[Niteshade]

Fisher has an interesting setup with some of their integrated amps: The power tubes will be positively biased because driver & preamp tube filaments are actually ran from the power tube's cathodes.

The cathode's voltage output follows the bias voltage closely.  Benefits? Noise elimination. Cons? Hot running power tubes.

[Niteshade]

I'm surprised that nobody has picked up on this topic- it's rather unusual to use cathode voltage to operate filaments in the preamp & phase splitter sections section. I would consider Fisher's idea to be a 1st generation rendition of this novel approach, therefore there is enormous room for improvement.  aa

[Niteshade]

It looks like this topic was clearly a dud. 

It is an unusual approach to design that Fisher most likely pursued to cut costs- but it makes sense.

Look here:

1. No selenium rectifiers (60's, remember)
2. No large electrolytic capacitors needed for low resistance DC loads.

I see LOTS of potential! I have to assume nobody else does. 

Just because it's something that was most likely forgotten due to complications or associated with cheap tube gear is no reason to shun it.

For example: The modern membrane-style personal computer keyboard was the most unreliable thing around until it was revisited a couple decades ago- THEN it became ***the most** reliable technology around!

[JoshK]

Do you mean like this?

http://www.tubecad.com/july2000/page4.html

I am just not certain from your description what your referring to.  Schematics or diagrams help.

Half way down the blog, here as well

http://www.tubecad.com/2006/05/blog0065.htm

Note: in the examples I linked to, the power tubes are still negatively biased.  Not sure how you are positively biasing your output tubes unless you are driving it in class B with another power stage. 

[JohnR]

I also don't see why running the driver filaments from the power tube cathode makes the power tubes positively biased. Which unit specifically are we talking about? It's possible I have a schematic.

[Niteshade]

Here's a partial schematic of a Fisher X-101C.

[JohnR]

If I'm reading this correctly, the grid is biased negative relative to the cathode by R62, R67, R74. The schematic in fact indicates an intended bias of about -11.5V.

It is an interesting choice though, I assume the general operating conditions must make everything quite robust, as pulling out any of V1-V4 or V7-V10 must throw all of the biasing off quite a bit.

[Niteshade]

I was referencing all voltages to chassis ground with the model I was working on.  In the past I removed this circuit, installed cathode biasing and DC filament voltage to those tubes. It's funny- now I want to make something like this for the fun of it and see how it turns out.

If I'm reading this correctly, the grid is biased negative relative to the cathode by R62, R67, R74. The schematic in fact indicates an intended bias of about -11.5V.

It is an interesting choice though, I assume the general operating conditions must make everything quite robust, as pulling out any of V1-V4 or V7-V10 must throw all of the biasing off quite a bit.

[JohnR]

Cool - let us know. It would be interesting to see what kind of minimal circuit might result.