| Products > Test Equipment |
| IC testers, do they work? |
| << < (3/9) > >> |
| TheDefpom:
I have a few different IC testers including the ones mentioned here, they seem to work fine, obviously they all cover different IC families or only partial families, but they do detect faulty devices a lot of the time, if you are ever not sure about a device, you can test a suspect device and then a known good device and check for the same result. The RCT version is extremely comprehensive and has new chips added frequently, the only catch is you have t build it yourself and if you want to update it you need to use a programmer. |
| Someone:
--- Quote from: Fungus on August 08, 2023, 06:34:30 am --- --- Quote from: Someone on August 08, 2023, 12:33:30 am ---Probably more correctly described as a IC identifier rather than a tester, there are no pass/fail criteria or performance measurement as would be associated with testing. --- End quote --- I imagine a chip has to be working for it to be identified correctly. --- End quote --- "working" enough to match logic patterns does not say it is working to datasheet specifications. Limited leakage/pull-up/pull-down at logic pins is often required for correct circuit operation as one example of a parameter which would need to be tested to confirm a chip will work according to its specifications. An ok/identifcation from these widgets is only a cursory check of: pattern at io pins matches pattern in library. Dictionary definition of test vs identification matches what I wrote. Or do you want to come up with your own new definitions of "test" and "identification". |
| David Aurora:
--- Quote from: Someone on August 08, 2023, 07:22:50 am --- --- Quote from: Fungus on August 08, 2023, 06:34:30 am --- --- Quote from: Someone on August 08, 2023, 12:33:30 am ---Probably more correctly described as a IC identifier rather than a tester, there are no pass/fail criteria or performance measurement as would be associated with testing. --- End quote --- I imagine a chip has to be working for it to be identified correctly. --- End quote --- "working" enough to match logic patterns does not say it is working to datasheet specifications. Limited leakage/pull-up/pull-down at logic pins is often required for correct circuit operation as one example of a parameter which would need to be tested to confirm a chip will work according to its specifications. An ok/identifcation from these widgets is only a cursory check of: pattern at io pins matches pattern in library. Dictionary definition of test vs identification matches what I wrote. Or do you want to come up with your own new definitions of "test" and "identification". --- End quote --- Nobody is actually using these for verifying specs. They are go/no testers. Almost like a diode test function on a meter- nobody is sweeping through semiconductors on a board with a diode test expecting to actually characterise things, it's a ballpark look for major issues. Same with these- if a supported chip fails, you know it's fucked and you bin it. If it passes then chances are high that it's not the droid you're looking for and you keep hunting. |
| Someone:
--- Quote from: David Aurora on August 08, 2023, 08:06:41 am --- --- Quote from: Someone on August 08, 2023, 07:22:50 am --- --- Quote from: Fungus on August 08, 2023, 06:34:30 am --- --- Quote from: Someone on August 08, 2023, 12:33:30 am ---Probably more correctly described as a IC identifier rather than a tester, there are no pass/fail criteria or performance measurement as would be associated with testing. --- End quote --- I imagine a chip has to be working for it to be identified correctly. --- End quote --- "working" enough to match logic patterns does not say it is working to datasheet specifications. Limited leakage/pull-up/pull-down at logic pins is often required for correct circuit operation as one example of a parameter which would need to be tested to confirm a chip will work according to its specifications. An ok/identifcation from these widgets is only a cursory check of: pattern at io pins matches pattern in library. Dictionary definition of test vs identification matches what I wrote. Or do you want to come up with your own new definitions of "test" and "identification". --- End quote --- Nobody is actually using these for verifying specs. They are go/no testers. Almost like a diode test function on a meter- nobody is sweeping through semiconductors on a board with a diode test expecting to actually characterise things, it's a ballpark look for major issues. Same with these- if a supported chip fails, you know it's fucked and you bin it. If it passes then chances are high that it's not the droid you're looking for and you keep hunting. --- End quote --- Except it's not even a go/no-go as they dont specify what has been tested. A diode test on a multimeter generates some measurements which can be directly applied to the end use (screening for matched forward voltage or checking for excessive reverse leakage are both practical on a multimeter) but a widget that says "diode*" * probably. Is not a test. Look at the good link above --- Quote from: ledtester on August 08, 2023, 03:08:15 am ---For a tester that makes analog measurements on the outputs, have a look at the Tauntek IC tester: http://tauntek.com/LogICTester-low-cost-logic-chip-tester.htm It's just a pcb + pre-programmed microcontrollers. You'll have to supply some commonly available parts to build it. List of supported devices: http://tauntek.com/ctchiplist.pdf A blog post about it: https://www.barbouri.com/2021/09/13/building-the-tauntek-logic-ic-tester/ --- End quote --- A basic tester that has some limits etc, very different from a device which tells you what it thinks the chip is (identifies). |
| David Aurora:
--- Quote from: Someone on August 08, 2023, 10:06:45 am --- --- Quote from: David Aurora on August 08, 2023, 08:06:41 am --- --- Quote from: Someone on August 08, 2023, 07:22:50 am --- --- Quote from: Fungus on August 08, 2023, 06:34:30 am --- --- Quote from: Someone on August 08, 2023, 12:33:30 am ---Probably more correctly described as a IC identifier rather than a tester, there are no pass/fail criteria or performance measurement as would be associated with testing. --- End quote --- I imagine a chip has to be working for it to be identified correctly. --- End quote --- "working" enough to match logic patterns does not say it is working to datasheet specifications. Limited leakage/pull-up/pull-down at logic pins is often required for correct circuit operation as one example of a parameter which would need to be tested to confirm a chip will work according to its specifications. An ok/identifcation from these widgets is only a cursory check of: pattern at io pins matches pattern in library. Dictionary definition of test vs identification matches what I wrote. Or do you want to come up with your own new definitions of "test" and "identification". --- End quote --- Nobody is actually using these for verifying specs. They are go/no testers. Almost like a diode test function on a meter- nobody is sweeping through semiconductors on a board with a diode test expecting to actually characterise things, it's a ballpark look for major issues. Same with these- if a supported chip fails, you know it's fucked and you bin it. If it passes then chances are high that it's not the droid you're looking for and you keep hunting. --- End quote --- Except it's not even a go/no-go as they dont specify what has been tested. A diode test on a multimeter generates some measurements which can be directly applied to the end use (screening for matched forward voltage or checking for excessive reverse leakage are both practical on a multimeter) but a widget that says "diode*" * probably. Is not a test. Look at the good link above --- Quote from: ledtester on August 08, 2023, 03:08:15 am ---For a tester that makes analog measurements on the outputs, have a look at the Tauntek IC tester: http://tauntek.com/LogICTester-low-cost-logic-chip-tester.htm It's just a pcb + pre-programmed microcontrollers. You'll have to supply some commonly available parts to build it. List of supported devices: http://tauntek.com/ctchiplist.pdf A blog post about it: https://www.barbouri.com/2021/09/13/building-the-tauntek-logic-ic-tester/ --- End quote --- A basic tester that has some limits etc, very different from a device which tells you what it thinks the chip is (identifies). --- End quote --- Again, you're looking at this from an overly ambitious perspective. My analogy with the diode test was precisely that when you're quickly sweeping through with one as a go/no go test you AREN'T using it to match forward voltages and so on. It's just "Is it blatantly fucked? Yes? In the bin it goes". Same with these cheap IC testers. As a quick example of both the analogy and the actual use case- on one job today a power supply board was shutting down. I had seen the exact same fault on another unit yesterday, so the first thing I did was probe the rectifier diodes I had seen fail before and found a shorted one as expected. I don't remotely care about the precise forward voltage when using the meter for this, all I want to know is if the diodes are basically OK or not. In the case of the IC tester, earlier tonight I had a channel strip from a mixing console up on the bench. On these particular strips I've seen one particular IC kill audio on a few of them, so if one goes down now I'll routinely pull that IC and throw it in the tester. If it fails I can replace it on the spot while I'm at the studio and the channel will likely spring back to life. No need to wire up the channel with a bench supply, set up a scope, etc., if it's as simple as a bad IC I know and it's sorted in a minute. |
| Navigation |
| Message Index |
| Next page |
| Previous page |