need help with cathode follower circuit

[morcey2]

frequencycentral why the 100k to the plate for a CF?

I figured the OP is looking for a non-inverting tube unity gain buffer, and the circuit I posted has worked to me in the past. I education is however incomplete.

What you have there is a grid-leak-biased cathodyne phase inverter.  The output will be clipped pretty heavily as there's no voltage swing in one direction. 

Here's an AC-coupled CF that should work quite well with 12V-36V B+ voltages:



The value for C2 can vary depending on the load being driven.  Something else you need to remember is that some effects like being driven from a higher-impedance source (fuzz-face, Rangemaster, etc) and don't sound right with a buffered signal. 

Matt.

[REGNAD]

heres what im looking at doing with the circuit. is c2 the only value that i adjust depending on impedances? and how do i determin the impedances of points a,b,and c? do i simply put my volt meter on the input jack of the pedal? if so, do i do it while the pedal is active?

[Brymus]

Wow this is why I read these threads and follow links.
Nice info guys.
Gus I really liked that interview you linked to I am guessing this is what started this idea-From Pete Cornich QnA...

"Tube Buffering? Does this enhance the tone at all?"

"Oh yes; when you plug your passive guitar directly into your tube amp it sounds great, doesn't it? Having a unity gain tube buffer as the input to my All Tube System allows the guitar to react exactly as if it were connected directly to the tube amp input (without the loading effects of long cables/effects pedals etc.). The effects outputs are connected to other unity gain tube buffers so that each effect "thinks" it is connected directly to the amp input allowing each effect to deliver it's full signal without the loading effects of long cables, other effects etc."

And what you pointed out about HIS use of the tube buffers in the MK2 setup,that there isnt one between each effect,they are used where they will do the most good.

RENGAD good luck ,let us know how it sounds when you finish,you(or Pete Cornish actually) may start a new trend,peace.

[REGNAD]

yeah, my initial thought was to use them between each effect, but i may deviate from that. if what i understand is correct, that the buffer allows each effect to react as if its the only effect in the chain i would think one at each effect would be ideal, but then again, i'm new to this. i dont, however, buy into the idea that this is in any way wasted efforts. hell, if my rig sounds better to me when im wearing ripped jeans than when im wearing Sunday's best then so be it!   i do need a little input on the C2 values and how the circuit components may or may not change depending on load.... Matt, you still out there?

[morcey2]

heres what im looking at doing with the circuit. is c2 the only value that i adjust depending on impedances? and how do i determin the impedances of points a,b,and c? do i simply put my volt meter on the input jack of the pedal? if so, do i do it while the pedal is active?

I'm still here.  Just had family stuff (shopping, dinner, tazering the teenagers, etc...) to take care of.

The impedance can't be measured w/ a meter unfortunately, but it can be calculated.  The impedance at 'A' and 'C' will be quite high, theoretically in the 6M-ohm area.  In practice, it'll be lower than that, but still high enough to ignore most of the time.  The output impedance at B will be very low, probably in the 400-500 ohm range.  That should be able to drive a load all the way down to the 5k area without any real loss of signal.  

C2 was a total guess for me at this point.  It all depends on the input impedance of the following stage/pedal/amp.  If it's driving a low-impedance pedal or something like the 'low' input on most fender amps, you may need to increase it to a 470n or even 1µ to drive it correctly.  

I build tube amps for quite a few people and really love the sound of tubes, but in a situation like this, I don't think there's that much to be gained by using a tube over a JFET, MOSFET, BJT, or even an op-amp buffer.  BUT, it'll work this way too and will sound good.

As for the Pete Cornich QnA quotes, he's mostly correct in the effect of a tube buffer isolating the guitar from loading effects of pedals, but the same thing can be accomplished with any half-decent buffer.  The fact that it's a tube has little if anything to do with it at that point.  The other thing to remember is that some effects depend on the loading of the guitar pickups to sound right.  

Most of the stuff I've posted is ripped off of merlin's site:
http://www.freewebs.com/valvewizard/accf.html

He explains it much better than I can.

Matt

EDIT:
Something else to consider is if the pedal itself has a buffer on either input or output.  If it does, another buffer driving it (or being driven by it) won't be of much use.

[REGNAD]

"The impedance can't be measured w/ a meter unfortunately, but it can be calculated.  The impedance at 'A' and 'C' will be quite high, theoretically in the 6M-ohm area.  In practice, it'll be lower than that, but still high enough to ignore most of the time.  The output impedance at B will be very low, probably in the 400-500 ohm range.  That should be able to drive a load all the way down to the 5k area without any real loss of signal."

so when you say the impedance at a and c will be high,is this the impedance of the input of the tube circuit? and i have zero understanding with the last sentence

[morcey2]

"The impedance can't be measured w/ a meter unfortunately, but it can be calculated.  The impedance at 'A' and 'C' will be quite high, theoretically in the 6M-ohm area.  In practice, it'll be lower than that, but still high enough to ignore most of the time.  The output impedance at B will be very low, probably in the 400-500 ohm range.  That should be able to drive a load all the way down to the 5k area without any real loss of signal."

so when you say the impedance at a and c will be high,is this the impedance of the input of the tube circuit? and i have zero understanding with the last sentence

And that's how it should be.  Don't question, just do it.    

Just kidding.  Short answer is, for maximum signal (voltage) transfer, you want a low impedance driving a high impedance.  A buffer usually has such high input impedance ('A' and 'C') that almost anything can drive it without significant signal loss.  It also has such a low output impedance ('B') that it can drive almost anything ('pedal') without significant signal loss. 

I'll try to find a link that explains it better than I can when I get to work.  And pretend to work.  Because they pretend to pay me. 

Matt

[REGNAD]

, yes sir. i gathered my components today, maybe ill get the circuit built today...

[morcey2]

, yes sir. i gathered my components today, maybe ill get the circuit built today...

I wish my teenagers listened that well.   

[Ben N]

Umm, someone correct me if I'm wrong here, but don't CFs have low input-Z compared to common-cathode configured triode amplifier stages? Looking at that schematic, I don't see how the input Z at points A & C can be any more than 500k. The typical 12ax7 amp input stage has an input Z of ~1M. Following Cornish's logic, the first stage the guitar sees should be a common-cathode amplifier, not a CF. (Then again, following that self-same logic, you only need ONE tube input buffer to interface with the guitar, and not a slew of them scattered along the signal chain.) Allow me to suggest that the input buffer should be a single 12AU7 stage, a 12AX7 stage with a split plate load for lower output-Z, or a two stage buffer: common-cathode direct-coupled to a CF, for both high input-Z and low output-Z.

BTW, morcey2, when you make reference to the results of calculations, would you mind spelling them out for the less techically adept among us? Perhaps if you could show how you found 6M impedances at A &
C then I would be enlightened.

[morcey2]

Umm, someone correct me if I'm wrong here, but don't CFs have low input-Z compared to common-cathode configured triode amplifier stages? Looking at that schematic, I don't see how the input Z at points A & C can be any more than 500k. The typical 12ax7 amp input stage has an input Z of ~1M. Following Cornish's logic, the first stage the guitar sees should be a common-cathode amplifier, not a CF. (Then again, following that self-same logic, you only need ONE tube input buffer to interface with the guitar, and not a slew of them scattered along the signal chain.) Allow me to suggest that the input buffer should be a single 12AU7 stage, a 12AX7 stage with a split plate load for lower output-Z, or a two stage buffer: common-cathode direct-coupled to a CF, for both high input-Z and low output-Z.

BTW, morcey2, when you make reference to the results of calculations, would you mind spelling them out for the less techically adept among us? Perhaps if you could show how you found 6M impedances at A &
C then I would be enlightened.

I'm just headed to my son's football game, but if you'll look at the valvewizard page above, the calculations are there.  The short answer is that it is a bootstrapped stage because of the negative feedback of the load resistor (33k in this case) and the input Z is many times higher than the grid resistor.   

More to follow....

Matt

[REGNAD]

...........yeah,.....soooo i should get a flux capacitor instead?

[Ben N]

Thanks, Matt--that Valvewizard page is indeed an eye-opener.

[REGNAD]

im sorry, what pin numbers am i looking at in this schematic?

Wow this is why I read these threads and follow links.
Nice info guys.
Gus I really liked that interview you linked to I am guessing this is what started this idea-From Pete Cornich QnA...

"Tube Buffering? Does this enhance the tone at all?"

"Oh yes; when you plug your passive guitar directly into your tube amp it sounds great, doesn't it? Having a unity gain tube buffer as the input to my All Tube System allows the guitar to react exactly as if it were connected directly to the tube amp input (without the loading effects of long cables/effects pedals etc.). The effects outputs are connected to other unity gain tube buffers so that each effect "thinks" it is connected directly to the amp input allowing each effect to deliver it's full signal without the loading effects of long cables, other effects etc."

And what you pointed out about HIS use of the tube buffers in the MK2 setup,that there isnt one between each effect,they are used where they will do the most good.

RENGAD good luck ,let us know how it sounds when you finish,you(or Pete Cornish actually) may start a new trend,peace.

[morcey2]

im sorry, what pin numbers am i looking at in this schematic?

thought I posted this last night. 



so for one half it would be plate=1, grid=2, and cathode=3
for the other half triode, plate=6, grid=7, and cathode=8.

Matt.

[REGNAD]

thx, what's pins 4,9, and 5 for? when do they get used?

[morcey2]

thx, what's pins 4,9, and 5 for? when do they get used?

They're for the heater.  if you're using a 12V heater, ignore pin 9 and run 12V between 4 and 5.  If you're using 6V heater, jumper 4 and 5, then run 6V between 4/5 and 9.

[REGNAD]

.........these connections arnt in the schematic. could you add this t the a ove schematic so I don't f things up?

[Ben N]

If you look at most tue amp schematics, the connections for the heaters will not be included, since they are assumed, plus they are not part of the audio circuit. At most, they will be presented as a separate circuit (which is what they are) coming off the specialized heater tap on the power transformer. Putting the heaters in the middle of the audio circuit just makes the scematic more crowded and confusing.

[REGNAD]

simply connect power like this? so whats b+ for?