DIY M50 capsule

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

kingkorg

Well-known member
Joined
Apr 15, 2017
Messages
2,972
Location
Norway
So, i found some info on m50 capsules. One of recent versions is basically km84 omni type capsule, which is again fairly simmilar to chinese SDC capsules.

Rode made Faulkner series m50 type mic which has the same backplate as chinese ones. Plastic backplate is sealed for omni, so i sealed mine with epoxy. No spacer between backplates in omni version.

I grinded off the front part of the capsule. I will test it with honeycomb type of grille in front eventually.

In this arrangement known 8k peak of chinese capsule disappears.

I have some shielding issues at the moment, but the measurements are pretty much spot on.

The sphere is just a rubber ball.

 

Attachments

  • 20190809_165154_crop_453x453.jpg
    20190809_165154_crop_453x453.jpg
    142.1 KB · Views: 83
Rode's Faulkner omni capsule construction. Identical to mine, except the honeycomb front.
 

Attachments

  • Screenshot_20190809-182320_Firefox.jpg
    Screenshot_20190809-182320_Firefox.jpg
    92.7 KB · Views: 132
Last edited:
Details on Neumann's capsule

http://www.saturn-sound.com/Curio's/neumann's%20kk50%20-%20later%20type%20k50%20capsule.htm


My initial measurements show that the capsule behaves exactly as it should.

Smooth boost of HF at 0 degrees.
Than it becomes hyper cardioid at higher frequencies at 90 and 180 degrees. Remains omni below 1k.

At 180 degrees it is basically flat up until HF roll off.

Without the ball the capsule has omni response at distance throughout the frequency range, and HF response is not as high as at 0 degrees with ball.

Again, this mic is to be used far from a source. HF boost of the capsule is required as high frequency range gets lost (converted into heat) at distance. At least this is Neumann's theory behind this mic.

Reflections sound darker, but flat(ish) and don't have resonant peak off axis and 180° unlike cardioids/hypercardioids.

Purple 0 degrees
Green 90 degrees
Yellow 180 degrees
 

Attachments

  • 20190809_184728-1600x1004.jpg
    20190809_184728-1600x1004.jpg
    437.5 KB · Views: 78
Just for comparison, this is a cardioid capsule under same conditions. Dip in the mid range, resonant peaks at higher frequencies that add up to front side response, and we get peaks at 0° at exact same places.

Opposte happens with this omni with the sphere.
 

Attachments

  • 20190809_185544-1600x953.jpg
    20190809_185544-1600x953.jpg
    355.3 KB · Views: 50
This is Neumann's published m50 response, which is terribly smoothed and not very detailed, but it does show rise at HF starting at about 2k and rises untill 15k. All of this is pretty close to my diy variant.

For about 20$ and some DIY, i am honestly stunned by the results.
 

Attachments

  • Screenshot_20190809-190825_Firefox.jpg
    Screenshot_20190809-190825_Firefox.jpg
    60.6 KB · Views: 46
I’m impressed! They’d be fun to play with!

I heard Ben at BeesNeez has been working on one... I’ve been waiting eagerly, as I have a pair of his first 49/50 bodies that have been waiting for them since I already have some 49-types in SYT-5 bodies.
 
One huge advantage of these capsules is that they are very easy to re-skin. No tools required, besides membrane material. I use two component epoxy glue from Bison for gluing to the ring.

So if anyone manages to source sub micron nickel foil...
 
I sourced very expensive 0.8-micron 99.1% pure aluminum with absolutely no pinholes, medical-grade, but they won’t sell it to me even when following their guidelines to allow me to purchase it!

Everyone except for one person have said it’s impossible to work with anyway. The one person was confident enough to try with me knowing it’s experimental.

Anyhow, maybe they’re saving me some hard-earned cash without even knowing it, but I still wish I could have tried and failed instead of not being able to try at all.
 
This guy has 1 micron nickel. The price is crazy, but i asked if he would sell a 10cm sample.


https://rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.com%2Fulk%2Fitm%2F223575617098
 
Interesting thread!
Which chinese capsule did you modify? Response charts look great.
All the pictures of the Rode TFM50 mic that i  found show another capsule type, with screws in front, like the kk50 aluminium capsule or the schoeps caps. Where did you find the picture above? To me, it looks like the (cardioid) capsule of the new TF5 (kinda kk84 clone)
 
I wasn't quite clear, i was talking about backplates on Rodes Faulkner omni capsule. It is the same as cardioid, except the omni has no holes in second backplate. It is the same backplate as the one found in these chinese capsules. Neumann's later version has the same construction, difference being groove pattern in backplate.

Not sure if they use the same backplate on Rode TFM50.

The capsules i used come from Pronomic SCM1. However the same ingredients can be found in MXL601, Takstar CM60, Behringer B5 etc. The only difference is front grille, and spacers. None of which used in my omni, nor im Neumann's or Rode design.

 
 
How many dB per division is your measurement?

Are these the same capsules people were looking to source in the Pronomic mic thread? Those newer ones that look like the ones from the iSK BM-80?
 
Sadly i can not give you the answer as my rig is not calibrated. I use it as a relative measurement tool (compared to other mics), and it's the reason why i provided regular cardioid responses to make the point.

However my subjective aproximation is that the high end boost of my omni capsule at 0° is tops at about 7db at 30 cm distance from the speaker.

Don't think it even matters, as no manufacturer ever provided true graphs.

According to Neumann this peak should be reduced at distance. I can not confirm this as at 10m response would fall apart. I will just have to listen to it once i finish the build with adequate source. Which will probably be a far mono drum room mic.

The graphs are calibrated through ecm8000 mic, so when you look at part of the response response and it looks flat-ish it means it is.

The capsule is the one with "the wrong" type of grille. Not the km84 lookalike.
 
This is a bit off topic, but how much of the presence boost with a capsule like this comes from an incorrect grille type? Even with mics from the same manufacturer (in my experience it's iSK) and similar capsule sizes, there's a much bigger boost with the slotted grille. Not that I trust their frequency plot too much, but the I think the BM-80 advertised a response similar to the iSK Pearl (where the biggest boost is around 4 dB in the high end), and the Little Gem/CM-10 has a 6 dB peak instead. I'm speaking about just the cardioid and super cardioid capsules of the Little Gem/CM-10 since the omni capsule is advertised as being very flat (I'm not sure I actually believe this, but I lack the equipment to do a proper measurement). There's a good chance the capsules are very different though since the BM-80 costs twice as much as the CM-20, and the same as the Little Gem/CM-10.

Also, is the self noise of the capsule on a different body still 20 dBA, or does a good amount of the self noise of the SCM1 come from the preamp in the body?
 
I really can not say and be sure, as a lot of it has to do with backplate hole size, spacer thickness between the backplates, etc. Cardioids are way more complex. Which is the reason you dont see a lot of truly flat cardioids.

Front grille has a lot to do with off axis response as well, and how non 0° sound comming from the front gets distributed.

That is why i might try this with a honeycomb style grille in front.

There is always this capsule as an option, which looks like the third version of m50 capsule. Costs a bit more, but doesnt need to be modified.


https://www.bhphotovideo.com/c/product/470261-REG/Rode_NT45_O_NT45_O_Omnidirectional_Replacement_Capsule.html
 

Attachments

  • Screenshot_20190810-054713_Firefox.jpg
    Screenshot_20190810-054713_Firefox.jpg
    367.9 KB · Views: 29
Obviously there are solutions like this, which are very simple to try out.

However i strongly believe these spheres should not be used with hyper or cardioid mics, as it leaves an empty resonant chamber behind the capsule.

Since the holes of backplates are not sealed i can not imagine what kind of havoc it would do to FR of the mic, let alone pattern response.
 

Attachments

  • Screenshot_20190810-065559_Firefox.jpg
    Screenshot_20190810-065559_Firefox.jpg
    126.5 KB · Views: 36
Back
Top