Violin microphone on the cheap


or

Nobody in their right mind would ever use DIN for a mic

A dissertation on orchestral amplification

Background

Kårspexet is an amateur farce group at KTH (The Royal Institute of Technology) in Stockholm. As such, we have a 20 piece orchestra to back up six actors with some semblence of music. This orchestra sounds great unplugged in an orchestra pit. Problem is, not all theaters in Stockholm have a pit that can accomodate an orchestra that large, drums and all. This is usually solved by placing the orchestra somewhere else and amplifying. At the theatre where the following was attempted, Göta Lejon, the orchestra was placed on the balcony.

Caveats

There are pros and cons to both acoustic and amplified. The orchestra pit was a good idea before the advent of the Senneheiser MKE-2 lavalier and wireless bodypacks. A lot of the orchestra can leak into the actor's microphones, especially cymbals and hihats and other percussion oddities. This is not a good thing. On the other hand, by tolerating some leakage and some clever plexiglass placement, keeping track of 8 channels is a lot easier than keeping track of 30. The orchestra sounds more natural, and the singing sounds great if the actors stay the hell away from the orchestra pit (also known as "closest to the audience, farthest from the scenery" and where all the dance numbers take place).

Micing the orchestra and placing it somewhere else/covering it up results in cleaner input in the actor mikes. This doesn't stop the problem with the drums and guitar amplifiers leaking in the orchestra mics.Guitar amplifiers can be turned down, but having a drummer not hit the cymbals and not bang the living daylights out of the toms is a rare sight.

The leakage is worst on the strings and light winds, as they can't produce the same quantity of moving air that a trombone can. You need a good condenser or an ElectroVoice ND-308 in front of each, and condensers are great at picking up just about everything. A lot of the music alternates between quiet and loud. This is called being dynamic and is unheard of in most rock'n'roll. Our music writers tend to write stuff that is dynamic. Plus instruments move around a bit, so you get varying levels when the orchestra won't cooperate (hint: they never do. Especially when on tour in Finland).

The solution

The obvious solution is to jam an MKE-2 or a DPA-3 into each violin (you can get away with a close-miced SM57 or Schoeps condenser on most of the winds). Given a good bugdet, this is what people do, but Kårspexet doesn't have that bugdet. Drat.

But wait! You can buy cheap electret microphone elements from your friendly electronics supplier, you say. All you have to do is jam them down the F-hole of a violin and you're set! Except for a few things:

Send in the clowns

Some people in Linköping did the thinking for us, but we did a lot of this our own way to suit ourselves. First things first, we need to mount the thing. Easiest is to hook something to the side.

So now what?

Here's where we started thinking. Our original inspiration hade a jelly jar next to each violin with a 9V battery and some drive electronics and ran it unbalanced into the recording console. There's definitely room for improvement, we said.

All mixers have phantom powering at 48 volts. All that count that is. This supply is great for powering condenser mikes, or active DI:s. Or electret mikes. Wait, we have an electret mic here. Hmm. *scratch head* *ponder* *google*. Hey! this circuit does what we want!.Ok, let's build four of them in a zinc shielded box with XLR:s on one side and 3-pin DIN on the other (Nobody in their right mind uses DIN anymore for microphones).

Channel one, component side.


Solder side of 4 channels


Four channels, mounted in the box.

Finished product

It came out to be two boxes: four channels for the violins, two for the violas (the cello has an AKG instrument mic in the bridge).

Mounted violin
(and our workbench).

Finished box
for the violins. P48; mmm... this runs on 48 volts only. Yummy.

Conclusion

Everything worked on the first try. The circuit works (even after my soldering this in on veroboard) and so does the microphone. The isolation is great; with Pink Floyd playing at 100 dB in our workshop we still only heard the violin. Worked great in the field as well, much like having a portable isobooth. It also sounds much better than a contact mic or having a mic in the bridge. Some EQ:ing required, but it sounds a whole lot better than the alternatives.

Credits

Martin Persson and Martin Björnsson in Linköping for the original idea, inspiration and drawings. ELFA for the stuff. Radionämnden for the workshop space, tools, zinc boxes and music (can't work without music). Tomi Engdahl and Epanorama.net for the circuit design. Most of the physical stuff was done by Måns Nilsson. Circuit building was done by Ian Delahorne (me).

Måns at work on the box

New version

We had a bunch of ideas (well, at least two) for the next version. We were out of Eurocard-sized veroboard, and veroboard really sucks anyway. Plus the element was too big.

I spent some time getting to know gEDA and PCB, and ended up with a reasonable PCB for the project. Drilling PCB's are so much fun, so next version will probably be SMD. Also, the new element is a lot smaller. I probably should add new pictures, and recording samples like I did to the theatre-sound mailing list. Oh, and the PCB worked on the first try as well


Ian Delahorne, <ian (kanelbulle) stacken.kth.se>
$Id: index.html,v 1.4 2005/08/13 23:34:34 ian Exp $