Neumann M49 Clone : D-M49c and D-M49b Tube Microphone Build Thread. (+Sample)

[poctop]

No necessarily. You need to measure the voltage drop. ie; the difference between V on each side of the 100k resistor. So if B+ is 120V on the B+ side of the plate resistor and you measure 50V at the plate then you get 70V of drop. I=E/R or 70mA=70/100k.

Cheers,
jb

that would be more something like 0.7ma not 70ma  ,
if you measure the voltage at the anode (A) or plate directly you should have somthing like in the 50V range,
or measuring drop at R10 will give you the relationship demonstrated above by JB,
Best,
Dan,

 

[OneRoomStudio]

I'm thinking about trying to build up a 49b using a 5718 rather than a 5480. (5718 datasheet: http://frank.pocnet.net/sheets/079/5/5718.pdf)

Apparently the 5718 was a favorite AC701k replacement tube of Oliver Archut's (see here: http://medias.audiofanzine.com/files/archive-ac701-substitute-471341.pdf), and others have confirmed that it sounds great. I have a GE and an RCA 5718 lying around, a Dale M7, and a donor body. Seems like a no-brainer. Anyone have any experience with a 5718 in this design?

 

[Melodeath00]

Nice find, OneRoomStudios! Thanks for posting

 

[adrian]

I'm Currently finishing  2  PSU and cables for this nice project, 8)
I have sommercable SC-OCTAVE TUBE wire, it have 5 AWG 26 (for audio and control) and 2 AWG 20 conductors,
I used the 2 AWG20 conductors for B+ and 0V,
is that OK?

 

[0dbfs]

I'm Currently finishing  2  PSU and cables for this nice project, 8)
I have sommercable SC-OCTAVE TUBE wire, it have 5 AWG 26 (for audio and control) and 2 AWG 20 conductors,
I used the 2 AWG20 conductors for B+ and 0V,
is that OK?

Use the larger wires for heater connections since they have the most current.
Cheers,
j

 

[adrian]

OK!
so between 0V (7) and H+ (4),

thanks a lot!!

 

[Phrazemaster]

Hey gang, I've run across several suggested mods to the M49 that supposedly improve bass/high freq response as well as correcting phase shift issues.

Here they are summarized; note they apply equally to B and C versions:

[list type=decimal]
[*]Remove R3 or make it much higher – eliminates 40Hz bass rolloff as well as corrects phase shift (Klaus)
[*]Remove C6 – eliminates 18kHz high end rolloff and phase distortion (Klaus and others)
[*]Make cap C4 2pf or eliminate  - improves output and freq. response (David Bock, Nyqvist here, and others)
[/list]

Here are some (but not all) relevant links:

Remove R3 or make it much higher:

http://www.neumann.com/forums/view.php?bn=neumann_archive&key=990963033&v=f (plus search this forum)

"...R3 can be gradually increased in convenient stages: first, leave R3 at 5 meg Ohms, and put a 1 meg resistor in series with it, for a total feeback resistance of 6 meg. Then increase the added resistor's value to 2, 3, and 4 meg. You will notice that at around 7 meg total resistance it becomes increasingly hard to hear a difference in the mic's low end response between this very slight feedback and no feedback at all (which would be resistors disconnected altogether) Kind regards, Klaus Heyne
P.S.: I would always try to run the mic as wide open as possible, i.e. R3 out or only minimally engaged; that's when phase shift is at its lowest."

Remove C6:

... Last but not least I removed C6. This opened the mic up added more air and seemed to balance the whole shooting match a bit better.

https://groups.google.com/forum/#!topic/rec.audio.pro/jQC3wLhKWo0

"3.  Neumann M49, cap. C6.
In this case cap. C6 together with the output impedance of the amplifying
stage (it can be considered as an an active resistance) establish first
order LP (low pass) filter that reduces signal at high frequencies.
Measuring the circuitry, we got the following results:
    2 kHz  - 0.04 dB
    4 kHz  - 0.12 dB
    8 kHz  - 0.43 dB
  10 kHz  - 0.65 dB
  16 kHz  - 1.48 dB
  20 kHz  -2.14 dB
31.5 kHz  -4.08 dB

Capacitance C6 brings the following phase distortion:
    2 kHz +  4.1 Deg.
    4 kHz +  8.2 Deg.
    8 kHz + 15.9 Deg.
  10 kHz + 19.5 Deg.
  16 kHz + 29.1 Deg.
  20 kHz + 34.3 Deg.
31.5 kHz + 44.5 Deg.

It is preferable to remove this capacitance."


Reduce or Remove C4:

http://prorecordingworkshop.lefora.com/reply/46983815/Anybody-have-a-BLUE-M49-schematic#.VKz5hMaoXbg (plus search this forum)

"@ryan the 8pF value is from the m49b, b for broadcast, so there's a filter as well as higher amount of NFB. You could go as low as 2pF or up to 4pF, I think 8 is ridiculous as the AC701 barely has enough gain to start with." (David Bock)

https://groups.google.com/forum/#!topic/rec.audio.pro/jQC3wLhKWo0

"2. Neumann M49, cap. C4. Cap. C4 (8 pF) is placed between tube's AC 701 plate and grid. So it is parallel to the tube's own capacitance Cag. This capacitance according to the tube's manufacturer's literature is 2.2 pF. As a result we get 10.2 pF. At stage's gain (20 dB) - in other words ? 10 times (it is a typical volume) we get 102 pF (10.2 pF x 10 = 102 pF) known as Miller's capacitance that is connected between tube's grid and cathode. This is a parasitical capacitance that causes attenuation of signal in cases of capacitive acoustical transducer (condenser mic capsule). It will reduce the signal dynamic range. So it is preferable to remove this capacitance."

These sound like no-brainers to me; what do you think?

Also, as a true newbie, is it really ok to simply omit these components from the PCB altogether? Or would a jumper be necessary in any case? As far as I can tell, I can simply omit them all with no problem, with the exception of C4 which I'm uncertain what would happen.

What say you?

Many thanks,

Mike

 

[micaddict]

Thanks, Mike.

Although a guy like Klaus does have a reputation, I'm not convinced his mods improve on what the original builders (talk about reputation) had in mind. E.g. from what I understand, there is some deliberate negative feedback going on.
Also, making it "better" can in fact compromize the character of the mic. There is a reason most prefer it over a TLM49.

But yes, at some points the original builders were forced (e.g. by the big broadcast clients) to implement things they rather didn't.
Even this could have added character we are now used to, but it may be an area to explore. Experimenting and discussing hurts nobody.

BTW, did you also find (in numbers that is) what Klaus does to U67s? Filtering, NFB and phase shift a-plenty in that one. Question is, do we want to miss that?

 

[0dbfs]

If you haven't already you guys should check out, build, and compare olivers m49 conversion ckt here (which eliminates the items mentioned or you can add them back in as needed/desired):
http://www.tab-funkenwerk.com/sitebuildercontent/sitebuilderpictures/TLM49ArchutModschematic.jpg

What I appreciate about the M49 approach is that there are several points that allow for fine tuning of the systems to more specific situations / needs / clients. In Oli's conversion ckt it's a pretty standard self-bias plate-out setup with no additional feedback (other than the self-bias) or RF filtering. Look at Max's MK7 to see how you may want to implement a fixed bias scenario into Oliver's M49 conversion ckt.

Neumann manufactured and delivered a consistent product across a global client base and this microphone system was able to be put into service in a wide range of environments in the field without "getting under the hood". The various RF and other filtering components allowed for a more consistent product in a wider range of environments (not necessarily optimizing a capsule/tube/trafo in a specific environment/application). When they went to the self-bias version they were also able to deliver a more consistent product because the bias implementation in the B version exposes more of the tube fingerprint.

If you want to deliver a more consistent product then perhaps sticking with the C version and including the FB/filtering will help with some consistency. For more individual, unique, or special needs however there are options. Who wants the same mic as everyone else anyway (we've already got essentially the same DAW/plugins)?

To summarize, tuning of the original mic/ckt "system" may include items like:
- Capsule selection (M7, pvc, mylar, kk47, c12, mfgr, other)
- Transformer & tube selection-and-setup (ratio, materials, etc)
- Output cap selection & value
- Fixed/Self bias implementation
- passive/regulated PSU implementation
- Grid bias point spec
- Grid/backplate/rear-diaphragm resistor value
- Heater voltage (ie; under-heating and to which degree)
- Cathode bypass cap value/selection
- 5M neg feedback/HPF selection/omission
- 2-8pf plate to grid cap spec
- 400-600pf Plate to gnd cap value/selection/omission
- Head basket geometry

Hope this provides some food for thought and don't forget to make more microphones!

Cheers,
jb

 

[Phrazemaster]

Thanks for the thoughtful replies guys.

Can anyone explain how the NFB works in this circuit?

Also if I wanted to leave out C4, can I just omit it entirely from the PCB?

Thanks!

Mike

 

[micaddict]

If you haven't already you guys should check out, build, and compare olivers m49 conversion ckt here (which eliminates the items mentioned or you can add them back in as needed/desired):
http://www.tab-funkenwerk.com/sitebuildercontent/sitebuilderpictures/TLM49ArchutModschematic.jpg

What I appreciate about the M49 approach is that there are several points that allow for fine tuning of the systems to more specific situations / needs / clients. In Oli's conversion ckt it's a pretty standard self-bias plate-out setup with no additional feedback (other than the self-bias) or RF filtering. Look at Max's MK7 to see how you may want to implement a fixed bias scenario into Oliver's M49 conversion ckt.

Neumann manufactured and delivered a consistent product across a global client base and this microphone system was able to be put into service in a wide range of environments in the field without "getting under the hood". The various RF and other filtering components allowed for a more consistent product in a wider range of environments (not necessarily optimizing a capsule/tube/trafo in a specific environment/application). When they went to the self-bias version they were also able to deliver a more consistent product because the bias implementation in the B version exposes more of the tube fingerprint.

If you want to deliver a more consistent product then perhaps sticking with the C version and including the FB/filtering will help with some consistency. For more individual, unique, or special needs however there are options. Who wants the same mic as everyone else anyway (we've already got essentially the same DAW/plugins)?

To summarize, tuning of the original mic/ckt "system" may include items like:
- Capsule selection (M7, pvc, mylar, kk47, c12, mfgr, other)
- Transformer & tube selection-and-setup (ratio, materials, etc)
- Output cap selection & value
- Fixed/Self bias implementation
- passive/regulated PSU implementation
- Grid bias point spec
- Grid/backplate/rear-diaphragm resistor value
- Heater voltage (ie; under-heating and to which degree)
- Cathode bypass cap value/selection
- 5M neg feedback/HPF selection/omission
- 2-8pf plate to grid cap spec
- 400-600pf Plate to gnd cap value/selection/omission
- Head basket geometry

Hope this provides some food for thought and don't forget to make more microphones!

Cheers,
jb

I kinda like that post.  ^

 

[Phrazemaster]

Just FYI FleA uses (b) version.

http://www.flea-microphones.com/fdq.html

 

[micaddict]

They do -c, too, upon request.

 

[Phrazemaster]

Thanks Henk.

Can anyone explain the function of c4??

If leaving out would I just wire jumper across the PCB?

Thanks

 

[0dbfs]

I believe C4 is an RF neg feedback path, 4-8pf plate to grid. Plate signal is opposite polarity to grid so it filters and cancels out RF as well as increasing miller C which acts as a C divider (ie; pad) WRT the capsule capacitance. Since the value is labeled as "selected" is anyone aware of what the selection criteria or "factory test setup" is or may be?

Thx,
jb

 

[Phrazemaster]

JB are you saying that C4 is needed to eliminate radio frequencies? (RF)?

Also any one know what David Bock meant about reducing C4 because it affected the tube gain?

Finally, if I want to leave it out it looks like I'll need to just jumper it, since other parts of the circuit still need to connect...right?

Thanks gents

 

[Phrazemaster]

OK JB, found a post from Neumann (with a correction; he talks about "R4" but mistakes it for C4):

"R4, the feedback capacitor on the tube, is typically 8 pF in the M49b, typically 0..2pF in the M50b.

Selected to obtain the correct amplification.

That's it."

From: http://www.neumann.com/forums/view.php?site=neumann&bn=neumann_archive&key=897298159

 

[Nyquist]

I used a 2pf on my 49B build. It increased gain and reduced the noise floor a few dB.

 

[poctop]

I would like to like to post but there Is no Like button ?
D

 

[Phrazemaster]

I thought something was wrong with my MIND! I've tried to like a couple posts lately from my iPhone but no button; it's not here on my mac now either! 

Thanks Nyqvist and JB and Henk!