So maybe you could elaborate the "older" circuits have not been driven away. Some specific products with pros and cons maybe. DPA and Schoeps seem to be highly regarded is there another story? I can't find any EW reviews that don't jive with my original opinions, accurate vs colored, highish noise floor, nothing about interface issues.
Last edited:
Well if you made any sense I could reply in a meaningful way. The sound qualities of a microphone have little to do with balanced versus unbalanced. In a studio not truly balanced is not an issue. Outside a studio where cable runs are longer and there are many more sources and higher levels of intereference balance and lack of any forms of rectification are far more important.
As usual you start with an insult even though you know very well we are not talking about 500 foot runs in an outdoor environment. I've told you several times I respect the issues of your business but they don't apply to most of what we are doing.
This might not be the ideal way to do it but I believe I can apply the correction with inverse FFT in Octave. I would prefer the .wav file from the frequency sweep along with the THD .wav file.
If you prefer this, the process is:
1: Record a frequency sweep with mic on channel 1, speaker current on channel 2.
2: Record THD without moving the mic. Both recordings should have close to the same length.
Easier for you, now it's up to me to get the processing right.
Last edited:
Sorry I didn't get what you needed earlier. I thought you wanted the woofer distortion measurement at a known voltage. The problem is that the W18 are already mounted in cabinets with crossovers, and current drive and current measurements would need to disconnect the drivers from the crossover, which, frankly, I don't want to do. All this work is time consuming, I did what I was willing to do and what time to spend. Sorry for misunderstanding.
Scott,
We really don't communicate. That is not an insult. It is commentary. I had to read your bit several times to get even a bit of feel for what was confusing the issues.
One issue was the different voicing of microphones. That is something I did not touch on as there are clear differences between microphones and the application of the differences is based on personal choice primarily.
Another is the type of condenser microphone, polarized, electret or RF. A clear difference.
Then there was the issue of transformer coupling. I addressed the later issues. You confused me by seeming to equate the accoustic issue with the physical construction ones.
So yes I often have few clues as to what you really mean.
Which has nothing at all to do with anything we are talking about. The issues here are pure instrumentation issues technical quantifyable ones. How would the standard Schoeps transformerless mic circuit prevent jn from making his measurements, what is the THD of a particular mic in this application, etc.
No difference at all here, both will do, subjective results recording music is OT.
Last edited:
Funny I thought the topic had wandered to insulting Tourny.
As to what would prevent JN or others from using a studio microphone from measuring intermodulation distortion? That would require selection of a microphone that is appropriate... No kidding.
Now what reviews of the various microphones has to do with that becomes the reference to authority argument. What is required is the reference to measured data and results.
I mentioned reciprocity. That calibrated not just frequency response but also distortion. Although it does require a bit of precision and number crunching to extract the results. A simple reciprocity set up would have at least one modified test transducer as the source, two others unmodified interchanged as source and receiver. Thus you could compare and contrast the results between modified and unmodified. The difference is of course the data you are looking for. The limits would show up when reversing the unmodified as source and receiver.
It is much easier and more clearly understood to have a perfect reference type of test. Even though it is clear there is no such thing as a perfect reference. To approach that one uses precision parts.
Now in the case of an electret microphone one really big measurement issue is temperature sensitivity. That is why such microphones when used professionaly for measurement are calibrated before each use. Usually not a recording issue and with a modicum of care should not be in this instance.
The next issue would be selection of capsule type and size. Type would almost certainly lead one to an omni directional capsule. Size becomes an interesting issue. A one inch capsule begins to show wavelength induced frequency rise as low as 5,000 to 6,000 hertz. However it will have lower noise and distortion than a smaller capsule.
I could go on about more of the issues in measurement microphones but as a practicing engineer the simplest solution is preferred. That would be to try the different microphones on hand and compare the results by measurement. A crude method compared to what can be done. But that is clearly what is happening.
So if I were going to do these measurements without a proper microphone I would either buy one on eBay, use it and resell it or buy a one inch electret capsule. For a preamp I would build a simple circuit or use another electret condenser microphone as an electronics donor.
As to what would prevent JN or others from using a studio microphone from measuring intermodulation distortion? That would require selection of a microphone that is appropriate... No kidding.
Now what reviews of the various microphones has to do with that becomes the reference to authority argument. What is required is the reference to measured data and results.
I mentioned reciprocity. That calibrated not just frequency response but also distortion. Although it does require a bit of precision and number crunching to extract the results. A simple reciprocity set up would have at least one modified test transducer as the source, two others unmodified interchanged as source and receiver. Thus you could compare and contrast the results between modified and unmodified. The difference is of course the data you are looking for. The limits would show up when reversing the unmodified as source and receiver.
It is much easier and more clearly understood to have a perfect reference type of test. Even though it is clear there is no such thing as a perfect reference. To approach that one uses precision parts.
Now in the case of an electret microphone one really big measurement issue is temperature sensitivity. That is why such microphones when used professionaly for measurement are calibrated before each use. Usually not a recording issue and with a modicum of care should not be in this instance.
The next issue would be selection of capsule type and size. Type would almost certainly lead one to an omni directional capsule. Size becomes an interesting issue. A one inch capsule begins to show wavelength induced frequency rise as low as 5,000 to 6,000 hertz. However it will have lower noise and distortion than a smaller capsule.
I could go on about more of the issues in measurement microphones but as a practicing engineer the simplest solution is preferred. That would be to try the different microphones on hand and compare the results by measurement. A crude method compared to what can be done. But that is clearly what is happening.
So if I were going to do these measurements without a proper microphone I would either buy one on eBay, use it and resell it or buy a one inch electret capsule. For a preamp I would build a simple circuit or use another electret condenser microphone as an electronics donor.
Last edited:
How would you extract distortion with reciprocity? The three cornered hat aspect of reciprocity will help in amplitude sensitivity but I don't see how you can derive some reference-less distortion number. I would have thought it would be a standard measurement if that's possible. Actually measuring sensitivity with sine waves needs a low distortion source or it won't be accurate as the harmonics affect the level.
Mems mikes have very stable and uniform sensitivity and good extended response. Weak on noise and dynamic range. And there a a whole bunch of issues about bottom port vs. top port (depending on mfr) and analog vs. digital. Still for voice microphones on phones they are a great choice. And small enough to use several for sophisticated noise cancelling.
Hmm I have a pair of these EMM6 mic's. About 0.5 inch, though.
Very flat FR
But would like to know THD at 80-90 dB SPL. Spec says 1% THD at 127dB.
So, what is the process?
View attachment EMM 6 microphone FR.pdf
Microphone distortion measurements - Gearslutz
THx-RNMarsh
Very flat FR
But would like to know THD at 80-90 dB SPL. Spec says 1% THD at 127dB.
So, what is the process?
View attachment EMM 6 microphone FR.pdf
Microphone distortion measurements - Gearslutz
THx-RNMarsh
Last edited:
I would never raise a red flag for a 0.3 ΔdB but that just me.
You too Dan will come to the point of deciding how to level match and how good you’ ll make it (when you finally proceed to the multi postponed tests) .
1948 ITU Conference, Copenhagen
http://search.itu.int/history/HistoryDigitalCollectionDocLibrary/4.65.43.en.100.pdf
1952 ITU Conference, Stockholm (again Germany was not a participating member)
http://search.itu.int/history/HistoryDigitalCollectionDocLibrary/4.81.43.en.100.pdf
UKW stands for: Ultra kurz Wellen (Very short Waves) 87.5 to 100MHz at the time
George
Hi George.
Pavel posed the test, I responded that I noticed difference and asked if the cause was distortion introduced by the test speaker.
Pavel responded that the louder recording was the original.
TBH the difference was fine, very fine but I did pick out a difference which turned out to be due to level mismatch....if the levels were perfectly matched I might not have picked a THD difference, certainly not so easily or definitively....bass harmonic distortion is not as readily detected as mids/highs harmonic distortions.
My testings will not require any level compensations so that error mode will be eliminated.
Dan.
Last edited:
Dan, I was requesting ABX result, which I have not got yet. Everyone can see in Audacity, Foobar, whatever, that one file is 0.3dB louder. So I do not get the subjective impression seriously, until the ABX protocol is posted (please kindly do not explain again and again why you do not use foobar ABX). Interestingly enough, "louder" was attributed to "dirtier", in other words distorted. Almost textbook reasoning, right? However, wrong, unfortunately, this time.
You can't do this without accounting for the phase and certainly not with summing all the harmonics into one THD number. This is an interesting side discussion but PMA has already shown a mic that has no measurable distortion at the levels needed. 1" clones of the classic capsules are available from Taiwan for $30 or so I'm sure they would do fine.
Last edited:
Please, could-you explain how is-it possible to have any acoustical source able to produce one or two acoustic sinusoidal signals (in the same time) with no distortion, so we could measure distortions of microphones and believe in the numbers ?
(Adding that those numbers do not help us a lot to know how our ears and brain will process the music recorded by this mic in a more or less pseudo realistic way).
Last edited:
Yes I forgot to mention 1st harmonic level only was being given in the example. My 10 resistor test measures distortion and also is based on reciprocity. Others use a bridge with reference resistors. That has the limit of showing when the resistors under test match the characteristics of the reference ones. As the reference ones always have some error you just show compliance with the reference not an absolute value.
Same issue with using a microphone capsule. You can get very good ones that are better than the louspeaker under test, but at some point the calibration of absolute accuracy comes into question.
Same issue with using a microphone capsule. You can get very good ones that are better than the louspeaker under test, but at some point the calibration of absolute accuracy comes into question.
Last edited:
Not so hard! You use two mechanical resonators at different frequencies. For example if you have an aluminum rod of say 3 cm diameter and 1 m long you can tap it on the end and get a reasonably pure sine wave. Another 10% longer will yield a lower frequency. Now there will always be some harmonics and frequency drift but the difference tone will stand out if the measurement system is at fault. The rods by themselves will approach zero difference frequencies. Of course there will be other items in the acoustic environment that will also get modulated and provide difference tones, harmonics and even rattles totally off test frequencies.
Last edited:
- Status
- Not open for further replies.