Capsule construction. the practical aspect

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

badger

Active member
Joined
Jun 5, 2004
Messages
32
Location
London
Just a few observations on general construction techniques.
Capsules should be regarded as examples of precision engineering,wether you decide to build K47,K67,resonant chamber or a mixture,the same principals apply. With regard to the back plate attention should be paid to drilling the holes with a high degree of roundness,making sure not to leave overhang on the diameter after lapping.This can be removed by gentle scrubbing with a fiber pot scourer.Dont forget that these cylindrical chambers constitute small air springs helping among other things to maintain equal compliance across the diaphragm. I gave details of precision lapping in a previous post.Make sure that the interference groove is acuratly turned to the specified dimensions.
The anular ring should be turned to the design thickness making sure that the dimension will not cause cavity resonance.The underside clamping face should be lapped after drilling to give a smooth inclusion free surface.The inside quarter of this face should be bevelled and polished to avoid any minute interference of the dynamic portion of the membrain.A light should be held at an angle to the tensioning jig to observe striations occuring at the periphery of the membrain.The inside diameter of the spacer should be cut very cleanly and lined up with the outside diameter of the interferance groove.The holes in the spacer should also be cleanly cut ,the best way is to make a small punch from a suitably sized drill shank.Use the smallest screw thread possible to avoid interference with the PET.If you use slotted screw heads,use a shrouded screw driver,I use 14 BA screws.Tighten the screws equal and opposite as you would a car cylinder head.
Look out for any striations occuring as you proceed,easing off screw tension.or re applying in a different orderwill usually rectify the problem.This is dependant upon you obtaining uniform tension in the frame in the first place!
If you are going to cement the PET to the anular ring apply a minute film of adesive to the outsid diameter only.A small bevel should turned in the ring for this purpose.I'm not going to tell you how to do this.mainly because it's too complicated.Have a lookat the Neuman Pics that Zebra posted you will get the idea.
BTW I dont think Neuman produce production capsules like this.Looks like a lab set up to me. Its pretty much what I used to use in the garage.
I did give a reasonably comprehensive description of how to heat treat and pre tension PET film in a previous post at the old place.I dont know if it got into the meta if not I could re post it if anyone is interested.
I think part of the answer is to strive to get a stable membrain seated with equal tension and a high degree of flatness accross the back plate.You may then have a chance of making a useable microphone.
I am not going in to design criterior or the math. Tim C and others are handling thatvery adaquately.
One observation on design.I think Gus got it right.use the math to get in to the ball park,then its hours and hours of trial and error.
Hope I'm not teaching my grandmother to suck eggs and I hope some of it may be of some use.no way definative,just what has worked for me over the years (mostly).
Andy.
 
thanks for the post. I am still collecting things for the capsule part. The capsule looks to be the hardest part of a close to total DIY build.
 
how are you guys drilling the holes? just with a drill press? :?
ive got a small milling machine, that does .002 increments. i can convert it to cnc for about 200$. thinking that would produce a cheap, precise way to cut capsule holes.
 
This is a bit long... I've built three or four capsules that technically work, and I did my own gold coating of mylar as well...

I use a mill with carbide drill bits (I get them from McMaster-Carr and they are called printed circuit board bits). I run it at 4000 rpm (as fast as it goes), and do about a 0.3mm to 0.5mm per second plunge. The mill at work is manual but has digital readout to 5 microns in the X and Y axis, and I lift the bed to drill the hole - it just has the graduated markings on the handle but that's good enough for 25 micron precision. I mount the capsule in a 5C collet for this operation (it's the same one I use in the lathe to face off the brass bar and also to do the final surfacing and machining in the diaphragm spacer). On the lathe I use a home-made high speed steel tool. Finally, to finish the process, I fine-tune the diaphragm spacer with 2500 grit sandpaper, a granite table and some nice micrometers. I don't use shims on my capsules to get the diaphragm spacing - I use a machined-in feature on the capsule. I find it a lot easier to get this feature machined in, and the Debenham capsule uses this method also.

You can build a thermal evaporator at home but it's a bit of a struggle. I eventually got mine to work quite well, but getting rid of vacuum leaks is a major pain in the behind. If you are interested in going down this path I can give you a few suggestions. PM me for that. Realistically, it'll cost somewhere around $1500 US, though, and a few cases of beer to friends that can weld or help with machining and that sort of thing. The evap supplies are easy to get from www.2spi.com.

I'd like to take apart an M7 but my budget prevents this.... I have some wacky designs as well, and some work and others sound about as bad as an Apex 430 capsule (and that's pretty bad). Other backplates that I have in process are based on staring at photos of M7's and KK47's and that sort of thing, and guessing at hole sizes. I find the two things that my bad capsules have are either 1. incorrect high-frequency response (either way too much or way to little), or 2. totally wacky patterns, like a figure-8 at 1kHz, cardioid at 500 Hz, and omni above 2kHz.
 
Great info. Andy. What do you guys use for a tensioning jig? I have a MXL 2003 capsule that had an accident... The diaphragm is shot. I want to rediaphragm it, probably with an unrolled cap b/c I have no access to an evaporating machine. I'd have to change it to edge terminated, but thats not the end of the world...Can anyone point me in the direction of how to construct a tensioning jig? Can't quite get my head around it at the moment. Thanks.


Jason
 
I did give a reasonably comprehensive description of how to heat treat and pre tension PET film in a previous post at the old place.I dont know if it got into the meta if not I could re post it if anyone is interested.
Please!

thanks

tim
 
Tim,
I think I found the relevant post.I will post it tomorrow.I'm not sure it will help your particular application,but it should be of general interest.
In the same thread I think Tim C described a simple device for determining the FA resonance of the diaphragm. if it's not in the meta perhaps Tim could re post.
Andy.
 
A few point s on diaphragm conditioning.This is not intended to be definitive,but a guide for further experimentation.
Most problems are caused by"cold flow relaxation",which occurs over a period of time or under certain temperature conditions.
Try the following.
Take a square of film approx 11cm sq attach one side to a clean plastic faced board with scotch tape,attach the opposing edge to a piece of wood [or similar ] approx 16cmx1.5cmx1.5cm in the same maner.Then carefully pull the wood untill you feel the film stretch very slightly,repeat the process with the other sides.Then apply the film to your tensioning gig and heat in a clean oven at 120-130degs for about 20 mins allow to cool and repeat the process. You should end up with a stable membraine. try not to touch the film with your fingers !

Food for thought.
Some time ago I experimented with a porous ceramic back plate. The random uniformity of the porousity can be calculated with respect to critical damping, there also being an intrinsic phase delay network !
The ceramic disk is first lapped,then evap coated and inserted into a brass or non conductive ring depending on system.
Worked quite well from what I remember.
 
Andy thats very cool about the ceramic.

Have you had a Oktava 012 capsule apart? The cardiod one uses what looks like sintered brass as the back vents.
 
Gus I dont know the 012 ,but I guess it's the same principal,if you use sintered brass no need to vap the face after lapping.I think there may be a problem with continuity.The damping frequencyis probably not critical and can be contained within useable limits.the problem lies in the delay network .To maintain polar conformity with respect to frequency,I think would be difficult given the random nature of the beast.But worth some thought.
 
Am quite new to microphone DIY so please forgive if this is not the right place for this elementary question :oops:
I have a c37 mic cap.
I am wondering simply if this cap can be mated with most other amp circuits or would it want to see certain sony-like specs out of an amp in order to operate?
In particular, i have an oportunity to buy a geffel m70 minus its cap and body.
If this is the wrong area to post such a question, perhaps someone might direct me to the right forum?
Thanks Any, All;
scott
 
This isn't exactly the best thread to put that in, but you'll get an answer anyways...

The C37 capsule is similar to most other capsules (60 volt polarization), so you can use it with most amp circuits. HOWEVER the C37 capsule's output is about 15 dB lower than most other capsules, so the amplifier tube or FET needs to be really, really, really quiet. Most 6AU6 tubes are far too noisy, and Sony must have selected them very carefully for use in their circuit.

As for practical capsule construction....

I just finished builiding backplates for 30 capsules. For building a lot of capsules, a CNC mill (to drill all of the holes) and an accurate CNC lathe are indispensible. I have the CNC mill but not the CNC lathe and it's almost enough to drive me to drink. If you pick a capsule that does not require turning upside down and aligning carefully to drill (ie pick something other than a Neumann M7 or K47) it's much easier!!!

-Dale
 
[quote author="tmbg"]lapping ceramic sounds like an unpleasant experience.

[/quote]

I like the porus backplate idea. but why not start out with a sintered metal piece instead of the ceramic? I have an AKG dynamic mic that uses a sintered windscreen. it is pretty darn acoustically transparent. If you want to go totally DIY, you could probably control the sintering to get just the acoustical resistance you want. by the way, brass sinters easily and the powders/ beads are readily available.

mike p
 
Back
Top