NOS = new old stock
TS = thread starter or topic starter, AKA (also know as) OP = original poster
And here's more!
TS = thread starter or topic starter, AKA (also know as) OP = original poster
And here's more!
Good topic. Having run into issues with measuring high (3%) 2nd harmonic distortion (on a Purifi driver for which I had 3 reference tests and should have <<1% HD at this level) when doing near field measurements with my Dayton EMM-6 i did some research and found this useful article.https://www.mtg-designs.com/tips-tricks-tests/measurement-mic-tests/hd-measurement-mic-tests
I have used both the Dayton and Behringer entry-level mics and to date they have served me well to understand basic behavior. However as I go into the final stage of prototyping for which both predictable FR response <0.5dB and distortion resolution (<0.5% eg lower than the actual distortion of the Purifi drivers) are key requirements I am about to order the Earthworks M23R.
I have used both the Dayton and Behringer entry-level mics and to date they have served me well to understand basic behavior. However as I go into the final stage of prototyping for which both predictable FR response <0.5dB and distortion resolution (<0.5% eg lower than the actual distortion of the Purifi drivers) are key requirements I am about to order the Earthworks M23R.
Larger capsules often handle distortion from nearfield testing better, it’s why I chose the OM1 (which is almost 1/2”).
http://www.lineaudio.se/OM1.html
Notably:
133db at only .5% THD
Self Noise 18db
-it’s not great, but for about $150 US shipped (when I purchased it) it’s tough to beat.
Last edited:
https://www.diyaudio.com/community/...surement-challenge.407527/page-5#post-7572239
These measurements were done with a dynamic microphone, about 2 inches from the diaphragm. None of the electret condenser microphones advertised above can do this. I repeat: none!
Total waste of money regarding loudspeaker distortion!
...and very bad advice!
Best regards
Bernd
These measurements were done with a dynamic microphone, about 2 inches from the diaphragm. None of the electret condenser microphones advertised above can do this. I repeat: none!
Total waste of money regarding loudspeaker distortion!
...and very bad advice!
Best regards
Bernd
Last edited:
please explain in more detail... It is not clear to me what point you are making...
What kind of microphone is a waste of money, and what advice is bad advice?
Thanks Scott - could you share your own distortion measurements if available - interested to learn more on the OM1
Indeed tests (like any) has its limitations and I am planning to do some of my own using the Purifi PTT6X as a reference since I can compare it to Purifi’s own as well as HifiCompass and AudioXpress measurements eg can broadly separate mic distortion from driver distortion for different levels. Once done I will share.
Member
Joined 2003
All measurements here were completed with OM1, known as "Omni1" today.
https://www.htguide.com/forum/forum...ng-discussion/950349-some-random-driver-tests
I don't think nearfield measurement is that useful for distortion, nearfield has frequency limitations based on driver diameter, which limits high frequency accuracy of fundamental and harmonics. For most drivers, 315mm distance is good balance between accurate frequency response, and reduced room interaction. Only large drivers may require further distance. 315mm also makes translation of SPL to 1m easy, just subtract 10dB.
Yes, unfortunately a distortion measurement of a driver really isn’t going to tell you anything about the mic. itself, at least not without a comparison with another mic. and even then you’ll just have a difference between the two and not an absolute value for the mic. in question.
Generally nearfield within the modal region, though what that proximity should be is trial and error testing for the particular environment.
Generally nearfield within the modal region, though what that proximity should be is trial and error testing for the particular environment.
Member
Joined 2003
It’s not perfect indeed but interesting to compare HD measurements for the same driver and voltages at both 315mm and 100m (<150Hz only) with different mic’s - and apart from this will be running some measurements at different distances incl
1m farfield with same PTT driver at different voltage levels.
Thanks for sharing the link and nice set of measurements on a wide range of tweeters and midrange-woofers. Seems indeed with the Omni1 at 315mm you are getting credible 2nd order HD results.
That is a brilliant idea !
Please don't let my comments distract you ;-)
Best regards
Bernd
Last edited:
@mainframe99
With regards to microphone for crossover design, you don't need anything fancy. People have been doing good designs with the something like the early ECM8000, and able to compensate for any mild deviation in the 2nd top octave by ear. More importantly is that you use dual channel measurements to capture phase information correctly when you take full polar measurements in at least 2 planes ie. horizontally and vertically. It doesn't matter if you have a laboratory grade scientific instrument if you are taking on-axis measurements only; it is far away from describing how a speaker sounds, one could argue that it's close to useless (ie. bad data is worse that no data)
When measuring for in-box frequency response, of course you need to be far away enough to capture the baffle diffraction loss appropriately (ie. twice the distance of the longest cabinet dimension)
eg.
reference:
https://www.ap.com/wp-content/uploads/2021/09/AppNote-Loudspeaker-EA-Measurements.pdf
But when measuring drive distortion there's no need to measure at some arbitrary distance like 1m. You'll can move your mic closer (or further) away to capture the SPL you're interested in, to improve your signal to noise ratio, or avoid overloading your mic/pre-amp signal path respectivly.
When you look at specifications of mic/mics capsules, you'll find that the sweet spot tends to be around 94dB.
So if you're you're measuring something that plays 130dB @1m, for example, you'll want to move your microphone a little bit further away, because the the microphone's own distortion may be significant enough to affect your result. So 2 or 4 meters may be helpful.
Conversely, if you're measuring at a very low level eg. 75dB, you might be swamped by the noise floor of your room, which could up upwards of 40dB on a good night. A good night for me is when the AC is off, the fan is off, and there's no extranous background noise.
If you can measure from 50cm, 40cm, or 31.6cm, this will increase your signal to noise ratio by +6dB, +8dB and +10dB respectively. Sometimes you'll need this to measure really low distortion devices.
Here's something I measured with a Sonarworks XRef20 microphone, which sells for about $159 AUD (100 USD/EU) into a Focusrite 2i2 Gen 3 audio interface which is around AUD$250.
Measured at 40cm, using a sweep length 1M sample, and 4 repetitions, with improves the signal to noise ratio.
PS. The frequency of the chirp of a common garden cricket in Australia is 7KHz (don't ask me how I know)-
With regards to microphone for crossover design, you don't need anything fancy. People have been doing good designs with the something like the early ECM8000, and able to compensate for any mild deviation in the 2nd top octave by ear. More importantly is that you use dual channel measurements to capture phase information correctly when you take full polar measurements in at least 2 planes ie. horizontally and vertically. It doesn't matter if you have a laboratory grade scientific instrument if you are taking on-axis measurements only; it is far away from describing how a speaker sounds, one could argue that it's close to useless (ie. bad data is worse that no data)
When measuring for in-box frequency response, of course you need to be far away enough to capture the baffle diffraction loss appropriately (ie. twice the distance of the longest cabinet dimension)
eg.
reference:
https://www.ap.com/wp-content/uploads/2021/09/AppNote-Loudspeaker-EA-Measurements.pdf
But when measuring drive distortion there's no need to measure at some arbitrary distance like 1m. You'll can move your mic closer (or further) away to capture the SPL you're interested in, to improve your signal to noise ratio, or avoid overloading your mic/pre-amp signal path respectivly.
When you look at specifications of mic/mics capsules, you'll find that the sweet spot tends to be around 94dB.
So if you're you're measuring something that plays 130dB @1m, for example, you'll want to move your microphone a little bit further away, because the the microphone's own distortion may be significant enough to affect your result. So 2 or 4 meters may be helpful.
Conversely, if you're measuring at a very low level eg. 75dB, you might be swamped by the noise floor of your room, which could up upwards of 40dB on a good night. A good night for me is when the AC is off, the fan is off, and there's no extranous background noise.
If you can measure from 50cm, 40cm, or 31.6cm, this will increase your signal to noise ratio by +6dB, +8dB and +10dB respectively. Sometimes you'll need this to measure really low distortion devices.
Here's something I measured with a Sonarworks XRef20 microphone, which sells for about $159 AUD (100 USD/EU) into a Focusrite 2i2 Gen 3 audio interface which is around AUD$250.
Measured at 40cm, using a sweep length 1M sample, and 4 repetitions, with improves the signal to noise ratio.
PS. The frequency of the chirp of a common garden cricket in Australia is 7KHz (don't ask me how I know)-
Last edited:
The D2 (in dB) follows the fundamental SPL in detail. So D2 is solely from the overloaded distorting mic, therefore D2 is wrong and useless.
Best regards
Bernd
(sorry for harshly and bluntly repeating myself)
I have no objections to the usual inexpensive 1/4" elektret condenser measurement microphone with a calibration file. However, it is impossible to complete all tasks with just one microphone,
if distortion is part of the task list.
Best regards
Bernd
Last edited: