Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: tubenit on December 14, 2018, 08:59:27 am
-
While I am aware there are other options (including perhaps better ones), I am wondering IF this idea would work to split the plate resistors as shown and whether that would increase clean headroom? I've done this before but not with a cathode follower attached.
Thoughts? I've got some time to experiment in the next handful of days and thought I might try out a few ideas.
With respect, Tubenit
-
Just guessing here but i think splitting that resistor may decrease the headroom by moving the bias point closer to b+ voltage. For most clean headroom i'm thinking grid voltage of cf should be close to middle of ground and B+.
If the stage after the cf is distorting you can probably increase the headroom by splitting the 100k cathode resistor of the cathode follower.
-
With the DC-coupled cathode follower, you probably mess-up the bias too much to be happy. The "follower" will work as a diode: no current gain and some distortion. I'd say "try it", but it also may bring the heater-cathode voltage so high it ruins a good tube.
-
Since you're experimenting you may want to try this...
EDIT... added bias resistor
-
Since you're experimenting you may want to try this...
Interesting idea! Couple of questions: is that pot safe with that kind of voltage on it? And will this increase headroom which would be the goal.
With respect, Tubenit
-
I would use a PEC or similar mil spec pot. Can't say about 'headroom'. It acts just like any volume control. It will allow you to set the split resistors ratio to anything and still maintain a 100K overall plate load. The cap and resistor added to the grid of the CF will eliminate the dc problems that PRR mentioned.
-
Oops! I forgot a resistor. Hopefully I got it right now.
-
Or you can use level-shifted DC coupling (and it you should be able to make it work without stressing the h-k insulation in the CF stage)
-
If you have time, this is a good read and you can implement the math to your own circuit to find out the input sensitivity of each stage:
https://ampbooks.com/mobile/classic-circuits/soldano-slo/ (https://ampbooks.com/mobile/classic-circuits/soldano-slo/)
one point i found from it is that the stage before cathode follower (if i understood it correctly) goes to overdrive before the cathode follower.
with respect
vms
-
one point i found from it is that the stage before cathode follower (if i understood it correctly) goes to overdrive before the cathode follower.
I don't know what the book said, but it's the other way around.
I thought that the early distortion of the DC CF was the crux of the biscuit. We use the DC CF to purposely get crappy performance. If you want clean performance, use the one sluckey posted. That one allows you to design the gain stage for max headroom and to adjust the CF for max headroom.
-
The Fender design is marginally biased and the CF does tend to diode-out first.
Many copies are worse-biased and mostly work as diode.
If the two stages are designed very carefully for highest output level, the THD in the driver will be large while the CF is clean essentially until it all wall-slams.
If I am interpreting what is really going on big-picture: there is a lossy tone-stack after. If we really wanted clean we would not build-up high level and then cut it down. In g-world we often DO want clipping. But you need to look at the whole signal chain and pick your battles.
-
Tubenit, checkout Merlin's Preamp book, Ch 5, The Cathode Follower. The Replies here are correct that the driver stage's supply voltage > the CF grid must balance with the CF's cathode voltage for proper bias. So you can't just cut the grid voltage.
Merlin's book, pp. 131, Fig. 5.20 & 134, Fig. 5.22, show a tapped, split plate resistance for the driver tube. But these examples also show a bypassed K resistor for the driver tube, thus increasing the driver's gain.
The point of all this is that these variations alter the ability to overdrive, along with frequency response & resulting overdrive tones of the CF circuit as a whole.
To complicate matters, a small bottle pentode could be used as the driver or the CF. And there are numerous ways to connect driver > CF.
-
I genuinely appreciate everyone's response. THANK you!
As I've reread the responses and thought about it. I think there are much easier ways (that I am already familiar with) to increase headroom in the amp I was considering so I plan to pursue another route.
with respect, Tubenit