2
\$\begingroup\$

I am looking for methods to verify whether a Piezo is damaged/defective or good. Are there ways to verify that using just electrical parameters like resistance, capacitance etc.

My issue is that it's not feasible to test dB levels using a handheld meter at the assembly line, so I was wondering if there are electrical parameters which we can test for other than sound level or current consumption (that actually need to turn on the device, specially the piezo unit).

Right now, the Piezo that I am using has a Vpp of 30 V and capacitance of 150 nF and being driven by an H-bridge.

  • A Good Piezo: Sound is sharp and unbroken
  • A Bad Piezo: Sound is distorted/broken (generally lower dB level than Good Piezo), sort of double sound, always an effect of damaged piezo while assembly
  • Damaged Piezo: No sound at all

I am more interested in differentiating Good vs Bad Piezo:

  • So, if I can measure some electrical parameters and define a ballpark for Good piezo
  • If measured values are out of that range, it is not good and can be discarded

Here is link to the product: AB4113B

Acceptable methods of testing:

  • Using Multimter/LCR
  • Oscilloscope/Function Generator
  • Any other parameter testing method that can be integrated in a test jig for production testing
\$\endgroup\$
6
  • 2
    \$\begingroup\$ Welcome! If testing sound pressure is not viable, what is viable? LCR meter? Impulse test? Oscilloscope+function generator? \$\endgroup\$
    – winny
    Commented Feb 2 at 9:43
  • 3
    \$\begingroup\$ Please edit the question with a link to the datasheet for the device in question. \$\endgroup\$
    – RussellH
    Commented Feb 2 at 9:50
  • \$\begingroup\$ A very warm welcome to the site. You'd have to edit your question (don't post info in comments) and clearly define what the threshold between 'damaged/defective' and 'good' is. Those words on their own can only describe the opinions of the inspector. You'd need to put measurable parameters on them. Thanks and, again, welcome. \$\endgroup\$
    – TonyM
    Commented Feb 2 at 11:06
  • 1
    \$\begingroup\$ Thank you for the comments and warm welcome! I can use Meters, but don't know how Oscilloscope or Function generator would be helpful here. As for the measurable parameters, that's what I am looking for. So, if the piezo is damaged what kind of measurable parameters would change? \$\endgroup\$
    – Anurag Kumar
    Commented Feb 2 at 12:09
  • \$\begingroup\$ Thanks for the edit, that's a great improvement to your question and I've upvoted it. One more thing to edit in: does a a damaged piezo sound any quieter than a good piezo? \$\endgroup\$
    – TonyM
    Commented Feb 2 at 12:31

1 Answer 1

Reset to default
1
\$\begingroup\$

I would suggest measuring the 'Q' and resonant frequency of the transducer, along with capacitance (should be within an acceptable range) and resistance (should be really high).

That should detect bonding and mounting issues as well as outright failure of the connections and that sort of thing.

I would also test 100% for SPL for a while initially and spot and lot checks later, even if it costs a bit more to move them to a place where they don't deafen the workers or supply PPE. It's very expensive to repair a reputation and it's possible some defects only show up at high SPLs.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks for your response @Spehro. Can you please explain more about 'Q', is there a general formula to calculate this, I am unable to find a clear detailed description about this. - Is it possible to measure resonant frequency without actually turning ON the piezo - I have measured capacitance and resistance of the various assembled units, but capacitance is varying over a 30nF range and resistance is varying from 3-500 MOhm for Good Piezo units \$\endgroup\$
    – Anurag Kumar
    Commented Feb 5 at 8:13
  • \$\begingroup\$ atcp-ndt.com/en/support/faq_trz/…. Here is a brief description from ultrasonic transducers. Q and resonant frequency can be measured at a lower intensity than full, with the caveat I mentioned in the answer. One way is to just send an impulse and measure the frequency and duration (number of cycles above a certain amplitude) of ringing. Of course it will make a little sound but not full volume. \$\endgroup\$
    – Spehro Pefhany
    Commented Feb 5 at 10:17

Not the answer you're looking for? Browse other questions tagged or ask your own question.