Topic: Recommended microphone for convolution measurements (Read 7702 times)

Matt · « **on:** December 20, 2012, 08:31:28 am »

I've got a little extra time next week, so thought it would be fun to finally get convolution setup on my home system.

I need a microphone and calibration file. I'd like to find something on Amazon so I can be sure to have it next week.

I have a Radio Shack decibel meter and also a Audio-Technica ATR-6550. I'm not sure if either of those help.

I thought about getting this Behringer ECM8000.

I'm not sure what to do about the correction file. Are there stock ones for common microphones that will get me pretty close?

Thanks for any tips.

BrianG · « **Reply #1 on:** December 20, 2012, 09:11:39 am »

Matt,

Have a look at Dayton Audio EMM-6 which is very similar to the Behringer but has individual calibration files available from their web site.

Usually available from Amazon.

Regards, Brian.

nwboater · « **Reply #2 on:** December 20, 2012, 09:14:36 am »

http://www.parts-express.com/pe/showdetl.cfm?partnumber=390-801

Dayton mic from Parts Express. After receiving the Mic you download the calibration file based on the serial #.

Rod

nwboater · « **Reply #3 on:** December 20, 2012, 09:27:59 am »

http://www.amazon.com/Dayton-Audio-Electret-Measurement-Microphone/dp/B002KI8X40/ref=sr_1_1?ie=UTF8&qid=1356017088&sr=8-1&keywords=emm-6+microphone

The same mic with downloadable cal file. "Sold by Parts Express and fulfilled by Amazon".

A lot of people using REW use this mic.

Rod

Matt · « **Reply #4 on:** December 20, 2012, 09:30:13 am »

Thanks for the recommendation guys.

I ordered one of these, so I'll be able to play next week

NoDigital · « **Reply #5 on:** December 21, 2012, 06:17:34 am »

For the record, here's one guy that recalibrates EMM-6 and calibrates ECM8000's.

http://www.cross-spectrum.com/measurement/calibrated_dayton.html

EMM-6's aren't always as calibrated as you might think and ECM8000's can have wild fluctuations (Upto 7db) beyond what the 'on the box' flat graph indicates. Another drawback with both these is the need for a 48V phantom supply.

Come Jan he'll be offering calibrated UMM-6 mic's which are USB based and therefore no need for any additional hardware variability from preamp, cabling, etc. I'm awaiting the UMM-6 availability so Santa's not delivering for me this year...

mojave · « **Reply #6 on:** December 21, 2012, 10:29:53 am »

Do you have phantom power available? If not, the easiest to use is probably the Icicle XLR to USB mic preamp. However, Bernt with Audiolense says you get the best results when using a single ASIO audio device for both the mic and output.

What software are you going to use for convolution? Did you trade licenses with Bernt?

I think it would be cool to have an advanced version of JRiver someday with Audiolense "built-in."

Matt · « **Reply #7 on:** December 21, 2012, 10:52:55 am »

Quote from: mojave on December 21, 2012, 10:29:53 am

Do you have phantom power available? If not, the easiest to use is probably the Icicle XLR to USB mic preamp. However, Bernt with Audiolense says you get the best results when using a single ASIO audio device for both the mic and output.

I don't have a phantom power source.

I was going to use an M-Audio 2496 card since it's in the computer anyway for doing other stuff. It only has RCA.

Is there a simple amplifier I could add?

I should add that I was going to order a kit from Bernt to avoid this complication, but I didn't think far enough ahead (those kits ship from Europe) and wanted to have it next week.

mojave · « **Reply #8 on:** December 21, 2012, 11:23:30 am »

I sent you an e-mail with a free solution.

nwboater · « **Reply #9 on:** January 01, 2013, 12:23:19 pm »

Quote from: Matt on December 20, 2012, 08:31:28 am

I've got a little extra time next week, so thought it would be fun to finally get convolution setup on my home system.

So Matt, did you get anywhere with this?

Rod

Matt · « **Reply #10 on:** January 02, 2013, 01:24:22 pm »

Quote from: nwboater on January 01, 2013, 12:23:19 pm

So Matt, did you get anywhere with this?

I've spent my time wrestling with a Steinberg (which I'll use for convolution measurement).

Slow and steady wins the race

Unless there happens to be something that's fast and steady in the same race.

mojave · « **Reply #11 on:** January 14, 2013, 03:47:03 pm »

There is currently a discussion on the Audiolense forum about how to best handle the LFE and rerouting. Since JRiver already went down this road with Room Correction, perhaps you could see how it compares in Audiolense and make some suggestions to Bernt.

mykillk · « **Reply #12 on:** January 14, 2013, 07:43:36 pm »

I just pre-ordered one of those UMM-6s. I can't believe I lucked out like this on the timing since the mics will be shipped sometime soon. Now I don't have to buy a USB based XLR adapter with phantom power

Any idea how one of these custom calibrated UMM-6s compare to the standard Audyssey mic?

Also, are there any free or cheaper alternatives for the room correction software? Audiolense is like $600...

mykillk · « **Reply #13 on:** January 14, 2013, 08:10:24 pm »

Quote from: mojave on December 21, 2012, 10:29:53 am

However, Bernt with Audiolense says you get the best results when using a single ASIO audio device for both the mic and output.

Doesn't ASIO4ALL allow you to choose different inputs and outputs under the same audio device?

InflatableMouse · « **Reply #14 on:** January 15, 2013, 01:49:08 am »

Quote from: mykillk on January 14, 2013, 07:43:36 pm

Also, are there any free or cheaper alternatives for the room correction software? Audiolense is like $600...

Check out REW.

)p( · « **Reply #15 on:** January 15, 2013, 02:15:09 am »

Quote from: mykillk on January 14, 2013, 07:43:36 pm

I just pre-ordered one of those UMM-6s. I can't believe I lucked out like this on the timing since the mics will be shipped sometime soon. Now I don't have to buy a USB based XLR adapter with phantom power

Any idea how one of these custom calibrated UMM-6s compare to the standard Audyssey mic?

Also, are there any free or cheaper alternatives for the room correction software? Audiolense is like $600...

I use the free drc software from here:
http://drc-fir.sourceforge.net/

mwillems · « **Reply #16 on:** January 15, 2013, 11:52:52 am »

Quote

Doesn't ASIO4ALL allow you to choose different inputs and outputs under the same audio device?

My experience trying to use ASIO4ALL as a wraparound bridge when doing convolution testing was very poor. I have an M-Audio external card with phantom power for my mic, but I need at least five outputs for playback on my system, and my output card is internal (an Asus) and has no phantom power. I tried to take measurements using ASIO4ALL to try and "bridge the gap" between the input external card and the internal output card and the resulting filters (when applied) made everything sound very, very odd when I got done (hollow sounding, tons of pre-echo, etc.).

To be fair, it may be my user error in some other part of my process that screwed everything up because I haven't been able to try it the "right" way yet. I still don't have a soundcard with both phantom power and enough outputs to do convolution on my setup (I've been too busy trying to finish up some other things to worry about convolution just yet). I'll probably have to invest in a freestanding analog phantom power amp at some point that I can route into my Asus (as even the USB mics mentioned up thread would be separate audio devices in windows).

That said, what I can report is that using ASIO4ALL did not produce satisfactory results for me.

mykillk · « **Reply #17 on:** January 15, 2013, 08:56:16 pm »

I could really use some examples of a proper setup for being able to do accurate time domain measurements with either an XLR or USB mic. Is it even possible with a USB mic? mwillems does not leave me hopeful. I might cancel my UMM-6 order and go with a full XLR setup if the price isn't exorbitant.

mojave · « **Reply #18 on:** January 16, 2013, 10:27:23 am »

Quote from: mykillk on January 15, 2013, 08:56:16 pm

I could really use some examples of a proper setup for being able to do accurate time domain measurements with either an XLR or USB mic. Is it even possible with a USB mic? mwillems does not leave me hopeful. I might cancel my UMM-6 order and go with a full XLR setup if the price isn't exorbitant.

If you aren't doing active crossovers but are doing bass management and EQ, then the UMM-6 will be sufficient. Active crossovers require exact timing between drivers so using ASIO with the same audio device for measuring/playback is recommended. However, if you are doing convolution of speakers that have a passive crossover in them, then you will be fine.

natehansen66 · « **Reply #19 on:** January 16, 2013, 12:09:28 pm »

Quote from: mykillk on January 15, 2013, 08:56:16 pm

I could really use some examples of a proper setup for being able to do accurate time domain measurements with either an XLR or USB mic. Is it even possible with a USB mic? mwillems does not leave me hopeful. I might cancel my UMM-6 order and go with a full XLR setup if the price isn't exorbitant.

I know what you mean. When measuring phase on separate drivers for crossover work you need to be able to loopback the signal to account for the time of flight between the 2 measurements.I don't think the USB mic can do that.

A good free software for doing linear phase correction is rePhase: http://sourceforge.net/projects/rephase/

mwillems · « **Reply #20 on:** January 16, 2013, 02:57:54 pm »

Quote

If you aren't doing active crossovers but are doing bass management and EQ, then the UMM-6 will be sufficient. Active crossovers require exact timing between drivers so using ASIO with the same audio device for measuring/playback is recommended. However, if you are doing convolution of speakers that have a passive crossover in them, then you will be fine.

Which is exactly my situation (active crossovers), I'm bi-amping using JRiver as my crossover. So if that's the determining factor, folks with passive crossovers in their speakers may have better results.

mykillk · « **Reply #21 on:** January 17, 2013, 06:07:41 pm »

Excellent. Thanks for the info guys.

So if I understand right I don't need the precise ASIO timing just to set the delay on each speaker so that they match up?

I was reading up on active crossovers and it seems pretty cool. But I don't have bi-ampable speakers so it's of no use to me right now. And I can see why the precise timing is so important. It would sound pretty weird to not have the tweeter and the woofer in your speakers matching up...

mwillems · « **Reply #22 on:** January 17, 2013, 08:58:00 pm »

Quote from: mykillk on January 17, 2013, 06:07:41 pm

Excellent. Thanks for the info guys.

So if I understand right I don't need the precise ASIO timing just to set the delay on each speaker so that they match up?

I was reading up on active crossovers and it seems pretty cool. But I don't have bi-ampable speakers so it's of no use to me right now. And I can see why the precise timing is so important. It would sound pretty weird to not have the tweeter and the woofer in your speakers matching up...

All speakers are potentially bi-ampable if you don't mind opening them up ;-) It's fairly easy to bypass a passive crossover if you're committed to the idea.

That said if I were dealing with a normal tweeter with a more or less flat frequency response, I'd probably have gone passive and skipped the biamping because passive crossover designs that just cross things over are easy to find or make. But my speakers are homebrew, and my HF drivers are compression drivers on a pair of old Altec sectional horns and there are so many issues I'd need to fix in my passive crossover (Horn compensation, Driver EQ, 20 dB of attenuation, delay to sync the woofers because the horn is almost two feet deep) making one would have been a huge pain (for someone without an EE background).

And bi-amping offers some concrete audio advantages, as well, if you have enough amp channels and enough DSP to do it. But it makes things complicated in other ways. When I build my center channel, it will probably not be bi-amped :-)

mykillk · « **Reply #23 on:** January 18, 2013, 06:41:22 am »

Quote from: mwillems on January 17, 2013, 08:58:00 pm

All speakers are potentially bi-ampable if you don't mind opening them up ;-) It's fairly easy to bypass a passive crossover if you're committed to the idea.

That said if I were dealing with a normal tweeter with a more or less flat frequency response, I'd probably have gone passive and skipped the biamping because passive crossover designs that just cross things over are easy to find or make. But my speakers are homebrew, and my HF drivers are compression drivers on a pair of old Altec sectional horns and there are so many issues I'd need to fix in my passive crossover (Horn compensation, Driver EQ, 20 dB of attenuation, delay to sync the woofers because the horn is almost two feet deep) making one would have been a huge pain (for someone without an EE background).

And bi-amping offers some concrete audio advantages, as well, if you have enough amp channels and enough DSP to do it. But it makes things complicated in other ways. When I build my center channel, it will probably not be bi-amped :-)

Oh man, don't tempt me to start busting open my speakers for rewiring! I could only imagine what my room would look like with two amps for every speaker

natehansen66 · « **Reply #24 on:** January 18, 2013, 08:06:46 am »

Quote from: natehansen66 on January 16, 2013, 12:09:28 pm

I know what you mean. When measuring phase on separate drivers for crossover work you need to be able to loopback the signal to account for the time of flight between the 2 measurements.I don't think the USB mic can do that.

I take that back. Holm Impulse can account for the timing differences when set up correctly with no physical loopback. I use REW for all of my measurements, but when I want to look at phase I use Holm.

Delay can be figured by looking at the impulse (or step response at lower freqs) when measuring 2 drivers/speakers simultaneously. I'll make a short measurement centered on the crossover region to limit the data in the window.

INTERACT FORUM

Author Topic: Recommended microphone for convolution measurements (Read 7702 times)