Good, budget lav for camcorder

[Rick Reineke]

A ground loop and EMI (electro-magnetic interference) are different. (cause and effect). A 'ground loop' typically has a primary 60Hz hum (in the US) with some lower volume harmonics. EMI is usually a broadband buzzing type sound with less content @ 60Hz. A ground loop's 60Hz hum is relatively easy to attenuate via notch filters and/or NR. Not so easy with EMI, which is usually caused by unbalanced audio runs and/or poorly shielded cables in close proximity to power cables and/or transformers. Audio 101: If audio cables have tobe crossed, it should be at right angles.

[Jon Fairhurst]

As a kid, guitar amps weren't all properly grounded, not having 3-prong plugs. It was really common to have ground loops with LF hum. "Here, hold my guitar..." Zap!

In my wife's case, it was EMI with HF buzzing over 60 Hz. Such is life with a long, unbalanced cable. I'll mention the "right angle" thing to her, but I'll recommend that the "keep 'em separated" thing be the main approach.

[Richard Crowley]

Some camcorders are notorious for leaking nasty RFI out of the LCD viewfinder display. IME, the solution was to very carefully experiment and route the audio cable AWAY from the LCD viewfinder screen. or in extreme cases to close the LCD screen and use an external monitor.

And some camcorder "line-lump" mains power bricks are notorious for creating problems exactly like that. In many cases the problem could only be solved by removing the mains power supply completely from the equation and operating on the camera battery.

Of course, this is because audio is always a secondary "feature" that is tacked-on to the design just before it goes to manufacturing. Audio has always been the poor step-child of video production. Which is why it is so common for people to use double-system sound.

[Rick Reineke]

"As a kid, guitar amps weren't all properly grounded, not having 3-prong plugs. It was really common to have ground loops with LF hum. "Here, hold my guitar..." Zap!"
- That and single-coil guitar pick-ups (common to Stratocasters) which were extremely prone to EMI.
When a buddy of mine moved his music recording studio to a new location (and built a new building), he had the entire building wired with a Equi=Tech balanced power system. It was very expensive but, it's a 'world-class' facility and eliminated most of the single-coil pickup EMI issues as well as the push noise floor further down. Prior to that, Strat players were instructed to, "always hold the guitar neck in this direction and don't move"

[Greg Miller]

Jon:

There are lots of lav clips available on eBay. I like the type that hold the mic in a formed circle made of spring metal ... no plastic. They come in specific sizes so measure your mic with a caliper or micrometer before you start looking.

I have had better luck getting rid of buzz with a two-part procedure.

First I look at the spectrum and identify the strongest harmonics (usually the lowest frequencies). I carefully remove (or at least lower) those with notch filters. The first few, at least, are below any important vocal content, so there's no adverse effect on the desired audio.

After that, I take a noise sample, and use adaptive filtering in the usual way. (As with any noise, this sometimes works best with two light passes, rather than one heavy pass.)

I find that if the adaptive filter doesn't have to "work hard" on the strongest lowest frequencies, it can have a much better result -- with less artifacts -- on the higher frequencies.

If that isn't good enough, there's an even more complex approach, but with a few more steps.

[Jon Fairhurst]

Thanks Greg,

That's quite helpful. Years ago, my son and his freinds shot a hilarious scene with great timing - and with an unbalanced cable near an AC cord. I rescued it but the sound wasn't great. As the composer, it meant wall-to-wall music to cover up the noise. With notches at 60, 120, 180, etc before doing the NR passes, I wonder if the results would have been good enough to not need music as a distraction.

Not long afterwards, he bought an M-Audio Microtrack II, which included phantom power and balanced input. It was great in theory and stunk in practice! The preamps were so poor that the hiss was nearly as bad as the AC buzz! (The amateur boom operator not being aggressive enough with getting the boom close didn't help.) I spent hours with NR and we still had to be really aggressive with sound design and music to get a good result. Spend too little money and you'll spend a lot of time!

Fortunately, my wife's project is more of a vlog. With a close lav, AC buzz, and simple aggressive NR, it's not pro, but it's better than most of the distant/cheap mic audio of its peers. With good cable path management, she'll be golden. :)

[Greg Miller]

If you need a long cable run, you could always try an inline transformer like this one located near the mic power module. Hosa XLR Male Lo Z to 1 4" TS Female Hi Z Microphone Input Impedance Transformer | eBay

In fact you could dramatically shorten the cable between the power module and the unbalanced male plug.

Of course there's a tradeoff. The transformer impedance is 50k ohms :: 200 ohms. (It's really designed for a very high impedance mic or perhaps a guitar pickup, not for a low-z unbalanced mic like your lav.) The voltage on the balanced side will be 1/15.8 the voltage on the unbalanced side. That's a voltage drop of 24dB. So while this may eliminate some noise pickup, you may pay for it with low levels at the recorder input, and a resulting higher noise floor from the recorder's mic preamp.

If I happen to spot an unbal :: balanced transformer with a better primary impedance, I'll let you know.

[Battle Vaughan]

Updated my email and managed to lock myself out of dvinfo.net for a couple of days. Jon, repeating my offer, I have a spare clip I can send you, email me if you wish....

[Jon Fairhurst]

Thanks for the offer, Battle. No need. We rigged up a simple solution that will do the job. :)

On the transformer thing, I guess we could use two - one near the mic and one near the camcorder to go unbalanced -> balanced -> unbalanced, but that violates the simplicity goal. Hopefully, good routing will do the trick. If not, NR to the rescue!

[Greg Miller]

No Jon, a second transformer won't work.

It would work if the unbalanced side had an impedance that was close to the impedance of the mic and the recorder input.

However, in this case the unbalanced side is 50,000 ohms. If you do the two-transformer configuration you suggest, and if the unbalanced "output" end is not connected to anything, then you would have about the same voltage there that you had at the mic. But your recorder input is probably a lot less than 50,000 ohms. That will load down the 50,000 ohm transformer output, which will result in a lower voltage, and also will have some adverse effect on the frequency response.

And, indeed, it violates the "simplicity principle."

*** Actually, looking back, I see that for some reason I assumed your recorder had a balanced input. I see now that it's a camera with an unbalanced input. So my original suggestion wouldn't really solve the problem. The configuration of the system is determined by the "lowest common denominator" which in this case is the camera's unbalanced input. So even if you used that transformer, the wiring from the transformer to the camera would still be unbalanced. :-(

And indeed, as I said above, the two-transformer scenario would work except that the impedance is drastically wrong. Transformers are wonderful devices and have a lot of uses, but unfortunately one cannot ignore impedances. The way to make the long cable run balanced would require a correct transformer at each end, and that would start getting too complicated for your wife's situation.

[Rick Reineke]

"The way to make the long cable run balanced would require a correct transformer at each end"
Greg is correct on this. No cheap/easy way around it, unless someone has the knowledge to assemble/solder small electronics components..

[Jon Fairhurst]

Of course, the nicer solution is to have my wife use one of my juicedLink preamps with a balanced run to a pro mic, leaving a short run with a hot signal into the camera. But she wanted ownership of a simpler setup at a good price. The AT mic seems to be the right solution.

This morning, she said that she wants to do another video. Fingers crossed that a good cable path does the trick!

There's the possibility that there is ceiling lamp wiring directly below her office. That could complicate her setup for sure. I'll keep my NR software at the ready!

[Battle Vaughan]

May I suggest she monitor with headphones to check the hum or buzz or whatever it was? If she is running the camera off a wall-wart and gets the noise, suggest she try running off batteries and see if it goes away. As I think I mentioned, I had a similar experience with a camera set up that way and the noise was from, I think, a ground loop with the mike and power supply....

[Greg Miller]

Monitoring, of course, is always a good suggestion ... in fact we might say it's a necessity. After all, you wouldn't shoot without using the viewfinder.

Be that as it may, my understanding of this setup is that the mic is a lav with neither the capsule nor the battery supply grounded; the camera is on a tripod which presumably is not grounded; and the camera itself may or may not be AC line powered. Since the only possible ground is the camera's power supply, I don't see how there could be any ground loop, since that requires having the circuit connected to ground at two or more points. (If you also had a mixer, or a video monitor, which was line powered, then there would certainly be the potential for a ground loop.)

OTOH, it's possible that the power supply itself might be producing some noise (strictly speaking, not a ground loop) into the system. If it's not an OEM supply, but rather was designed for some other gear (e.g. a battery charger, game console, answering machine, etc.) then it might not be quiet enough to run a piece of high-gain audio equipment. (To quote Mr. Vaughan himself, "Don't ask me how I know this.")

But, again, I agree 100% that monitoring is a must!

[Richard Crowley]

Originally Posted by Greg Miller

Monitoring, of course, is always a good suggestion ... in fact we might say it's a necessity. After all, you wouldn't shoot without using the viewfinder.

But, again, I agree 100% that monitoring is a must!

My signature line over on another forum is...

"Recording audio without metering and monitoring is exactly like framing and focusing without looking at the viewfinder."

I continue to be flabbergasted at people who come to these forums complaining about things they should have caught before hitting the Red Button by simply monitoring properly.

"I forgot to remove the lens cap when I shot all the scenes last week. Is there any way of recovering that footage?"