Insight into "U47-inspired" with EF12 tube?

[soapfoot]

there are great solid-state mics.  I'm not interested in making one right now, however.  Maybe some other day.  Right now, I'm interested in making a tube mic inspired by a U47.

 

[e.oelberg]

What about a solid state circuit?

that is what I was thinking about. Grosser is using fets in his VF14 substitute anyway.
did you experiment with solid state amplifiers to get a u47 inspired sound ?


nicholas

 

[soapfoot]

Thanks to the links here and Google I've been doing a lot of reading and learning.  I've also sent some e-mails to a few experienced people trying to dig up a legible, correct, legitimate schematic for a U47 EF12 conversion.  I've seen a few "inspired by" ideas, one illegible schematic, etc.  But nothing stands out head and shoulders above the rest, and nothing seems like a real "conversion" of the original circuit for the new tube-- the STU47 from this board looks very appealing, but it seems like a "tribute" in the form of a ground-up design.

I've come up empty-handed on the "conversion" schematics, which is what I'm really after, so maybe I'll take a different tack.  I'm experience- and training-wise under-equipped to do work on figuring out the "conversion" entirely myself, but I've been trying to do what I can.

Looking at the data sheets, I can already see that the filament voltage of the EF12 is 6.3V.  I've seen it recommended (per Oliver Archut) to underheat it slightly, to about 5V or so.  Seems that this would be most elegantly accomplished through a separate conductor in the tube mic cable, it seems.  This much makes sense, and while it's obvious, it's at least a start.

Does anyone know what else would need to be changed from the U47 schematic to make the EF12 work?  Even bits and pieces would be helpful as I try to piece this together.

Here's a Philips datasheet for an EF12:
http://tubedata.itchurch.org/sheets/030/e/EF12.pdf

This sideways thing is the best I could come up with for VF14 data:
http://tdsl.duncanamps.com/pdf/vtubes.pdf

Also in the course of my detective work, I found that there was one mic in Neumann's history for which both VF14 and EF12 schematics exist: the CMV-5B.  Apparently, both versions were at least drafted, though nobody has seen a VF14 version.  Nevertheless, the schematics exist.

EF12: http://www.elektroakustika.cz/images/cmv/cmv5/schema.gif
VF14: http://recforums.prosoundweb.com/index.php/fa/16142/21957/

I've mostly been a kit builder and have done some rudimentary design stuff, but mostly with guitar amps.  I do not have any formal electronics training and am totally self-taught.  While I can identify the purpose of a fair amount of the components on the schematics, I cannot parse exactly what was done to convert from VF14 to EF12 operation, nor if I would be able to apply those changes to the U47 circuit.

I'm sure Gunter Wagner's design is proprietary, though I haven't asked him.  Just seems like he wouldn't want to freely give out that info.  That said, I'd be real curious to see how he's running the EF12 in his mic, and if it required major departures from the original circuit.  Most agree that his mic sounds very good.

Thanks for any help anyone can lend.

 

[zebra50]

As you know, the BIG change is having 6V heaters instead of 20V. With the original U47 supply you just have 105V for your B+, and a big resistor to drop it for the 20V heater & cathode bias. To get this down to 5 or 6V at relatively high current compared with VF14, you need to scrub off a stupid amount of heat.

So any sensible (to my mind at least) mic design with EF12 has a separate 6V supply. And at that point you've taken a big step away from being just a 'conversion'.

Max's MK47 design uses the heater supply to bias the cathode, which is closer to the U47 philosophy. I haven't really studied his new design but I imagine there is much to be learned there too. My mic is a bit simpler, with straightforward heater and cathode resistor, and a bit less like a real U47 design.

 

[milesjackson]

Looking at the data sheets, I can already see that the filament voltage of the EF12 is 6.3V.  I've seen it recommended (per Oliver Archut) to underheat it slightly, to about 5V or so.  Seems that this would be most elegantly accomplished through a separate conductor in the tube mic cable, it seems.  This much makes sense, and while it's obvious, it's at least a start.

Underheating a vf14 with a specified filament voltage of 55V down to 36V in the original U47 circuit is not identical to underheating a ef12 from 6.3V down to 5V.  I've been down this "I'm gonna make a u47 with the parts I can get" road, and I have to say I gave up on that.  Build a good microphone, and don't worry about whether or not it vaguely resembles the original U47 circuit.

 

[Skylar]

Sorry to interject.

I'm beginning to realize that there's nothing special about the VF14 (or that whole family of tubes for that matter).
You can get great sounds from many, many tubes.

The secret is the capsule...


If you follow the history of what Gus has gone through in his experiments over the years and start doing the same thing, you find yourself coming to the same conclusions.
As much as we love tubes, solid state starts to make real sense after a while.

 

[gary o]

Am I correct in saying that different ways to power the heaters makes no different to the sound of the mic (longs as we give heater the right voltage a certain percentage less than usual as I understand it)and as long as the cathode is biased correctly without the heater connected ......maybe a dummy load would be used

So if I understand correctly in a real U47 we could disconnect the heater, replace with dummy load, externally feed the VF14 with the 36V & the mic would still sound the same ?

 

[MagnetoSound]

... as long as the cathode is biased correctly ......

... we could disconnect the heater, replace with dummy load, externally feed the VF14 with the 36V & the mic would still sound the same ???

Yes.

 

[pasarski]

...But a noisy heater supply would make the mic noisy in U47 circuit.

 

[rodabod]

... And you'd need to keep the cathode resistor the same value, thus meaning you'd need an external bias voltage source.

 

[gary o]

Hi again so getting LT from HT supply is a good way to get less noise in yr LT then.....do other mics do this ?

So I hope Im still right to say that we can just adjust the cathode bias part of real U47 circuit to be correct and supply the different LT for a different tube Im talking about the E14 that im told is same as the V14 apart from heater & then mic should sound the same as with V14.... I no other DIYers & designers are doing this but I seem to read that it might not same sound the same I think it should be but not sure thats why I ask.....thanks for your patience

very interested to learn about this interesting type of circuit

cheers

 

[gary o]

... And you'd need to keep the cathode resistor the same value, thus meaning you'd need an external bias voltage source.

Hi Roda......please excuse my lack electronics basic knowledge......can we not adjust the mic circuit to bias tube from the 105VDC going into mic.....Im keen to understand this ......again thanks for patience.

 

[soapfoot]

Im talking about the E14 that im told is same as the V14 apart from heater & then mic should sound the same as with V14.... I no other DIYers & designers are doing this but I seem to read that it might not same sound the same I think it should be but not sure thats why I ask.....thanks for your patience

I've been doing a lot of reading, and though it seems that way on paper, according to experts like Klaus Heyne and Oliver Archut, the EF14 is not the same as the VF14 (with different heater voltages).  The numbers are the same, but the cathode material is different, and the internal construction of the tube is different.  Supposedly, the sound is quite different.

Several different builders have said they actually prefer the sound of the EF12 over the EF14 in a modified 47 circuit.  And since I happen to already have one, I'd like to figure out how to implement that.  However, I haven't yet figured out how they're doing that, and nobody seems too keen to share (understandably, for those who make their living doing this).

As for the underheating to 5V, this was a recommendation of Oliver Archut over on the Neumann pinboard.  I believe it was not necessarily for the sound, but to keep the tube living longer.  

As for the solid-state thing, I understand that for many this does make a lot of sense.  However, I love tubes and am doing this primarily as a learning experience.  I'm very interested in learning how to make the tube work.  Not even getting into how it may or may not sound to use a solid-state device, or how much "sense" it makes, I'd rather use the tube at this time.

...hopefully I'm not the only one who's interested in figuring this out.  

 

[zebra50]

... And you'd need to keep the cathode resistor the same value, thus meaning you'd need an external bias voltage source.

... which is sort of what Max does in the MK7...

and nobody seems too keen to share

Well, we're all sharing! Really all the info is here now - 6V heater (or five if you prefer), and cathode bias from heater supply. Everything else as it was in the 47, then tweak til it sounds good.

 

[soapfoot]

Well, we're all sharing! Really all the info is here now - 6V heater (or five if you prefer), and cathode bias from heater supply. Everything else as it was in the 47, then tweak til it sounds good.

And I do thank you!

So is that all there is, really?  That sounds easy enough.

So that brings up a couple of questions--

You say "cathode bias from heater supply," and I think we're getting into something I don't understand.  To me, "cathode bias" has always meant "place a resistor between the cathode and ground, of a particular value that will cause the appropriate positive voltage to develop on the cathode, making the grid appear negative by comparison."

So what does it mean to get cathode bias "from the heater supply?"  Do you mean take a small positive voltage from a 6v DC heater by way of a resistor, and apply that directly to the cathode, as opposed to using a resistor to develop the bias?  And at that point does the cathode "float," with no ground connection?

Bias should be approximately how many volts negative?

And finally... "tweak til it sounds good..." what would be some places to start tweaking, in your experience?  In other words, can we start a discussion on what sorts of things, when changed, have what sorts of effects?

Thanks for your patience.  I've never delved this deeply into microphone design before.

 

[rodabod]

Hi Roda......please excuse my lack electronics basic knowledge......can we not adjust the mic circuit to bias tube from the 105VDC going into mic.....Im keen to understand this ......again thanks for patience.

Yeah, you probably could do this. Cost you about (at a guess) 40mA if you form a potential divider with the 29R resistor (in order to drop ~1.1V)

 

[MagnetoSound]

So what does it mean to get cathode bias "from the heater supply?"  Do you mean take a small positive voltage from a 6v DC heater by way of a resistor, and apply that directly to the cathode, as opposed to using a resistor to develop the bias?  And at that point does the cathode "float," with no ground connection?

Yes. The U47 is a fixed bias amplifier, although the (positive) bias is applied at the cathode, rather than (negative) at the grid, as is more commonly the case.

Bias should be approximately how many volts negative?

Bias is given as 1.1v for the VF14. The sweet spot may well be elsewhere for other tubes (1.6 or so for EF86). Remember, grid is negative wrt cathode, and vice versa.

 

[soapfoot]

That's very helpful, thank you.  Makes a lot of sense.

Something else I've observed from looking at schematics and data sheets--

In the VF14, each grid, cathode, and metal case have their own individual pin connections on the tube.

In the EF12, the third grid, the cathode, and the metal case are all tied together at one pin connection.

The U47 schematic shows that the second and third grids are tied to the plate (triode operation).

How would I connect the EF12 for triode operation?  I'd assume that the second grid would be connected to the plate?

 

[zebra50]

http://recforums.prosoundweb.com/index.php/m/0/32588/16/3507/#msg_32588

(In some circles cross posting is considered bad form!)

For tweaking, plate resistor, output capacitor value & type, and polarisation voltage spring to mind.

 

[soapfoot]

Thanks for your thoughts.  I'm aware of that forum's rules against cross-posting.  I figured my reply would be within the 'spirit' of the rule-- instead of creating a new thread or bumping an old thread, I replied with similar thoughts to a thread that was already on "page 1,"  to share information I had gathered in the course of doing this that I thought would be relevant there.

If you feel like that's bad form, I will delete my post over there.  It wasn't my intent to "spam," but to hopefully steer a thread which seemed to have drifted off-topic.

Edit: I've added a note to my post over there mentioning this conversation, and stating that I will happily remove my comments there if Klaus feels like this qualifies as "cross-posting."  I know he runs a tight ship, and I don't wish to run afoul of anyone.