Mystery Circuit

[metrologist]

Hi All. While rummaging through some of my old boxes of stuff, I came across 4 of these boards. No idea what they were for or how I got them, but I think they may have some application as a brushless motor driver or something. I'm puzzled by how they would work or what voltage to supply. It appears the Source of the IRF840 fet also connects to the neg lead of a tantalum, if I've drawn this correctly.

[Cliff Matthews]

Just op amps as FET drivers.. But they don't look well isolated from one another. Paste without nylon washers still makes all those heat sinks conduct and two channels look close to shorted.

[CJay]

Just op amps as FET drivers.. But they don't look well isolated from one another. Paste without nylon washers still makes all those heat sinks conduct and two channels look close to shorted.

It may actually be OK*, hard anodisation is quite a good insulator on aluminium but I'm more inclined to think they've just been knocked about a bit in storage.

*As in the heatsinks may not conduct if they touch each other, I realise the FETs aren't isolated from then because there's no insulating washer on the screw

[metrologist]

OK, I'm not that savvy on FETS. What I don't understand is why the Source pin is connected to what I think is ground. There are many pins bridged and going to the track that runs around the perimeter of the board and some of the central tracks. It looks like the ground plane to me. Also, the Drain jumper wire is very small.

On the board, there are two adjacent pins - red one goes to the non-inverting input of the op-amp, the other blue goes to the FET drain output. I thought, cool, I put in a small signal and get out a big one. That does not happen.

I applied 5V to the other lone pin and connected all the ground pins together as ground. When I apply 5V to the "input" red pin, the "output" blue just stays low with less noise.

Then I thought perhaps the Source for the fet would be a separate + supply, but that does not work with the rest of the circuit (the tantalum would be reverse polarity). After that, I thought maybe whaever device is connected to power and return goes through the fet (from drain to source).  Maybe I just need more than 5V?

[langwadt]

I'm guessing linear driver for unipolar stepper motor, coils go between supply and drains, opamp sets gate drive to sorta constant current

[Gyro]

It's not a "Mystery Circuit".

You need to trace it out and draw a proper schematic, not lines between package pins, veroboard bottom photos etc. You have to benefit of being able to buzz out and verify connections, which we don't, just peering at photos.

Once you have drawn the schematic then, if its function isn't obvious, people will be able to have a decent chance of telling you what it is rather than guessing.

[metrologist]

OK, sorry. I'm just not very good at this. I did find a free online schematic app at digikey. I probably have some mistakes in this. What would I connect to the 5 pins to test this?

[Gyro]

Well you're getting better at it  . Practice makes perfect.

I suspect that there are some errors, yes. Incuding passive component values would help identify these better.

So what do we know so far?

- The circuit has an open drain output.
- The Mosfet is being driven by a non-inverting amplifier circuit. R7 is there to prevent oscillation of the Opamp / mosfet gate.
- Pin 1 looks like ground, but so does Pin 4 as C3 look is a decoupling capacitor for the opamp supply.
- Pin 2 is the analogue input with R8 and R9 forming an input potential divider.
- Pin 3 is your positive supply for the opamp.
- R5 and C2 look right as the opamp feedback path
- Really not sure that the bottom of the schematic is correct (R2 - including wiper connection, R3, R4, R10 and C1).
- It's possible that one of the Resistors (R4?) might be being used as a current sense resistor, you need to add values to make this clear.

It's unlikely that the opamp is there just to buffer the mosfet gate. My guess, when the schematic is clearer is a set of variable current sources (assuming that one of the resistors is in series with the Mosfet source).

Keep at it. 

[metrologist]

The mosfet source pins are all connected to the outer track that then has the large white wire (circuit foil). I'm puzzled because the - side of the tantalums are also connected to that track.

Also, the pot has one end and center pin connected together. I might be able to buzz some of the circuit tonight and verify the connections.

As I mentioned, I did connect 5V to pin 3 and grounded pins 1 and 4. There was a repeating pulse on pin 3 and a larger pulse on pin 5, and this was intermittent when I applied power, so I assume some circuit oscillation. Then I connected 5V to pin 2 and pin 5 stayed at flat line. This made me wonder if I had a negative voltage on pin 5, would current then flow from pin 5 to 1 and be limited by the voltage on pin 2?

Hoping this can be used or modified to drive a 3-phase brushless motor (3 wires).

[CJay]

Yuo're aware that Tantalums are, in contrast to 'normal' polarised capacitors, usually marked with a stripe on the positive connection right?