Identifying Sanded Down Opamps

How do you trace a circuit from a stompbox you have? How to remove that ugly goop in your boutique stomper? All secrets are releaved here...
Post Reply
User avatar
george giblet
Resistor Ronker
Information
Posts: 440
Joined: 30 Dec 2008, 17:27
Has thanked: 245 times
Been thanked: 351 times

Post by george giblet »

A common thing done by boutique pedal manufacturers is to sand down the face of the opamps to remove the part number.

There are a few tricks you can use to reduce the unknowns:

1) BJT input vs JFET/MOSFET input.
Key parameter: opamp input bias current.
You can measure the voltage across the 1M or 470k bias resistor on the input stage.
If the voltage is zero it is likely to be a FET input opamp.
If it is say in the 2mV and above, say upto 300mV, it is likely to be a BJT input opamp.
Something like a NE5532 will have a very high input bias current.
You can use a larger resistor to get a more accurate figure.

2) Input protection diodes
Some opamps have input protection diodes (eg. NE5522) which can be measured with a multimeter.
They appear across the + and - inputs.

3) Opamp supply current.
This is a weaker test and it is complicated by the fact datasheets are done at certain supply voltages,
which aren't 9V or 18V, and the current at other supply voltage may vary. Datasheets might show maximums
when your opamp is closer to typical.

4) Output swing and Output swing under load.
You actually want the saturation voltages ie.how close the output swings to the rail.
You can add a load to observe behaviour.

5) Output current limit for positive and negative shorts.

6) Slew-rate. Sometimes the shape of the waveform produced under slew-rate limit
can give a hint to the type of opamp.

7) Bandwidth. Build a circuit, or use the actual pedal, and measure the bandwidth to get an idea of than opamp bandwidth.

8 ) There's other parameters like Common mode voltage range or what the waveform looks like when
the inputs are biased close to the rails which can be unique to each opamp.

If you compare against datasheets you need to be aware these figures may vary with supply voltage
and the parameters will not be the same at a different voltage. Also be away of min/max specs vs
typical. Your opamp is likely to be closer to typical. Using the graphs in the datasheets is often
better than the tabulated data because it is more like typical data. (you might need older versions of
datasheet as the graphs often don't appear in later datasheets)

Once you have narrowed down the list. You can check its behaviour using above tests to see if it is similar to the
sanded-down opamp.

Anyway the main point is here is not the specifics. The idea is each opamp has a certain key parameters that makes it unique. You can extract info from outside the box to reduce the choices from completely unknown to a small set of candidates. A final ear test is obviously required!

User avatar
lcv
Information
Posts: 28
Joined: 28 Jul 2009, 08:24
Has thanked: 6 times
Been thanked: 5 times

Post by lcv »

Decap the package and inspect the die directly is an option too
Lcv

User avatar
Dirk_Hendrik
Old Solderhand
Information
Posts: 4208
Joined: 03 Jul 2007, 08:44
Location: Old Amsterdam
Has thanked: 234 times
Been thanked: 896 times
Contact:

Post by Dirk_Hendrik »

lcv wrote:Decap the package and inspect the die directly is an option too
Lcv
Inspect for what??
This is only fun when you decap a lot of opamps and make a reference library for them.....

Or... let's say I consider Georges suggestions more suitable for now.
Sorry. Plain out of planes.

http://www.dirk-hendrik.com

User avatar
lcv
Information
Posts: 28
Joined: 28 Jul 2009, 08:24
Has thanked: 6 times
Been thanked: 5 times

Post by lcv »

Hi Dirk
You are right, the option I mentioned is more related to professional hacking, so it's not very practical.
You'd need chemicals, access to a microscope and some intelligence work (e.g. dice pictures).
Obviously for an analog IC designer visual inspection of a die is very informative. On a good day you can even find plain written information on it.
Thanks
Regards,
Lcv

User avatar
ppluis0
Diode Debunker
Information
Posts: 927
Joined: 14 Jul 2010, 18:33
Location: Argentina
Has thanked: 128 times
Been thanked: 310 times

Post by ppluis0 »

george giblet wrote: 3) Opamp supply current.
This is a weaker test and it is complicated by the fact datasheets are done at certain supply voltages,
which aren't 9V or 18V, and the current at other supply voltage may vary. Datasheets might show maximums
when your opamp is closer to typical.
But we simply can plug several types of opamps onto a breadboard, supply them with 9Vdc and measure and take note of the current consumption of each one. 8)
After that, compare with the supply current of the sanded device under test at the same voltage and look for similarities.

Cheers,
Jose

Post Reply