Replacing GOhm resistors with diodes, the theory behind it

[abbey road d enfer]

Would simulation likely do a good job analyzing the in circuit behavior?

IO on't know. Someone has to do it but I'm too lazy. I'm just good at criticizing other's works.

I have not quite figured out how to simulate the capsule behavior (changing capacitance with amplitude) from the previous THD discussion, but it seems like these two phenomena of changing capsule capacitance and changing effective resistance could interact.

I believe it's feasible. To represent the capsule, I would use a capacitor of value C=(1/K+P).Cnom, where P is the instantaneous acoustic pressure and K a coefficent that could be adjusted for giving a believable sensitivity.

 

[kingkorg]

Would simulation likely do a good job analyzing the in circuit behavior?

I have not quite figured out how to simulate the capsule behavior (changing capacitance with amplitude) from the previous THD discussion, but it seems like these two phenomena of changing capsule capacitance and changing effective resistance could interact. I am curious to know if the magnitude approaches anything like audible. If the resulting distortion is 130dB below signal levels it becomes just an academic curiosity.

I found it easier in the end to just do acoustic testing as there were also many discussions about capsule distortion. I'm having issues getting any audible distortion. It seems THD is dominated by biasing, rest of the circuit, output transformer, usual stuff. I am guessing that is why specified THD published by most manufacturers is gotten by just injecting signal into the circuit. To me it doesn't seem anything significant is happening THD wise in capsule to gate section. Haven't tested with diode gate biasing.

 

[terry setter]

There is no need for bootstrapping. High ohmic value of reverse bias is an intrinsic characteristic of a PN junction.

Can you be more specific here?
"lower noise of a diode pair" compared to what?
"increase in shot noise" resulting from what?

What type of tubes can operate with such high valu grid resistors? You typical 12AX7 or EF86 gets crazy with these values. Not to mention leakage from coupling caps.f

Biasing how? You mean as a replacement for the grid resistor?

Hi, sorry to present such an unclear description of my points. I'll try again:
In condenser mics, the impedance of the capsule is so high (multiple gig Ohms) that using "small" resistors like the typical 100K you might find on a typical circuit using a12AX7 or EF86 prohibits the capsule from being able to deliver a usable frequency response or any significant signal. The older "classic" tube mics used between 60 and 150 Meg grid resistors. Noise drops and low frequency response extends as the grid resistor value is increased, but at around 3G, the Shot noise (which, in this case I understand to be a function of electron movement related to grid leak) ramps up significantly and you lose the benefit of the high value grid resistor. There are a number of successful tube mics on the market that use 1G or more on the grid. Here's a link about Shot noise:
https://www.electronics-notes.com/articles/basic_concepts/electronic-rf-noise/shot-noise-what-is.php
To bias the tube in many of my plate-loaded microphone designs, I use a diode (or, more often two) beneath the cathode. The cathode goes to the positive side of a diode. Then the negative side of that diode goes to the positive side of a second diode - if higher bias voltage is desired - and the final diode's negative side goes to ground. No resistor, no bypass cap(s). It's way quieter than using resistors and it adds no phase shift as is associated with bypass caps when used in conjunction with a resistor on the cathode of a tube. Each diode is good for .5-1.0V of bias voltage on the cathode, depending on the particular diode selected. UF4007s are good for .5V each, and Shottky diodes fall somewhere between .8 and 1.3V each. Don't try this with LEDs. They're noisy and also they are impacted by light entering the LED.

When I started reading this thread, I was thinking there might be a way to use a diode/s in place of the high value grid resistor, but it seems that buying a two or three gig resistor will be way simpler, as well as possibly cheaper, and not lift the resistance so high that my circuit loses noise performance.

 

[abbey road d enfer]

When I started reading this thread, I was thinking there might be a way to use a diode/s in place of the high value grid resistor, but it seems that buying a two or three gig resistor will be way simpler, as well as possibly cheaper, and not lift the resistance so high that my circuit loses noise performance.

The reason for replacing the Gohm resistor(s) is that they are generally expensive, often bulky, and hard to find. Unless trying to use diodes made out of unobtainium and forged by vestal virgins, diodes are cheaper.
However, grid current in vacuum tubes, which is much higher (about 100x) than leakage gate current in FET's, makes it impractical. You are right, it results in increased noise, which is not compensated by the low-frequency shift of the KTC noise spectrum.
In addition, the use of a very large grid resistor results in significant downward shift of the operating point of the tube, which would limit headroom.
Typical grid current of an audio tube is about 100nA, which would suggest about -100V into a 1G resistor, which clearly puts the tube out of its comfort zone.

 

[thor.zmt]

Would simulation likely do a good job analyzing the in circuit behavior?

That will depend entirely on how realistic the simulation is.

I have not quite figured out how to simulate the capsule behavior (changing capacitance with amplitude)

Capacitance change is modest. If we have 60V bias, a 10% change in capacitance will give 6V out.

For 1% it would be 0.6V. And for 60mV peak (~40mV RMS) we get 0.1% capacity change.

I feel a "generator" in series with an equivalent capacitance is a model mostly adequate.

I am curious to know if the magnitude approaches anything like audible. If the resulting distortion is 130dB below signal levels it becomes just an academic curiosity.

Realistically, taking into account the distortion of the human ear, I fully expect the distortion to be sufficiently low.

Don't forget, every J-Fet contains a reverse biased diode anyway.

Thor

 

[abbey road d enfer]

That will depend entirely on how realistic the simulation is.

Of course. The simulation is not the problem, it's the model.

I feel a "generator" in series with an equivalent capacitance is a model mostly adequate.

This wouldn't work for simulating the distortion due to capacitance variation. You need a non-linear cap, expressing the fact that C is in the form 1/(1+x), x representing the instantaneous sound pressure.

Realistically, taking into account the distortion of the human ear, I fully expect the distortion to be sufficiently low.

A similar reason is often given for neglecting head amp noise.
Actually, a lot of discussions are about the frequency response of microphones, particularly how to tame the HF rise due to diaphragm resonance and diffraction, which is not such a big issue because it can easily be corrected electronically.
It's too bad IMO that it occults other performance aspects.
The fact that a nuisance can be deemed negligible is not a good reason for not investigating it.

 

[kingkorg]

FWIW. I did some tests recently with low THD HF driver Uli suggested in another thread up to 130db 1k-24k.

I have tested at different pol. voltages Rode NT2a edge terminated capsule, K67 from Takstar, Primo em200, Sennheiser mkh800 capsule (this one polarized with 20vdc). Maximum THD i got at 130db was 0.5% throughout the frequency range. At this level i believe all the THD came from the driver itself. Test circuitry was basically free of thd at these levels.

 

[Khron]

FWIW. I did some tests recently with low THD HF driver Uli suggested in another thread up to 130db 1k-24k.

I have tested at different pol. voltages Rode NT2a edge terminated capsule, K67 from Takstar, Primo em200, Sennheiser mkh800 capsule (this one polarized with 20vdc). Maximum THD i got at 130db was 0.5% throughout the frequency range. At this level i believe all the THD came from the driver itself. Test circuitry was basically free of thd at these levels.

Might want to consider feeding a known signal amplitude into the circuit, determine the exact gain, and work out what voltage the capsule output is that corresponds that. And if or whatever correlation between that and final output THD might be...

 

[kingkorg]

Might want to consider feeding a known signal amplitude into the circuit, determine the exact gain, and work out what voltage the capsule output is that corresponds that. And if or whatever correlation between that and final output THD might be...

It should be piece of cake, as it's unity gain. I was expecting higher THD values because of some Sennheiser papers on MKH series. Since it was only 0.5% thd @130db in total, I gave up on further testing, i personally don't care to investigate further. I can't hear that.

 

[kingkorg]

Other than that, a great quote from the Neumann guy. His whole reply is a must read:

"For the U87 and 94 dBSPL the change in capacitance calculates to be 0.07 pF, 1.74 pF at 122 dBSPL and 5.5 pF at the maximum 132 dBSPL. Even though the capsules are acoustically identical, for the U87A these values are 0.06 pF at 94 dBSPL 0.08 pF at 117 dBSPL and 2.6 pF at the maximum 127 dBSPL."

Reply #4:

https://repforums.prosoundweb.com/index.php/topic,36194.msg531542.html#msg531542

 

[Ralf 71]

Hi, sorry to present such an unclear description of my points. I'll try again:
In condenser mics, the impedance of the capsule is so high (multiple gig Ohms) that using "small" resistors like the typical 100K you might find on a typical circuit using a12AX7 or EF86 prohibits the capsule from being able to deliver a usable frequency response or any significant signal. The older "classic" tube mics used between 60 and 150 Meg grid resistors. Noise drops and low frequency response extends as the grid resistor value is increased, but at around 3G, the Shot noise (which, in this case I understand to be a function of electron movement related to grid leak) ramps up significantly and you lose the benefit of the high value grid resistor. There are a number of successful tube mics on the market that use 1G or more on the grid. Here's a link about Shot noise:
https://www.electronics-notes.com/articles/basic_concepts/electronic-rf-noise/shot-noise-what-is.php
To bias the tube in many of my plate-loaded microphone designs, I use a diode (or, more often two) beneath the cathode. The cathode goes to the positive side of a diode. Then the negative side of that diode goes to the positive side of a second diode - if higher bias voltage is desired - and the final diode's negative side goes to ground. No resistor, no bypass cap(s). It's way quieter than using resistors and it adds no phase shift as is associated with bypass caps when used in conjunction with a resistor on the cathode of a tube. Each diode is good for .5-1.0V of bias voltage on the cathode, depending on the particular diode selected. UF4007s are good for .5V each, and Shottky diodes fall somewhere between .8 and 1.3V each. Don't try this with LEDs. They're noisy and also they are impacted by light entering the LED.

When I started reading this thread, I was thinking there might be a way to use a diode/s in place of the high value grid resistor, but it seems that buying a two or three gig resistor will be way simpler, as well as possibly cheaper, and not lift the resistance so high that my circuit loses noise performance.

Hi out there, I‘m a newbe in this Forum. But this topic is exactly what I like to talk about. In order to get rid of this Gig-ohm thing I use a complete different schematic. Please excuse my bad English, I have to get netter by time. I use the valve itselve to do the job: I don‘t use a cathode R and I don‘t use a bias R . Just let the electrons float freely to the anode where there is an R of somewhat 47k to 200k. The backplate of the capsule is directly connected to the grid . The capsule bias is achived by a voltage divider from the anode source. The second half of the ECC is used to give a symetrical output. I use B+ up to 270Vdc.

 

[abbey road d enfer]

Just let the electrons float freely to the anode where there is an R of somewhat 47k to 200k. The backplate of the capsule is directly connected to the grid .

Assuming I understand correctly that you have no galvanic connection to teh grid, it raises a number of questions.
What is the operating current of the tube? Is it stable? Is it consistent from tube to tube?
What about noise?

 

[Ralf 71]

Well, this is the 3rd try to send the document. I hope it will work this time.
Btw. How do you folks put your grafics here, I guess I need some help with this?
Anyway, I hope that you now can read the schematic and the grafs of my mic.
I wonder what you might say an answer is highly apreciated, thanks
Ralf

 

[k brown]

Well, this is the 3rd try to send the document. I hope it will work this time.
Btw. How do you folks put your grafics here, I guess I need some help with this?
Anyway, I hope that you now can read the schematic and the grafs of my mic.
I wonder what you might say an answer is highly apreciated, thanks
Ralf

Admittedly I haven't followed this thead that closely, but what's the aim here compared to the simple 'Royer/Altec/C-37A-style' grid leak method of biasing a capsule in a tube mic, that uses no gigohm resistor(s)?

 

[Ralf 71]

well the aim is to get rid of the big resistors and this thing works great

 

[kingkorg]

Tim Campbell told me that increasing the Polarizing Voltage, also reduces the Frequency responce + changes the sensitivity of the capsule.
This makes sense, as the increased Voltage also increase the pull towards the opsite polarity (no matter what that is - it's the difference that counts) .... Actually bending the membrane.
We came across this when debating a FM Microphone - where it isn't the change in charge but in the capacitance, that gives the Modulation => signal out .... and why that is to be prefered to the traditional application.

This is true for edge terminated capsules with more relaxed diaphragms such as CK12. However it is not so true for something like k67 where you have center termination, and the capsule is more forgiving regarding tension. I have used above 100vdc with some k67 without any change in tonality.

 

[abbey road d enfer]

Anyway, I hope that you now can read the schematic and the grafs of my mic.

The generator should be in series with C3.
R7 doesn't do much here, just bothers signal.
Diode D1 does not do what you think it does.
Grid bias of VT1 is governed by plate voltage of VT2.
Cathode voltage is naturally higher than its grid, so the diode is reverse polarized, which does almost nothing to the operating point.
I bet if you disconnecte D1 and R1, there is no significant change.
C6 at 1nF results in frequency response down 3dB at 3kHz, -10 at 10kHz. That would make a pretty dark microphone.
The cathode resistor of VT1 being 100k, loading it with 10k is not right.

 

[terry setter]

Admittedly I haven't followed this thead that closely, but what's the aim here compared to the simple 'Royer/Altec/C-37A-style' grid leak method of biasing a capsule in a tube mic, that uses no gigohm resistor(s)?

I can't speak for others, but the reason I do it is go get higher PV on the capsule, for more output. Also, more predictable results across various tubes.

 

[Ralf 71]

Ok, thx for your thoughts. U are proberbly right with the generator, I tried it that way but didn‘t get the right reading of the frequencies.l think this might be caused by the tiny capacitor. The circuit around D1 is just for savety purposes. The anode voltage is on before the heater is fully on.
The 10k at the output represents the smallest possible resistance that the output should see(just for testing).As far as the 1n is concerned please Look at the last graf I put on the sheet. The mic sounds exactly like what you expect from that. Please keep in mind that this kinda mic is working real fine the last 15 years. I did many vocal and instrumental recordings with it.
And... it is quiet too, as far as I can tell.
Greetings from Germany
Ralf

 

[k brown]

well the aim is to get rid of the big resistors and this thing works great

That was my whole point - the Royer/Altec/C-37A doesn't use any gig-value resistors, and is waaay simpler.

I've built it with 5840, 6AU6 and 6AK5.