6N6P with elevated cathode: where to connect screen-pin?

[planet IX]

In the shown application the 6N6P has its cathode potential at roughly 180V (both cathodes actually).
Were am I supposed to connect the screen-pin to? Is it okay when I still connect it to GND?
Or should I connect it to an elevated potential? For example put a 180V Zener between it and GND?

Thanks, Boris

 

[lcsaszar]

The pin 9 (shield) should be connected to GND.

 

[Steve Morley]

IX,
Assuming you are referring to the shield connected to pin 9, it can be connected to ground though a cap if you don't want to have it at DC ground potential. The cap will allow AC nasties to be sent to ground.

S.

 

[6A3sUMMER]

I agree with Icsaszar,

If you only connect pin 9 through a capacitor to ground, then the DC voltage of the shield will be . . .
Whatever it wants to be, moment by moment, according to any stray electrons in the tube.

I prefer to connect such elements to ground.
Your Mileage May Vary.

 

[planet IX]

Thank you all. Actually my intention is to connect the screen to ground. I only was not sure if the potential difference between cathode/heater (180V) and screen (0V) could cause any harm to the tube...

 

[triode_al]

In that design, the screen often goes to a voltage divider, say at about 60 v. With a cap to earth.

 

[6A3sUMMER]

Elevating the screen (shield), seems very surprising to me.
I believe it shields one triode from the other, so that it lowers the capacitance from triode 1 to triode 2.

Any free electrons would be attracted to an elevated screen (like the action of a plate).

This is a dual triode tube.
It is not a Pentode, and it is not a Beam Power tube, and it is not a Beam Tetrode tube.

Just my opinions.

 

[kevinkr]

I always ground the shield in the 6N6P. The electrodes it's seeing are the plates, not cathodes so cathode potential is probably irrelevant here.

Make sure you are not violating filament/cathode insulation ratings, you may need to float the filaments at mid-point between ground and cathode potential, unless you have dedicated filament windings and use one tube for voltage amplification and the other as a cathode follower. (assuming stereo here)

 

[kodabmx]

Ground it, float it, in my experience it doesn't make a lick of difference for audio.

 

[6A3sUMMER]

Capacitance between 2 equal flat area conductors facing each other, spaced by distance D,
is proportional to 1/(Distance squared).

Between two flat area conductors, Let D = 1, you get 1 / (1 squared) = 1

Then put a floating flat area conductor between the two other flat area conductors, and you have two series capacitors, so for the total capacitance divide each capacitance by 2.
you get:
(1 / (1/2 D squared)) / 2 = (1 / (0.5 squared)) / 2
= (1 / (0.25)) / 2
= 4 / 2 = 2
2 times the capacitance with the shield in place versus no shield in place.

True, the plates are not flat conductors, they are shaped (round). But the same principals apply.
You probably should ground the shield. If not, ask the manufacturer to take the shield out on their next run of the production line.

Just my opinion.

 

[kodabmx]

The Caa of E88CC tube is 0.045pF... would double that make any practical difference, especially at AF? I really don't think so.

http://frank.yueksel.org/sheets/030/e/E88CC.pdf

 

[6A3sUMMER]

The E88CC is classified as an RF double triode.

I bet the measurement of the plate to plate capacitance is specified with the shield pin 9 grounded (not floating).

 

[triode_al]

In that design, the screen often goes to a voltage divider, say at about 60 v. With a cap to earth.

oops wrong 🤢🤢 I agree the shield (not called screen to avoid confusion) goes to ground [My error, being a couch potato, is I thought the reference was the heater. oh my.]

1. The shield might be needed in very high radio frequencies, but in audio we never see those, our impedances are too high, so our bandwidth is already very very limited. I mean, you can't get 30dB gain and >200kHz bandwidth.
2. The reduction in plate plate capacitance has no impact on the signal, specifically not in the above circuit: the top plate is already grounded (for AC of course).

 

[Depanatoru]

Shielded dual trides could be made with much less distance between anodes ( so higher capacitance ) , I wouldn't risk left it floating even for audio ,,,
Improved versions of 12AX7 like ECC808 and others , all have shields and are made just for audio .

 

[planet IX]

Again, thank you all. I mistakenly used the term screen when I meant shield.
But now it is clear that the shield (Pin 9) goes to GND.

Cheers, Boris