Partial feedback DHP SE amp

[Fuling]

Another thing you can try is to run the driver plate circuit from an ultralinear tap (if you have one available) from the OPT

I´ve thought about that, but in this case it´s impossible due to the lack of UL tap🙁

Adding or avoiding a 12K shunt (feed resistor vs. CSS) makes little difference.

I´ll go for CCS then, since resistor loading will give (too) low gain and swing capacity at this low B+. As a matter of fact I´ve already put together a little veroboard with two MJE350 based CCS´s.

A little status report on the project:

The top plate of the chassis is almost finished, only a few more holes to drill before polishing/brushing/painting or whatever I choose to to.
The OPTs and PSU chokes should arrive tomorrow, and with a little luck I have all the other parts at home already.

CCS board and filament PSU board are ready for testing.

Looking at the Hammond page, I see that there will be a 16 ohm tap at the secondary available for cathode feedback if this plate to plate stuff fails. 16 ohm is less than the 37 ohms that was suggested earlier, but more that 8 ohms, right?🙂

C3m in pentode mode could be an alternative if the PC86 fails...
Well, seems to be plenty of ways to skin this cat!

 

[mark_titano]

> Which are the differences between plate-to-plate and plate to grid feedback?

Plate to grid needs an extra capacitor, which complicates the frequency response calculations. Plate to plate often does not need an added cap. It also applies feedback around the grid coupling cap: for small feedback, that means you can uses a slightly smaller cap, for large feedback you have to do a full stability analysis.

Otherwise they do the same thing.

> can work with fixed bias? Any alteration in the math?

Yes. No.

PRR thankyou very much 🙂

In plate to plate feedback, however, I have to set the value of Rfdb thinking not only to AC condition but even to the different DC bias between driver and output tube, right?

I've read about the idea of using a ccs as anode load for the driver. What do you think about the possibility to use a mu-follower load (mosfet or tube)? In this way you can connect the feedback loop to the anode of the driver tube and take the output from the cathode of the upper tube/mosfet.
Now it should be possible to obtain an high impedance node for the feedback loop and a low impedance output to drive the power tube. But maybe there is something I've missed...

Mark