After three years and 12,000 miles on my converted pick-up, I would like to share some of my experiences with the discussion list. Maybe something I say here will help someone just starting to avoid some of the mistakes I have made. 
[B]Sizing the controller [/B]–
One of the benefits touted for electric motors is that they can give you maximum torque right from the beginning as low rpms. However, none of the books on EV conversions (or other information) make it clear that the relationship between controller, motor, and battery pack is much more complicated. I need about 32 hp (24 Kwatts) of power to go up the hills on my commute. This means that the motor would be drawing about 200 amps from my 120 volt battery pack – quite reasonable for flooded lead acid batteries. So a controller rated at 400 amps maximum should be OK, right? WRONG! What no one told me was that at low rpms the motor voltage is much lower. I have an Impulse 9 motor and have measured that at 2000 rpm it only is running at 44 volts. This means that the controller has to be putting out 545 amps to get the necessary power (32 hp).
This was my first mistake. I purchased a Curtis 1221 controller rated at 400 amps. I had real trouble starting up hills at lights because I did not have enough current to the motor at the low rpms to get the power I needed even though the controller was putting out the 400 amps. At 2000 rpm I was getting only 23 hp (about 17 Kwatts) to the motor. When starting, the power was even less. So I was very glad I kept my manual transmission. It allowed me to drive around with the Curtis, albeit at a slow speed. I was able to get a high enough rpm in first gear to get enough power to the motor to get going.
My suggestions: The rating of a controller should be about 3-4 times higher than your expected current from the battery pack when going up hills or starting. I needed 200 amps from my battery pack. I found that a controller rated at 600 amps (Kelly) was just barely adequate at one at 700 amps (Logisystems) and 800 amps (Kelly) did the job. I now have a Net Gain rated at 1000 amps and that works the best. If you decide for a direct drive and eliminate the transmission you will probably need to double that rating.
[B]Five controllers blow up –[/B]
I replaced the underpowered Curtis with a Kelly KDH12600B (600 amp controller). This was adequate but blew out while driving after about 3 months. The folks at Kelly replaced it without any charge and let me upgrade to a KDH12800B (800 amp). Unfortunately this one lasted only one month and blew out. The folks at Kelly replaced it with another 12800B. This one lasted only 1 week before blowing up. So, the folks at Kelly replaced it with a KDH14650B. This one ran OK but I was reluctant to use it for very long. So, I purchased a Logisystems 700 amp controller. This one lasted for 1 year before blowing up. Logisystems repaired it under warranty but the repair only lasted four months before blowing up again.
Warning…when these controller blow while driving it can be a harrowing experience. There is a big bang and everything dies. You just hope there is a place you can pull over before the car stops.
So why did all of these controllers keep blowing up on me? Either I have bad “karma” with EV’s or there is something in my system that is causing failure of the capacitors. :eek: All I have been able to find out is that the electrolytic capacitors used in these controllers have weak points. The gearing or driving style may cause duty cycle to hover around certain values that can maximize ripple currents the capacitors must handle, and therefore work them to death. Anyone have other ideas?
[B]Where am I now?—[/B]After the Logisystems controller blew up the second time I decided to purchase a NetGain controller rated at 1000 amps. The newer version of the NetGain controller and the Soliton controller both use thin-film capacitors rather than electrolytic ones. These are supposed to be less prone to problems. We shall see. The NetGain controller however was only two thirds the price of Soliton so I went with that one. I have had the NetGain installed and running for five months now without blowing up, and it is providing me with more than enough power. I also purchased the computer module (called the Interface Module) that allows me to monitor over a dozen different factors and alter controller defaults. Five of the most important variables (current to motor, voltage at motor, current from battery pack, voltage at battery pack, controller temperature) are all on one screen so I don’t need five dials. It also stores the data on a disk that I can download to a spreadsheet. This is really cool!
However, the NetGain has not been without a problem. I would call this more “teething” problem since it involved the internal software (called firmware) rather than the hardware. Nothing as drastic as getting stranded on the side of the road. When I first installed the NetGain I would get sudden power drops I can best describe as “hiccups.” The folks at NetGain figured out that it was a result of a communications snafu between two circuits in the controller caused by electrical noise from the high current wires. I guess I do have bad “karma” for EVs. This noise problem had not shown up in all of the previous controllers they sold. I sent the controller back to NetGain for a firmware fix and that solved the problem. [I also got their latest version of the mother board in the process that further reduced the footprint of the controller by 1 inch.]
So now, finally, I have a controller that has the power I need and hopefully will not blow up on me. I should also mention an additional benefit of the NetGain controller. It uses a Hall effect throttle rather than pot box and has a much better response when starting up. All of the previous controllers caused a sudden “jump” in the car when I stepped on the throttle. The Logisystems was the worst. The NetGain however is very smooth and does not do this.