DIY Mic (WM-61A) vs. RS Meter - my measurements

[nelmr]

Okay, I finished my measurement mic. I bought some helping hands with alligator clips and this helped a lot. I was able to successfully solder my first capsule on the first try, albeit my soldering could have been a better - but it works so that's what counts right?

Anyway, I hooked it up to my laptop, set the SPL to be calibrated to 75dB and measured the 10-80hz response of the bass in my room with this mic. I then did the same with the radio shack meter. I have the DIY mic cable taped onto the RS meter so that the DIY mic is about an inch above the RS meter's. There is no enclosure to my DIY mic. Just the cable + mic capsule.

Below is the results of my test. The note about the RS meter being calibrated or not deals with using a calibration file or not for that meter (obtained from Home theater shack for the new model RS meter). Obviously no calibration file has been used for the DIY mic. Please note that I repeated this test twice to verify the findings:

Right Front Crossed Over at 200HZ, 24 second sweep 10-80Hz:

Right Front Crossed Over at 60Hz, 24 second sweep 10-22,000Hz. Averaged to 1/12 octive

I'm surprised to find very similar plots for the calibrated RS meter vs. the DIY Panasonic WM-61A mic below 2khz. Note that the RS meter calibration file only went up to 1khz. So that's the reason red and green lines join then after. IG, Soundhound, I am curious about your response. Both charts are made with my soundcard calibrated down to 10Hz and up to 20khz. Here is the soundcard response without calibration:

 

[Flint]

Here's what I see... in your bass only chart I see a difference of only 1 to 1.5dB between the DIY mic and the calibrated RS-Meter, but in the fullrange chart I see differences as wide a 3dB in that same range. A difference is 3dB is very siginificant for a measurement. Why the disparity?

Also, the fullrange measurement really shows why you need a proper measurement mic. Our ears are vastly more sensitive the most minute differences in output in the 800Hz to 5kHz range, and the RS-Meter is not even close to being accurate in that super-critical range.

 

[soundhound]

Also, the fullrange measurement really shows why you need a proper measurement mic.

You just won't let it go, will you. First you dismissed my electronic crossover board (without even looking at the schematic of the circuit no less) then you doubted "amateurs" ability to solder a DIY mic, now you are dismissing the DIY mic's response as out of hand. Would you feel this way if you had recommended this DIY mic? I think Not.

For the record, the DIY mic capsule is within a dB or two over the entire spectrum compared to my Bruel & Kjaer professional measurement microphone, and that microphone is within a fraction of a dB of flat over far more than the audio range. The DIY mic is plenty flat enough for measurements like these; differences from small mic placement would swamp any frequency response irregularities between the DIY mic and the Bruel & Kjaer mic.

Please restrain your ego. You are not the sole source of knowledge on the planet. Please remember that.

 

[Towen7]

FWIW, I read the comment as an endorsment of the DIY mike by pointing out the inaccuracies of the RS meter at the mid range.

 

[Flint]

Indeed, I was very much endorsing the DIY mic as the absolute best option here.

 

[soundhound]

The differences in the plots are probably due to the frequency response shaping in the RS meter's circuit. Remember, that meter has "C" weighting, which has some rolloff in the high and low ends of the spectrum, so that automatically makes it unsuitable to use for frequency response measurements. The roughness in the high end is probably representative of the response at that point (but see below). I know this mic capsule to be pretty darn good across the range.

Taping the DIY mic capsule to the RS meter is not wise, and will introduce response problems because of diffraction from the RS meter's casing. The proper way to do this is to have each mic occupy the same space independently. I use two mic booms and swing one into place, then the other. The mic and the RS meter should be at least a foot or more separated from each other to avoid diffraction, and this will show up especially in the high frequencies.

 

[nelmr]

Taping the DIY mic capsule to the RS meter is not wise, and will introduce response problems because of diffraction from the RS meter's casing. The proper way to do this is to have each mic occupy the same space independently. I use two mic booms and swing one into place, then the other. The mic and the RS meter should be at least a foot or more separated from each other to avoid diffraction, and this will show up especially in the high frequencies.

I understand. I'll make sure I leave just the DIY mic by itself when I make measurements with it. By the way here are the pics of how I had the DIY mic taped to the RS meter and some closeups that you can see my soldering job (which works, but isn't the best):

 

[soundhound]

Yeah, the DIY mic capsule is way too close.

The bottom line in the soldering is that it works. Less than great solder joints may be more prone to failure from vibration, but your mic isn't going to see violent movement.

For what its worth, even I have messed up on solder joints on this mic capsule, mostly with solder flowing too much and shorting the two wires together. Even though I generally solder under a magnifier for small stuff like this, it is still harder to solder this mic capsule than wires on a circuit board......now soldering teeny tiny surface mount parts is another story, and I use a microscope and specialized soldering equipment for that task. Surface mount parts like 0201 resistors which are the size of grains of sand are particularly fun.....

 

[soundhound]

FWIW, I read the comment as an endorsment of the DIY mike by pointing out the inaccuracies of the RS meter at the mid range.

Had the comment not been part of a general pattern of dissing anything I've designed or promoted, I could see your point. Unfortunately, it is part of a pattern which needs to stop.

 

[nelmr]

So would this be better for measuring response at the couch. I tapped the cable to the couch and put a soft foam pad ontop of the tapped area. Would it be perferable to just remove the foam?

FYI: Yeah there is a coffee table in front of the couch. I'll remove it when making room response measurements. Actually, it would be interesting to see the effects of it being there. The center channel would be most affected, but probably still for the mains.

 

[soundhound]

So would this be better for measuring response at the couch. I tapped the cable to the couch and put a soft foam pad ontop of the tapped area. Would it be perferable to just remove the foam?

FYI: Yeah there is a coffee table in front of the couch. I'll remove it when making room response measurements. Actually, it would be interesting to see the effects of it being there. The center channel would be most affected, but probably still for the mains.

That placement would be acceptable as the couch and foam act to keep back reflections down. The thing is to keep the mic at least a foot from anything else which could reflect sound waves. At a foot spacing from a reflective object, the interference would be in the vicinity of roughly 1000Hz, so the farther away from objects the better. The high frequencies are particularly susceptible to interference because of the short wavelengths involved.

BTW, normally with an omnidirectional mic, the angle of incidence to the sound source is supposed to be 90 degrees (i.e. with the mic pointing straight up to the ceiling) in order to keep high frequencies from being overly emphasized as the wavelength starts to become a significant part of the diameter of the mic capsule. However since the frontal surface area of your mic is so small, there is really no problem with using it as you are. My 1/2" Bruel & Kjaer mic has to be used at 90 degrees for flat response.

 

[nelmr]

BTW, normally with an omnidirectional mic, the angle of incidence to the sound source is supposed to be 90 degrees (i.e. with the mic pointing straight up to the ceiling) in order to keep high frequencies from being overly emphasized as the wavelength starts to become a significant part of the diameter of the mic capsule. However since the frontal surface area of your mic is so small, there is really no problem with using it as you are. My 1/2" Bruel & Kjaer mic has to be used at 90 degrees for flat response.

Actually when I soldered the capsule the wire wasn't perfectly straight. The mic capsule is about 45 degrees relative to the wire. So In the pic it actually is pointing straight up to the ceiling.

Thanks soundhound for your help with this.

 

[nelmr]

Also the WM-61A capsule has generic calibration files available from 300hz to 22000 hz online. It seems that this capsule may only be 2.5dB down in the higher freqs. Here is my comparison using this file.

I got the file from here:

http://www.claudionegro.com/download.html

 

[nelmr]

And here is the freestanding measurement vs. the taped to the RS meter measurement.

NOTE: The placement isn't exactly the same so that probably explains the differences in the bass, plus the dishwasher is on now. Lastly, both measurements were without a calibration file.

So the high frequencies seem more tamed compared to the taped-to-the-rs-meter one, as soundhound predicted.

 

[Flint]

Wow! I post a great "there you go" post supporting Altec's mic idea and he atttacks me twice! WTF?

 

[PaulyT]

I tried to find a chick on a bike, but this is as close as I could come...

Anyway, don't make the admins get involved, we're a grumpy bunch.

 

[nelmr]

Wow! I post a great "there you go" post supporting Altec's mic idea and he atttacks me twice! WTF?

IG, I think Soundhound was misunderstanding your use of the words "proper measurement mic", when it seems he thought you were implying the DIY mic is not considered as such, but to me it sounds like you were saying it would fall into that category. The implication was that the RS meter was not a "proper measurement mic", not the DIY mic.

 

[soundhound]

Also the WM-61A capsule has generic calibration files available from 300hz to 22000 hz online. It seems that this capsule may only be 2.5dB down in the higher freqs. Here is my comparison using this file.

I got the file from here:

http://www.claudionegro.com/download.html

If you were pointing the mic at 90 degrees incidence to the speaker, then that might explain the discrepancy at the high end. Your mic does not need to be at 90 degrees to the source; point it directly at the speaker. Only mic capsules which are larger need to be at 90 degrees.

 

[soundhound]

Wow! I post a great "there you go" post supporting Altec's mic idea and he atttacks me twice! WTF?

Live with it.

 

[soundhound]

I tried to find a chick on a bike, but this is as close as I could come...

Anyway, don't make the admins get involved, we're a grumpy bunch.

I have never spoken out one way or another on Flint's projects or ones he recommends. Never. Yet he makes snide remarks on mine. I didn't start it, but am only calling out what needs to change. His ego knows no bounds.

Frankly, if the admins wish to ban me, please go ahead. It will give me more time to attend to my other interests.