If you did, we wouldn't be discussing this matter.

A programmer-based, or parallel-port-based programmer is nowhere near fast enough to determine that a device is functioning nominally within spec's. All it can do is to verify that it appears to satisfy a truth table programmed into the support software. That truth table can be obtained from any family-logic databook.

A 74C273 will look just like a 74AC273, from the standpoint of the device-under-examination-socket. If you had a PC that literally ran at a high enough rate that it could change each and every device pin concurrently at a 250 ps step, you could, perhaps, detect timing variations that exist between technologies. If your circuit or programmer had the ability to distinguish between totem-pole outputs and tristate outputs, which most do not, then you could possibly go a bit deeper into determining whether the device meets its spec's. All those things can be done, but no programmer I know of can do them, and there are more ...

Now, I disagree with Erik's position, and it's not in any way certain that he's wrong about what he says, that external damage is likely to cause propagation delay changes. I've actually got the ability to stimulate a circuit at a high rate and quickly sample its response to such stimuli, yet I've never seen a part that satisfied its truth table that didn't work as well as any other in its socket in the circuit-under-observation.

Just the same, if you want to build or buy incoming-inspection equipment, there is stuff out there that you can obtain. Automatic test equipment is VERY costly. I've spend some time, over the years, examining the synchronous outputs from various counters. If there were any place in which a variation in propagation delay due to some form of external damage were to be critical, I'd guess that's where it would be. I've used the 200 MSamples/second pattern generator and 500 MHz oscilloscope channels on my logic analyzer to look at such things often enough. I've not even seen variations in rise and fall times with that setup unless loading was changed.

However, Erik's right in that unless you absolutely can't obtain purportedly "good" IC's from a supplier, you should use "new" parts rather than sifting through endless supplies of discarded parts. I've used discarded parts that were on rejected or junked PCB's at times when certain part numbers were in short supply due to long lead-times, sometimes nearly a year. I've never been disappointed with parts that satisfied their databook-based truth table. That doesn't mean that it can't happen.

RE