Reccomend a Wired Lav NOT a Sony That Doesn't Pick up Ultrasonic Motion Detectors?

[Ken Reeser]

This is actually a follow-up to a post of mine from 2008:
http://www.dvinfo.net/forum/all-thin...buildings.html

The cause of the problem was, as Ty Ford suggested, the ultra-sonic motion detectors used in some rooms. Moreover, the problem was most bothersome when a small room is equipped with a detector made for a large room, which emits a much more powerful signal. The problem is worst when the detector is right over the participants.

(To sum it up for those who won't read the previous post: my microphones were picking up this signal and this made the VU meter on my Shure mixer swing wildly and hover in the upper ranges, causing the limiter to kick in intermittently, playing havoc with my recording quality.)

Once I confirmed this, I discovered that my hand-held AKG C1000 did not pick up the signal at all, even when pointed directly at the device. Obviously a microphone has to be sensitive to the frequency in order for it to be picked up and passed onto the mixer. This gives me hope that perhaps other brands of lavaliers might not be affected by this particular high frequency, yet still be as good as or at least comparable to the Sony 44b's and 55b's I have been using.

So I am seeking a recommendation, but only if you have some specific experience with this problem and know that a particular lavaliere won't be affected by the signal AND that it's a good microphone.

Anyone? Thanks.

[Greg Miller]

Do you have any idea, or is there any way to find out, what range of frequency(ies) are used by the detectors?

I would think a good universal solution would be to build a high-cut filter which can be inserted in line between the mixer and the mic. This would be similar to the low-cut filters for reducing room rumble and wind noise, except that it would cut the highs rather than the lows. Ideally, the filter would hopefully pass everything up to around 15 kHz, and then drastically cut everything above that.

Of course to design a filter, one needs to know where the offending frequencies are located. If they are very close to the audio band (for example, if they're around 25 kHz) then designing a sharp enough filter would not be trivial. OTOH, if they're at least an octave above the audio band (say around 45 kHz or higher) then the filter design becomes simpler.

Indeed, there might be some mics out there that are not capable of picking up these detectors, but I imagine this is a rather unusual question, so most mic specs might not predict this with any certainty or accuracy. It's certainly nothing I've run into, so I can't give any advice about specific mics. But I suspect lavs, which are typically small diaphragm condensers, might very likely have some response up above 20 kHz; clunky old large-diaphragm dynamics are probably much safer in general.

Sorry I can't advise re: specific mics, but I would be interested to know what frequencies we're talking about.

[Ken Reeser]

Originally Posted by Greg Miller

...Of course to design a filter, one needs to know where the offending frequencies are located. If they are very close to the audio band (for example, if they're around 25 kHz) then designing a sharp enough filter would not be trivial....

As a matter of fact the frequency is exactly 25kHz.

So, add to that the fact that building any kind of filter would be difficult for me, perhaps that's not such an easy solution. I have looked for just such a solution, and the internet is full of schematics, designs and forums dedicated to building such things. I can't understand more than three words in ten. If there were such a design available that listed the parts in normal English, or at least by names I could look up and order, plus a visual diagram of how all the parts are connected together (not schematics), then maybe. But I see nothing of the kind.

However, I have discovered this little item:

http://www.bhphotovideo.com/c/product/275823-REG/PSC_FPSC0010E_ALPF_Low_Pass_Filter.html

Which is the one and only low-pass filter commercially available as an in-line XLR filter, which is exactly what I need. However it does not filter from, say, 20 kHz and above, but at 7 kHz. It says the cutoff is -3dB. Is that drastic enough? It sounds like that ought to do the trick, especially since it was designed to filter high frequency noise. And my research tells me that most human voices fall within the 1kHz to 5kHz range, so there should not be any effect on the voices of my subjects. In theory.

It might help you to know that the full audio range is not important, just the full human conversational voice range.

I am on the verge of buying one of these and testing it, but if it fails I'd like to know if there's a simple answer in getting a different mic that doesn't pick the sound up in the first place. Hence the purpose of this thread. But if you or anyone has comments on the filter idea, please do.

[Greg Miller]

For speech intelligibility, you only need frequencies up to about 3500 Hz. However, the human voice certainly does contain frequencies above that, and above 7000 Hz as well.

If you apply a sharp cutoff at 7000 Hz, and compare the audio to unfiltered speech, you certainly will hear a difference.

On the other hand, if you start a [i]gradual rolloff[/u] at 7000 Hz, the difference might be nearly imperceptable.

So we need to know the actual corner frequency, and slope (dB / octave) of this filter. I wonder where you found the 7000 Hz figure that you quote. I saw absolutely no specs on the B&H website, and a quick internet search didn't find any specs.

By all means, if you find a link to any specs or to the manufacturer, post them, and let's give it some more thought. And you could also call B&H and ask them if they'd let you return it if it doesn't solve your specific problem.

[Ty Ford]

Ken,

Thanks for the nod. :)

Are the 44b and 55b two or three wire mics? I'm wondering, if they are two wire mics, maybe the addition of the 3rd wire would allow common mode rejection to cancel out the 25kHz, or shield it from the audio. Perhaps it's also a function of the transmitter input topology. What transmitters are you using? If the transmitter has an unbalanced input, it might be allowing the 25 kHz in. I guess if the diaphragm is picking up the 25 kHz as audio, that may not be a solution.

The only 3 wire lavs I know are the Sanken COS-11 and Sennheiser MKE-2. Might give them a try. Maybe rent a couple and see.

Or find someone who can build you some inline Low Pass filters that would cut off at 15 kHz-16kHz before the transmitter. I used to have problem with LF on an Audio Technica U100 wireless. They were so wide open, that with Countryman B6 mics, the LF from HVAC systems and other large industrial electromechanical devices would generate enough LF to make the compander wiggle. A mic with a more restricted LF response scraped off the LF before it got to the transmitter. You may be having the same problem on the other end of the frequency spectrum.

Regards,

Ty Ford

[Greg Miller]

Based on Ken's original description, I'm guessing that the mics are picking up the acoustic signal (just slightly above the range of average human hearing) and then converting that to an electrical signal... that's what mics do.

If that is the case, then balanced or unbalanced won't make any difference. And in fact the lavs that he has been using (Sony ECM-44B and ECM-55B) are already balanced. This just reinforces my belief that the problem has an acoustic origin, and not a problem with unwanted electrical pickup.

And I don't believe he's using any transmitters; he specified "wired lavs" in his question, so it's not a question of transmitter electronics.

[Ken Reeser]

Thanks for the all the ideas!

Originally Posted by Greg Miller

... So we need to know the actual corner frequency, and slope (dB / octave) of this filter. I wonder where you found the 7000 Hz figure that you quote. I saw absolutely no specs on the B&H website, and a quick internet search didn't find any specs.....

The website I got the 7kHz from is this:
Adapter Barrels

And I will be writing an email to them right after this. But all they say in the text is:

"Low Pass Filter: Provides high frequency cutoff, -3dB @ 7 KHZ, to reduce high frequency noise."

Nothing about roll-off, corners, etc. Not explicitly. Still, I think I will give it a try, as it does seem to be designed to do what I want, albeit at audible frequencies. Oh, by the way, its a special order, so no returns.

Originally Posted by Greg Miller

Based on Ken's original description, I'm guessing that the mics are picking up the acoustic signal (just slightly above the range of average human hearing) and then converting that to an electrical signal... that's what mics do....

And I don't believe he's using any transmitters; he specified "wired lavs" in his question, so it's not a question of transmitter electronics.

That is correct: the 44b and 55b are wired mics, no transmitters involved.

Originally Posted by Ty Ford

Ken,

...Or find someone who can build you some inline Low Pass filters that would cut off at 15 kHz-16kHz before the transmitter.....

Ah! That would be excellent! But I have no idea where to go to find someone who can build something like that. And it sounds expensive. Still, if you know of anyone or anywhere, I'd look into it.

Maybe this is what I'm looking for:

Youspice: 20 KHz Second Order Low Pass Filter

Again, terms like "second order" mean nothing to me, but it's one of many schematics I found in my search.

[Richard Crowley]

So the turnover frequency of the FPSC0010E is specified as 7 KHz. If we assume a simple 3dB per octave from a passive single-pole filter, that yields around -10dB at 25 KHz. Quite possibly not enough attenuation @ 25KHz to eliminate your stated symptoms. That Youspice circuit has a much higher turnover frequency and an equivalent slope, so you can see that it is only ~5dB down at 25KHz. The PSC adapter barrels appear to be made for line-level use and may not operate as expected at mic levels (or impedances).

Assuming you can't turn off the detector, I would try using a different room (without a detector) or a different microphone (many are not nearly that sensitive at such a high out-of-band frequency). Or you could use a light stand to hold something over the emitter (like a big foam hemisphere or a thick hat, etc.) IMHO, using a low-pass filter is a fiddly and dubious way to solve this problem.

Note that this situation is possibly an environmental health hazard as this high-level noise is impacting your ears even if you don't perceive the sound. That can't be good for your ears.

[Ty Ford]

Richard,

I like the hat idea!

Get a bumpy foam mattress sheet. G-tape it into a cylinder and g-tape it to the emitter.

Regards,

Ty Ford

[Ken Reeser]

Originally Posted by Richard Crowley

So the turnover frequency of the FPSC0010E is specified as 7 KHz. If we assume a simple 3dB per octave from a passive single-pole filter, that yields around -10dB at 25 KHz. Quite possibly not enough attenuation @ 25KHz to eliminate your stated symptoms. That Youspice circuit has a much higher turnover frequency and an equivalent slope, so you can see that it is only ~5dB down at 25KHz. The PSC adapter barrels appear to be made for line-level use and may not operate as expected at mic levels (or impedances)....

Hmmm... well, I'm not sure what -10dB actually means to the signal I'm getting from the mics, but it sounds like a heck of a lot. And even a little would help alleviate the real problem of the limiter kicking in because it is detecting a screaming match in the room.

I have emailed the maker with my concerns. But in the meantime, I think it is safe to assume that this device performs some useful function. So if it doesn't reduce high-pitched noise significantly without causing too much change in the quality of a person's speaking voice, then what does it do?

If, however, the filter is designed for line-level and not mic level, perhaps that would be a problem. But what are you reading that suggests this is the case?

As for working in a different room or rigging up a cone of silence in the middle of the ceiling - the most immediate and obvious effect would be to turn the lights off - neither of these are real options to me. So far the limiter seems the best bet. Still accepting suggestions for different lavaliere mics, though.

[Ken Reeser]

Okay, got a reply to my inquiry from Mr. Ron Meyer at the PSC company (at least I'm assuming he's a Mr.; odds are he is, but I might be wrong), who says:

"I think our barrel would work great for your application. It rolls of at 6dB per octave with a -3db point at 7 Khz. At 14 Khz the signal will be down approx 9dB and at 25Khz it will be down approx 14dB. There is very little audio energy from the human voice at 7Khz......almost none. Keep in mind the telephone system runs out at about 3.5Khz."

So, It's off to the internet with me and my credit card, assuming we've passed the sabbath. I must say that was very nice of him or her to reply on a weekend.

I'll let you know how it works out. That might take a while. It may be weeks before I get back into such a room again. But I'll be ready.

[Greg Miller]

Filter performance is affected by the source impedance (the mic) and by the load impedance (the mixer/recorder input). So unless all these impedances are specified, it's really hard to predict how a given passive filter will work.

(By the way, a single pole filter has a slope of 6dB per octave, 20dB per decade.)

Still, $36 is trivial, so if you have the time and inclination, it wouldn't hurt to try a test.

To be really scientific, I'd tape the mic to a desk stand (so the position remains fixed), turn on the recorder and observe levels (from the ultrasonic system), then put the filter in line and observe levels again.

Of course another approach would be to have the building electrician disconnect the ultrasonic system and bypass it so the lights stay on.

[Ken Reeser]

Whether or not the filter works will be immediately apparent, believe me.

It also seems apparent that many people don't understand what an ultrasonic motion detector is, how they are installed and why. These things are designed to be on all the time to turn the lights out and save electricity. If they could be turned off, tenants would leave them off and they wouldn't do any good. And no, they aren't a hearing hazard.

I can't call the building maintenance department and demand that they disable the ultrasonic motion detector. I can't attack it with a screwdriver or redesign it so it can be turned off. Nor can I move the meeting to another room, which, in all likelihood, would also have a motion detector anyway. And try to remember that motion detectors detect motion in order to decide whether or not to keep the lights on. Blocking the motion detector is not an option.

Can't turn off, disable or block the motion detector; it's here to stay. Can't move the room. The solution has to be something I can do to my equipment, is least expensive and most permanent. That means either A) New mics, B) New mixer, or C) Low-Pass filters.

While I am still looking for A), a lavaliere microphone of good quality that will not be affected by the signal, C) became my best option after I finally found such a filter. At $38, it is by far the cheapest solution, and I could get by with two sets of three, and could use them on whatever mic or mixer I used, now or in the future. Call it fiddly if you like, it is actually a precision solution that hits the trifecta: inexpensive, little added time and goes wherever my kit goes.

Knowing the limits of the problem is the key to finding effective solutions.

[Richard Crowley]

Did you ever identify exactly which model lav microphone you are using? I tried to scan back through all the dozens of messages and didn't see it. I continue to be rather surprised that your microphone is so sensitive at 25KHz. I don't think I have ever used any lav mic that was that sensitive to ultra-sonic frequencies. How do people use ANY microphones in those buildings? Telephones, cell phones, etc.

[Greg Miller]

Originally Posted by Ken Reeser

It also seems apparent that many people don't understand what an ultrasonic motion detector is, how they are installed and why. <snip> If they could be turned off, tenants would leave them off and they wouldn't do any good. <snip> I can't call the building maintenance department and demand that they disable the ultrasonic motion detector. I can't attack it with a screwdriver or redesign it so it can be turned off.

I'm not sure what has prompted this heated-sounding response.

I certainly do know what they are. I know they are used to control lighting, HVAC, and security systems. I know that there is not a simple switch where a tenant can turn them off. I didn't suggest that, nor did I suggest your "attacking" or redesigning them. I did suggest having a building electrician temporarily disable them.

I used to work in a large conference center with a lot of kinky electronics. If someone was coming in to shoot video, and asked for our cooperation, we gave it to them. If the air handler was too noisy, we'd temporarily turn it off. If they got flicker from fluorescents, we'd turn them off. If they needed special power or a feed of some sort, we'd take care of their request.

Certainly if you're just renting a hotel room for the normal day rate, it might be unreasonable to expect help from the building engineers. But if there's a room where you frequently shoot video, and the ultrasonics are causing problems, I would hope (at least from my past experience) that a reasonable request might get results from the house staff. There is no "tenant operable" switch, of course, but it would be relatively trivial for a building engineer/electrician to temporarily disable/bypass the system.

Originally Posted by Ken Reeser

While I am still looking for A), a lavaliere microphone of good quality that will not be affected by the signal,

The key is the mic's inherent frequency response. Dynamic mics have much higher mass, given the moving coil. That will limit the high frequency response. It's a real challenge to get a dynamic with good, flat HF response. A dynamic mic will be bigger and heavier than a newer electret mic.

The electret has extremely low mass (only a thin metallized diaphragm) so the response is going to extend out a lot further, and roll off a lot more gradually, at the high frequency end. Now if the manufacturer anticipated problems with ultrasonics, they might have designed the mic's integral electronics to have a steep rolloff at the HF end of the spectrum. But if they did not intentionally include sugh a filter in the mic's electronics, then an electret will likely be more problematic in your situation.

The filters you found are not quite ideal, from a theoretical perspective (see comments in my previous posts) but let's hope that they solve the problem for you! I will be interested to hear your findings, especially if you perform a semi-controlled test as I described earlier.