Measuring harmonic distortion in mics?

[ChrisSead]

Hey everybody,

Currently writing my uni dissertation about effects of choice of microphone on vocals in a final pop mix (some very interesting stuff, for those who care - TLM 103 beat a U87 Ai in a blind test for this particular song).

Was looking to gather some more data to explore the result, and of course something often stated around here is that mics like the U87, with an output transformer, will have some kind of 'mojo' thanks to harmonic distortion that isn't present in transformerless mics. Anybody know a reliable way of measuring this? Just needs to be relative between the mics, as in I need to be able to definitely say the U87 is more harmonically rich than the 103 (if that's even true).

Or failing that, anybody know of any reliable sources that look into this? Measured THD published from Neumann as far as I'm aware is talking about distortion when the microphone is being overloaded, so it doesn't seem helpful to me when talking about recording vocals -- unless there's something I'm not understanding there. Also tested the C414 XL II and Rode NT1-A so any data on those would also be a great help.

Thanks!

 

[RuudNL]

An acoustical measurement will probably produce unreliable results, because you will be measuring the distortion of the speaker (in fact the whole reproduction chain) AND the distortion produced by the microphone itself.
IMHO it would be more reliable to inject a signal directly in the circuit just after the microphone capsule.
Although the distortion of the microphone capsule is not included in this case, I suppose this will produce a more reliable result. It is is important to keep the output level of the microphone within a realistic range, anyway below the specified maximum output.

 

[kingkorg]

You can use REW software, this way you will get detailed per harmonic and thd profile. I would add not all transformers add distortion, some might perform just as transformerless.

Thd change is also very dynamic and non linear so you would want to perform the test at different levels.

 

[ChrisSead]

An acoustical measurement will probably produce unreliable results, because you will be measuring the distortion of the speaker (in fact the whole reproduction chain) AND the distortion produced by the microphone itself.
IMHO it would be more reliable to inject a signal directly in the circuit just after the microphone capsule.
Although the distortion of the microphone capsule is not included in this case, I suppose this will produce a more reliable result. It is is important to keep the output level of the microphone within a realistic range, anyway below the specified maximum output.

I see, I thought that might be the case. Do you think a preamp, speaker etc. would produce such a significant amount of distortion compared to the microphone that it would be impossible to compare different microphones? Because my first thought was as long as every microphone was testing with the same system, you'd at least be able to relatively compare them -- but would the preamp/speaker distortion kind of just overpower that?

You can use REW software, this way you will get detailed per harmonic and thd profile. I would add not all transformers add distortion, some might perform just as transformerless.

Thd change is also very dynamic and non linear so you would want to perform the test at different levels.

Very good point about testing at different levels, thanks for that

 

[RuudNL]

Don't underestimate the distortion of a loudspeaker!
(Put a squarewave in and you will get something close to a sinewawe out...)

 

[MicUlli]

Probably some german PhD thesis may help you:

https://www.google.com/url?sa=t&rct...krofonen.pdf&usg=AOvVaw0j9BnM4CXVzs2kwDO9VOwm
https://www.google.com/url?sa=t&rct...ument_10.pdf&usg=AOvVaw2v_3dt1B9yP0xUoApr-mod