electric car drive

[franklin2]

I need some expert circuit advice. I am building an electric car. I will be using a Chevette. I have scrounged a GE 9.9kw motor and an ancient SCR drive that was in a forklift.

The SCR drive is of unknown origin from an old Clark forklift. From what I have been able to research, it is probably some form of Jones chopper. It has two large SCR's, a smaller SCR, a transformer, two large capacitors and other bits and pieces. The one important piece of the drive that I can't figure out is the epoxied electronic control module. I went to the local Clark dealer, but they can't help without knowing the model number of the forklift.

I was able to figure out the chopper part of the circuit though, and was able to test it with a small dc motor with 12 volts going into the chopper circuit. I found in this configuration, all I needed to do was to take the gate of the largest SCR and touch it's anode, and the motor would run. Take it off, the motor would stop.

So I had the idea I could build my own PWM circuit to drive the gate, thus making my own drive. I first built this circuit(low side version)

http://www.solorb.com/elect/solarcirc/pwm2/index.html

and it worked great by itself with the small test motor.

I then tried to mate this PWM circuit to my forklift drive. Here is what I did.

**broken link removed**

I first hooked the little transformer driving the primary and hooking the secondary to the gate. This pulled the drive signal to nothing when I hooked it to the gate. I measured the current of the gate when it is shorted to the anode, and it's about 50ma. I then hooked it in the configuration shown in the above diagram. The motor will run in this configuration, but will stop at either end of the speed control travel, and has very limited and unstable speed control in the middle.

I was wiring it this way according to a diagram I saw driving a Jones chopper. I assumed I needed the transformer for isolation. I will be running the PWM circuit on 12 volts, and the traction motor will probably have anywhere from 48 to 96 volts, depending on the top speed of the car. I am suspicious of the transformer not being correct. Do you think I could wind my own? Or have a source for one that would work correctly? I do need it don't I?

I have a 2 year electronics degree, but that was 20 years ago, and as you can see, I know enough to get myself in trouble. If you guys can offer any suggestions(besides spending around $1000 for a controller) I would be grateful.

Dave F

 

[Oznog]

I am suspicious of the transformer not being correct. Do you think I could wind my own? Or have a source for one that would work correctly? I do need it don't I?

I have a 2 year electronics degree, but that was 20 years ago, and as you can see, I know enough to get myself in trouble. If you guys can offer any suggestions(besides spending around $1000 for a controller) I would be grateful.

Well I don't wholly understand how the SCR commutates in this DC circuit, I assume there's something about the motor's current draw I don't understand. Anyways, an optoisolator is another way to deliver the small current to trigger an SCR. How much trigger current does it need? The opto can only deliver a few mA, and it comes at a high cost since it would be sourcing from a 96V rail. The transformer may remain the best idea.

 

[ante]

Hi Dave,

The SCR circuit does not work that way. Since an SCR run on DC just turns on and than conduct indefinitely it can’t work alone. One of the SCRs is the main SCR this will conduct the motor current. Another is the reversed to shut down the first once for every pulse. To shutdown the main SCR it needs reversed polarity at a higher voltage this it where the transformer and the big capacitor comes in to play. When the pulse opens up the main SCR the motor current also runs trough the primary side of the transformer (step up) and the pulse generated in the secondary is charging the cap. Now the shut down SCR fires discharging the cap (reversed polarity) through the main SCR thus shutting it down. This is how it commutates. Sometimes a third SCR is conducting the charge current to the cap. The control module supply both these (or three) pulses for the SCRs and at the same time checking lots of things like if a contactor is welded and the average voltage and motor current and so on. It’s not simple to mimic these signals since there are some unknown factors here. I don’t know Jones systems I have worked with several others like General Electric, Sevcon, Zapi, Cable Form, Sigma and Curtis. All the newer once are MosFet drives using PWM and some are inverter types (AC-Drives). Do you have a schematic for this Jones drive?

 

[franklin2]

Hey guys, thanks for the replies. I manually drew all the connections, and then made a homemade schematic of the drive. It seems to have all the features you were describing for the commutation circuit.


**broken link removed**

I hooked a potientiometer to 1,2,3, and jumped 8,9. The potientometer idea didn't work, which I was suspicious of since I could not get any sort of voltage out of 1,2,3 terminals to power the pot. So I started touching 12 volts to 1,2,3 and got some crazy reactions of the motor. I got nothing if 8,9 wasn't jumped, so I figured it must be some sort of interlock.

After this is when I found if I put power on the gate that goes to scr1, then the drive would run. Take it off, the drive will stop. That's is all I have accomplished, besides getting the smaller PWM motor control to work by-itself.

I am going to do some research on the opto-isolater. Is there any other device that would be fast enough to switch the scr on and off? I don't know what frequency the PWM controller I built runs at. I wonder if that is part of my transformer problem? The wrong inductance which is fighting the signals from the PWM?

Could someone also explain why Curtis controls sells a controller for a permanent magnet motor(cheap), and a controller for a series wound motor(the type of motor I have). I downloaded a Curtis manual for a series controller, and see where they have the seperate connection from the field. They where talking about "plugging"(being able to throw the drive in reverse to help stop the motor), but I would not use this feature in my car since it has brakes. If I could use this cheaper type of control, I would buy it and forget about this old forklift drive I have.

 

[williB]

was this the pdf you were looking at ?

https://www.electro-tech-online.com/custompdfs/2004/10/Curtis_manual.pdf

 

[williB]

I am not at all positive on this , mind you...
but the scr drive you have might be for a hydraulic system..ya know, to power the electric motors that pump hydraulic fluid....

can u take any pictures of the scr drive.?
you might be able to take it apart and scrounge the scr's out of it..

 

[williB]

I would not use this feature in my car since it has brakes.

they use this feature to recharge the batteries ..

 

[ante]

The voltage for the pot is usually in the range from 5V to 15V depending on the manufacturer. There is sometimes different demands that has to be fulfilled to “release” traction. This means that there might be no voltage in the pot circuit unless everything is OK and set to go. As I mentioned before the control checks certain thing like welded contactors, selected direction (which way to go), shorted SCRs, open motor circuit, that the cap is charged and so on. That was just a few of the possible checks that the controller performs before startup. Another one that can also be the issue in this case might be that a microswitch which is placed in the throttle unit, is supposed to initiate the circuit just as the pot starts to move (this is shown in the Curtis manual). Another question is should it run with a pot? Some controller’s uses inductive or opto “throttles” for their systems. These systems usually run at 1 – 4 kHz and the newer MOSFET pwm:s at around 15kHz. There are different options on the systems, some have a “by pass” contactor that shorts the main SCR when running at full speed to reduce the voltage drop (loss) over the SCR. There are also some with a regenerative braking system, which charges the battery when braking, or running downhill. The PM motor is easier to mess around with since there are only two connections, which makes the system simpler. Maybe you should check with Zapi and Sevcon too for best prize.

 

[franklin2]

I am not at all positive on this , mind you...
but the scr drive you have might be for a hydraulic system..ya know, to power the electric motors that pump hydraulic fluid....

I am sure it's the traction motor drive. I have the other large panel that has all the contactors for the pumps, forward-reverse, and main contactor.

As I mentioned before the control checks certain thing like welded contactors, selected direction (which way to go), shorted SCRs, open motor circuit, that the cap is charged and so on.

Yes, I am finding out more and more about this stuff. I have already figured out on the other main board I have, it has another control module and contactor that is a called the "high speed" module. I have figured out from reading and looking at the board, that this is the field weakening circuit. Basically it's a bar of metal that bypasses some of the field current, making the motor run faster with light loads. Who knows, the speed pot may have went into that module first, and then that module tells the scr module when to go.

It's a mess without the proper wiring diagram. That's why I was going to make my own circuit. Granted it would not have the safety features you describe, but I figured it may be better than the alternative which I am seriously considering. That is take the contactors I have, and hook them to micro-switches which are mounted sequentially to the gas pedal. Push on the pedal a little bit, and I will send one battery (12V) to the motor. Push a little more and I will send 2 or 3 batteries (24v or 36v) to the motor, etc. They say the drawback to this system is the battery position needs to be rotated because the starting batteries get more use than the full pack. I am also afraid I will mash the pedal to the floor from a standstill, and break something in the driveline. Maybe I could rig up a timer of some sort.

I am not picky. I just want something that will work. I am the type that will go into a project with not much investment. If it works out, and I use it alot, then I tend to spend more money on it later on upgrades. This sort-of handicaps me in the beginning, but I can't justify spending a bunch of money on something that does not pan out.