Mosfet Cathode Follower

[bool]

+1

[MetalGuy]

* so if I want a 16:1 global attenuation, a simple solution to lower the output impedance of the preamp could be to keep R19 and R20, lower R20 value to 39k, and take directly the output from R19/R20 junction (maybe keeping C16 and R26), with no use of mosfet cathode follower.
The output impedance would be around 33k, which is not great, but not too bad.

Also note that your volume pot when maxed is not 1M anymore thus presenting the tone stack with a different load.

* if I want an output impedance below 10k, the mosfet cathode follower is necessery, but I'm confuse with the R21-R22 values (R.G. solution vs Merlin's one)

In cases like this there's only one solution: try both versions and decide for yourself although the noise levels Merlin is talking about will be undetectable for your years. Considering the tubes' thermal noise that is unavoidable I seriously doubt you'll hear any difference if at all.

[PRR]

> the thermal noise from those resistors in parallel with whatever other resistances are there adds to the input noise, doesn't it?

Those parallel resistances *load* the 10Meg thermal-hiss source, reduce its hiss voltage.

There is a little hiss generator inside any resistance. But this generator has an internal impedance. This is used in the usual network calculations to determine the total hiss.

And two such hiss-generators in parallel are *exactly* equivalent to one resistor of the equivalent parallel value.

Using ruff numbers.

Hiss of 100K over much of the audio band: 4uV
Hiss of 10Meg over much of the audio band: 40uV
Parallel them.

Roughly speaking, the hiss of the 10Meg with 100K across it is loaded-down by a factor of 0.01. The hiss of the 100K with 10Meg across it is loaded-down by a factor of 0.99. The parallel resistance is 99.0099K. The effective hiss of the combination is the *same* as a 99.0099K resistor.

Any other result violates Conservation Of Energy. Taken to an absurdity, we could find a 1,000,000,000 Meg resistor (actually 3.6^17 Ohms), connect it in place of our utility power, and live on hiss. While motors might not like it, incandescent lamps and most heaters would not mind random hiss power.

[Sheldon]

I think a miss the point somewhere.

Here is the circuit I drew with R.G. recommandations :


Here is the circuit I drew with Merlin recommandations :


There is discussion about the adding noise or not with 10M resistors.
But what is the improvement with removing R19 and R20 and instead attenuate the signal more by adjusting R24/R25? Obviously the aim is to have something as transparent as possible and with no influence on the previous stages (tone stack...)

Thanks a lot, I know I ask a lot of questions but I'm quite a newbie, and I'm passionated about learning things and understanding!

[Sheldon]

@ Merlin : The cathode follower I try to transform in mosfet cathode follower is quite similar to the Orange Rockerverb one you describe in your "Designing Tube Preamps for Guitar and Bass" book

[merlinb]

@ Merlin : The cathode follower I try to transform in mosfet cathode follower is quite similar to the Orange Rockerverb one you describe in your "Designing Tube Preamps for Guitar and Bass" book

Yes, it will work fine.

[Sheldon]

thank you
The last thing I don't understand is why is it better to make a 16:1 attenuation after the cathode follower rather than making a 4:1 attenuation before the cathode follower and another 4:1 attenuation after it, like in the original Orange schematic?
With the voltage divider, the tone stack will see an impedance from 400k to 1M.
Without the R19-R20 voltage divider, the tone stack will see a 1M impedance no matter where the volume pot is.
Will it lead the tone stack to work differently?

[merlinb]

The last thing I don't understand is why is it better to make a 16:1 attenuation after the cathode follower rather than making a 4:1 attenuation before the cathode follower and another 4:1 attenuation after it, like in the original Orange schematic?

Because doing the attenuation after the CF gives you lower output impedance, lower noise, fewer parts and a simpler circuit.

With the voltage divider, the tone stack will see an impedance from 400k to 1M.
Without the R19-R20 voltage divider, the tone stack will see a 1M impedance no matter where the volume pot is.
Will it lead the tone stack to work differently?

No. Jeez if it matters that much to you, just make R21 and R22 equal to 1.5Meg again -same result (1.5M||1.5M = 750k. In parallel with 1M pot = 428k load). You're way too worried about tone stack loading.

[Sheldon]

thanks a lot to everyone for your precious help!

[Sheldon]

I have an additional question : are 0.5W resistors ok for R21 to R25 or is it better to use 1W or 2W resistors?

[merlinb]

I have an additional question : are 0.5W resistors ok for R21 to R25 or is it better to use 1W or 2W resistors?

0.5W is fine for all except R24, which probably ought to be 1W or more.

[Sheldon]

even if the C voltage of the schematic is 200V ?

[merlinb]

even if the C voltage of the schematic is 200V ?

Do the math. The MOSFET gate is biased to 100V. The MOSFET source will be a little below this, but ignore that. The drain current will therefore be 100V/(22k+1.8k) = 4.2mA.
Power in R24 = 0.0042^2 * 22000 = 0.39W.
A bit too close for comfort in a 1/2W resistor.

[Sheldon]

Sorry, my question wasn't clear, I was asking for R21 and R22 with 200V since the voltage wasn't written on my schematic
in R21, R22 if I'm not wrong : I=U/R so I=100/10000000=0.00001A
P=R*I²=0.001W

[merlinb]

Sorry, my question wasn't clear, I was asking for R21 and R22 with 200V since the voltage wasn't written on my schematic
in R21, R22 if I'm not wrong : I=U/R so I=100/10000000=0.00001A
P=R*I²=0.001W

Exactly.

[Sheldon]

thank you

[Sheldon]

Hello, I'm testing this cathode follower, except I have R21=R22=2.2M instead of 10M


I have a power issue.
What should be the current consumed by the B+ supply of the ZVN0545A?
In fact, I have a nixie power supply :

Rsense (R30-R31) is set to have a maximum output current around 10mA.

C is the B+ of the mosfet cathode follower
D and E are the B+ of two 12AX7 tubes

When I disconnect the cathode follower, so with only the two 12AX7, it works, the Nixie power supply deliver 220V and the B+ of the tubes are around 205V. The tubes consume around 1.5mA per tube.
When I connect the cathode follower, the Nixie output voltage falls to 160V. So it means that the cathode follower consumes more than 7mA, but I expected it to be around 1-2mA.
Can you please help me?

[merlinb]

What should be the current consumed by the B+ supply of the ZVN0545A?

This was answered in post #32. It should be about 4mA, so maybe you wired something wrong, or maybe your nixie supply isn't as good as you think it is!
If you want to reduce the MOSFET current, increase R24 and R25. For example, 47k and 3.3k should result in about 2mA.

[Sheldon]

Thank you for your precious help, the problem was with the precision of the Rsense resistors in the Nixie power supply, the output current was too limitated.

My preamp seems to work, except a hiss issue (ground?) when I turn up the gain pot above the middle :
http://www.diystompboxes.com/smfforum/index.php?topic=110111.msg1013584#msg1013584