DIY sphere stereo mic

[k brown]

Without the dough for a Schoeps KFM 6 (~$8,000), KFM 360, or even a T.H.E. BS-3D (~$2,500) - all three now discontinued anyway - I decided to make a budget version based on Hudson Fair's ('Plush' on Gearspace) opinion that the KFM 6 sphere would have been more useful if it were the smaller 7" size of the KFM 360 (see his post below).

Found a dense solid-foam 7" ball on Amazon ($15) that is way heavier and 'deader' than the cheap styro wig heads many use for 'binaural' mic projects. Reviews on Amazon say that any kid that gets whacked on the nose with one of these 'play balls' is gonna have a nosebleed! This is no Nerfball!

Mounted a pair of Audio Technica ES945 boundary mics (cheap on eBay) in it, that have a frequency response very close to the Schoeps MK 2XS used in the KFM 6. Removed the circuit board and XLR insert from the 945s, hardwiring the cables to the FET of the capsules. Can use with any of several AT power modules designed for the 2SK660 FET (most of them are). The foam is firm enough to hold the threaded mic bodies in place by friction alone.

The ball was only available in red, so covered with gunmetal grey Plasti Dip (still needs a second coat).

At about 1lb.,the finished mic is very lightweight compared to any of the commercial models (KFM 6 - 2 1/4 lb., Neumann KU 100 - 3 1/4 lb., BS-3D - 3lb.) - as to whether the ball is acoustically inert enough not to color the sound; only tests will tell - certainly far superior to styro!!

One hitch is that the ball doen't show up on Champion Sports' website, so the ones on Amazon may be limited stock clearance.

By the way - there was a minor error in 'Plush's' post: The recording angle of a 7" sphere is 120 degrees (same as 60cm spaced omnis) , not 110 degrees (the 8" KFM 6 angle was only 90 degrees).

 

[k brown]

Despite the size of the grilles, the ES945s actually contains these little beauties, which many have compared favorably with DPA 4060:

 

[k brown]

A useful addition:

Going off the notion that the KFM 360 is essentially a baffled pair of MS mics with 'half-omnis' as their Mids (which when matrixed, creates a pair of cards facing forward, coincident with another pair that face the rear [other patterns can, of course be produced, but I don't know how useful they'd be]), I tried strapping a pair of EM21 cards below the omni capsules; this is the equivelent of just the front half of the KFM 360s four outputs. By cross-fading between the omnis and the cards, one can get any pattern from omni, through subcard to cardioid, to suit the room or the situation (which of course, if the outputs are recorded to four seperate tracks, can be done in post).

Indeed, across the front hemisphere, the cards image nearly identically to the built-in omnis, but around the back, of course, the level drops way off - thus producing a directional sphere mic.

For test and photo purposes, the cards are held in place by a sort of hair-net thingy that came with one of the mannikin heads I got for binaural experiments. Still have to rig up a way of cleanly and securely attaching them in a way that's removable when not needed.

Though the Schoeps photo seems to show the 'external' mics as pointing up, they are actually side-address figure 8s, so the diaphragms are oriented the same way as the cards are on my mic.

 

[ccaudle]

Audio Technica ES945 boundary mics

While searching for some of those on eBay I found that the Shure MX390 looks almost exactly the same, same threaded housing, looks like same grill.
Searching for MX390 at the Shure website I could not find anything, but MX395 looks like the current equivalent.
This is the eBay picture. Looks the same as the AT to me. Anyone run across them before and know if the capsule quality is equivalent?

 

[k brown]

I have a couple of the MX-series lavs; I'd say they are equivilent quality overall, but the Shures are even brighter, and I never had good luck EQ-ing them to be useable.

 

[k brown]

Got the sphere in front of an all-volunteer/community orchestra (and all that that implies!) the other night.

https://www.idrive.com/idrive/sh/sh?k=x2z9t8k4b7

 

[ccaudle]

Got the sphere in front of an all-volunteer/community orchestra (and all that that implies!) the other night.

https://www.idrive.com/idrive/sh/sh?k=x2z9t8k4b7

I am not able to download that file. Is it working for anyone else? Firefox 104.0.1

 

[k brown]

Is your browser preventing pop-ups?

 

[ccaudle]

Is your browser preventing pop-ups?

I have it set to block unrequested pop-ups, but pop-ups from clicked links get opened in a new tab.
However, the link is working this morning. Yesterday it was just displaying an XML file with complaints about some error in the XML.

 

[ccaudle]

Got the sphere in front of an all-volunteer/community orchestra (and all that that implies!) the other night.

https://www.idrive.com/idrive/sh/sh?k=x2z9t8k4b7

I like it. Reminds me of the recordings I used to make with baffled omnis. Makes sense that sphere and Jecklin disk style would be pretty similar.

 

[k brown]

The big challenge with the Jecklin is using a material for the disc that blocks as much sound as possible, while simultaneously reflecting as little as possible. Though I've never tested it, it's hard to believe that the Haun disc made of thin metal, covered with thin foam, are really optimal. I got the best results using 1/4" hardboard covered with real lamb's wool, not artificial. Doesn't look so great, though.

The sphere avoids these issues.

Next iteration will use different mics - these little ATs are just too bright and brittle; require too much EQ in post, even used in the diffuse field.

 

[ccaudle]

I got the best results using 1/4" hardboard covered with real lamb's wool

I used thin hardboard from a clipboard covered with about 2in/5cm thick open cell foam (was grey anti-static foam used for packing some kind of electronics in a past life). Not very wooly, but because of the thickness cut a lot of reflection.

these little ATs are just too bright and brittle

Interesting, with the Shure earphones I was using to listen I did not find the recording overly bright sounding at all. Of course you did mention EQ in post, so I guess you did a good job with that. I know the Schoeps were equalized for diffuse field, since they were pointing more than 90 degrees away from the front sound sources, so you want a bit of treble lift on-axis.

Next iteration will use different mics

Any likely candidates? Does Primo have easily avaible 1/2" omni capsules?

 

[k brown]

For now, I'm kludgeing my Oktava MC012s onto each side, the way the fig8s are on the KFM360.

For an imbedded build, I'd recommend the Primo EM23 (no built-in FET, though).
https://micbooster.com/microphone-capsules/56-primo-em23-microphone-capsule.html

 

[MicUlli]

Yes, Primo offers the very famous EM273. EQ can be done acoustically by using a 3mm pipe...

 

[MicUlli]

I do not recommend the EM23 for this purpose because of the very steep rolloff starting at 6 kHz under 90°

 

[k brown]

I do not recommend the EM23 for this purpose because of the very steep rolloff starting at 6 kHz under 90°

Good point! EQ-able, though.

That falloff is inherent in capsules of that size - same size as used in the Neumann and Schoeps sphere mics.

 

[ccaudle]

the ES945s actually contains these little beauties

What size is that capsule? About 10mm diameter? Or more like 6.35mm/0.25in?
If it is under 8mm I would assume that has better high frequency response, but also higher noise floor.
From the specs I found it appears those have around a 30dB SPL equivalent noise floor. That seem about right?

 

[k brown]

ES945 boundary mic and ES943 lavalier both use the same 8.4mm capsule. For some reason the response graphs are more different than what would expect just from the presence of the grille on the 945.

Noise is surprisingly low.

 

[k brown]

Windscreen update:

Inspired by this Youtube vid:
I ordered the muffs ($15 !) and added a layer of Scotchbrite inside. Holding the mic right up against a 24 inch room fan on it's highest speed: ZERO HF or LF wind noise! . . . Zero.

https://www.amazon.com/dp/B00GKFATHI?psc=1&ref=ppx_yo2ov_dt_b_product_details
I actually don't intend to use the sphere much for outdoor recording, but wanted to pass this along for anyone else who might find it useful. I seriously doubt that the foam 'phones' ($470 !!!) Neumann sells for the KU100 dummy head could possibly work any better (if even as well!).

Or for that matter, the foam ring Schoeps sold for the KFM 6 (still can't believe that the only wind protection Schoeps ever sold for that $8,000 mic was a silly band of foam).

Also updated the mic capsules. Managed to kludge a pair of ATM10a capsules into the same housings the 945s came in - much smoother, richer sound. The current model, AT8010, has even flatter response. Not concerned about the larger diameter of the capsules; the Neumann and Schoeps sphere mics had even larger capsules.

 

[k brown]

Went outdoors to test in actual wind, and what do you know, I ran into a jazz street band I'd recorded over 15 years ago, playing in the exact same spot!

Sax, flute, elec guitar and elec upright bass. There's no PA; only the bass and guitar are amplified. Here's a radio station podcast about the spot:

https://www.kalw.org/show/crosscurrents/2020-03-12/groove-on-down-to-tunnel-jazz
They setup in front of a large pedestrian underpass, and on my honor, I swear the only reverb you hear is coming from the tunnel behind them. Only processing was some spatial equalization to open up the bottom, and hi and lo EQ to compensate for the earmuffs, and a considerable peak in the bass amp - but NO reverb.

https://www.idrive.com/idrive/sh/sh?k=j6i4x7y0w9
(to download, you may have to enable popups)