DriveEV Logo
JeepEV: Electric Vehicle Conversion Project
Motor Controller

< Part Two >



Part One: Part Two: (Update)

« March 5th, 2005 »

Decision to get a new controller

     After driving my Jeep Cherokee around for over 9 months with the Raptor 600 motor controller powering the vehicle, I decided a controller upgrade was in order. Why? Well, it all comes down to vehicle weight and aerodynamics verses available power. When I first started the Jeep EV project, I thought that a 600-Amp motor speed controller would provide more than enough power for driving my Jeep. I based this expectation on other EV conversions (mainly light trucks) which used the same components I did in the Jeep, and got reasonable performance out of the vehicle. I also based this decision on advice I got from my parts supplier, who assured me that a 600-Amp controller would be sufficient. However, what he (and thus I) failed to realize was that the 4x4 Cherokee when all said and done is a heavier, less aerodynamic vehicle than most small trucks. The result was that the Raptor 600 ended up being undersized for the Cherokee. Yes, the Jeep drove fine around town and eventually got up to highway speeds with the Raptor in control, but it suffered from less-than-stellar acceleration performance, and struggled to get up large hills.

     The only way I could get better performance short of heavily modifying the Cherokee was to upgrade the motor controller to a more powerful unit, and so I did. See, basically the more electrical power you can throw in to the motor, the more mechanical power it will produce. To get technical for a moment: Power equals Voltage multiplied by current, or Watts=Volts*Amps. The most current I could hope to get out of the Raptor 600 was 600 Amps in both the battery and motor loops. Thus, the theoretical maximum power the Raptor 600 could output from my Jeep's 160 volt battery pack was about 96 kiloWatts (kW). Since there was no easy way for me to raise the voltage of my battery pack (short of using different batteries), the only way I could pass more power to the motor was to get a controller with a higher current rating. The new controller I purchased (to be identified below) is capable of 1000 Amps in the battery and motor loops. So the theoretical maximum power it can output from the Jeep's 160 volt battery pack is 160 kW, a decent step up from the Raptor 600 controller. Note that with both controllers, I probably haven't and won't ever easily achieve this maximum power output in real life due to many factors (such as battery voltage sagging, battery current limits, etc...). None the less, you can see that upgrading to a 1000-Amp controller still makes a difference.

     If I haven't put you to sleep yet, here's some information on my new motor controller ;-) . I purchased a Zilla series motor speed controller from Cafe Electric to use in the Jeep. The model I bought is the Z1K-LV, a 1000-Amp, 72 to 156-Volt nominal controller. If you compare my new Zilla controller to my old Raptor controller, you immediately notice three things. First, the Zilla controller is roughly half the size of the Raptor controller. This is because it is water cooled and does not need a huge heat sink built in like the air-cooled Raptors. Second, the Zilla controller consists of two boxes; the larger Zilla power stage, and the smaller Zilla "Hairball" interface computer. The Raptor was only one unit. Lastly, the Zilla is painted a nice funky green color, instead of a funky blue-purple color like the Raptor :-).

Motor Controller 2 Page Photo Motor Controller 2 Page Photo Motor Controller 2 Page Photo
The DCP Raptor 600 compared to my new Cafe Electric Zilla Z1K-LV.



« March 5th, 2005 »

Installing the new Zilla Controller

     As the main power stage unit of the Zilla controller is smaller than my old Raptor controller was, I had to modify the controller mount I made to fit the Zilla. I thought about making a whole new mount for the Zilla, but decided it would be easier to just modify the existing one. The first thing I did was take the Raptor and the mount I made for it out of the Jeep. I set the Zilla on the mount and determined that all I needed to do was add another piece of flat metal across the inside of the mount to support the back of the Zilla. This was a relatively quick process of grinding some of the paint off the mount, cutting a piece of flat steel to the right size, welding it in the existing part of the mount, and repainting the mount. I then drilled four new holes so I could mount the Zilla to the controller mount with four 1/4" bolts.

Motor Controller 2 Page Photo Motor Controller 2 Page Photo
The existing controller mount is removed from
the Jeep and modified to fit the Zilla.

     Once I had modified the controller mount, I reinstalled it into the Jeep. Then I bolted the Zilla to it. I purposely mounted the Zilla as far forward on the mount as possible to leave some space for the water cooling pump and reservoir to sit behind it. Once I had the Zilla bolted down I connected it up. There are only five electrical connections to the Zilla power stage. The motor positive (+) and negative (-) and battery positive and negative connect to the four large buss bars which stick out the front of the controller. The fifth connection is a Category 5e computer network cable which connects the Zilla power stage to the Hairball Interface. Cafe Electric did an excellent job of color coordination; the Category 5e cable matches the green color of the Zilla exactly!


Motor Controller 2 Page Photo Motor Controller 2 Page Photo Motor Controller 2 Page Photo
The Zilla power stage is mounted in place of the old Raptor.



« March 5th, 2005 »

Installing the Zilla Hairball Interface

     As mentioned earlier, the Zilla controllers consist of two units. The Hairball interface is the smaller, rectangular unit of the Zilla controller package. The Hairball Interface is where all the low voltage wiring connects to the Zilla, and where the Zilla's main computer resides. This includes the 12-Volt power and ground connections, the "run" and "start" inputs from the ignition switch, the inputs from the motor speed sensor(s), the outputs for a "battery" and "check engine" indicator lamp, and many other signal wires (many of which are for optional features that you don't have to connect/use if you don't want to). Also, the Hairball has a 6-pin phone jack which you use to connect it to a computer or PDA (like a Palm Pilot) through a standard serial port. From the computer or PDA, you can do neat things like change all system settings on the Zilla, view current system status (includes Data Acquisition), and check for any error codes. Cool!

     After searching for a suitable spot for the Hairball Interface, I decided the best spot to mount it would be up on the firewall, at the passenger side of the engine (now motor) compartment. I chose this spot because it would put the Hairball in a visible location, making it easy to access when I need to connect my computer to it. Unfortunately, I discovered that I'd have to mount the Hairball upside down to ensure that the wires connected to its terminal strip would not hit the underside of the hood. So I set the Hairball up in its future mounting spot and marked the location of its mounting holes on the metal of the firewall. I then drilled small pilot holes in the firewall, and used four self-tapping sheet metal screws to secure the Hairball Interface to the firewall. With that the Hairball was mounted. It does look a bit strange that the Zilla logo on the Hairball is upside down, but this was one of those situations where form had to follow function.


Motor Controller 2 Page Photo Motor Controller 2 Page Photo Motor Controller 2 Page Photo
The Zilla Hairball Interface is mounted to the Jeep's firewall.

     Next I began wiring up the signal wires to the Hairball. Due to the fact that the Hairball is mounted far from where the Raptor used to be mounted, most of the low voltage signal wires I needed to connect to the Hairball were also no where near it. This resulted in my having to redo some of the low voltage wiring, to route the necessary wires up to where the Hairball is mounted. It wasn't difficult to re-route the wires, just a bit time consuming (but what isn't?). Once I had the Hairball all wired up, I got the wiring put into split loom tubing and all cleaned up.


Motor Controller 2 Page Photo Motor Controller 2 Page Photo Motor Controller 2 Page Photo
The necessary signal wires are re-routed to the Hairball.



« March 5th, 2005 »

New Throttle Switch installation

     The last significant modification that had to be made to the Jeep in order to begin using the new Zilla controller was to replace the throttle switch. The DCP Raptor controller used an inductive throttle switch, which was basically a tube wound full of wire that you pull a metal slug through to change the switch's inductance and thus tell the controller to "go". While inductive throttle switches probably last longer than any other type, they aren't very common and are usually purpose built for the controller they are used with. The Zilla controller, on the other hand, uses a standard 0-5000 ohm potentiometer (variable resistor) based throttle switch. The most common resistive throttle switch used in on-road EVs, golf carts, etc.. are the Curtis "Potbox" switches. I ended up purchasing a Curtis PB-6 potbox which contains a microswitch in addition to the 0-5K ohm potentiometer built-in.


Motor Controller 2 Page Photo
The Curtis PB-6 Potbox throttle switch.

     I installed the new Potbox throttle switch in almost the exact same place as the old throttle switch. The installation was fairly straight forward. I removed the Raptor's old throttle switch and mounting bracket, and installed the Potbox to the firewall using four self-tapping sheet metal screws. I then used Zip ties to connect the Potbox arm to the end of the accelerator cable which is near by and now about four inches to short. Although I've discovered that zip ties work quite well as an accelerator cable extender, this is only temporary as I'm sure they weren't designed for this application ;-)


Motor Controller 2 Page Photo Motor Controller 2 Page Photo Motor Controller 2 Page Photo
The Potbox throttle switch is mounted to the Jeep's firewall.



« November 15th, 2005 »

Results of the Zilla upgrade

     Once I had the Zilla all hooked up, of course the first thing on my mind was to go for a test drive around the neighborhood. So, I got the batteries all charged up and went for a drive. The very first thing that I noticed was that with the Zilla installed, the Zilla and the electric motor *sound* very different than before. From a stop, the motor and Raptor controller used to make sort of a light groaning/whining sound as the motor came up to speed. Not the Zilla. The Zilla and the motor now make this low sort of hissing/blowing air sound from a stop. I've been told that this sound is the result of the Zilla rapidly modulating the PWM switching frequency at low motor speeds, whereas the Raptor controllers have a fixed frequency that doesn't change. But regardless of that, it is really cool to put on the throttle and hear this responsive pssshhhshshhssh sound as you crank up the amps and take off into the night (or whatever ;-). The next things I noticed during my first drive with the Zilla is that my Jeep has a *lot* more power and "punch" now, and that accelerating from a stop is much smoother than it was with the Raptor controller.

     So, if you haven't figured it out yet, I really like my little Zilla controller! It wasn't a cheap upgrade by any means, but I feel it was well worth the money spent because my Jeep's performance is now as good as I think it should be. Also, it is really neat to be able to hook a computer up to the Zilla and set all the settings and see exactly what the controller is doing. Cafe Electric really makes a good product and the Zilla is a work of art. I just wish I had gone with a Zilla in the first place and saved myself some time and hassle playing around with [arguably] inferior motor controllers!


Remember, More photos are in the Photo Gallery!