DIY RCA BK5 Ribbon Mic- small signal output

[jannae66]

Hello dear DIY community,

First, I would like to thank you all for the many great contributions here in the forum. These have helped me a lot with my ribbon mic mods. Now I would like to start my first thread and maybe I will help someone with it.

I have a few questions regarding my DIY RCA BK5 mic. But first I wanted to tell you a little bit about the origins of the built.

After modding a few Ribbon mics, I tried to build my own one. And because there is no real cheap RCA BK5, I decided to build this. I´ve read a lot of things about the BK5, studied drawings and pictures of it and it took a long time before I even started.

It took me a lot of time to choose the different materials and to make the individual components, but it was a lot of fun. I think I built it every now and then for half a year and improved a lot of things. Like the clamps. After using acrylic with a copper strip and getting a signal that was way too weak, I made clamps out of copper that gave a much stronger signal.

The mic body is from a broken Stagg condenser mic. The labyrinth (1 meter long) made of aluminum, the motor made of acrylic glass, phase shift openings with mesh, 4 neodymium magnets (25mm x 4mm x 4mm), 1.8 micron corrugated ribbon foil, thick copper wire, blast shield made of two layers of perforated grid (2mm holes), plastic and fabric. As a test, the transformer is a cheaper one from a T-Bone RB500. A Lundahl LL2913 has already been ordered.



In my opinion, the microphone sounds really great. Whether electric guitar, acoustic guitar or vocals. Really nice forward sound with a minimal proximity effect, for a ribbon. And a realy tight pattern. But my signal is about 10 -14db weaker compared to my modded Beyerdynamic BM 85 (Beyer M500) or Bumblebee RM-6.



Now here´s my questions and maybe someone can give me a few tips:

Would it help a lot if I included a magnetic return circuit? As marked in red in the picture. I only have access to 6mm gauge cold rolled steel. In the bottom plate would be a hole for the labyrinth path.

Would it help if I integrated 1 or 2 additional magnets on each side of the magnetic return circuit? Marked in black as shown in the picture.


How much db increase can I promise from it?

Unfortunately, I can't do the magnetic return circuit around the copper brackets, because I don't have any space in the head basket.



Thank you very much and best regards!!!



Jan from Franconia, Germany

 

[dmp]

Cool project! The Bk5 is a favorite mic of mine. And I've been wanting to DIY a unidirectional ribbon for sometime. Super impressed.
In the originals, the magnets are attached to a inner body I believe, which provides the return path. Pic shows one of mine that I took apart to re-ribbon. I don't think adding magnets is right, you want a complete path of a higher permeability back around. Could you replace the acrylic with a steel part? Perhaps you could do dimensioned drawings and have it machined or 3D printed from an online company? (I'm not sure if this is possible).
Have you ensured the magnets have the field pointing in the correct direction? I made this mistake once. Sometimes the long rectangular neodymium magnets have N-S pointing out the ends.
If you can get a good SNR I wouldn't worry too much ultimately if the output is low - a vintage BK5 has a pretty low output and needs a good preamp. I have both a M500 and BK5 so I could compare levels, but I also would expect they would be similar.

How many dB of rejection do you have from the rear?
FIY, your labyrinth looks more like the 77 or ku mics, I think the BK5 had a tube. It also has vents, as shown in the diagram.

 

[Marik]

Hello dear DIY community,

First, I would like to thank you all for the many great contributions here in the forum. These have helped me a lot with my ribbon mic mods. Now I would like to start my first thread and maybe I will help someone with it.

I have a few questions regarding my DIY RCA BK5 mic. But first I wanted to tell you a little bit about the origins of the built.

After modding a few Ribbon mics, I tried to build my own one. And because there is no real cheap RCA BK5, I decided to build this. I´ve read a lot of things about the BK5, studied drawings and pictures of it and it took a long time before I even started.
View attachment 106136
It took me a lot of time to choose the different materials and to make the individual components, but it was a lot of fun. I think I built it every now and then for half a year and improved a lot of things. Like the clamps. After using acrylic with a copper strip and getting a signal that was way too weak, I made clamps out of copper that gave a much stronger signal.

The mic body is from a broken Stagg condenser mic. The labyrinth (1 meter long) made of aluminum, the motor made of acrylic glass, phase shift openings with mesh, 4 neodymium magnets (25mm x 4mm x 4mm), 1.8 micron corrugated ribbon foil, thick copper wire, blast shield made of two layers of perforated grid (2mm holes), plastic and fabric. As a test, the transformer is a cheaper one from a T-Bone RB500. A Lundahl LL2913 has already been ordered.

View attachment 106153View attachment 106152View attachment 106139View attachment 106140View attachment 106141View attachment 106142View attachment 106143View attachment 106150View attachment 106144View attachment 106147View attachment 106148View attachment 106149View attachment 106145View attachment 106146

In my opinion, the microphone sounds really great. Whether electric guitar, acoustic guitar or vocals. Really nice forward sound with a minimal proximity effect, for a ribbon. And a realy tight pattern. But my signal is about 10 -14db weaker compared to my modded Beyerdynamic BM 85 (Beyer M500) or Bumblebee RM-6.



Now here´s my questions and maybe someone can give me a few tips:

Would it help a lot if I included a magnetic return circuit? As marked in red in the picture. I only have access to 6mm gauge cold rolled steel. In the bottom plate would be a hole for the labyrinth path.

Would it help if I integrated 1 or 2 additional magnets on each side of the magnetic return circuit? Marked in black as shown in the picture.
View attachment 106135

How much db increase can I promise from it?

Unfortunately, I can't do the magnetic return circuit around the copper brackets, because I don't have any space in the head basket.



Thank you very much and best regards!!!



Jan from Franconia, Germany

Awesome work!

Indeed, the return circuit would be the first thing to look at for some good at least 3dB signal improvement. Judging from the ribbon size you could probably use a higher ratio transformer. If you can measure output impedance that would be great way to determine the actual ratio. I would make the horn with smooth (not stepped) walls. The last thing to try is your labyrinth stuffing looks a bit too dense, loading the ribbon too much and absorbing a lot of energy. I’d try some loose acousta stuff. From the beginning try putting it only into the straight passages (not the elbows). All that should considerably increase the output.

Hope it helps.

Best, M

 

[jannae66]

Cool project! The Bk5 is a favorite mic of mine. And I've been wanting to DIY a unidirectional ribbon for sometime. Super impressed.
In the originals, the magnets are attached to a inner body I believe, which provides the return path. Pic shows one of mine that I took apart to re-ribbon. I don't think adding magnets is right, you want a complete path of a higher permeability back around. Could you replace the acrylic with a steel part? Perhaps you could do dimensioned drawings and have it machined or 3D printed from an online company? (I'm not sure if this is possible).
Have you ensured the magnets have the field pointing in the correct direction? I made this mistake once. Sometimes the long rectangular neodymium magnets have N-S pointing out the ends.
If you can get a good SNR I wouldn't worry too much ultimately if the output is low - a vintage BK5 has a pretty low output and needs a good preamp. I have both a M500 and BK5 so I could compare levels, but I also would expect they would be similar.

How many dB of rejection do you have from the rear?
FIY, your labyrinth looks more like the 77 or ku mics, I think the BK5 had a tube. It also has vents, as shown in the diagram.

Hello dmp,
thanks a lot for your appreciation and your hints! Unfortunately, I´m not familiar with 3D-drawings and I wanted to keep the costs as low as possible. But I was thinking about 3D-printing the labyrinth (if there will be a second mic) because that took really long to make. The magnets are definitely in the right direction.
For electric guitar the output is really no problem but when recording acoustic or vocals it would be good to have more. Even with a Fet-Head there is too much background noise that I don´t get with other ribbon mics, which have better output.
Unfortunately I have no possibility to measure the rejection from the rear, or I don´t know how to do it without a proper absorbed room. Do you have any suggestions for it?

Yes, the labyrinth is more a 77dx or ku4 style, because for the original BK5 labyrinth I would have to drill out a big cavity in the middle for the transformer and make the labyrinth surrounding it (see picture) and that wouldn´t be efficient material wise.

I think you mean the phase shift openings with the vents? I have two holes under the clamps for it. Inspired by the AEA KU5A. You see it in the vid when you slow down on 2:11 min:

You tube: Introducing the AEA KU5A Supercardioid Ribbon Microphone​

there´s also a small hole under the bigger hole in the KU5A, but I don´t know what advantage that would have. I don´t have a development team to figure that out.

I couldn´t exactly figure out how it is made on the original. Is there a huge gap between the magnets only covered with cloth? What would be the benefit? I thought that it´s too much of a open area, that would be too sensitive for incoming noise.

 

[jannae66]

Awesome work!

Indeed, the return circuit would be the first thing to look at for some good at least 3dB signal improvement. Judging from the ribbon size you could probably use a higher ratio transformer. If you can measure output impedance that would be great way to determine the actual ratio. I would make the horn with smooth (not stepped) walls. The last thing to try is your labyrinth stuffing looks a bit too dense, loading the ribbon too much and absorbing a lot of energy. I’d try some loose acousta stuff. From the beginning try putting it only into the straight passages (not the elbows). All that should considerably increase the output.

Hope it helps.

Best, M

Hello Marik,
I´m really happy that you wrote back! Your professional background is a real inspiration, cause you managed to sell highly professional mics (what I´ve read) with a great price point all over the world.
I only have a 20 min drive to Thomann. I must check out your products there!
So thanks a lot for your appreciation and your hints, also!

If I have time next week, I start to implement the return circuit and rebuilt the horn. Unfortunately I don´t have such sensitive multimeter to measure the ribbon impedance. Or do you have a hint, how to do it another way?
Do you mean by acousta stuff some lower density foam?
Do you think that the Lundahl LL2913 would be a good match or is there a better transformer for my built?

I also have a fine metal mesh in front and back of the ribbon for protection. Do you think this is also reducing output?


So, everybody thanks a lot! It´s really great to have such a forum, where you can get so many tips from people that know their business- and in such a quick way!
I´m only doing it for my own as a hobby and the gap between guys like you and me regarding the development of a mic is too big, but very inspirational. I´ve built guitars, amps, cabinets, guitar effects, I´m recording and mixing my band, so I really love to learn new things and make things on my own. I think the biggest influence is my father, cause he had to do everything by himself to save money. Maybe he only passed on the will to save money to me. I´m lucky to have a good job as a teacher now, but you see that the lack of money can really open up your horizon.



Best regards,
Jan

 

[k brown]

Great project - best of luck!

Made me think: has anyone tried making a cardioid (or even varipattern!) ribbon mic by placing an SDC omni capsule on top of a ribbon motor and combining outputs (like the Altec 'bird cage', but with a condenser, rather than dynamic omni)?
https://coutant.org/altec639/

 

[jannae66]

Great project - best of luck!

Made me think: has anyone tried making a cardioid (or even varipattern!) ribbon mic by placing an SDC omni capsule on top of a ribbon motor and combining outputs (like the Altec 'bird cage', but with a condenser, rather than dynamic omni)?

Thank you!
Marik also developed such a mic: Samar Audio Design TF08 | RecordingHacks.com

 

[k brown]

Thank you!
Marik also developed such a mic: Samar Audio Design TF08 | RecordingHacks.com

He'd be the man to do it! Stellar!
Nice write-up on the prototype, too. Samar Audio Design Hybrid | RecordingHacks.com

But has anyone done a DIY project (or, to derail the thread even further [sorry] a DIY of something like the Josephson C700 8+omni condenser; what would be the advantage of an 8+omni varipattern over a dual diaphragm LDC - other than sep outs of the two capsues?)?

 

[Murdock]

I would also be very interested in a ribbon and omni scd mic!
The "advantage" would be two "perfect" transducer combined, in the sense that a ribbon is a "natural" velocity transducer and the single small diaphragm condenser is a "natural" pressure transducer. Whereas the dual ldc diaphragm is "only" a compromise. I think this subject was discussed in the "Why you should never use multi pattern mics" thread.
I used quotation marks because each has it's own pro's and con's. So I wouldn't say it is better per se.
But would be very interesting nonetheless.
The question is how to combine the two signals as the output level and impedance is quite different I would guess.

 

[dmp]

I have a 639 birdcage and while it is a cool mic, it isn't in the same realm as a BK5 or KU RCA, imo
The condenser is so different from the ribbon that they combine into something much different than a single ribbon sound.
The unidirectional RCAs with the labyrinth sound amazing.

 

[dmp]

For electric guitar the output is really no problem but when recording acoustic or vocals it would be good to have more. Even with a Fet-Head there is too much background noise that I don´t get with other ribbon mics, which have better output.
Unfortunately I have no possibility to measure the rejection from the rear, or I don´t know how to do it without a proper absorbed room. Do you have any suggestions for it?

You can get a sense of the rejection just using a test tone. For instance, I have a guitar tuner that does a 440 Hz tone. Put the test tone in front of the mic, measure the dB, then behind the mic, and measure the dB. The difference gives you a rough idea of the rejection.

 

[dmp]

So for reference, I just measured this with a excellent condition BK-5
The rear is -19dB compared to the front with a 440 Hz test tone about 3 inches away (from the front and from the back).
And the output level of the BK-5 is about 0.5-1dB higher than a M500 with a test tone comparison but they are very close.

 

[Marik]

If I have time next week, I start to implement the return circuit and rebuilt the horn. Unfortunately I don´t have such sensitive multimeter to measure the ribbon impedance. Or do you have a hint, how to do it another way?
Do you mean by acousta stuff some lower density foam?
Do you think that the Lundahl LL2913 would be a good match or is there a better transformer for my built?

I also have a fine metal mesh in front and back of the ribbon for protection. Do you think this is also reducing output?

Jan,

For the impedance measurement you need an LCR meter. Many modern ones are pretty cheap. Ideally, just get something, which measures at different frequencies, say 100Hz, 1kHz, 10kHz, and 100kHz. Or you can use an old method of 6db signal drop.

Acousta stuff usually used in acoustical filling for the speakers--white mineral cotton. Sometimes something similar used for the pillow filling... not sure how it is called.

The original BK5 ribbon is very small (on par with Beyer ribbons) and much more prone to resonances. You are using a bigger ribbon, which is good in terms of damping, so you don't need as much load. You already have considerable load on the ribbon from the labyrinth. My feeling you could move the mesh a bit further from the ribbon and use only in the front. That would give you protection, but without extra loading the ribbon, which means some extra output.

Speaking of labyrinth, I think with all H. Olson's genius, he just miscalculated it, missing the fact the filled elbows do not pass the signal like a straight line and the signal just fires back at the elbows. Interesting, the Beyer originally was making their ribbons with labyrinth (M320/360), but then changed the design to a straight tube like in M160/260/500. While the tube is short, with acoustic stuffing it is efficient enough to emulate infinite load. Likewise, instead of labyrinth that would be worth trying a straight tube. Not practical for a commercial design, but for DIY you could take say, 1 meter tube hanging out the mic (or you could coil it inside the mic body) which would be a much better idea and easier implement than a labyrinth. Just stuff it lightly with acousta stuff, making denser towards the end and that would have by far better properties than a labyrinth.

Best, M

 

[jannae66]

Jan,

For the impedance measurement you need an LCR meter. Many modern ones are pretty cheap. Ideally, just get something, which measures at different frequencies, say 100Hz, 1kHz, 10kHz, and 100kHz. Or you can use an old method of 6db signal drop.

Acousta stuff usually used in acoustical filling for the speakers--white mineral cotton. Sometimes something similar used for the pillow filling... not sure how it is called.

The original BK5 ribbon is very small (on par with Beyer ribbons) and much more prone to resonances. You are using a bigger ribbon, which is good in terms of damping, so you don't need as much load. You already have considerable load on the ribbon from the labyrinth. My feeling you could move the mesh a bit further from the ribbon and use only in the front. That would give you protection, but without extra loading the ribbon, which means some extra output.

Speaking of labyrinth, I think with all H. Olson's genius, he just miscalculated it, missing the fact the filled elbows do not pass the signal like a straight line and the signal just fires back at the elbows. Interesting, the Beyer originally was making their ribbons with labyrinth (M320/360), but then changed the design to a straight tube like in M160/260/500. While the tube is short, with acoustic stuffing it is efficient enough to emulate infinite load. Likewise, instead of labyrinth that would be worth trying a straight tube. Not practical for a commercial design, but for DIY you could take say, 1 meter tube hanging out the mic (or you could coil it inside the mic body) which would be a much better idea and easier implement than a labyrinth. Just stuff it lightly with acousta stuff, making denser towards the end and that would have by far better properties than a labyrinth.

Best, M

Marik,
thank you very much for your hints. So you mean, all the hard work I put into the labyrinth was for nothing and doesn't do much? That is hard.
Why is the AEA KU5A than still made with the superior labyrinth design? Only to stay true to the RCA design and accepting the negative things?

I've also thought about constructing a spiral labyrinth before that. It was already done that way with the RCA 77A, but then changed.

So I thought the later KU4 labyrinth design would be better. Damn.
The spiral shape would work well as a 3D print. I'm going to implement all the other tips first and maybe someday I'll make a spiral labyrinth.
Thank you again and best regards,
Jan

 

[flashlamp]

did you use a lathe or just a drill press?

 

[jannae66]

did you use a lathe or just a drill press?

Hi flashlamp,
just a drill press and dremel. wish I had a lathe...
best regards,
Jan

 

[dmp]

So you mean, all the hard work I put into the labyrinth was for nothing and doesn't do much? That is hard.

If the rejection from the rear is good and the off-axis frequency response sounds good, then your labyrinth is good. If the rear load is imperfect, it will affect these characteristics. Have you done any measurement of the rear rejection yet?

 

[jannae66]

If the rejection from the rear is good and the off-axis frequency response sounds good, then your labyrinth is good. If the rear load is imperfect, it will affect these characteristics. Have you done any measurement of the rear rejection yet?

No, I'm sorry. I disassembled the mic before I could do a test.

 

[Marik]

Marik,
thank you very much for your hints. So you mean, all the hard work I put into the labyrinth was for nothing and doesn't do much? That is hard.

As mentioned earlier, first try to fill the labyrinth much looser and without the filling at all in the elbows. Then try the straight tube coupled to the motor loosely filled (and experimenting between filled/not filled) and see if there are any changes. That will give you an idea for the optimal arrangement.

I've also thought about constructing a spiral labyrinth before that. It was already done that way with the RCA 77A, but then changed....
The spiral shape would work well as a 3D print. I'm going to implement all the other tips first and maybe someday I'll make a spiral labyrinth.

The spiral labyrinth you mentioned has the same 'elbows'. Earlier I mentioned coiling the tube exactly for the reason to eliminate those elbows.

If the rejection from the rear is good and the off-axis frequency response sounds good, then your labyrinth is good. If the rear load is imperfect, it will affect these characteristics.

The main purpose of the labyrinth is to provide an infinite load to the ribbon to get an optimal output and uniform frequency response. Obviously, since the output is low the load is not optimal.

Best, M

 

[dmp]

The main purpose of the labyrinth is to provide an infinite load to the ribbon to get an optimal output and uniform frequency response. Obviously, since the output is low the load is not optimal.

I would suggest verifying the mic works well without the labyrinth installed, with good level, first. This will verify the output level and characteristics (fig8 of course), and eliminate many potential issues with the magnets, ribbon, and transformer.
Then install the labyrinth and evaluate the rear rejection and off-axis response.
I think it is may not be the labyrinth that is causing the low output.