PCB with integrated ribbon motor, transformer, pre-amp, XLR

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

rubinstu

Active member
Joined
Jan 10, 2019
Messages
37
While working on my DIY ribbon mic, particularly on a custom PCB for an integrated pre-amp, I thought of an idea. I can't believe that I'm the first person to consider this, but I did not see any examples from a cursory Google search.
I'm thinking of making a single PCB that integrates all of the components of a ribbon mic. It would have cutouts to hold the magnets and a slot for the ribbon, some mechanism to secure the magnets and ribbon, mount for the transformer, pre-amp circuitry, and an XLR connector.
I think a single part like this could actually be smaller and easier to work with mechanically. (I hate all those loose wires flopping around, soldering wires to PCBs, etc.) The whole thing can be shoved into a tidy housing as one piece.
The PCB itself can offer some EMI shielding with a ground plane.
You can design just about any shape you want and not have to worry about how it will be machined.
FR4 is fairly stiff stuff, but obviously not as strong as many metals.
Issues with resonances?
If this exists already, I'd love to know!
Thanks
 
Are you going to have any steel as your return path for the magnetics ? I have worked on a few ribbon mic designs where we have used a PCB instead of magnet wire to speed up assembly time.
 
Cool idea - it would probably be specific to a particular mic body, yes?
The internal shape & resonances are important once you are trying to get to 'greatness'
May also be a cool route to a unidirectional ribbon, which has been on my dream to-do list for sometime.
 
Are you going to have any steel as your return path for the magnetics ?
@a soBer Newt, I hadn't really thought about adding any steel for a "return path". Honestly, I'm not 100% sure what you mean in this context. If you mean to "shape" the field, then then answer is no. With the two magnetic bars closely spaced and "north facing south", there should be a pretty uniform field in the path of the ribbon. Around the outside of the magnets, I expect a pretty traditional "looping" pattern.
1675445029091.png(Ref: SchoolPhysics Magnetic fields)

But if I'm missing some important detail here, please let me know!

...it would probably be specific to a particular mic body...
@dmp, Maybe. I was actually thinking that it could be made so you can put them in different geometries, but with the same PCB. For example, mayb two can "snap" together one way to make a Blumlein configuration, or another way to get two ribbons in parallel. Or just use one for a basic setup.
 
@a soBer Newt, I hadn't really thought about adding any steel for a "return path". Honestly, I'm not 100% sure what you mean in this context. If you mean to "shape" the field, then then answer is no. With the two magnetic bars closely spaced and "north facing south", there should be a pretty uniform field in the path of the ribbon. Around the outside of the magnets, I expect a pretty traditional "looping" pattern.
View attachment 104352(Ref: SchoolPhysics Magnetic fields)

But if I'm missing some important detail here, please let me know!


@dmp, Maybe. I was actually thinking that it could be made so you can put them in different geometries, but with the same PCB. For example, mayb two can "snap" together one way to make a Blumlein configuration, or another way to get two ribbons in parallel. Or just use one for a basic setup.
For a ribbon motor for maximum sensitivity and highest possible output you’d want a ‘proper’ steel frame. That is, the field formed not only between the closest to each other S/N poles, but also between outer poles. That outer field partially cancels the main one. The steel frame will increase distance between outer poles, which together with return path would significantly reduce parasitic cancelling field, effectively increasing output—in the ribbon mics every dB counts.

Best, M
 
@Marik, I think I mostly get the gist of your comment, but I'm still not 100% sure I got the details. My VERY simplistic interpretation is that one may want "shape" the magnetic field so that it's optimally dense through the path of the ribbon, and minimize the "stray" flux.
I had not really thought about the stray path adding to any kind of cancelation, but I think I can get it. I think the analogy may be that you need a speaker in a box or on an infinite plane, otherwise the air pressure from the speaker ends up just sucking air from behind the speaker, canceling the sound wave. I think you're saying the same effect happens with the mic's magnetic field.
So, with that in mind, would a horseshoe or split-ring magnet be more affective because the magnetic fields are highly contained and focused?
It just keeps getting more interesting! Thanks!
 
Here is a quick 2D Simulation using FEMM Mag1 shows a simple ribbon design with a steel return path, Mag2 shows what happens with out that return path. You can see the difference in magnetic density in the two graphs. If you haven't read this BBC article it's a good read and has some interesting insights.
 

Attachments

  • Mag2A.PNG
    Mag2A.PNG
    12.7 KB · Views: 0
  • Mag1A.PNG
    Mag1A.PNG
    11.8 KB · Views: 0
  • Mag2.PNG
    Mag2.PNG
    27.8 KB · Views: 1
  • Mag1.PNG
    Mag1.PNG
    17.1 KB · Views: 1
shows a simple ribbon design with a steel return path
This is really interesting - something I hadn't thought of before. I used to have a B&O ribbon mic that just had two rectangular magnets, held with plastic end pieces (no return path) - I never liked it. But both these mics look to have a circular return path below the ribbon gap? (will read the BBC article later)
 

Attachments

  • coles4038.jpg
    coles4038.jpg
    30.6 KB · Views: 4
  • bk11.jpg
    bk11.jpg
    21 KB · Views: 4
Here is a quick 2D Simulation using FEMM Mag1 shows a simple ribbon design with a steel return path, Mag2 shows what happens with out that return path. You can see the difference in magnetic density in the two graphs. If you haven't read this BBC article it's a good read and has some interesting insights.

Thanks for that link to FEMM. I was looking for something like that last year, but I did not find the right search term to get me to FEMM.
I had an idea to do something similar to the OP, but using a 3D printed motor structure rather than PCB material. After some research I decided that it would not reach the performance I wanted without a steel structure to provide magnetic return path, and I put it aside at that point since I did not have access to any machine tools that could handle steel. It will be nice to have a simulation tool to actually evaluate how thin of a steel structure can be used before it starts to saturate.
 
How about an adhesive backed layer of mu metal over the PCB to create the return path for the magnets?
 
I think the analogy may be that you need a speaker in a box or on an infinite plane, otherwise the air pressure from the speaker ends up just sucking air from behind the speaker, canceling the sound wave. I think you're saying the same effect happens with the mic's magnetic field.
So, with that in mind, would a horseshoe or split-ring magnet be more affective because the magnetic fields are highly contained and focused?

Rubinstu,

The analogy with a speaker is a bit different, as it is frequency dependent. Because of the wavelength the effect is pronounced on the low end. The magnetic field in ribbon microphone is uniform across entire bandwidth.

The problems with the horseshoe are 1. it disturbs the fig8 pattern. You could do cardioid, or omni, but it considerably complicates construction and because of the air loading you get less output, 2. I never seen horseshoe neodymium magnets, so Alnico will have considerably lower field, 3. Custom horseshoe Neodymiums will be horrendously expensive, and 4. For horseshoe you will need to make poles, which will also dissipate some energy.



I used to have a B&O ribbon mic that just had two rectangular magnets, held with plastic end pieces (no return path)

DMP,

Most likely you are talking about BM6, or BM7 (together they formed stereo BM5). They actually, do have quite effective return path, which is the steel body itself--quite ingenious solution. The main problem all those old Alnicos lost most of their field long ago.

Best, M
 
DMP,

Most likely you are talking about BM6, or BM7 (together they formed stereo BM5). They actually, do have quite effective return path, which is the steel body itself--quite ingenious solution. The main problem all those old Alnicos lost most of their field long ago.

Best, M

Yes, it was a BM5. Replaced the weak Alnicos with a neodymium. And it had high output. Interesting that the cylindrical steel body would be effective since there is some air gap between the body and magnets. If I remember right, iron has a permibility ~5000 times higher than an air gap.

The problems with the horseshoe...
Isn't the coles magnet pictured above horseshoe? Or is the horseshoe a steel return path?
 
Here is another where the magnets are joined to a steel horseshoe for the return path (I think).
I would like to try building a unidirectional ribbon like this. Possibly an epoxy board for the overall structure and attaching the ribbon clamps, with magnets and some kind of return path.
I don't see why the two ribbons would require separate magnets. build with a single pair of rectangular neodymium magnets?
 

Attachments

  • IMG_20200304_212941.jpg
    IMG_20200304_212941.jpg
    4.4 MB · Views: 3
Yes, it was a BM5. Replaced the weak Alnicos with a neodymium. And it had high output. Interesting that the cylindrical steel body would be effective since there is some air gap between the body and magnets. If I remember right, iron has a permibility ~5000 times higher than an air gap.

There is a slot machined in the body to make sure the magnets fit tightly, so for all practical purposes the air gap is minimal...

Isn't the coles magnet pictured above horseshoe? Or is the horseshoe a steel return path?...

Here is another where the magnets are joined to a steel horseshoe for the return path (I think).
I would like to try building a unidirectional ribbon like this. Possibly an epoxy board for the overall structure and attaching the ribbon clamps, with magnets and some kind of return path.
I don't see why the two ribbons would require separate magnets. build with a single pair of rectangular neodymium magnets?

Those huge (particularly Coles) Alnicos are unavailable and most likely are custom made. The second design is particularly inefficient, as such big chunks of metal dissipate a lot of energy. For the unidirectional the U shaped return steel channel with facing each other Neodymiums at the ends as poles should be stronger and by far cheaper. Then you can port it and shape response in a few different ways--labyrinth, absorbing materials, etc.

Best, M
 
The second design is particularly inefficient, as such big chunks of metal dissipate a lot of energy.
The BK-11 picture above? I think the BK-11 had a higher output than the Coles.
But interesting that it lost the continuous U shaped steel return path of the 44 (top and bottom!); the BK-11 was the successor to the 44.
 

Attachments

  • RCA44.jpg
    RCA44.jpg
    109.4 KB · Views: 3
The BK-11 picture above? I think the BK-11 had a higher output than the Coles.
But interesting that it lost the continuous U shaped steel return path of the 44 (top and bottom!); the BK-11 was the successor to the 44.

Reminds me Chevy 57--huge, inefficient and.. powerful))). BK11 has longer front to back path (=higher output, less top end), more narrow gap (more field), and the ribbon is longer (the more the area, the more the output). One of the reasons to drop the 44 was financial (Olson always was conscious of sales), as two huge horseshoes were considerably more expensive than two tapered with huge steel returns.
 
Wow, this topic certainly took an interesting turn! I'm going to have to checkout FEMM. You have to respect a piece of software that has this in the Overview section:
Some familiarity with electromagnetism and Maxwell’s equations is assumed
Respect!
 
@a soBer Newt, I played a little with FEMM last night. Cool stuff! Would you mind posting your source file(s) for the model you made? I'd really like to see the details. Thanks!
 

Latest posts

Back
Top