The Electrical System:
June 10, 1998
The 12 Volt System: July 17, 1998
Happy New Year! Well the Escort is covered in snow, but I still do what I can in the way of design work.
There has been an interesting discussion on the EV mailing list about EV safety. A lot of the discussion has been focused on the best way to kill power to the controller, if it fails in the "ON" condition.
One of the things that surprised me, is that not everyone is using two main contactors. Bob Batson at EVAmerica has a good write up on EV safety. One of the design critera he recomends is redundancy. That is, if a safety component fails, there should be a back up way to perform the function.
Therefore, it makes most sense to use two contactors instead of one. It especially makes sense, if you're using a precharge resistor with your controller. If you use a precharge resistor and only one contactor, your controller is always on. Also, by having two contactors you can completely isolate the motor and controller from the battery pack (unless of course you're using a precharge resistor).
Bob Batson recomends that the primary contactor be energized by the ignition switch, and the secondary contactor by the potbox switch. The potbox switch is open when your foot is off the gas pedal, but closes as soon as you press on the gas pedal. Thus the secondary contactor clicks on when you put your foot on the gas, and off when you take your foot off the gas.
This way, if the controller fails in the "ON" state, once you take your foot off the gas, power to the controller is killed. You might say, if your contactor is wired through the ignition switch, why not just turn the ingition switch off? On all cars made since about 1970, when you turn the key off, it locks the steering column. So if you cut power by turning the ignition key off you'll lose the ability to steer.
Then someone on the EV mailing list said he disengaged one of his contactors with his brake light switch. That set me to thinking. Why should you disengage your secondary contactor just because you take your foot off the gas? A lot of times you might just be coasting. Why not make the brake pedal the failsafe mechanism for the contactor?
If you look at the schematic above you will see a lot of safety interlocks. First there's the intertial switch to cut power in the event of a crash. Second there's the charger relay, to prohibit driving the car while the battery charger is still plugged in.
From there, the primary contactor's coil directly connects. The secondary contactor has some more complicated relay logic. First it goes through the normally closed (NC) contacts of a relay that is activated by the brake light switch. Everytime the brake pedal is pressed, it will cut all power to the coil of the secondary contactor.
Now the circuit gets a little complicated. To start with, it splits up. One branch goes down to the potbox switch. From the potbox switch it goes to the KSI relay (which provides power to the KSI input on the Curtis controller). From the potbox switch is goes to the coil of a DPDT (Double Pole, Double Throw) relay. One set of contacts in this relay provides power to the coil of the secondary contactor. The other set of contacts act to latch this relay on once it's powered up.
The KSI input turns on the controller. It's also used for high pot protection (HPD). That is, when the KSI input receives power your foot can't be on the accelerator. If it is, the controller will not turn on until you take your foot off the accelerator. The prevents the car from ever lurching forward on startup.
To power up this relay, the potbox switch must close; however, once the the relay is energized, it stays on due to the second set of contacts. The only way the relay is turned of, is if the brake is pressed, activating the brake light switch, and killing all power to that part of the circuit.
There's actually a cleaner way to do it that cuts out two relays. Use a 4000 series CMOS D-type flip-flop. Have it be set everytime the potbox switch closes and reset every time the brake pedal is pushed. I decided to use relays since they're more familar to most people. I'm leaning toward using the 4000 series CMOS instead of the relays. The design and a description of it can be found by clicking here(2/5/00).
As I said, the Escort looks a little forlorn now. Sitting in the back yard covered in snow. I ordered the MDO board, main contactors, and other miscellaneous parts for the control board after Christmas and they arrived before the New Year. My used Curtis 1221C controller is in the mail and should arrive some time at the start of this week. I can plan the layout of my control board, and drill holes and screw things into place.
The motor bracket is still not done. I guess since it's only a $25.00 job my welder friend doesn't have much incentive to do it. I visited him today and dropped off the plans for the controller board bracket. Maybe the two of them will come to enough money to get him interested.
Designing a Controller
Board Bracket
Installing the Right Hand Motor Bracket
Conversion Index