CVS loaded tube

Thread Starter

coinmaster

Joined Dec 24, 2015
502
I'm trying to allow my amplifier output stage to have variable operating points without the need to manually change resistors.

I think I've got it figured out for the input stage but the output stage is another matter.

For a WCF I'm thinking of putting a CCS on the cathode and a CVS on the anode, the CCS setting the current and the CVS setting the voltage. Unfortunately it's not working out the way I'd hoped.

Draft21-1.png

The far left is the original circuit, the middle is the CCS on cathode and the far right is the CVS. Everything is dandy until I change the amplitude of the signal then everything gets thrown off.
What am I missing? Why does a change in AC amplitude screw over the operating points?
 
Last edited:

Thread Starter

coinmaster

Joined Dec 24, 2015
502
I'm not sure what I did wrong, but I'm also not sure that the CVS I have on there is what I want to be using because while the voltage should ideally remain where I want it, the current doesn't.

Is there something that can be used to adjust voltage and current individually to where I want them? Sounds like an adjustable power supply but I need something fit for a triode load.

I saw this and it seems to be an adjustable current/voltage but I can't really tell.


Any ideas?
 

Papabravo

Joined Feb 24, 2006
21,159
Would you mind elaborating on why you think this is a worthwhile enterprise? Sounds like there are some problems with the approach.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
Because shifting the load line out of the linear range adds different tones to the music. Usually designers spend many hours finding the "sweet spot". My favorite amplifier has its operating point in a very strange spot on the load line in fact and sounds mesmerizing. So lets say I have a software with preloaded info about the loadlines of my tube.

Now lets say I have a little joystick that moves a virtual pointer on the overlay.
I want the operating point of my tubes to match the location of the pointer in real time.
A complicated task? Perhaps, but I have many uses for such a device and I will try very hard to make it work.

With a single triode I don't think the issue would be that hard, a simple CCS on the cathode with a mosfet gyrator load should allow me to set the operating points because the current through the tube should match whats on the cathode no matter what voltage I apply to the anode.

But the WCF is a bit more complicated. The current on the anode of the top triode doesn't always stick with the current applied to the cathode of the bottom trode. So I have to figure out how to adjust and regulate both the voltage and the current for the top triode.
This could be easier to achieve if I used the top triodes grid to adjust the impedance of the tube but unfortunately I have a DC offset servo biasing that grid so the output of the WCF is DC nulled allowing OTL/OCL output so I can't really mess with the grid.
That does mean that the servo should be matching the impedance of the top triode with the bottom triode though.

In order to do this I may have to figure out why the bottom triodes grid is biased at
-100v while the top triode is biased at around -20v on the grid. I still haven't been able to figure that out especially since -100v grid voltage should be cutting off the tube according to the load line at the given voltage.

In any case Ohms law states that you can arrive at different voltages and currents just by changing the input voltage and resistance, I am frustratingly not at a point where I can figure out how to make such a circuit on my own.
 
Last edited:

Thread Starter

coinmaster

Joined Dec 24, 2015
502
I think the first step is to figure out how to make a CCVS. It should be possible, if you change the voltage and resistance you should be able to come up with any result you want.
I would need a circuit that would allow me to change the voltage and keep the current and also change the current while keeping the voltage.
Like changing the x and y axis on the load lines.
 

Thread Starter

coinmaster

Joined Dec 24, 2015
502
Maybe a logic circuit that can do ohms law calculations to adjust the v/i/r ? Seems like the most straightforward solution.
But then again ohms law won't give the appropriate resistance for a plate load, you would have to use a different formula for that but the formula for calculating plate resistors is only an approximation since tubes don't perfectly match their datasheets, meaning there is no calculable solution to define an exact resistance.

Perhaps it doesn't matter, you may not be able to predict the result but could use the formula to get an approximation and adjust the dial until it physically reads the current/voltage you want.
But then again for a WCF I don't think the normal formula applies , so perhaps there is a way to calculate the impedance of the divider that the WCF creates and then use that to calculate the resistance but I'm not sure how to calculate active tube impedance.

Or maybe it's just better to stick with ohms law calculations and just offset it.
What do you think?
 
Last edited:
Top