| Products > Test Equipment |
| IC testers, do they work? |
| << < (7/9) > >> |
| David Aurora:
--- Quote from: Someone on August 09, 2023, 12:19:09 am --- --- Quote from: David Aurora on August 09, 2023, 12:04:52 am --- --- Quote from: Someone on August 08, 2023, 11:55:12 am --- --- Quote from: David Aurora on August 08, 2023, 10:49:00 am --- --- Quote from: Someone on August 08, 2023, 10:06:45 am ---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. --- End quote --- If you are looking for shorts, why bother with diode test? Just use continuity mode and listen for beep. You framed the discussion around diode test.... which is not comparable to these automated IC "testers". I'll suggest they're much closer to continuity beepers in information content than a diode test mode. So when you say a "tester" which model/style are you referring to? what information does it provide? is its documentation available online? There are a range of products out there with wildly different capabilities, the device you refer to may well have much more intelligence than the example the OP is asking about. What passes as "ok" in these cheap testers doesnt always function in real circuits, thats the biggest problem as it provides a false sense of security/validation. --- End quote --- This is the weirdest hill to die on. First off, about your great advice about how I should use continuity mode for the tests described above- you realise a diode or transistor can fail open as well as short, right? A quick diode test will give you a ballpark idea of a working device and you can move on from there. The fact that you don't get this ought to automatically disqualify you from this conversation because this is real troubleshooting 101 shit. These devices are NOT analogous to continuity testers. I can put 3 different IC's in them one after another and it will tell me what they are or if they're cooked. Absolutely nobody in their right mind has any expectation that they will pick obscure/intermittent faults that might cause them not to match the data sheet specs, this is just some bizarre bee you have in your bonnet about them. Nobody is using these things to launch rockets, it's just a go/no go tool to speed up troubleshooting. You still need to use your brain and judgement beyond that point if there are doubts about a part. With regard to your argument that it's not a real world test- welcome to planet earth. Do you think a multimeter is a real world test for resistors/capacitors/diodes/etc? Of course it isn't. It doesn't need to be. It'll tell me if a part reads roughly what it should under simple, low voltage conditions and in most cases that is perfectly fine to get the job done. There are other more advanced tools available when required, and the same goes for IC testing. --- End quote --- Lol, I point out a comparative example/analogy and you nail that to the nth percent rather than stick to the thread topic? As above, if they are a go/no-go tester then where is the threshold for pass/fail? what do they actually test? You and Fungus seem to be ignoring that while claiming they are perfectly good testers. I say it isn't a tester because there is no documentation or evidence of what has been tested. I put all that context in my opening post yet you come back and try to argue... something? --- 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 --- Everyone else in the thread is more realistic and agrees they are some basic tool which is of little confidence. Many posters offered slightly more expensive tools that offer documentation and defined test capability along with increased functionality. So how about instead of just making noise you actually lay out your explanation of what they test, and why that is useful. Saying they tell you if the item is "working" is some non specific threshold, untrustworthy, particularly these extreme low cost items where the underlying implementation is subject to change (as it was never specified or documented) as long as it still meets the advertised function of telling you which logical pinout the chip is (identification). --- End quote --- Mate... are you serious? There are people here going "I use this effectively for troubleshooting" and your argument is what exactly? That even though it does what we need, we shouldn't use them because you have a weird phobia? I go back to what I just said about real world testing and thresholds- almost none of the tools we use daily test components under real world conditions. No multimeter out there is going to red flag a noisy plate resistor or excessive leakage on a capacitor, and by the same token these IC testers aren't expected to tell you if the high/low threshold is wonky by a few millivolts in a part. What is it about this you don't understand? |
| Someone:
--- Quote from: David Aurora on August 09, 2023, 12:31:47 am --- --- Quote from: Someone on August 09, 2023, 12:19:09 am ---I put all that context in my opening post yet you come back and try to argue... something? --- 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 --- Everyone else in the thread is more realistic and agrees they are some basic tool which is of little confidence. Many posters offered slightly more expensive tools that offer documentation and defined test capability along with increased functionality. So how about instead of just making noise you actually lay out your explanation of what they test, and why that is useful. Saying they tell you if the item is "working" is some non specific threshold, untrustworthy, particularly these extreme low cost items where the underlying implementation is subject to change (as it was never specified or documented) as long as it still meets the advertised function of telling you which logical pinout the chip is (identification). --- End quote --- Mate... are you serious? There are people here going "I use this effectively for troubleshooting" and your argument is what exactly? That even though it does what we need, we shouldn't use them because you have a weird phobia? I go back to what I just said about real world testing and thresholds- almost none of the tools we use daily test components under real world conditions. No multimeter out there is going to red flag a noisy plate resistor or excessive leakage on a capacitor, and by the same token these IC testers aren't expected to tell you if the high/low threshold is wonky by a few millivolts in a part. What is it about this you don't understand? --- End quote --- You're the one who keeps quoting my comments with some sort of bizarre disagreement to it. I'm pointing out your disagreements have no substance. So what exactly are these devices testing? I put it really simply above (quoted it again, and included it in the context with this post just so there can be no misunderstanding) and now you appear to agree with that opening statement, but are still arguing what exactly? |
| alm:
--- Quote from: David Aurora on August 09, 2023, 12:19:47 am ---Not true, they definitely can test *some* op amps. On mine it's only a handful, but I keep wondering if newer firmware has added any more. --- End quote --- What is not true? That I don't think they have any value for testing analog ICs? Can you read my mind? I didn't say they couldn't do any test on op amps, just that I don't think it's a very useful test. Never mind other analog ICs like motor drivers or tape head amplifiers. What do you mean it supports a handful of op amps? I can imagine doing a basic dc test for single and dual op amps, the same you can do with a breadboard, a few resistors and an LED. But why would it need to support more than a generic single, dual and quad op amp? Is it going to test parameters that are different between a ua741, op07 or tl081? Slew rate? GBW? Offset voltage? Bias current? That would be useful for checking for counterfeit ICs, especially if they could give quantitative results. But I highly doubt the cheap IC testers are going to do those tests. |
| tautech:
--- Quote from: David Aurora on August 09, 2023, 12:31:47 am ---No multimeter out there is going to red flag a noisy plate resistor or excessive leakage on a capacitor, and by the same token these IC testers aren't expected to tell you if the high/low threshold is wonky by a few millivolts in a part. --- End quote --- Oh a MC meter can in resistance mode. Used that many years back to find faulty ignition caps. Electronic ignitions saw the end to that need. |
| David Aurora:
--- Quote from: Someone on August 09, 2023, 12:55:32 am --- --- Quote from: David Aurora on August 09, 2023, 12:31:47 am --- --- Quote from: Someone on August 09, 2023, 12:19:09 am ---I put all that context in my opening post yet you come back and try to argue... something? --- 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 --- Everyone else in the thread is more realistic and agrees they are some basic tool which is of little confidence. Many posters offered slightly more expensive tools that offer documentation and defined test capability along with increased functionality. So how about instead of just making noise you actually lay out your explanation of what they test, and why that is useful. Saying they tell you if the item is "working" is some non specific threshold, untrustworthy, particularly these extreme low cost items where the underlying implementation is subject to change (as it was never specified or documented) as long as it still meets the advertised function of telling you which logical pinout the chip is (identification). --- End quote --- Mate... are you serious? There are people here going "I use this effectively for troubleshooting" and your argument is what exactly? That even though it does what we need, we shouldn't use them because you have a weird phobia? I go back to what I just said about real world testing and thresholds- almost none of the tools we use daily test components under real world conditions. No multimeter out there is going to red flag a noisy plate resistor or excessive leakage on a capacitor, and by the same token these IC testers aren't expected to tell you if the high/low threshold is wonky by a few millivolts in a part. What is it about this you don't understand? --- End quote --- You're the one who keeps quoting my comments with some sort of bizarre disagreement to it. I'm pointing out your disagreements have no substance. So what exactly are these devices testing? I put it really simply above (quoted it again, and included it in the context with this post just so there can be no misunderstanding) and now you appear to agree with that opening statement, but are still arguing what exactly? --- End quote --- Sorry mate, again- what is your point with all of this? I'm telling you I have one and it is helpful at work. Are you arguing that it isn't helpful? Are you only really arguing about the semantics of what we call these things? Or just that the results aren't as reliable as tests done on the test setups manufacturers use to get their specs? You keep ranting on about the finer details of the tests despite the fact that literally nobody cares about the finer details of the tests, only that we get a ballpark idea of whether or not a part works or not. Again, nobody is using these to test ICs going into space. Let's try another example and see if this one clicks. Those cheap 3 terminal multifunction testers that do LCR/transistors/etc. I use one all the time for checking go/no go status, checking pinouts, etc. It doesn't test under real world conditions, I don't know or care what exactly the test method is, and I don't trust or care about the particular measurement numbers it spits out. If those details matter I have dedicated, proper equipment for detailed measurements, but it's very rarely needed in practice. In most cases the 5 second test on the $10 tester tells me what I need to know in the moment and I can get on with the job. |
| Navigation |
| Message Index |
| Next page |
| Previous page |