Home Theater Forum and Systems banner

201 - 220 of 368 Posts

·
Registered
Joined
·
208 Posts
There's also the used-equipment inventory at Modal Shop.
Neat, love used equipment sources--good call, thanks.

Also remember that the noise floor of the mic will have to be at least 8 to 10 dB lower than the noise level to be measured, so to measure noise levels at 20 dBA, the mic will need a noise floor of 8 to 10 dBA.
OK so to verify RC30, the maximum acceptable noise level per the Nyal/Hedbeck paper, because the A curve is ~0dB at 4k, and the RC curve is at 20 there, to be able to verify that requirement you'd need a mic with 10-12 dBA noise floor? Am I incorrect because the RC measurement is in an octave band which will contain less energy and therefore correspond to a higher dBA noise floor requirement? I note that the chart in the link shows 40 dBA as the equivalent sound level of RC30--if true that would be a significantly easier requirement to meet (maybe I need to just read the ARTA manual finally, LOL).

With the LinearX mic you do have to pay extra for the clamshell case and mic clip. LOL. It also looks like it requires +9V so you have to build special cables rather than simply being able to use the phantom power on the common interfaces....
 

·
Premium Member
Joined
·
397 Posts
OK so to verify RC30, the maximum acceptable noise level per the Nyal/Hedbeck paper, because the A curve is ~0dB at 4k, and the RC curve is at 20 there, to be able to verify that requirement you'd need a mic with 10-12 dBA noise floor? Am I incorrect because the RC measurement is in an octave band which will contain less energy and therefore correspond to a higher dBA noise floor requirement? I note that the chart in the link shows 40 dBA as the equivalent sound level of RC30--if true that would be a significantly easier requirement to meet (maybe I need to just read the ARTA manual finally, LOL).
This is a hard question to answer since RC (as well as NC and NCB) are defined in the frequency domain, so the applicability of any particular mike will really depend on the specific characteristics of the mic as well as the spectrum of the background noise. The overall equivalent A-weighted level can very wildly for the same RC level. For example, if the background noise was a pure tone at 31.5 Hz at 60 dB, the A-weighted equivalent for that tone is 20 dBA, but it would still exceed RC 30.

All I can really say for sure is that microphone self-noise typically follows a 1/f pattern, so the self-noise may be more of a problem in low-frequency applications compared to mid- and high-frequency applications.
 

·
Registered
Joined
·
208 Posts
This is a hard question to answer since RC (as well as NC and NCB) are defined in the frequency domain
OK, I can see your point here.

Pros at this can't all just buy the ultimate B&K rig I'd imagine (although Nyal uses some type of B&K rig, don't know if it's "ultimate" -- but B&K's app note on KYDG says KYDG balked at the cost). And I wouldn't think they would all just buy a mic advertised for this use and hope. If there isn't a way to pre-calculate it, do pros do a test that verifies the self-noise of their chain?

All I can really say for sure is that microphone self-noise typically follows a 1/f pattern, so the self-noise may be more of a problem in low-frequency applications compared to mid- and high-frequency applications.
That seems promising, given that's effectively the shape of the allowed noise curves under most systems--especially RC. Thanks for that, I was wondering what the typical spectral shape looks like since I don't think anyone's mentioned it yet.
 

·
Premium Member
Joined
·
397 Posts
OK, I can see your point here.

Pros at this can't all just buy the ultimate B&K rig I'd imagine (although Nyal uses some type of B&K rig, don't know if it's "ultimate" -- but B&K's app note on KYDG says KYDG balked at the cost). And I wouldn't think they would all just buy a mic advertised for this use and hope. If there isn't a way to pre-calculate it, do pros do a test that verifies the self-noise of their chain?
Yes. The simplest way is to get a dummy mic that matches the capacitance of the microphone you're going to use (something like this) and determine the equivalent noise floor. The more difficult method is to put the mic in a quiet environment and determine the noise floor directly. For my noise floor measurements, I have an old hearing-aid chamber that I bought off eBay a few years ago, that provides isolation of about 15-20 dB. Using it in my basement at night, I can get levels down to below 16 dBA (that's as low as my Larson-Davis, NTi and Sencore meters go with the mics I own). Building a heavier-duty enclosure with mounting it on a vibration-isolated platform, you might be able to approach 5 dBA or so. And of course there are anechoic chambers such as the Orfield facility with free-field background levels down to 0 dBA or below.

That seems promising, given that's effectively the shape of the allowed noise curves under most systems--especially RC. Thanks for that, I was wondering what the typical spectral shape looks like since I don't think anyone's mentioned it yet.
If the noise source is a properly installed and balanced HVAC noise, the spectrum should resemble NC/NCB/RC curves. If the HVAC is improperly installed (unbalanced, or generating structure-borne noise) or it's a different source (outdoor traffic for example), the spectrum would change accordingly.
 

·
Registered
Joined
·
208 Posts
Yes. The simplest way is to get a dummy mic that matches the capacitance of the microphone you're going to use
OK gotcha. Now this is starting to make sense. One more question, that page says "The method is not usable for measurement of the thermodynamic noise of the microphone cartridge." I presume this is based on the presumption that is an inconsequential contributor to the total self-noise of the chain (so the preamp is virtually all of the self-noise), and that there are studies somewhere substantiating it; is this correct? Or is it just sort of standard practice so no one worries about it anymore?

For my noise floor measurements, I have an old hearing-aid chamber
Ahh, makes sense again. I saw a page somewhere discussing mic testing in a chamber, and I'd envisioned it before that. That there's a commercial product only makes sense.

If the noise source is a properly installed and balanced HVAC noise, the spectrum should resemble NC/NCB/RC curves.
LOL, I knew that--I was wondering about the spectrum of the self-noise, not the spectrum of the typical expected environmental noise/etc. I should have been more specific.

Thanks for your insights here.
 

·
Premium Member
Joined
·
397 Posts
OK gotcha. Now this is starting to make sense. One more question, that page says "The method is not usable for measurement of the thermodynamic noise of the microphone cartridge." I presume this is based on the presumption that is an inconsequential contributor to the total self-noise of the chain (so the preamp is virtually all of the self-noise), and that there are studies somewhere substantiating it; is this correct? Or is it just sort of standard practice so no one worries about it anymore?
Good question. Consulting my copy of Beranek's Acoustical Measurements, the self noise (in rms volts) caused by thermal fluctuations can be calculated using the equation

e = sqrt (4 * k * T * R)

where k= Boltzmann gas constant of 1.37e-23, T= absolute temperature in Kelvin, R is the resistive component of the mic impedance (usually in the range of a couple of hundred ohms)

For a temperature of 296 K (room temp), R set to 1000 ohms, that gives an rms voltage of 4e-9 volts. For a 10 mV/Pa mic, that corresponds to an RMS pressure of 4e-7 Pa, which is about -33 dB. So it looks like it is an inconsequential contributor.

Ahh, makes sense again. I saw a page somewhere discussing mic testing in a chamber, and I'd envisioned it before that. That there's a commercial product only makes sense.
You could just as easily build one using heavy wood and foam, it was just easier for me to buy it on eBay, since I don't think I paid more than $50 for it.

LOL, I knew that--I was wondering about the spectrum of the self-noise, not the spectrum of the typical expected environmental noise/etc. I should have been more specific.
Oops. :)

Here are some noise floor plots using an NTI XL2 and a dummy mic:





Here are some noise floor plots using an NTI XL2 and a PCB mic in my chamber (note the different y-axis scale):





The chamber measurements were made in the middle of the day, so they're probably a few dB high because of outdoor activity.
 

·
Registered
Joined
·
208 Posts
So it looks like it is an inconsequential contributor.
Does this only go for pro-style mics, which I presume don't have active components in the capsule? Because there are lots of pages about the Panasonic WM-61a (supposed that was/is used in the ECM8000) that supposedly has a FET in it, the (infamous) Linkwitz mod, etc.

You could just as easily build one using heavy wood and foam, it was just easier for me to buy it on eBay, since I don't think I paid more than $50 for it.
Fair enough. There's one there now too but it's $150 start price for the auction. Not that I really need such a thing, LOL.

Thanks for the plots, I always like data. :)
 

·
Premium Member
Joined
·
397 Posts
Does this only go for pro-style mics, which I presume don't have active components in the capsule? Because there are lots of pages about the Panasonic WM-61a (supposed that was/is used in the ECM8000) that supposedly has a FET in it, the (infamous) Linkwitz mod, etc.
That's a good point about the FET, pro measurement mic capsules (AFAICT) are relativey "pure" with just the diaphragm and a charged backplate. The pre-amps obviously have more electronics, but they are designed to be low noise (pre-amps go for about $300 - $500 new, $50-$150 used, plus you need a power supply).

The noise floor for WM-61a's that I've measured have been around 25 dBA.

Fair enough. There's one there now too but it's $150 start price for the auction. Not that I really need such a thing, LOL.
I'm guessing that one actually works, unlike the one I bought.

Another option would be to check it a local cabinet maker, furniture maker or woodshop. A local cabinet maker made the enclosure of my reference speaker (12-inch cube) using scrap pieces of 1/2-inch mdf for $50.
 

·
Registered
Joined
·
208 Posts
Let me just say thank you very much for all your responses. They were highly enlightening for me and I think I understand the basics of this area much better now.

I did think of one more pertinent question... for a hobbyist wanting to measure the background (or projector, pro amp, etc) noise in their HT, do you think (for absolute level) comparison to RS meter (eg I have the old analog model) is sufficient or would I need to consider getting a meter with calibration? Would it be better to just invest in an SPL calibrator? I read that meters are better than they used to be and that calibrators are a throwback (and double check) today.
 

·
Premium Member
Joined
·
397 Posts
I did think of one more pertinent question... for a hobbyist wanting to measure the background (or projector, pro amp, etc) noise in their HT, do you think (for absolute level) comparison to RS meter (eg I have the old analog model) is sufficient or would I need to consider getting a meter with calibration? Would it be better to just invest in an SPL calibrator? I read that meters are better than they used to be and that calibrators are a throwback (and double check) today.
If you're just looking at overall dBA or dBC levels, the Radio Shack meter isn't actually too bad. However, if you're trying to get levels at particular frequencies or particular frequency bands, stay away from the RS meters as their frequency responses are all over the place, especially above 1 kHz.

Now with card to getting a meter with or without calibration - if you get a meter that is ANSI or IEC rated, generally the frequency response will fall within the rated tolerances even if you don't have a specific calibration curve for the meter. Just keep in mind the the Type 2/Class 2 tolerances are +/- 1.5 dB from 100 to 1250 Hz, +/- 2 dB from 40-80 Hz & 1600-2000 Hz, and become higher beyond those limits (up to +5/-infinity below 20 Hz and above 8000 Hz) so it's up to you to decide if those tolerances are acceptable.

As for getting an acoustical calibrator, again that depends on how precise you are trying to get. Out-of-the-box most ANSI/IEC rated meters (even the cheap ones) will be within 3-4 dB of the absolute level - most of the differences are do to unit-to-unit setting variations as well as factors that can affect ambient air pressures (such as elevation). If you want more precision, an acoustical calibrator is a good investment, you can get decent Type 2 calibrators from Amazon. A calibrator is also good for checking the meter if you drop it, which might shift the calibration set screw found on lower end meters.

Again remember that the noise floor for your meter needs to be at least 8-10 dB lower than the level you're trying to measure. The noise floor on Type 2 meters (including the Galaxy ones I sell) are around 30 dBA, so if you're trying to measure noise from quiet equipment, you might have difficulty with those meters.
 

·
Premium Member
Joined
·
19 Posts
Herb,
I have a pair of B&K 4007’s with the original factory Calibration charts, dated Jan. 1993.
I’m considering having you calibrate one of them for me so I will have a cal. file for REW.
Could you please advise of the current pricing and approximate return shipping cost to Canada.
I trust you didn’t indulge in too much turkey today. Thanks for all your comments and Cheers.
 

·
Premium Member
Joined
·
397 Posts
Herb,
I have a pair of B&K 4007’s with the original factory Calibration charts, dated Jan. 1993.
I’m considering having you calibrate one of them for me so I will have a cal. file for REW.
Could you please advise of the current pricing and approximate return shipping cost to Canada.
I trust you didn’t indulge in too much turkey today. Thanks for all your comments and Cheers.
A basic calibration (on-axis frequency response only) is $55 per mic. Shipping is included for US customers, for Canada, I'll have to ask for $20 USD shipping. Two off-axis angles, polar response and sensitvity/noise floor are available for $10 extra each. $75 per mic gets you everything. Fee is payable with check or through PayPal. If you want to proceed, PM me and I'll send you shipping info.

edit: one thing to mention is that during my last job, we found that B&K mics were extremely fragile and we had around a 8-10 of them fail on us over the course of my tenure there (8 years). Depending on much you value these mics, you may want to consider if you really want to subject them to international shipping (or at least insure them).
 

·
Premium Member
Joined
·
19 Posts
edit: one thing to mention is that during my last job, we found that B&K mics were extremely fragile and we had around a 8-10 of them fail on us over the course of my tenure there (8 years). Depending on much you value these mics, you may want to consider if you really want to subject them to international shipping (or at least insure them).
Thanks for the quick reply. I will have to consider your concern about possible damage and get back to you.
 

·
Premium Member
Joined
·
397 Posts
edit: one thing to mention is that during my last job, we found that B&K mics were extremely fragile and we had around a 8-10 of them fail on us over the course of my tenure there (8 years). Depending on much you value these mics, you may want to consider if you really want to subject them to international shipping (or at least insure them).
As an FYI in case anyone is interested, we eventually phased out the B&K mics for GRAS microphones and those things were about as bulletproof as you could get - I think we lost one during my tenure. My 7-year old ACO Pacific 7052 has also been a champion, it gets used nearly every day, and its last calibration (last June) showed no problems.
 

·
Registered
Joined
·
335 Posts
I posted elsewhere, but if I have my CM-140 calibrated would it give me better results when setting speaker and subwoofer levels with pink noise? Or would the standard CM-140 suffice for that role? My Radioshack meter was stolen, so I need another meter, but if the Cross Spectrum calibration offers better results then I would do it. Just need some advice here.
 

·
Registered
Joined
·
208 Posts
I posted elsewhere, but if I have my CM-140 calibrated would it give me better results when setting speaker and subwoofer levels with pink noise? Or would the standard CM-140 suffice for that role? My Radioshack meter was stolen, so I need another meter, but if the Cross Spectrum calibration offers better results then I would do it. Just need some advice here.
Technically the noise a processor emits for the manual level setting operation is white noise, and no a "calibrated" meter will not really give better results. That's because (a) for a meter calibration does not change the frequency balance the meter senses, it only corrects the absolute level it reads and (b) the absolute level is not important when setting the channel levels as the most important thing about them is the relative level. That is unless you really care that 0 is "reference level"... but very few people care to listen at reference level and there's no reason to attempt to lose your hearing. You are best off considering the volume setting during regular listening to only be a relative indication and not worry about what the number is; just set it "by ear" to a level that allows you to hear most of the detail in a soundtrack while being comfortable.
 

·
Premium Member
Joined
·
397 Posts
Technically the noise a processor emits for the manual level setting operation is white noise, and no a "calibrated" meter will not really give better results. That's because (a) for a meter calibration does not change the frequency balance the meter senses, it only corrects the absolute level it reads and (b) the absolute level is not important when setting the channel levels as the most important thing about them is the relative level. That is unless you really care that 0 is "reference level"... but very few people care to listen at reference level and there's no reason to attempt to lose your hearing. You are best off considering the volume setting during regular listening to only be a relative indication and not worry about what the number is; just set it "by ear" to a level that allows you to hear most of the detail in a soundtrack while being comfortable.
It depends on how Vaughan100 is using the meter and the type of "calibration" being discussed. In terms of just a simple level calibration (my Verified products) and the meter being used as a stand-alone device, aackthpt's post is entirely correct. My "Verified+" offer frequency correction curves in narrow-band and 1/3 octave band levels. The 1/3 octave band correction levels can be used to correct the response of the meter when used with pink noise (although doing it manually would be tedious). If the meter is connected to a reliable sound card (or a decent computer-on board audio), the narrow-band correction files can be used with programs like REW etc to correct for the meter response.

The calibrations do provide increased accuracy, but the difference is not as dramatic as it is for microphones like the ECM8000.
 

·
Registered
Joined
·
208 Posts
In terms of just a simple level calibration (my Verified products) and the meter being used as a stand-alone device
Thanks for the clarification; I was indeed making that assumption. I would never think of someone hooking a meter up to REW when just setting levels (seems like unnecessary effort) but I also guess I hadn't considered that someone might be setting levels between mains and sub by using sweeps or tones+RTA window, in which case calibration would definitely be desirable!

Oh well, we all know what "to assume" does.... :whistling:
 

·
Registered
Joined
·
335 Posts
I'm talking about setting levels to 75 dB for each speaker and subwoofer. Would the standard CM-140 suffice or would the calibrate (verified) offer better results? The whole reason for NOT using my ears is because the ears are usually unreliable for setting levels, hence why I need to verify what the levels are.

I just need clarification on the meter thing. If I'm burning cash for no reason then I'll just stick to the standard CM-140. I won't be using REW, just the internal pink noise in the AVR.
 

·
Premium Member
Joined
·
397 Posts
I'm talking about setting levels to 75 dB for each speaker and subwoofer. Would the standard CM-140 suffice or would the calibrate (verified) offer better results? The whole reason for NOT using my ears is because the ears are usually unreliable for setting levels, hence why I need to verify what the levels are.

I just need clarification on the meter thing. If I'm burning cash for no reason then I'll just stick to the standard CM-140. I won't be using REW, just the internal pink noise in the AVR.
The standard uncalibrated CM-140 tends to be within 1-3 dB of the actual SPL value out-of-the-box. That said, the verified meters I sell have been adjusted to the correct value and sell for about the same price you can get a CM-140 elsewhere (and a little cheaper with the HTS discount).
 
201 - 220 of 368 Posts
Top