2
$\begingroup$

I would like to test the frequency and transient/step response of the Zoom H5 recorder preamps. If possible, I'd also like to test the preamps for the built-in (exchangeable) XY capsule module, but I'm not sure yet how to go about this properly.

To do the tests, I've ordered a BNC-to-XLR adapter in order to connect a digital function generator to either one of the two XLR mic inputs. The idea is to record sine frequency sweeps, square waves, and short pulses and analyze the waveforms on the PC using e.g. Audacity or some Python scripts. Question:

  • What input signal amplitude should/could I use in order to get reliable measurement data and also not to damage the recorder, and to what impedance do I have to set the signal output of the function generator?
  • How could I do the same measurements on the built-in mic input connector on top of the recorder, i.e. which connectors would I have to connect e.g. using small clamps?

Information on the recorder:

https://zoomcorp.com/de/de/handy-recorder/handheld-recorders/h5/

The function generator:

https://www.keysight.com/us/en/product/33210A/33210a-waveform-and-function-generator.html

And the adapter cable:

https://www.perakabel.de/bnc-stecker-auf-xlr-buchse-kabel-75-ohm.html

Improve this question
$\endgroup$
1
  • $\begingroup$ Measure THD and noise at different input levels, too. $\endgroup$
    – TimWescott
    Commented Dec 9, 2021 at 15:51

2 Answers 2

Reset to default
1
$\begingroup$

A log sweep sine at levels where the built in VU meter does not indicte clipping seems like a good idea. If possible, use the highest available samplerate and a lossless audio format.

Measuring microphones is a lot harder I think. You might want to find an anechoic chamber, or approximate free space or half-free-space and use a flat response loudspeaker as input. Moving loudspeaker and microphone close to each other reduce the relative influence of the room, but could introduce issues with loudspeaker polarity pattern and tweeter/woofer placement. Room reverberation can be reduced by time gating the impulse response, but that typically leads to issues in estimating low frequency response.

What are you trying to learn from this?

Improve this answer
$\endgroup$
2
  • $\begingroup$ Hi Knut, thanks for your answer. So I can probably start by setting the preamps to max. gain, selecting the highest sample rate, then increasing the signal amplitude up to -12 dbFS, and recording to WAV. I don't however, want to measure microphones, but characterize the actual recording hardware. The motivation...I guess a mixture of curiosity and the desire to have some quantitative data for the device. The recorder is meant for a professional classical musician and has the advantage of two XLR mic inputs, so if the built-in XY capsules are not satisfactory, one can upgrade... $\endgroup$
    – Ferdie
    Commented Dec 9, 2021 at 11:20
  • $\begingroup$ ...with external mics. And having some concrete data on the preamps might aid in the selection process and maybe also in explaining why the recorded audio sounds the way it does (i.e. some better physical understanding), especially since Zoom doesn't seem to publish much detailed information on their hardware... Also, I'm curious how far into the ultrasonic the Zoom H5 can go $\endgroup$
    – Ferdie
    Commented Dec 9, 2021 at 11:24
1
$\begingroup$

I would like to test the frequency and transient/step response of the Zoom H5 recorder preamps

Why? Preamps are relatively simple contraption and chances are they are NOT the limiting factor in the quality and noise floor of the H5.

What input signal amplitude should/could I use in order to get reliable measurement data and also not to damage the recorder, and to what impedance do I have to set the signal output of the function generator?

The recorder is speced up to 22 dBu if you engage the pad. That's roughly 10V RMS. I'd go with about half of that. Maybe 8V one sided peak (RMS is less important). Impedance anywhere between 100 Ohms to 600 Ohms is fine.

You can measure with any signal that's broad band enough. It will be much easier if you can get a numerical representation of the measurement signal. This way you can get the transfer function and impulse response by spectral division.

How could I do the same measurements on the built-in mic input connector on top of the recorder, i.e. which connectors would I have to connect e.g. using small clamps?

I would tread very carefully here: this is not a standard interface and not designed to connect third party hardware too. I', guessing these are electret microphones which means that connector also provides power to the microphones. Connecting this to a signal generator poses a risk to both devices. I wouldn't do this unless there is really compelling reason and not without taking a good look at the schematic first.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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