Triode-connected 6BW6 cathode follower. Gain will be low and distortion possibly high because of the low cathode resistor - this could be regarded as a textbook example of a poorly-designed CF making an elementary mistake. Fortunately line level signals may be low enough in voltage to get away with it.
Gotcha, I did say the resistors needed tinkering....
Low gain, GOOD, line level voltages (in & out), yes. NON inverting, bonus.
I assume the 'elementary mistake' you speak of is the low Rk then? If poorly designed, how to improve upon it?
Andy.
Low gain, GOOD, line level voltages (in & out), yes. NON inverting, bonus.
I assume the 'elementary mistake' you speak of is the low Rk then? If poorly designed, how to improve upon it?
Andy.
Yes, low Rk. Gain will be around 0.5? You can't simply increase Rk as that will affect the bias.
Change to the classic valve CF: two cathode resistors, one to set the bias and the other (larger value) to act as tail. Grid resistor goes to their junction, instead of ground. You would need to add an input coupling capacitor.
6BW6 is a strange choice. It has lowish gm but takes high current, as it is a miniature version of the 6V6. You would get excellent current drive capability, which will be good if you have a few km of interconnect cable at the output but unnecessary otherwise.
Change to the classic valve CF: two cathode resistors, one to set the bias and the other (larger value) to act as tail. Grid resistor goes to their junction, instead of ground. You would need to add an input coupling capacitor.
6BW6 is a strange choice. It has lowish gm but takes high current, as it is a miniature version of the 6V6. You would get excellent current drive capability, which will be good if you have a few km of interconnect cable at the output but unnecessary otherwise.
Okies, thanks for the tips - I think I know what you are describing as I have seen that in other circuits. What would be a better choice of valve do you think? Need a gain approaching unity as the goal is a buffer circuit.
Getting there I think!
Andy
Getting there I think!
Andy
Not to blog-flog, but you may want to read my article on cathode follower design:
The Heretical Preamp
The Heretical Preamp
Ohh, I shall personally peruse your site SY, thank you!
Andy.
p.s. Sorry to rain on your parade SY but I very much doubt you are "meaner than my ex-wife"!😛
Andy.
p.s. Sorry to rain on your parade SY but I very much doubt you are "meaner than my ex-wife"!😛
Something akin to this then?
R values will need head-scratching...
A
An externally hosted image should be here but it was not working when we last tested it.
R values will need head-scratching...
A
Yes, that would be better. I would not use a 6BW6 for this, but then I would not use a buffer anyway! Half an ECC88 would make a better buffer.
Something akin to this then?
R values will need head-scratching...
A
You must bypass the g2 supply (to the cathode)
7N7
No need, unless you particularly want to run the valve in 'pentode' mode. If you want that, then you will need to signficantly increase the value of R4 as otherwise it would put too much AC load on the cathode.
Voltage on g2 determines pentode behaviour so it is important that g2 voltage is "tracked". However with a 100-ohm resistor you would need about 560uF!! Why only 100 ohm? You must determine from the curves the likely g2 current and choose a resistor to drop the voltage to your desired value. In recent pentode applications I have been running g2 at about 190V with something like a 20k resistor and 4.7uF by pass.
Which valve are you using/proposing to use?
EDIT Ah! I see triode mode!! (blush) Why use a pentode then?
7N7
It will work either way. You have shown a circuit which uses it in triode mode. You could choose to change it to pentode mode. This would not be a correction but a modification. The circuit as shown does not need a cap between g2 and cathode; indeed adding one would introduce a fault unless you also increased R4.
Why use a pentode then?
Hmmm, allow me to embarrass myself too then.
I started with a Triode based design. But all I want is a buffer really and the design posted above that I built provided too much gain and inverted the signal (which I would really rather not have happen). Therefore more 'stuff' would need to be added and I am trying to keep things as simple as possible.
So then a college here at work suggested a Pentode wired as a Triode. This has the lower gain I want, will provide simplicity and also not invert the signal.... *I think*!
I have not seen any such Triode-based designs that can do that for me simply (and by that I also mean not having to provide plus & minus supply rails or >300V B+ rails etc etc).
I could of course be talking utter b0ll0cks. It has been known. On >1 occasion. Very steep learning curve for me all this. But FUN t'boot.
Andy
Hmmm, allow me to embarrass myself too then.
I started with a Triode based design. But all I want is a buffer really and the design posted above that I built provided too much gain and inverted the signal (which I would really rather not have happen). Therefore more 'stuff' would need to be added and I am trying to keep things as simple as possible.
So then a college here at work suggested a Pentode wired as a Triode. This has the lower gain I want, will provide simplicity and also not invert the signal.... *I think*!
I have not seen any such Triode-based designs that can do that for me simply (and by that I also mean not having to provide plus & minus supply rails or >300V B+ rails etc etc).
I could of course be talking utter b0ll0cks. It has been known. On >1 occasion. Very steep learning curve for me all this. But FUN t'boot.
Andy
You could ask your "college" why he thinks a pentode running 40mA (and 4mA/V) will make a better driver for a few metres of interconnect than a triode running 10mA (and 12mA/V). This may be a silly question if the aim is merely to impress someone with some valve 'jewellery', which I think is what you said early on. Maybe you could use a real 6V6 - that would look more impressive than a 6BW6.
Running at 20mA to 25mA and 6BW6's are cheap (and he saw some in my collection of valves) and he doubted very much if anyone would notice the difference between using a Triode and a Pentode (wired as a Triode). Oh & as he's shown me, it doesn't invert either.
It was only a suggestion at the end of the day which I thought was worth bringing up to see the pro's and con's of using such a topology.
Circuit now updated with hopeful more sensible values:
It was only a suggestion at the end of the day which I thought was worth bringing up to see the pro's and con's of using such a topology.
Circuit now updated with hopeful more sensible values:
Gain will now be lower as the transconductance has gone down but the cathode load has also gone down.
"Gain"? I thought a cathode follower had no (or very little) gain?
Voltage to current converter is how I have been thinking of this & in doing so I'm now probably going to open up a can of whoop-a$$. 🙄
Andy
Voltage to current converter is how I have been thinking of this & in doing so I'm now probably going to open up a can of whoop-a$$. 🙄
Andy
Gain can be less than unity. That might appear to be a perverse use of the term gain, but otherwise we might have to talk about 'amplifiers' which attenuate the signal. The voltage gain of a CF is roughly given by Rk/(Rk + 1/gm) for lowish Rk (like this design) or (1-1/mu) for highish Rk.
A CF is a voltage-to-current converter with high amounts of negative feedback.
A CF is a voltage-to-current converter with high amounts of negative feedback.
- Status
- Not open for further replies.