Don, On the "ol" treadmotors and controllers, I have one and was wondering... The motor has a pair of tach wires, and with those disconnected from the controller, the motor won't run. Is there a way of using the controller for a variable DC source without having an input from a tach? I have other DC motors without tach wires and one of these controllers would be handy to have for them. Thanks. BTW, I did the modifications (removing T3 and jumpering chip 6 to chip 8. Works great. I'm also assuming that reversing the brush connections would reverse the rotation (currently clockwise looking at end of motor shaft). Thanks again. Ken.
I'm not the one that dissected the old treadmill controller, though I think I do have one here somewhere. If you can find a schematic somewhere, I might be able to help you better.
If that is one of the controllers that uses the TDA1185 chip, it is possible to do open-loop control (no tacho input) with that chip. I don't know how much surgery would be required without looking at a schematic of how it's used in your controller.
Best speed control is achieved with a tach. It wouldn't be hard to add a tach to a motor that doesn't have one. Just use a gear or a toothed wheel you make by notching it with a mill, drill, file, whatever, put the gear or wheel on the motor shaft and sense the notches with a Hall geartooth sensor.
I was afraid of that -- but thanks for finding it. Wayne Cook drew that schematic by tracing the board. The way T3, T2 and ZD1 are shown don't make any sense to me
The TDA1085 has a provision for shutting down if the DC voltage on pin 12 (tach input) is above a certain level, which it would be in this circuit if no tach is connected. Try connecting the tach input (pin 12) to pin 8, which is the "ground" pin for the chip.
I think you'll find that this circuit will work a lot better with some sort of speed feedback. It's not intended to be an open-loop control. It compares "true speed" (voltage on pin 4) with "set speed" (voltage on pin 5) with an internal error amplifier. Even a slight difference between these two voltages can swing the output drive from min to max. So there needs to be some speed feedback so the chip can adjust output to make the actual speed (voltage) very near to set speed (voltage).
You might be able to trick it into being just a phase angle control (like a lamp dimmer) by putting a 10K pot from pin 9 to pin 8 with wiper to pin 16 and use that for your speed control. That would override the error amplifier. I'm guessing that 10K might be about right. You could probably use the existing 10K pot. You might need to ground pin 4 by connecting it to pin 8. If this works, your motor controller has considerably beefier output devices than a lamp dimmer and is much better suited for driving inductive loads like a motor. Most of the rest of the stuff on the board to the left of C16 then becomes irrelevant.
Specsheet for the TDA1085 at
Dang, I thought these had gone out of production years ago!
Don, I printed out your response for further investigation. Thanks for your help. BTW, I've had these motors and controllers laying around for several years, so that chip may in fact be out of production. I originally got them from Surplus Center when they were the "hot" item.
They never made any sense to me either but the end function is to require the speed control pot to be turned to zero before the motor would start up. Cutting T3 off the board will disable this function.
IRRC the tach on the original motors created a pulse. I don't know if this makes any difference on the above.
That it does -- for sure. I guess since these are treadmill motors, it only makes sense that if you're standing on the belt and turn on the switch, you wouldn't want the belt to take off at about 8 or 10 MPH from a dead stop..... (I know, I know, sense of adventure, etc.,)
The original motor has a "cap" on the end with fingers to pick up pulses. I haven't disassembled one to see what's there but the "magnet or other electronics" is inside the center of the "cap" and mounted around/to the shaft. It probably has about 8 "fingers" which would produce the pulses mentioned above. Thanks. Ken.