IC Testing General Best-Practices

[jmcmahon]

I'm working on a few guitar pedal circuits that include discontinued components (BBDs, corresponding clock chips, some companders, etc.). I'm happy to second-guess myself in terms of how I have things set up, but I'm increasingly confident some of the new-old-stock parts I got from eBay or Amazon merchants are defective. So I have two questions:

1. Are there general starting points for testing IC's, or will pretty much everything be entirely case by case? If the former, where do y'all usually begin?

2. Are IC failure modes generally obvious, or could all the testable voltages, resistances, etc. look good and the chip still be junk?

[bob91343]

See data sheets for test circuits.  Before you do that, measure each pin to see if it appears to be normal.

There are no rules.  Too many possibilities make it impractical to generalize.

[Berni]

To properly test a chip it has to be put into a test circuit specific for that chip.

However a lot of the time the chips are blown badly enough for them to show other problems. Easiest way is to get a multimeter in diode test mode and probe across the power pins. In the correct polarity (red on Vcc) the voltage drop should be higher than a diode(or out of range) while swapping the probes around should show a voltage drop similar to a diode. This is the internal ESD protection diode. Also blown chips often like to draw too much power and get hot when powered in circuit, so if a chip is getting unusually warm compared to how warm you think it should get according to the current consumption in the datasheet that could also be a hint.

Similar trick works for testing IO pins on chips. Put the red probe on GND and poke pins to see that they show a normal diode drop. But this is mostly only reliable on digital chips, some analog chips(opamps and similar) might not have ESD diodes for performance reasons. But if you see a dead short on any of the pins that could hint towards something being wrong with it.

[Vovk_Z]

1. Are there general starting points for testing IC's, or will pretty much everything be entirely case by case? If the former, where do y'all usually begin?
2. Are IC failure modes generally obvious, or could all the testable voltages, resistances, etc. look good and the chip still be junk?

The most reliable way to make sure IC are ok - not to bay them on Ebay, etc.
The most obvious failure - when there is some large DC voltage at IC output (when it hasn't be there).
I test IC in real circuit - if it (device) works as it was intended to (with right frequency range, without excessive noise etc) - then it is ok

[jmcmahon]

Thanks for the confirmations and tips, all. Most of the chips in question do have reputable (or at least less shifty?) sources but between what I assume is a combo of tariffs, supply chain disruptions, and DIY builders stuck at home looking for things to do, many of them have been out of stock for a while now. I'll just hold off on buying any further such chips, I don't need even more variables in an already complex equation.

For the time being, I'll consider them troubleshooting exercises that might prove functional!