How to get schematics out of circuits

After my last blog, the hacked amp, I was asked by quite a few people how I got the schematics.
In this blog, I'll try to give a deeper look on reverse engineering circuit boards.

So if you have a old crappy electronic device and you take out the PCBs, what is first?
You look for the datasheeets for all the chips on there.
I didn't knew I was holding an amp in my hand when I was looking at the board.
I thought it was kind of an digital to analog converter with low output power.
But I googled for the datasheets and found the MS6038 datasheet.
After I found it was a small amplifier, I thought, hey this would be rather useful in my workshop.
By accident, there were no piezos on the mainboards but normal 8 ohm speakers.
 So I pulled out the the pin assignment table from the datasheet

Pin configuration out of the datasheet

Like you can see, it ain't very different to the pin assignment of a dual operational amplifier.
Now, you could look at the application unit in the datasheet, but it isn't necessary the same as on your PCB.
And what I did, was that I took a photo of the board and used a image manipulator like GIMP or Paint.NET
to assign pin names.

Pin configuration in the circuit

When I was done, I added names and descriptions for the rest of the PCB.



I found the places where I soldered the VCC and GND(Vss) cables by following the tracks, scraping off the solder-mask at a position where I had enough space to solder it and checking it with my multimeter in continuity test mode. The old socket for the headphone plug had three solder points. I metered them all to get the ground point.

I knew there were electrolytic capacitors in the signal way so I couldn't test it.
But if you have a two-channel amplifier, it is pretty sure you will have a two-channel output.

Now I looked on the trimmer circuitry. It has only two outgoing wires and they even had tinned test points on them. A little step under the yellow text you can see my first trial to get contact to the tracks, but I had problems with shorting them out to ground.

The last step was soldering cables to all the points I figured out and testing it.
I used a square wave generator built of three parts: a 555-timer, an one-kiloohm-resistor and a 100nF capacitor. It generates a signal around 700Hz and allowed me to test the amps functionality with my oscilliscope.
You needn't do that, I just wanted to see the level of volume raising.
The output volume is about 140 percent of the input.

Then I wondered why I didn't destroy the amp due to my eight-ohm-speakers
while my amp was designed for 32 ohms.
As I metered the resistance of the output, I found it being at  24 ohms. 8 + 24 = 32, so everything was okay!


So now you know my procedure of refactoring a PCB.
Whats next?
Try it out yourself.

And I have something coming up built with the cd-tray.

\apexys\toan\11212010\1912