R. Williamson capsule

[strangeandbouncy]

Marik,


In 1995 I built a capsule from engineering drawings from a 1960's magazine. i don;t have the relevant info to hand, but perhaps it is the same one. Sean Davis was kind enough to diaphram it for me, and i can assure you that it is EXACTLY what lab mic sharks might wanna build in 2005!!!!!! I mounted it in a defunct KN86 body, and used it for vocals, room ambience, floor tom, Sax, BV's, mid in m/s set-up, in fact, i used that mofo everyday, til a client decided that THEY wanted it more than me. My intention was to mount it in a defunct U87 body with M269-type electronics, but it was so good I simply couldn;t afford to take it out of commission! B@@tard F@@ker Thief! Ho HUm. If i can find the drawings, I;ll see who wrote the article.

BTW the venerable Neumann M7 capsule (my favorite in any guise, be it M49/ U47/ Sela(check these swedish m49ish mics!)/M269/ or modern Microtech Gefell) that is still available in Gefell mics and that has not changed since 1932 . . . . I believe that it was a copy of the western electric capsule from 1928, so - 77 years old . . . so where does age come into capsule design? when a capsule is right it's right . . . . .


Anything sounds good if you built it yourself, but this puppy REALLY kicked BUTT.

Andy P

 

[Marik]

[quote author="strangeandbouncy"]Marik,


In 1995 I built a capsule from engineering drawings from a 1960's magazine. i don;t have the relevant info to hand, but perhaps it is the same one.[/quote]

Andy,

Don't take me wrong. I did not mean dissing. I am sure, if implemented right, this might be a kick butt capsule. I was probably speaking my personal "non-desire" of doing something what somebody has already done. I find it much more interesting and rewarding to make my own mistakes than to repeat somebody elses. :grin:

What I meant is that some smart folks here could do a design, which might be equally good, or even exceed the Williamson's one. Anyway, I beleive, Mr. Williamson would also be interested to read some wealth of the information we have here.

Basically, my problem with this capsule is that I think that two through-holes only would not make an even distribution of back pressure throughout whole surface of the diaphragm, and cardioid pattern might be compromised. Feel free to correct me.

 

[strangeandbouncy]

Marik.

Fair play, old Bean! Still can;t find my original artical. If i remember, there were more than 2 through holes. . . .

All the best,


Andy p

 

[dale116dot7]

Through holes / Blind holes

Debenham - 13 / 46
Williamson - 2 / 11
Shulien - 2 / 11 - stepped backplate
M7 - 30 / 61

The pressure does not have to evenly cancel out all over the diaphragm - the distribution does not have to be good. What does have to happen, however, is the sum of pressure differences across the diaphragm does. In other words, the diaphragm can be allowed to flap around as sound waves make it from the vent hole and around, so long as the net change in capacitance (roughly proportional to deflection) is zero for sounds from the back. I think Stephen Paul called this 'monkey motion'.

In my reading of the Shulien article, it seems as though his capsule is somewhat underdamped as he needed an EQ network to tame it - and with the step in the backplate that makes sense to me.

Debenham seems to have put the most work into a capsule design, has anyone here built this capsule? It's on my list of things to do.

 

[Marik]

[quote author="dale116dot7"]
The pressure does not have to evenly cancel out all over the diaphragm - the distribution does not have to be good. What does have to happen, however, is the sum of pressure differences across the diaphragm does. In other words, the diaphragm can be allowed to flap around as sound waves make it from the vent hole and around, so long as the net change in capacitance (roughly proportional to deflection) is zero for sounds from the back. I think Stephen Paul called this 'monkey motion'.
[/quote]

Thank you Dale, it is indeed very useful information! I will try to put together some backplates, soon.

Andy,

Make sure to post the article if you find it.

 

[dale116dot7]

Sony C37, 4 through, 16 blind. The backplate assembly is a bit challenging because the diaphragm tension can be adjusted in place. There are the following pieces from what I could see:

1. Capsule housing
2. metal washer between housing and diaphragm
3. fibre washer between diaphragm and clamping ring
4. diaphragm, glued to fibre washer.
5. diaphragm clamping ring (threaded, mates with housing)
6. metal backplate
7. plastic backplate holder, threaded into diaphragm clamping ring
8. plastic diaphragm spacer in centre of metal backplate
9. backplate locking ring, threads into capsule housing
10. shutter on a screw
11. holder for shutter and screw

One nice thing with this assembly is that each piece is not particularly difficult to machine or make, there are just a lot of them.

 

[adrianh]

You people are killing me here.
You make your own back plates you break drill bits.
You get to touch a Tele U47 and commit how simple it is.
OK folks we need capsule drawings!
I know about the 3 capsule articles.
Lets trade drawings of stuff I have not seen.

 

[dale116dot7]

Ok, I'll bite
http://www.10000cows.com/M7CapsuleDirections.pdf

 

[chrissugar]

Thank you Dale, that is very usefull info.

chrissugar

 

[Gus]

Dale thanks for the PDF.

A question, I broke a .6mm drill in the backplate. Someone posted here or at the old place a chemical that would disolve the steel drill and not the brass. Does anyone remember what is was?

 

[dale116dot7]

Aluminum sulphate dissolved in water. I haven't tried it though, I usually use carbide bits and if they break (which is very seldom), the bit mangles the backplate pretty bad so I start over.

 

[Gus]

Do you use a mill or drill press? I was using a drill press?

 

[scott_humphrey]

[quote author="dale116dot7"]Ok, I'll bite
http://www.10000cows.com/M7CapsuleDirections.pdf[/quote]

Thanks, Dale!

Here's a picture of the L-a-w-s-o-n M7 capsule:

 

[bill raymond]

I'm a bit hesitant to add my comments here and look really foolish; I don't have the audio/engineering background that so many of you all do--I'm just someone who likes to jump in over my head and make things, even if I don't really know what I'm doing: I've been working on a Williamson capsule with the Schulein modification plus a modification I came up with all on my own which I will offer up for criticism: I hoped to overcome the problems Schulein had with his stepped backplate by gluing a piece of acrylic to the stepped portion of the backplate and facing the whole thing on the lathe so that I would have a uniform gap of 2 mils between the diaphragm and backplate while preserving Schulein's attempt to provide greater sensitivity. In light of what has been said above I'm beginning to doubt that it will be the improvement I had hoped for. However, I'm going to try to complete the capsule anyway in the hope of learning about these things, and perhaps give a try at a double diaphragm capsule as well.

 

[dale116dot7]

I'm spoiled - I have access to a pretty good mill (and also a lathe) with digital readouts - it makes it actually really easy to make capsules. It takes me about four hours to make an M7 backplate on the mill - 153 holes. I have yet to break a bit while drilling a capsule mounted properly. The only breakage I have is if I don't tighten the capsule down tight enough, or if I don't put enough drilling oil on the capsule. I am thinking about getting the Taig CNC mill later this year - that would speed up making capsules just a bit.

 

[gyraf]

Bill,

I'm not sure that a regular lathe will have the needed precision to cut anything to 2mils precision and absolutely flat..

But other than that, it might just work...

Jakob E.

 

[Marik]

[quote author="gyraf"]Bill,

I'm not sure that a regular lathe will have the needed precision to cut anything to 2mils precision and absolutely flat..

But other than that, it might just work...

Jakob E.[/quote]

How about a mill with a rotary table?

 

[gyraf]

I don't know the basic precision of those - but 2mils is not much over a 32mm capsule diameter...

Jakob E.

 

[bill raymond]

I'm not sure that a regular lathe will have the needed precision to cut anything to 2mils precision and absolutely flat..

But other than that, it might just work...

Well, I'm going to give it a go anyway. Williamson apparently did his machining on a regular lathe without lapping the backplate, so maybe I'll be OK, and 2 mils is twice the spacing that Williamson used, so perhaps there's room for some error here. At the least, it's an opportunity to learn.

 

[PRR]

> Schulein's attempt to provide greater sensitivity.

Schulein was confused. Shows what happens when a good academic tries to be practical.

Williamson's design can be criticized: the capacitor near the edge adds little to the output, but a lot of dead capacitance. Some diaphragms are un-metalized around the edge for this reason; others use a backplate smaller than the diaphragm.

For a WE-640 type diaphragm (which this is NOT), which deflects in a parabolic/catenary shape, the backplate should be about 2/3rd the diameter of the diaphragm to optimize the output.

That seems to be what Schulein was thinking, only he went half, and did some other odd things too.

In the Williamson, the small backplate gap means it does not generally move as a catenary, but flatter, with more of the diaphragm contributing to output. The optimum might be 9/10, something like that, but that's about the same as full-diameter. And a reduced backplate adds significant complication compared to Williamson's simplicity.

Sensitivity is also related to stiffness: less stiff, more output. However less stiff means a lower top-resonance: Schulein seems to have gone much too far: lost both ends of the band, and needed a midband notch. Good experiment, bad recording tool.

As Bill realized, you can get the smaller effective backplate without the far-too-large back-volume if you fill-in some of what Schulein cut away.

Schulein's curves also suggest a massive air-leak, much more than just excess volume. Perhaps he should have hired a fine machinist instead of assigning a grad EE student....?

I still think, if you deviate much from Williamson's 1 mil back-gap and total enclosed volume, you are wandering off into the woods. There are many trails out of that forest, but many of them are dead-ends. For a NON-resonant response, you either need very small gap or very small-diameter holes to get the air resistance up. Williamson's insight was that small drills break, while a small air-gap may be manageable on the kitchen table.