Electronics > Projects, Designs, and Technical Stuff
Is it possible to destroy components with a capacitance meter?
niino:
I tried to check the output caps of a CPU VRM. A rough total value would have been good enough, so I tried to measure in-circuit. For some reason I put the probes on reverse and held them there for several seconds - I just had my brain turned off while doing this. Is it likely this destroyed the CPU and/or other components (aside of the caps themselves)? I am especially worried about the CPU.
TheMG:
Probably not. The test voltage and current of a capacitance meter is very low. In fact, many capacitance/ESR meters use an AC signal and are non-polarized.
This would be no worst than for example using the diode check feature on a DMM in reverse bias.
Very very very very unlikely to cause any kind of damage. Maybe there's some extremely sensitive components out there that might be damaged but you certainly won't find any of those on a PC motherboard.
niino:
I just had the time to measure the voltage my meter applies to the circuit when connecting it the same way as before. It was -1.1 mV DC.
I found a datasheet of a CPU from the same era made using a similar process (a 90nm Pentium 4) which had an absolute minimum rating of -0.3V, so I guess I didn't do any damage. However, no such documentation is available for the CPU in question.
pigrew:
--- Quote from: niino on July 11, 2020, 05:33:16 pm ---I tried to check the output caps of a CPU VRM. A rough total value would have been good enough, so I tried to measure in-circuit. For some reason I put the probes on reverse and held them there for several seconds - I just had my brain turned off while doing this. Is it likely this destroyed the CPU and/or other components (aside of the caps themselves)? I am especially worried about the CPU.
--- End quote ---
Yes. I've used LCR meters that can output 10 Vpp. I also have an LCR meter that can apply 1000 V for leakage testing....
However, your meter is probably somewhat low power. What model meter are you using? Also, power rails usually have a large capacitance, so they will also limit the AC bias.
The polarity matters mostly if you are using guarded measurements, or with mains-connected meters. The NEGATIVE usually is constant-voltage (sometimes a virtual ground), while the POSITIVE has large AC voltage.
Kleinstein:
Digital chips often have a limit of some -0.3 V, as this is the point where substrate diodes start to conduct. The real limiting point is often the reverse current. So as long as the current is limited to a safe limit (e.g. -1 mA) the chip may see a little more than the -0.3 V, but no more than the current limit that may be at some 10 mA. Without supply voltage even more current could be tolerated, as there would be no latch up.
The problem however could be static, especially if the chip is powered on.
Navigation
[0] Message Index
[#] Next page
Go to full version