wh when the pack had dropped to 63v under load. The aggressive duty cycle gave me 17.4 miles at 61whm, I expect more range from my "eBike" preset, and less from my "Unlimited' preset. The Kill-a-watt said the subsequent charge cycle used 1,450 wh, which would indicate that the cheap chinese charger is about 74% efficient, better than I expected. Anyhow, here are a few pics from after the ride, I guess the only major change is the front brake.