Small diaphragm capsule design help needed

[Jonte Knif]

But imagine if you would a one whole earth away! That would solve the problem.
Any way, thanks, I would be knocking on your door already.

I found an Optimum 180x300, used, with all bells and whistles. FOUR chucks. The original Chinese, and then Roehm 3-jaws, and 4-jaws adjustable plus 4-jaws self centering. I hope it is in condition. The chucks already cost more than a new machine. If I decide to buy it, I get it in 2 weeks. The guy updated to a bigger Optimum. I'm pretty sure the Roehm chucks will help a bit with the accuracy. And I can not see why my capsule design would actually need super fine tools, because the only really critical phase is lapping. I've made some initial plans for the tensioning jig. I'll post them as soon as I have drawn them for you all to comment.
For lapping I've got several suggestions how to deal with it. The latest came from a pro, who said that guaranteed 0,5u flatness and 0,1u roughness will cost 5e per piece, which is OK. How ever I told him that I prefer to have the tools myself.

I also got _much_ cheaper prices for all the parts form a proto shop. So I could proceed without a lathe, but I got the lathe fewer, so I guess I'll by it anyway.

The DPA microphone handbook is actually pretty good!

-Jonte

 

[leswatts]

The Opti 180x300 Vario looks like a nice little lathe. P5 bearings...that's good.

After your lathe fever is cured...then you must contract CNC milling machine fever.

Lapping is pretty easy...unless you want to do the concave backplate bit! even that's not too hard.

Les
L M Watts Technology

 

[PRR]

> Lapping is pretty easy...unless you want to do the concave backplate bit! even that's not too hard.

I don't know why you'd want that.

It sure is not "hard". Thousands of beginners have hand-ground concave glass telescope mirrors with only two pretty-flat glass disks, a little tar, and some abrasive powder. Al or Brass may be easier.

Amateur Telescope Making edited by Albert G. Ingalls
http://www.willbell.com/tm/tm7.htm

In fact: measuring a concave curve by mechanical means (straightedge and feeler-wire?) is fussy, but easy with optics. Put a bright LED on the far side of the room. Or a pinhole in a window shade. Use a toothpick to find the focused spot. I think the radius is twice(?) the focal length. For telescope mirrors with long focus, you must allow for non-infinite source distance (a collamated laser-pointer might be a new-tech trick). There's also a knife-edge trick which is more precise and also reveals non-spherical errors.

But I doubt a mike plate needs such precision. I bet catching an LED-spot on a toothpick and eyeballing the distance would be more than precise enough.

 

[audiox]

so I guess I'll by it anyway.

If you don't need it in this project, you need it in the next one for sure...

By the way, I know a guy who designs mechanical parts for medical equipment. He lives in Sweden at the moment but he has worked in Finland too. If you want, I can ask where he has manufactured his prototypes.

 

[Marik]

I found an Optimum 180x300, used, with all bells and whistles. FOUR chucks. The original Chinese, and then Roehm 3-jaws, and 4-jaws adjustable plus 4-jaws self centering. I hope it is in condition. The chucks already cost more than a new machine.

Hi Jonte,

I had a similar 7"x12" lathe and grew out of it about in a couple months--it is very light, unstable, prone to vibrations, no rigidity whatsoever (no amount of adjustment would take off cross-slide play), limited amount of cutting threads, impossible to cut bigger mic bodies, small spindle taper, etc. etc. etc. I hated it--there is nothing to talk about precision. I ended up buying a much bigger and heavier 10"x26", which I'd consider a minimum size for precision work on parts of microphone size. It took me two full weeks to adjust and set it up.

For precision cutting backplates you don't need chucks, but will need to look for collets, so look for a lathe which would have a bigger spindle taper. Ideally, that would be the lathe which accepts 5C size. I like those emergency (and/or step) collets, as you can cut them yourself to exact dimentions you need and then get perfect repeatability, but also there are some other different options (including Bison adjustable chuck/5C adapters, which are horrendously expensive).

What I can promise, if you are serious about capsules and mics, sooner or later you will find you cannot live without lathe...  

Best, M

 

[Jonte Knif]

Marik,

Well, that was a serious warning. That's exactly what I've read about the Chinese lathes that come from factories working alone without western control. The web is full of horror stories about people spending a week first cleaning, then greasing, then adjusting then getting disappointed after all. I haven't even considered these machines.

Was your lathe something like that or "western branded" like Grizzly or the two German brands Optimum and Quantum? They all come from the same factory and look pretty much the same too.

I called the guy who is selling his machine and it is actually not the Opti 180*300, but somewhat bigger Quantum 210 x 400, weighting 85kg. The seller said he has turned max 180mm steel pieces but it does not do it easily. For mic body size objects he said it will be "perfect" and the Opti 180 x 300 would not be adequate.

http://www.quantum-machines.com/products/lathes/d-210-x-400/index.html

That is really the maximum size I can take to my workshop, not because of space, but extremely nasty stairs and tight corners. I have my electronics workshop in a different place (at home) but I have rented this basement room for woodworking years ago.

I'll visit the guy in the evening. I have a reason to believe he is pretty honest, because the machine is already sold, as an exchange for a bigger Opti (400kg) he has bought. So I will be dealing with the lathe shop, not with this man. If they have not formed some kind of conspiracy (not common in Finland) I will get pretty objective info from person who used the machine for 5 years. I have to wait a couple of weeks for him to finish still one piece and put the lathe into selling condition. (cleaning)

-Jonte

 

[leswatts]

> Lapping is pretty easy...unless you want to do the concave backplate bit! even that's not too hard.

I don't know why you'd want that.

I don't know why I'd want that  either...but for type II mics some think it would be advantageous:

http://www.phys.unsw.edu.au/music/people/publications/Fletcheretal2002b.pdf

I've done the mirror walk around the barrel foucault test bit. It's pretty neat that you can do 1000 angstrom precision with almost no tools...just hand work. This is similarly true for making flat surfaces.

I'll be interested in the outcome of Jonte's lathe fever. I don't know that much about the quality of the small chinese lathes. We use american or british light industrial tools, which despite the name requires a forklift truck to move around. Not good for home use.

I can say one thing. A quick change gearbox is a good good thing to have!

Les
L M Watts Technology

 

[burdij]

This is in the nature of a buying tip when you go to your local lathe shop (is that like a Best-Buy where, instead of plasma TVs and laptops, you have rows of lathes, followed by rows of CNC mills, etc?) looking for a new lathe.

The stability problems with small lathes are related mainly to the way the tool post/cross-slide assembly is mounted to the main y-axis table. Here is a diagram of the toolpost and cross-slide as viewed from above:

The ring or plate that permits the rotation of the cross-slide assembly with respect to the main y-axis is retained in the cheap lathes by two bolts shown in red. When a tool is mounted in the end of the toolpost, due to leverage of the rotating part rotating CCW, or toward us in this view, the toolpost tends to bend downward, rotating toward the work piece. The thing to look for in a good lathe (or the first thing that needs to be changed in a cheap lathe) is to have a retaining ring or plate that has four bolts positioned as shown in blue. If the only lathes in your price range are of the two bolt type, don't despair. This is a relatively easy part to replace. Places like www.littlemachineshop.com have them ready-made for a lot of models.

 

[Marik]

Jonte,

My lathe is Grizzly:

http://www.grizzly.com/products/g0602

Generally, look for the biggest lathe you can fit. Of course, in Europe you can much easier find German, Swiss, or British made ones and that would be terrific, if you can afford them. On the other hand, when setup right Chinese made ones also can be very useful. To reduce vibrations and bed torsion distortions you will need to bolt it down to something big, heavy, and rigid. I actually separated motor from the bed and bolted both  to a stationary concrete pedestal. They are connected together only through the chain type belt (again, to reduce vibrations).

Another thing, make sure it can cut 60UNS thread--many mics (esp. measurement, like B&K, etc.) use it.

> Lapping is pretty easy...unless you want to do the concave backplate bit! even that's not too hard.

I don't know why you'd want that.

I don't know why I'd want that  either...but for type II mics some think it would be advantageous:

http://www.phys.unsw.edu.au/music/people/publications/Fletcheretal2002b.pdf

Thank you for the article!

I can say one thing. A quick change gearbox is a good good thing to have!

No kidding!
Mine has a semi-quick gear-box (i.e. for some threads still needs to manually change gears, which is PITA), but much better than my previous small jobby. Of course, CNC will take care of it.

Best, M

 

[Jonte Knif]

Burdij,

I have no clue, I just came from the visit and didn't know to look for that detail. Got the message a bit too late. But a doubt that in this machine you would find anything really premium.

Some minor issues: it doesn't have electronic speed control. Only 6 speeds with belts. I don't know about the thread sizes. The guy never used them. The brochure for year 2005 model (this was 2004) says: 10 - 44 TPI.  Does that cover 60UNS? No idea.

The quality seemed to be quite good. Everything moves smoothly, it cuts nicely,  and main lead screw was in perfect condition. The man has taken care of it. The western chucks looked pretty great but the Chinese wasn't crap either.
The quick change tool holder was of high quality. He had collets, but unfortunately they fit into his new machine too so I won't get them

He had a pretty big steam locomotive under construction, and I can imagine that he needs a bigger lathe. The last work with this lathe will be the engine pistons after which I can take it or leave it.
Hmmm....I could take this one, and get a lot of valuable western tools (some of which I will never need), or spend the same money to a new machine with no real tools at all.

Today I received alu foils from Lebow. 1.8u and 2.5u. It comes to one ones surprise that they aren't too easy to handle. The skinning jig has to be pretty well designed. But I think I can use the wax paper, to which the foil is "attached", to my advantage in some way or another. Ideas? Compared to 3u mylar foil this stuff is extremely weak.

-Jonte

 

[PRR]

Here's Olson&Massa's non-math discourse. They start along a different path than I did, but of course come out of the woods in a similar place.

This is 1934... the Mylars and such were not readily available.

 

[leswatts]

Here's Olson&Massa's non-math discourse. They start along a different path than I did, but of course come out of the woods in a similar place.

Thanks for that!

Yesterday I had some time (waiting for a graphics overlay for a preamp prototype) so I looked in detail at that Fletcher paper on concave backplates.

His mathematics show that with flat backplates there is a potential energy maximum at an intermediate position of the diaphagm.

Potential energy maxima are unstable, so if backplate distance is brought closer than that point the diaphragm will spontaneously "suck in" to the backplate and stay there unless the charge is bled off.

He asserts that this is the primary distortion mechanism in an omni condenser capsule, causing both odd and even harmonics.

He further asserts with (cosine) power series expansions that this distortion mechanism is COMPLETELY ELIMINATED with the properly curved backplate.

BTW I think this has been a good thread. Later I want to have a thread on ribbon mics. I have searched here some, but I would like to go into much greater detail, particularly with the physics. I don't know what I can add to 75 years of development...but we'll see!

Les
L M Watts Technology

 

[PRR]

> I looked in detail at that Fletcher paper on concave backplates.

Yes, it seems he spotted what everybody else skipped-past for 80 years. (In part because dumb old flat-plate mikes ARE good-enough for very many uses.)

And the amount of curve, relative to computed benefits, is quite small. And very non-critical. OTOO the curve on the foot of a 1940s-1990s US car-engine tappet/lifter. With the LED test, you offset the reflection a bit and look for a soft focus about a cubit out, done. In soft brass, you might get there by putting toothpaste on your thumb and rubbing for a couple minutes.

 

[Gus]

Jonte

What type of motor AC or DC?
If AC is it a 3 phase?
If so look into AC drives that can power the motor and make it a variable speed motor.

 

[Jonte Knif]

Single phase, so I guess there is nothing I can do? Or? I don't understand motors.

PRR, thanks for the paper!

-Jonte

 

[Marik]

He asserts that this is the primary distortion mechanism in an omni condenser capsule, causing both odd and even harmonics.

I will need to read this paper ones I have little more time. There was another paper from D. Lavry, which shows that primary mechanism of distortions in the capsules is due to non-linear nature of capacitance/voltage transfer.

Later I want to have a thread on ribbon mics. I have searched here some, but I would like to go into much greater detail, particularly with the physics. I don't know what I can add to 75 years of development...but we'll see!

It's been awhile since there were any "in-depth-threads" on ribbons, which is even more strange, considering last few years so many ribbons came to the market... ???

Best, M

 

[gyraf]

AIP handbook uploaded to the group gmail account.

I'll be back soon - had too much work lately...

Jakob E.