Hello,

Been lurking this group for a while and finally got around to actually building something! About 2 years ago I had some extra MDF lying around so I decided it would be fun to make my very own Binaural Head. I based the shape off of the Neumann KU-100 and used a projector to draft some stencils approximating the size of my own head (it ended up being a tad over-sized). After rough cutting/routing the 9 layers I needed to make up the width of the head I ended up with this monstrosity



After some diligent work with an angle grinder, sanding flap disk, and an old can of urethane clear coat the head was looking pretty good.



I then entered the final semester of my EE degree and became very busy with my senior design project so I put the head project on hold. Months turned into years and the head got kicked around my house as my attention turned to more pressing matters. All the while I could feel the hollow, blank stare of the binaural head-to-be piercing into my core.

Well, free time finally made itself available to me once again and I decided to turn this 10 lb mannequin head into a proper microphone. The original plan was to base the impedance converter on a Calrec schematic given to me by a friend. After a few failed attempts I decided instead to modify the popular OPA design that originates from this forum. Since I had already purchased the transformers I opted for a paralleled amplifier configuration. Douglas Self covers this technique in Small Signal Audio Design and details its improved load driving capabilities and reduced self-noise compared to a single opamp. Now, I am not sure whether the noise performance is actually improved when the source impedance is so high. Someone here may have a better idea but I would think that this configuration would reduce the input Z of the amplifier and cancel out any reduction in noise you typically get from a parallel amp combo (corrections welcome).



The capsules I am using are the Omni caps that came with my pair of Lauten LA-120's (good mics btw). They happened to be about the same diameter as a 7/8" Ryobi forstner bit I had so I simply marked their location and drilled holes on both sides of the head. I carefully disassembled the caps and soldered the shield of some mic cable I had to the threaded brass ring that holds in the back plate. After re-assembling the caps I soldered the twisted pair to the back plate connection. I cut this lead to be as short as possible and pressure fit the capsules into the head.



After laying out the circuit in Kicad I etched and drilled a pair of boards. I used some metal 1-gang outlet boxes from HD and some scrap metal from an old receiver as shielding. I drilled a pair of holes in the back of the neck (using the same 7/8" bit) and wired up some Neutrik XLR bulkheads I had laying around.



Overall I'm pretty satisfied with the result, I included a short recording of myself on drums as an initial test. I plan on dusting off my 3D printer and printing out a pair of human ears to help achieve that HRTF stereo image goodness. Update to follow.

On Nov 15, 2023, at 13:52, Chris (SoundLock) <chris@...> wrote:

?

Thiersch has been re-diaphragming Neumann capsules for decades and I have never had to reject a capsule. It's far from art, it's pure engineering.

Anyway none of this has anything to do with what I asked!

Which is about a cheep temporary k87 with isolated backplates.

On 15 Nov 2023 15:40, Jules Ryckebusch <ryckebusch@...> wrote:

I agree with Matt on this one. ?Rebuilding capsules is an art more than science. ?I know of people that do this. ?It is on par (although easier) with rebuilding a car engine. ?The current capsules made including Mic Parts are done on precision CNC machinery in Clean

Rooms with ISO 9001 quality systems. Then precision tensioned as part of the manufacturing process . ?A lot of people haven’t gotten their head around that yet. ?

Best Regards,

Jules Ryckebusch?

On Nov 15, 2023, at 09:22, Homero Leal <homero.leal@...> wrote:

?

Matt's capsules are great. I have used the RK87 and the RK12, and they sound beautiful.

Another option may be from Arianne Audio:?. 47 Flat is great, but I haven't personally tried the 87 yet.

And, if you just want a cheap option, buy a single side U87 style capsule from aliexpress or ebay, use a 56p C0G capacitor for the back side, and use the mic only as cardioid while the reskinned?capsules arrive.

Regards!

HL

El mié, 15 nov 2023 a la(s) 08:47, Chris (SoundLock) (chris@...) escribió:

Hi Matt, having a non original capsule massively devalues the microphone hence the reskin, which doesn't.

I've been doing microphone repairs a long time and Thiersch return capsules as good as the day they were made.

Yeah specially looking for K87 with isolated backplates not K67 (U67) or K870 (U87A).

Cheers

Chris?

On 15 Nov 2023 14:32, recordinghacks <matt@...> wrote:

> Considering they are just to keep the mics going they don't want?
> to spend a fortune which puts the micparts RK87 out of budget.?

We sell RK-87s as an *upgrade* option to many U87i / U87 Ai owners (and TLM103 owners too).?

The cost of an RK-87 is likely to be significantly lower than the cost of refurbishing the Neumann capsule. Seems like testing one might save your client a fair amount of money.

Assuming the refurbishing process would be done by a 3rd party (because, so far as I know, that’s not a service Neumann has ever offered) the refurbishing process won’t deliver an original Neumann capsule anyway, by which I specifically mean those refurb capsules would almost certainly sound different than the originals. At that point, testing alternatives seems like a good idea. YMMV.?

Neumann does not sell bare capsules any more, at least not in the US, although I’m sure they offer repair services for their microphones.


> JLI has some 47's. Not sure if that'll work for you. I didn't see the 87's,?
> but I just did a quick search.
>

K47 capsules would work in the sense of being electrically compatible, with the exception that a U87 i (without DC converter) would only be able to create a Cardioid pattern with a single-backplate K47. But more to the point, the U87 never had a K47 capsule, and a K47 capsule would not sound right and probably would sound quite bad given the bandpass filtering in the U87 circuit.

Also worth noting is JLI’s misleading use of the “RK47” product name; they are not selling genuine RK-47 capsules, but are merely taking MicParts’ product name. Left as an exercise for the reader is to speculate whether that’s done in hopes of profiting from MicParts’ reputation, or out of ignorance regarding the names of historical European capsule designs.

—
matt.

On Nov 15, 2023, at 09:22, Homero Leal <homero.leal@...> wrote:

?

Matt's capsules are great. I have used the RK87 and the RK12, and they sound beautiful.

Another option may be from Arianne Audio:?. 47 Flat is great, but I haven't personally tried the 87 yet.

And, if you just want a cheap option, buy a single side U87 style capsule from aliexpress or ebay, use a 56p C0G capacitor for the back side, and use the mic only as cardioid while the reskinned?capsules arrive.

Regards!

HL

El mié, 15 nov 2023 a la(s) 08:47, Chris (SoundLock) (chris@...) escribió:

Hi Matt, having a non original capsule massively devalues the microphone hence the reskin, which doesn't.

I've been doing microphone repairs a long time and Thiersch return capsules as good as the day they were made.

Yeah specially looking for K87 with isolated backplates not K67 (U67) or K870 (U87A).

Cheers

Chris?

On 15 Nov 2023 14:32, recordinghacks <matt@...> wrote:

> Considering they are just to keep the mics going they don't want?
> to spend a fortune which puts the micparts RK87 out of budget.?

We sell RK-87s as an *upgrade* option to many U87i / U87 Ai owners (and TLM103 owners too).?

The cost of an RK-87 is likely to be significantly lower than the cost of refurbishing the Neumann capsule. Seems like testing one might save your client a fair amount of money.

Assuming the refurbishing process would be done by a 3rd party (because, so far as I know, that’s not a service Neumann has ever offered) the refurbishing process won’t deliver an original Neumann capsule anyway, by which I specifically mean those refurb capsules would almost certainly sound different than the originals. At that point, testing alternatives seems like a good idea. YMMV.?

Neumann does not sell bare capsules any more, at least not in the US, although I’m sure they offer repair services for their microphones.


> JLI has some 47's. Not sure if that'll work for you. I didn't see the 87's,?
> but I just did a quick search.
>

K47 capsules would work in the sense of being electrically compatible, with the exception that a U87 i (without DC converter) would only be able to create a Cardioid pattern with a single-backplate K47. But more to the point, the U87 never had a K47 capsule, and a K47 capsule would not sound right and probably would sound quite bad given the bandpass filtering in the U87 circuit.

Also worth noting is JLI’s misleading use of the “RK47” product name; they are not selling genuine RK-47 capsules, but are merely taking MicParts’ product name. Left as an exercise for the reader is to speculate whether that’s done in hopes of profiting from MicParts’ reputation, or out of ignorance regarding the names of historical European capsule designs.

—
matt.

On 15 Nov 2023, at 05:29, Joe Todd <jjthaden@...> wrote:

?Thanks, Marcel. Are you saying a 5-cluster's S:N will be 2.25 x higher than a single one? Or under my scenario a same-area large capsule?

Let's simplify the scenario: two small capsules vs. one larger one with twice the diaphragm area. As before, same per-area signal in response to a test tone. And now version A: same self noise per capsule, either size. Version B: Large capsule has 3 dB higher noise.

Higher S:N for the capsule pair?

On Tue, Nov 14, 2023 at 11:00 PM, Jerry Lee Marcel

<jerryleemarcel@...> wrote:

If all the 5 capsules have the same weighting in the array, S/N would increase by 2.25 (about 7dB). However, proper arrays have different weighting and phasing for each mic.

On Tue, Nov 14, 2023 at 7:16 PM, Jules Ryckebusch

<ryckebusch@...> wrote:

Hey Joe, the theory goes like this: the signal to each microphone is correlated. ?For two capsules the signal adds by 6dB. ?But the noise is not correlated so it adds by 3dB. You have to keep doubling the number of capsules for the math to continue to work. ? Adding more capsules adds some complexity but the theory shows an improved signal to noise ratio.?

Best Regards,

Jules Ryckebusch?

On Nov 14, 2023, at 18:09, Joe Todd?wrote:

?I've read that an advantage of a cluster of capsules is a decrease in noise (or is it just increased S:N?).? If lower, why? Can the array's noise be lower than that of a single capsule of that model?

On Tue, Nov 14, 2023 at 11:00 PM, Jerry Lee Marcel

<jerryleemarcel@...> wrote:

If all the 5 capsules have the same weighting in the array, S/N would increase by 2.25 (about 7dB). However, proper arrays have different weighting and phasing for each mic.

On Tue, Nov 14, 2023 at 7:16 PM, Jules Ryckebusch

<ryckebusch@...> wrote:

Hey Joe, the theory goes like this: the signal to each microphone is correlated. ?For two capsules the signal adds by 6dB. ?But the noise is not correlated so it adds by 3dB. You have to keep doubling the number of capsules for the math to continue to work. ? Adding more capsules adds some complexity but the theory shows an improved signal to noise ratio.?

Best Regards,

Jules Ryckebusch?

On Nov 14, 2023, at 18:09, Joe Todd?wrote:

?I've read that an advantage of a cluster of capsules is a decrease in noise (or is it just increased S:N?).? If lower, why? Can the array's noise be lower than that of a single capsule of that model?

Got it, thanks Jules!? I'm actually familiar with that concept from another field (mass spectrometry). So now this followup: If a cluster of 5 small capsules around a center one has the same total useable diaphragm area as one large capsule, if a large and small capsule have the same intrinsic noise level, and if when exposed to a test tone the large one generates the same voltage ~per unit area~ as a small one, then would the small-capsule cluster still have markedly better signal to noise ratio?

On Tue, Nov 14, 2023 at 7:16 PM, Jules Ryckebusch

<ryckebusch@...> wrote:

Hey Joe, the theory goes like this: the signal to each microphone is correlated. ?For two capsules the signal adds by 6dB. ?But the noise is not correlated so it adds by 3dB. You have to keep doubling the number of capsules for the math to continue to work. ? Adding more capsules adds some complexity but the theory shows an improved signal to noise ratio.?

Best Regards,

Jules Ryckebusch?

On Nov 14, 2023, at 18:09, Joe Todd <jjthaden@...> wrote:

?I've read that an advantage of a cluster of capsules is a decrease in noise (or is it just increased S:N?).? If lower, why? Can the array's noise be lower than that of a single capsule of that model?

Thanks in advance

John Thaden

On Tue, Nov 14, 2023 at 7:16 PM, Jules Ryckebusch

<ryckebusch@...> wrote:

Hey Joe, the theory goes like this: the signal to each microphone is correlated. ?For two capsules the signal adds by 6dB. ?But the noise is not correlated so it adds by 3dB. You have to keep doubling the number of capsules for the math to continue to work. ? Adding more capsules adds some complexity but the theory shows an improved signal to noise ratio.?

Best Regards,

Jules Ryckebusch?

On Nov 14, 2023, at 18:09, Joe Todd <jjthaden@...> wrote:

?I've read that an advantage of a cluster of capsules is a decrease in noise (or is it just increased S:N?).? If lower, why? Can the array's noise be lower than that of a single capsule of that model?

Thanks in advance

John Thaden

I think that is wrong....Mic Scharf? ...27k goes from pin 2 to ground pin 1....pin 3 goes to plus on capsule...pin 2 goes to minus on capsule...Thats how all mine our wired and work good....