Brian are you seeing this through the run, or just from a standing start? I could see this through a run because I think system reactance would come in to play at differing loads and RPM's. As you know with audio, Impedance changes with frequency. As far as the spikes could that be from the pulse from the non charged coil?

The other thing is that these things are slow (only 10 Hz) they are a great tool, but for really measuring and looking at data it's like having a high speed camera @ 1000fps versus a home video camera @ 30fps. I really like mine, but it's going to be subjuct to errors just like everything else, and at only 10Hz sample rate I think it can "grab" data at slightly different times and give you less accurate readings, especially as through as you are looking at the data. This wasn't a controlled bench test so I think that more dynamic factors can happen here.

Jeff

Was at a standing start. Actually, it was when I drove my truck under the chain link fence. The tires got jammed and I was too lazy to go get it so I tried to back out. Jammed tires + WOT = high current. :smile:

A little update on this topic:

Still no response from Eagletree systems (and I checked my spam box). So, I guess they simply don't want to answer or they never got it.

Anyway, I recently took other Eagletree graphs of different vehicles and have logged currents as high as 168A on the same unit. So now I'm not sure what happened on the original test. The only two explanations are 1) the lipo cell resistance increases dramatically above a certain current (unlikely because cell resistance simply doesn't change that quickly), or 2) battery voltage tends to react a little slower than current spikes, so maybe the majority of the current spike happened in between samples, while the battery voltage dip averaged out due to it's slightly slower response.

Isn't there another data logger available that reads more frequently than the ET and medusa. I remember reading about one but it was being used for a scuba scooter. Very basic with near no versatility, it just took readings at specified intervals and the memory cleared upon upload of data to computer. I'll see if I can find it.

I'm not sure what kind of sensor the medusa uses but I sent an email to them and there reply was the shunt is used to measure the current. They said it would be OK to shorten the sensor wire e.i. cut off the bullet connector and solder it directly to a motor phase. I still haven't used it yet :whip:

Found it- the BNB DPR 150. Some are happy with it some are not. Someone reckoned it can sample up to 50 times per second though I haven't checked on that.

Has BNB gone out of business? I can't get to there website.

I realize the ET is between the batteries and caps but is it possible the caps are holding the voltage up hence the delayed drop in the batteries?

Maybe for a couple microseconds. 1000uF doesn't do a whole lot to store a charge when you are drawing well over 100A.