Troubleshooting a circuit

My problem isn't with home entertainment but this is as close as it gets. I have a controller that regulates the speed of a DC motor on a salt spreader for my truck. It isn't very complicated but the manufacturer wants $235 for a replacement. The problem is that when I start the spreader it goes to full speed almost as soon as I begin to increase the dial. It is numbered from 1 to 9 and by 2 it is on full blast. I have no schematic. I have some basic knowledge of electronics but am not sure even where to start. If someone knows where to look for more info or how to begin troubleshooting, I would appreciate any help at all.

Ken (from over in heating and cooling)

 

The motor control maybe more complicated than it seems if the diagrams on this page (scroll down a bit) are typical of their construction. Looking at these examples it may be best to just pay the money for a new one.

I suppose there's nothing to lose by having a look at the control since it doesn't work properly anyway but unless something obvious is apparent, electronic fault finding may not be worth trying.

 

This circuit does use the 3523 controller like the one in one of those diagrams but it appears in my circuit they control the positive side of the motor instead of the negative. There isn't much that could have happened to this control. It was only used a few times before I got it and even though it is 10 years old, it probably has about 15 minutes of use on it. I have no knowledge of any wiring mishaps but anything is possible. Thanks for the link to the web page. I'll study it a little more and see if I can make heads or tails out of it.

Ken

 

I think that the circuit should be very similar to the one that is in your unit. It looks like the IC converts the control signal into a series of pulses (PWM) which is then turned in to two anti-phase signals that are fed to the output MOSFETs in a so called "push-pull" configuration. This is where two transistors are used in anti-phase and is very commonly used in switching or power amplifier design.

If you could get an oscillscope you should be able to find the fault within a few minutes - just by tracing the pulses through the circuit. You're right about the cost - you should be able to fix it for a few $ at the most.

I'm not sure how I'd go about fixing it without a scope as there are many things that could have failed. Most likely something in one leg has failed keeping it full on rather than pulsing. Do the transistors get hot - if so it could point to this condition.

 

I don't have a scope but I can probably find someone who does. I'll check the temperatures tomorrow.

 

Couple of tips when you do the troubleshooting.

See if you can download a datasheet for this IC off the web. Place the control in the "off" mode and do some voltage checks with a DVM and make sure it corresponds to what the datasheet for the IC says.

You can also check the voltage on the bias circuits (like the 1K/4.7K divider). I'm assuming that you can either make up a circuit by looking at the circuit board or get a hold of one. If the PCB is not a multilayer one you can draw one up by looking at the tracks.

I personally have never encountered this IC but looking at the circuit you need to start at pins 11/14 and work you way through to the 2 MOSFETs with the scope to see where the pulses disapper.