EL34 SE pin 1 ground ?

[Wilder Amplification]

Not much more to say really! Essentially, the cathode current of a pentode is substantially constant for a given bias voltage. Most of it comes from the anode, but some comes from the screen, and while the anode voltage is greater than the knee voltage the ratio of anode current to screen current is more-or-less constant (typically about 5-7 for tetrodes, 8-10 for pentodes). This ratio is determined by the suppressor voltage. As you make the suppressor more negative you increase the ratio (no, I don't know by how much- I have never looked into it!) so you get less anode current but more screen current.
In terms of what comes out of the speaker, I would guess it is approximately equivalent to lowering the screen voltage by some proportional amount.

As Joey observed, although it reduces peak anode and screen currents, the quiescent screen current increases. With a new build you can take this into account by just using larger screen stoppers (1.5k?) or something. Eventually I will look into exactly how more "at risk" the screen is, but that's an experiment for another day.

On my 100 watt hot rod 2203 clone I have the negative suppressor grid setup. My quiescent measurements aren't any different than when I had pin 1 tied to the cathode -

Va = 460V
Vg2 = 455V
Biased at 35mA x 2 on each side of OT primary

On 1K screen resistors I'm seeing a 5 volt drop, which translates to 5mA per screen at idle. That's about what they idled at prior to converting to the negative suppressor grids.

[Steve Conner]

Well, I believe it's the other way around. The quiescent screen current is hardly affected by the suppressor grid, what's affected is the division of the electron stream between screen and plate at peak signal conditions. This is when the plate is at its lowest voltage, and is the least attractive target for electrons, so the suppressor makes the greatest difference.

I found that the lowest peak signal screen current was had by putting the suppressor at 0V as the tube designer intended. Making it negative reduced plate current and increased screen current, and oddly making it positive did exactly the same thing! But the effect wasn't very large in either case.

[Joey Voltage]

I found that the lowest peak signal screen current was had by putting the suppressor at 0V as the tube designer intended. Making it negative reduced plate current and increased screen current, and oddly making it positive did exactly the same thing! But the effect wasn't very large in either case.

Interesting.

[ThermionicScott]

This has been informative! Has anyone played around with injecting signal onto the suppressor grid (for more linearity, tube life, etc)?

- Scott

[Wilder Amplification]

This has been informative! Has anyone played around with injecting signal onto the suppressor grid (for more linearity, tube life, etc)?

- Scott

Yes...they don't like that too well.

[bloom]

and what about pin 6? Can I just tied it to pin 1, grounded all together? Pin 6 is not explained on the data sheet for EL34...

[bloom]

never mind, pin 6 is not used on EL34.. Thanks for the discussion, it really clarifies for pin 1.