New ribbon mic materials and construction

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

leswatts

Well-known member
Joined
Apr 29, 2009
Messages
269
Location
Tiger, Ga USA
I'm continueing (I hope) some discussion on alternate  or new ribbon materials that started on a gearslutz thread that Marik and AEA participated in a bit. GS is the wrong place to talk about microphone engineering.

I've been working on this a little, and today researched the "Roswellite" C&T/Shure microphones some. Looked at the patent applications a bit. I'll keep an open mind , but some things are starting to look a little bogus.
Carbon nanotubes? Hmm. I think.....not.  So what is the stuff?

I'm looking at ways to increase blast durability of ribbon mics by other means.
Anyway, there hasn't been a good ribbon mic discussion here in a while, and i'd love to talk about it some.

Les
L M Watts Technology
 
Well, a good many looks but no responses.

I can say that I found more about "roswellite" on my own.

Before Shure purchased it, there was a lot of ad copy about carbon nano tubes, "nano enabled",
and "shape memory".

Here is what I found:

Single walled carbon nanotubes do have conductivity about as good as aluminum if you consider a single tube across its length.

They have been made as long as 18 mm in laboratory conditions. But in the commercial sense the technology results in short (perhaps on the order of 100 micron) fibers in a powder form. They can be had for a few hundred dollars a gram.

But to transfer charge from one nanotube to another requires quantum mechanical tunneling, and this destroys the metal like conductivity previously mentioned.

The best single wall nanotube composite resistivity I found was 0.15 ohm-cm. Compared to aluminum, this is basically an INSULATOR.

After looking at patent applications I think "roswellite" is this:

Thermally formed corrugated PET film about 2.5 micron thick with aluminum sputtered or evaporated on one or both surfaces. In other words, capacitor film.

I'm not saying it doesn't work well, but the "carbon nanotube" and "nano enabled" copy I think was a bit over the top. BTW Shure seems to make no such claims.

Good luck for them trying to get the patent for aluminized mylar diaphragms approved! Long public domain I think.
 
I wish I could contribute to this more as I like ribbons a lot and have done some (home) research on ribbon transformers and active ribbon circuits.

Bob Crowley used to post here once in a while and he has his own microphones blog:
http://microphonium.blogspot.com/

As you may know, Crowley & Tripp is or was a subdivision of Soundwave Research (http://www.soundwaveresearch.com/)who do nano stuff also in the medical field. So there may be more to Roswellite than just a "kind of capacitor film". But I don't know. I'm not a nano guy.

I suppose you could just contact Bob and ask nicely, he seemed easy enough to talk to when he posted here.
 
Hi Les,

I've been researching nano-tubes for awhile. The good thing about them is they formed by "casting" and "remember" the shape very nicely. The tubes are incredibly strong on themselves, but the problem is they don't form a strong "fabric". In essense, it is like bowl of spaggeti--they all fall apart, hence the need for some "unifying" substance. The Al sputtering does the trick, but as I wrote in another message on GS, the sputtered Al has worse conductivity. However, I'm sure, this still works nicely with very efficient motors.

I agree with you on the coated Mylar application. This has for ages been used in some mics and in many ribbon speakers (which are the same transducer as a ribbon mic). 

I can write and invite Bob to participate, if he can. 

Best, M
 
As you may know, Crowley & Tripp is or was a subdivision of Soundwave Research (http://www.soundwaveresearch.com/)who do nano stuff also in the medical field. So there may be more to Roswellite than just a "kind of capacitor film". But I don't know. I'm not a nano guy.

I suppose you could just contact Bob and ask nicely, he seemed easy enough to talk to when he posted here.

Yes. I too worked (a lot) in conductive nanoparticles for non audio applications.

Bob wouldn't be able to say much since he sold the technology to Shure.

And I couldn't Talk to Shure much even though I used to be an audio research engineer there. I'm more of a competitor now.

Bob Has caimed carbon nanotube ribbon microphone diaphagms on one application.

But the technology does not exist as far as I know. That's fine... he could own it if it ever does exist.
Right now, the numbers don't add up.

However his method patent application for "shape memory" ribbons is here:

http://www.freshpatents.com/Methods-for-forming-and-using-thin-film-ribbon-microphone-elements-and-the-like-dt20070927ptan20070223773.php

Convinced?

Les
L M Watts Technology
 
Marik,

Crossed in the mail...

Yes, I'm familiar with bulk fullerene conductive material. Again, electrons would have to tunnel to the unifying substrate, and there goes the conductivity. The best I found in any modern research paper was 0.15 ohm-cm....what's that....100,000 times less conductive than bulk aluminum? Useless in a ribbon mic...

Les
L M Watts Technology

 
Useless in a ribbon mic...

I should say as a conductive element...might have structural advantages.

And I'd love to talk to Bob... He's a smart guy. He has a lot more patents than me.But for the previous mentioned reasons prob can't

But check out that method application....

Les
 
leswatts said:
Useless in a ribbon mic...

I should say as a conductive element...might have structural advantages.

And I'd love to talk to Bob... He's a smart guy. He has a lot more patents than me.But for the previous mentioned reasons prob can't

But check out that method application....

I will let Bob know. Not sure though how much he can tell, esp. now, after Shure took over.

Yes, I am familiar with that method application. I was referring to it when was writing about "the coated Mylar application" in ribbon speakers.

Best, M

 
IMHO, The most promising application for CNTs is in transparent conductors, with potentially better performance and mechanicals than ITO and PEDOT-PSS.  For conventional conductor replacement, blahh....hype.  I'll be going to the IDTechEx Printed Electronics conference in December, if there are any developments worth mentioning, I will (but I doubt there will be).  Roswellite, perhaps as real as the Roswell event itself.

Also, Mark, check your PMs.

-Chris
 
I didn't get the impression from what I have read about the Crowley and Tripp microphones that the Roswellite ribbon was conductive as a result of the use of carbon nanotubes. I would think it would be as conductive as a carbon graphite fishing rod or a carbon graphite golf club, i.e. not very. I think the use in this case gives the ribbon lateral and tensile strength, much like an incredibly thin layer of fiberglass. The conductivity is rendered by depositing a conductive film on the surface of the ribbon. Of course, I don't have any "inside" information.

With regard to new technology in ribbons for ribbon mikes, I was able to audition two new active ribbons from Audio Technica, each of which claimes a patented process of etching the ribbon, probably from a thicker base material to create a series of micro-ribs that give the ribbon lateral stiffness so that it won't sag or become mis-aligned in the pole structure. The two models are the AT4080 and the AT4081. On vocals, these are outstanding and the extended frequency response in the highs while not offering the same transparency that a condenser would have, will certainly extend their use. The more expensive model (the AT4080 at $999) was noticably better but on a male voice both performed better than an RE27-type dynamic which I have thought always has a really "boxy" and closed-in sound. I think Shure will get a real run for their money out of these.

EDIT - Or possibly they won't. I just looked at the Shure site and it says that the Crowley and Tripp microphones were acquired from Soundwave and immediately discontinued. What's up with that?
 
I just looked at the Shure site and it says that the Crowley and Tripp microphones were acquired from Soundwave and immediately discontinued. What's up with that?

Really???

I just thought they stopped reselling the C&T line, but used the aluminum/mylar in the KSM Shure branded line.

By my rough calculations a bit of PET wouldn't degrade output so much, since it's half the density of aluminum. So all other things being equal a 2 micron PET/2 micron AL might lose what...20 log(1.5) or 3.5 db over just a 2 micron aluminum only? Actually it would be a little less due to air mass loading. This also assumes that the applied metalization could be made to be as conductive as wrought material. (might not be able to if vapor deposited)

I do note on the KSM that it was felt the output should be equal or even a little greater than an SM57. Seems they sacrificed a little on the high end to get enough pressure gradient to do that. Also, output imppedance is rather high.

As far as the nanotubes...I think they will have some very good applications.

Depending on the chiral vectors, they can be made to have either metallic or semiconducting properties.
So they may be used for high performance transistors and such. They have excellent thermal properties.

Les
L M Watts Technology
 
Whoops. Sorry, didn't see that they had moved the mikes to their "China" brand line. Looks like they eliminated some models though.
 
Since nobody posted the patent, which includes nano-tubes, here it is:

http://v3.espacenet.com/publicationDetails/description?CC=WO&NR=2006047048&KC=&FT=E

Best, M
 
Okay, just about done with my review of possible ways to further improve a 75 year old  ribbon technology.

1) Durability
Obviously a major consideration.

Material: Aluminum is the lowest resistivity density product material I can find that's stable at room temperate in air. However the commonly used pure aluminum is exceptionally weak.

Going to aluminum alloys could improve strength by three to five times. These alloys have about 25% higher resistivity density, so there would be a 1 to 2 db sensitivity reduction for equal noise voltage.
Also, they MUST be protected from corrosion by a passivating layer or they will quickly weaken and imbrittle.

As far as composites...I calculated something like a 2 db sensitivity loss (for same noise) by using a carrier like PET the same thickness as an acopanying aluminum conductive member. The greater elongation of things like PET would offer much greater blast protection. I note that the aluminum would still be taken to plastic deformation, so prolonged large excursions would still destroy the conductivity of the ribbon...but good protection from an occasional blast.

I have long, long used PET/AL for ribbon speakers, as have many.

Ribbon forming: Corrugations have been typically formed by cold working foil in a "gear" type device. This cold forming would introduce significant residual stress in the material...weakening it.
This can be cured by either a post form heat treatment or stretch forming . I estimate doing so could increase durability 50% or so.

Further, I suspect that a sine wave corrugated end clamped ribbon will have significant stress concentrations, probably near the clamp points. I'm trying to do a dynamic finite element model to see if other corrugation patterns could offer improvement.

2) Sensitivity/high frequency response

I group these together since they are so fundamentally related.

Magnets: Obviously Neo magnet technology has simplified ribbon microphone design, although high flux density can be obtained with other materials. Open circuit voltage is directly proportional to the gap flux density.

The great advantage of Neo I think is it's high coercivity. This makes it resistant to it's own self demagnitizing field so much that it can be used without a magnetic return circuit fairly well. Marik has seen this. I do calculate that a return circuit still will give a bit more flux density...perhaps 25-50% depending on the design.
It's quite unlike alnico or ferrite though, where a path and concentrating pole pieces are absolutely required.

It looks to me like about 5x5 mm magnet cross section with a 5 mm gap seems optimum. My calculations show about 0.4 Tesla with no return circuit, perhaps 0.5 to 0.8 with. Much less and I can't match a dynamic microphone sensitivity... much bigger (magnet size) and high frequency response suffers.

Extending high frequency response Like everyone else I guess, I want response well above the often seen 8-10kHz rolloff, without reducing sensitivity to get it. I can think of nothing better than the waffle plate resonators... not only do they offer a peaking mechanism, but also phase shift, increasing delta P if used on only ONE side. So what about a single side resonator or even two on each side tuned differently?
----------------------------------------------

So that's about it. Here's what i'm doing then: Doing a dynamic FEA of corrugated ribbon structures If I can. (I have GRAPE and LISA right now). Getting some solutionized 2024 T0 foil as thin as I can, and attempt to roll it further. Thinnest I have found so far is 25 micron.I can muster foil rolling in my shop If I get proper heat treat capability. It will require several anneals to get to 2 microns. I'll talk to foil rollers to see if there's some brick wall problem I don't know about.

I thought I'd post these thoughts if anybody cares. Comments welcome.

Les
L M Watts Technology

 
Hi Les,

Just a few thoughts.

leswatts said:
1) Durability
Obviously a major consideration.

Material: Aluminum is the lowest resistivity density product material I can find that's stable at room temperate in air. However the commonly used pure aluminum is exceptionally weak.

Going to aluminum alloys could improve strength by three to five times. These alloys have about 25% higher resistivity density, so there would be a 1 to 2 db sensitivity reduction for equal noise voltage.
Also, they MUST be protected from corrosion by a passivating layer or they will quickly weaken and imbrittle.

As far as composites...I calculated something like a 2 db sensitivity loss (for same noise) by using a carrier like PET the same thickness as an acopanying aluminum conductive member. The greater elongation of things like PET would offer much greater blast protection. I note that the aluminum would still be taken to plastic deformation, so prolonged large excursions would still destroy the conductivity of the ribbon...but good protection from an occasional blast.

I have long, long used PET/AL for ribbon speakers, as have many.

Generally speaking, if the HF response is the goal then IMO the 2db sensitivity loss is A LOT for a ribbon mic, but again, the question is where to make compromises, which is the main part of the design.

leswatts said:
The great advantage of Neo I think is it's high coercivity. This makes it resistant to it's own self demagnitizing field so much that it can be used without a magnetic return circuit fairly well. Marik has seen this. I do calculate that a return circuit still will give a bit more flux density...perhaps 25-50% depending on the design.
It's quite unlike alnico or ferrite though, where a path and concentrating pole pieces are absolutely required.

It looks to me like about 5x5 mm magnet cross section with a 5 mm gap seems optimum. My calculations show about 0.4 Tesla with no return circuit, perhaps 0.5 to 0.8 with. Much less and I can't match a dynamic microphone sensitivity... much bigger (magnet size) and high frequency response suffers.

According to my calculation for N50 Neodymium 0.4T for your magnets/gap sounds about right. With backing yoke it jumps to some 0.68T. If you deside to use a return circuit most likely it will have more losses than yoke so it would be safe to assume then you get some 0.5T--0.6T in the gap. Considering there might some saturation, most likely it is rather 0.5T.

Let me see if I have magnets of that size (I should) and double-check the reality.

leswatts said:
Extending high frequency response Like everyone else I guess, I want response well above the often seen 8-10kHz rolloff, without reducing sensitivity to get it. I can think of nothing better than the waffle plate resonators... not only do they offer a peaking mechanism, but also phase shift, increasing delta P if used on only ONE side. So what about a single side resonator or even two on each side tuned differently?

Your effective path distance is 22.5mm, which corresponds to a null @ ~15.2KHz. Since the top end starts dropping at least at half wavelength it will start at 7.6KHz. Since we have about 2.5mm depth (ribbon recessed to half of the magnet) we are getting some diffractions there, so the null most likely will shift up… probably not by much, though… too lazy to calculate.

If you ask me, I hate the idea of ONE side resonator. When we talk a “native” ribbon we are thinking of perfect Fig8. Anything else and we cannot use it in MS.

leswatts said:
So that's about it. Here's what i'm doing then: Doing a dynamic FEA of corrugated ribbon structures If I can. (I have GRAPE and LISA right now). Getting some solutionized 2024 T0 foil as thin as I can, and attempt to roll it further. Thinnest I have found so far is 25 micron. I can muster foil rolling in my shop If I get proper heat treat capability. It will require several anneals to get to 2 microns. I'll talk to foil rollers to see if there's some brick wall problem I don't know about.

If you find something worthwhile please let me know—I will be more than happy try and test it.

Best, M
 
Mark,

I'm muddling through getting a corrugated ribbon mesh model in LISA. It doesn't take DXF files well, which makes it a pain.

If it turns out a non constant corrugation scheme might have merit, I can make dies for it on the cnc.
I do that a lot for the diaphragm dies used on the little punch press.

2 db loss is a good bit, yeah. The actual ribbon mass increase for equal noise calculates to 3.5db, but the effective mass must include air load, which is substantial I think. I can calculate it. But i'm guessing
20 log (1.5 Mrib +Mair/Mrib+Mair) will be about 2 db. Thinner the ribbon-the less it matters. I haven't calculated optimal ribbon mass/thickness considering johnson noise level, (small) pressure noise, and open circuit output. I just know that increases in output with mass and thickness reduction has to start going away as an acoustic impedance match is approached.

I think the magic number might be 1-2 micron more for other practical reasons...as thin "silver leaf" users will attest to.

Single waffle plate? I don't know, it was just a thought. Should be tuned to only affect the response around the null. I can model it. Not sure what the results would be...might go cardioid at very high frequency.

I'm thinking of ribbon computer/robotic autotune as well...I have that system set up for the sixth order energy harvesting microphones we developed. We tune 3 mechanical components to Q and frequency.
Tuning ribbons to frequency and symmetry (no rocking modes) seems not too hard. Just money.


Les
L M Watts Technology
 
Les,

You also need to consider a couple more things. Thicker ribbons much more prone to resonant modes, so you will need some additional damping, which of course, reduces the output even further (and affects the sound). On the other hand, they are thicker and have lower DCR, so to some extent you could make it up in the transformer. Moreover, you could combine acoustical and electrical damping, but since the resistance noise is dominating, you really need to watch that out, so the output transformer design is crucial and cannot be taken out of equation.

Best, M

P.S. Sorry, what is LISA?
 
You also need to consider a couple more things. Thicker ribbons much more prone to resonant modes, so you will need to damp them more,

Yeah, I imagine the transverse mode HF resonance can be an issue at some point. Fortunately it would be partially rejected since it involves  some areas moving forward in the gap while other parts move back.
Same thing with the transverse resonance harmonics. Looking at the BBC data you see big repeated impedance spikes, but not necessarily big response spikes at the same frequencies. Such motions I think would induce local eddy currents though. FEA may shed some light on this for me.

P.S. Sorry, what is LISA?

LISA is a low cost finite element program capable of doing dynamic analysis. For our well funded previous microphone project we used supercomputer FEA that even modelled the air loads, but I had to pay $2000 a pop for analysis runs. But it wasn't for pro/consumer audio. Audio gets little or no r&d funding, so I can't afford to do that. But I can do simpler, still useful runs on LISA and the like.

I see the ribbon as a spring design, and with most spring designs one wants every bit of the structure to store strain energy equally. Like a good leaf spring. I suspect corrugated microphone ribbons don't do this.
I think some parts will have a lot of strain energy, while other parts are dead weight mostly. Optimizing via FEA therefore might dramatically improve durability by spreading out the stress field. We'll see.

If you want, you can play with LISA as well. A demo is downloadable here:
http://www.lisa-fet.com/

the output transformer design is crucial and cannot be taken out of equation.
For sure. If one wants to make commercial ribbons one has to be a transformer maker I suspect.
Can't be throwing big money at sowter, lundahl, etc. I have some experience in that area from Shure.
Not that i'm a transformer guru or anything. But I have designed a good many magnetic circuits.

I have a question for you: Termination. I would think getting long term low ohmic contact with aluminum might be an issue. What materials have you seen used in the clamp area? Do some (Coles) actually use an aluminum compatible solder? Are some gold plated? (yes aluminum and gold can do bad things together)

Les
L M Watts Technology
 
Bleh...can't edit posts here I guess...when I said "transverse resonance harmonics" I meant longitudinal.

Also...Brian at Edcor told me that they have to hand stack nickel lams...and that just kills the price. I've dealt with mu-metal and such, and I know the slightest little bend or distortion wrecks the permeability unless you do another hydrogen anneal.

Les
 
Hi Les,

It seems we are having a nice dialog here while not many other folks are interested to participate  ;D ;D ;D

leswatts said:
P.S. Sorry, what is LISA?

LISA is a low cost finite element program capable of doing dynamic analysis. For our well funded previous microphone project we used supercomputer FEA that even modelled the air loads, but I had to pay $2000 a pop for analysis runs. But it wasn't for pro/consumer audio. Audio gets little or no r&d funding, so I can't afford to do that. But I can do simpler, still useful runs on LISA and the like.

If you want, you can play with LISA as well. A demo is downloadable here:
http://www.lisa-fet.com/

Thanks, I will look at it. Right now it is only $50...

leswatts said:
the output transformer design is crucial and cannot be taken out of equation.
For sure. If one wants to make commercial ribbons one has to be a transformer maker I suspect.
Can't be throwing big money at sowter, lundahl, etc. I have some experience in that area from Shure.
Not that i'm a transformer guru or anything. But I have designed a good many magnetic circuits.

I make my own and spent a few years to find what works, what sounds good, and what can be efficient enough to match the ribbon well... and I don't believe it is EI  ;).

leswatts said:
I have a question for you: Termination. I would think getting long term low ohmic contact with aluminum might be an issue. What materials have you seen used in the clamp area? Do some (Coles) actually use an aluminum compatible solder? Are some gold plated? (yes aluminum and gold can do bad things together)

This is one of the big issues. I actually wanted to mention it in the previous message, but somehow it slipped my mind when was writing. Every manufacturer deals with the problem their own way. Most use either brass, or Silver plated clamps. I've never seen gold plated clamps, or soldering. I actually tried solder myself. It works well, but can be tricky to find the right temperature (little hot and that thin foil melts right away, little cold and you can't deal with the solder), so it is important to find the right technique.

The worse contacts I saw were in Oktava (which often just lose the contact) and Chinese ribbons. In fact, with much thinner ribbons I am getting less overall (ribbon+termination) resistance than some 4-6 micron Chinese stuff  ::).

Best, M

 

Latest posts

Back
Top