U47EF Oliver Archut version

[rock soderstrom]

sounds wise its still way off version 1, when i plug V1 in, if i put it about a foot away pointing at my mouth and play acoustic guitar, its almost a perfect balance of voice and guitar mix wise and sounds full bodied and no weird mirophonics.
with V2 in the same position the guitar doesnt come through, is very distant the overall sound, sounds thin in comparision and metallic and ringy.
i suspect the capsule so i'll change it for another one i have and see what i get.

Capsule is undoubtedly important for the sound. Which capsules do you use at the moment?

 

[Spencerleehorton]

Changed the capsule and it sounds much better, im running through a API312 preamp and have had to give it two more clicks of gain, now on 7 rather than 5 out of 12 clicks of gain to get the same acoustic guitar and vocal level compared with version 1.
starting to sound pretty good now apart from the ringy microphonics which some notes resonate.
Ive ordered another EF800 which will probably end up in both of these mics.

 

[Spencerleehorton]

ive got no idea what these capsules are to be honest, i know they are not great and were cheap, i think most of mine are from microphone parts a few cheap chinese ones, which i think this other one was.

 

[rock soderstrom]

starting to sound pretty good now apart from the ringy microphonics which some notes resonate.

...I also had a 5654 yesterday for the first time that was completely unusable because it was constantly singing along. Maybe I put too much B+ on it.

 

[Spencerleehorton]

Relay on top and you can see the rubber rings on the tube to help stop the ringing

 

[RuudNL]

Do these rubber rings really make a difference?
Because it is not the glass that is responsible for the ringing, but the inner construction of the tube.

 

[Spencerleehorton]

Ah right well makes sense, didn’t think it was making any difference!!
I have four EF80 and they all are quite microphonic, I chose this one as it was the least.
Will change for the EF800 when it arrives.

 

[Roman Beilharz]

For some reason people publishing these various u47 schematics, replicas, etc, believe 1G resistors are an upgrade. Same goes for higher value plate>transformer coupling caps.

These determine crucial low cut points in the circuit, and can affect the sound tremendously. There is also the issue of grid current with tubes, one more reason not to use 1G. 1G will let subsonics straight through the tube and potentially overdrive it, the transformer too.

If you are after accurate replica i would stick with original values. Changing these crucial component values will impact the sound way more than any specific tube, or capacitor type used.

It's kind of difficult to go against Oliver's schematic, but there are very clear reasons why you wouldn't want 1G there. I have no idea why he went for it.

@kingkorg
Could you be so nice to elaborate, why the 1G resistors will "let subsonics straight through the tube" (as for why this happens)? This is over my head, but I would really like to get it.

And if subsonic overloading was/is actually the case within the circuit, could it be an intended effect to create some sort of desired artifacts?

Thank you, my best

Ro

 

[RuudNL]

Let's assume that the capacity of the microphone capsule is 60 pF.
With a 1 G.ohm resistor the -3 dB point is 2.65 Hz.
Since transformers tend to produce higher distortion at low frequencies and high levels, this is unwanted.
With a 100 M.ohm resistor, the low -3 dB point shifts to 26.5 Hz, still low enough for all purposes.

 

[rock soderstrom]

Let's assume that the capacity of the microphone capsule is 60 pF.
With a 1 G.ohm resistor the -3 dB point is 2.65 Hz.
Since transformers tend to produce higher distortion at low frequencies and high levels, this is unwanted.
With a 100 M.ohm resistor, the low -3 dB point shifts to 26.5 Hz, still low enough for all purposes.

+1

The coupling capacitor from the anode to the output transformer also plays a role. The larger it is, the more low-frequency signals reach the transformer with the same consequences.

 

[rock soderstrom]

And if subsonic overloading was/is actually the case within the circuit, could it be an intended effect to create some sort of desired artifacts?

As Ruud has already described, too many low-frequency signal components can make the output transformer sweat, the distortions increase. This is personally evaluated differently. Some are happy and call it transformer sound (only to a certain degree), others want to avoid it at all costs, as it is a colouring of the signal.

 

[kingkorg]

@kingkorg
Could you be so nice to elaborate, why the 1G resistors will "let subsonics straight through the tube" (as for why this happens)? This is over my head, but I would really like to get it.

And if subsonic overloading was/is actually the case within the circuit, could it be an intended effect to create some sort of desired artifacts?

Thank you, my best

Ro

+1 what Ruud and Rock said.


There's also grid reverse current issue many mention which becomes a problem with 1G. I have never put this to a test in practice tbh, but the theory seems to be solid.

Edit:
Sony (c800g)has avoided this issue with higher plate voltage (more headroom for the tube) and super chunky transformer. Still not sure about reverse grid current.

 

[Roman Beilharz]

Thank you all for shedding some light onto this so quickly!! I am impressed.

I am about to get there and I will try to do my homework to truly understand the coherences, but the practical consequences got clearer. Have a great day everyone!

 

[kingkorg]

Thank you all for shedding some light onto this so quickly!! I am impressed.

I am about to get there and I will try to do my homework to truly understand the coherences, but the practical consequences got clearer. Have a great day everyone!

Capsule (basically 60pf capacitor) and 1G resistor form simple high pass (lo cut) filter. If you use one of those online RC filter calculators it will became obvious what is going on.

 

[Spencerleehorton]

So what’s the calculation so I can work out where I’m at with 72pf capsule and 200M resistor?

 

[rock soderstrom]

So what’s the calculation so I can work out where I’m at with 72pf capsule and 200M resistor?

This formula gives you the lower cutoff frequency at which the level was attenuated by -3dB.



11.02Hz for your numbers.

Here is a online calculator

http://www.sengpielaudio.com/calculator-RCpad.htm

 

[Spencerleehorton]

What does the dot represent?
6.28 is 2 pi
R = 200,000,000 ohm
C = 72pf

 

[kingkorg]

What does the dot represent?
6.28 is 2 pi
R = 200,000,000 ohm
C = 72pf

Multiplication.

 

[Khron]

C = 72pf

That's 72 x 10^(-9) 10^(-12), by the way, just for the sake of clarity.

 

[Spencerleehorton]

Ah ok I get it now

6.28 x 200000000 x 0.000072 = 90432
1/ 90432 = 0.00001106