G2 PS PT little question

[kagliostro]

Some time ago I was planning an SE amp with 1625 (or 807) tube

on the datasheet with 350V on plate and 250V on G2 (Grid voltage -18) consunption of G2 is 2.5mA (54mA Plate)

assuming to solve the problem to feed G2 at a 100V lower than B+ using this method (a small transformer connected to filament winding)



which will be the smaller, in VA, transformer that can be used ? A 3VA transformer will be enough ?

I did a bit of math (always unsure about my math) I guessed that a trasformer rated for 1VA will be enough ( )

Many Thanks for any help

Franco

[thetragichero]

are you set on using a transformer for this?
how about another branch off the plate node? assuming plate node is 350v you need to drop 100v at 2.5ma. that's 40k dissipating 250mW. let's say 39k 1W resistor?
maybe i'm a cheapskate but that's adding a cents-costing part (plus electro capacitor) instead of dollars-costing transformer plus diode bridge plus plus plus....

[kagliostro]

Thanks for the answer

... assuming plate node is 350v you need to drop 100v at 2.5ma. that's 40k dissipating 250mW. let's say 39k 1W resistor?

Unfortunately that will not work, 100V is a too large amout and it will not be result in a stable G2 voltage (consider that there is a dinamic on the functioning)

Franco

[tubeswell]

Add up the V.A of each secondary winding and then divide that ‘V.A sum’ by the PT mains voltage to get the V.A of the Primary. (The primary V.A is the PT V.A and the mains fuse wants to be a SloBlo 2x the current of that)

The V.A of the boost PT works the same way. (The secondary load determines the current that the primary needs to handle). If ‘D’ is for g2 of the 801, use a factor 2x greater than the maximum current that g2 will see (to get the ‘secondary’ V.A). (Also, you may want to think about more filtering for the g2 supply. YMMV)

[kagliostro]

Ciao Tubeswell

Many Thanks

On 807 datasheet (1625 is the same) with 350V Plate & 250V Screen current on screen is indicated as 2.5mA



but on 6L6G datasheet (considered identical to 807 tube, only Top Cap difference) is specified 2.5mA with zero signal and 7mA at max signal (18V AF peak)



as you say I must use the 7mA doubled and consider 14mA for math

so ... 250V  x 0.014A = 3.5 W, at least a .... 4 VA 230V/18V transformer if exist (the PT I've has 14V for Heaters and so I need a 230V/18V transformer to obtain 250V DC rectified)

Please, spent a bit of your precious time and write which will be your math

--

... Also, you may want to think about more filtering for the g2 supply ...

The PS I posted was only an example, but ... yes, my fault, I didn't considered a better filter

Many Thanks

Franco

[sluckey]

Maybe something like this...

[kagliostro]

Ciao Steve

true, even that could be a solution, I hadn't thought about it

I think I've an example of G2 feeded near that way, just a moment ......

Find !





Franco

[sluckey]

One slight difference... Your circuit will always put 150V on the screens regardless of what the plate voltage is. In my circuit the screen will always be 100V less than the plate. IE, if plate rises to 400V screen will rise to 300V. If plate drops to 300V screen will drop to 200V. One thing about my circuit... I forgot to calculate the wattage rating of the 10K. It will actually dissipate 6.25W with voltages as shown. So use at least a 12.5W resistor for safety.

[tubeswell]

Ciao Tubeswell

Many Thanks

On 807 datasheet (1625 is the same) with 350V Plate & 250V Screen current on screen is indicated as 2.5mA



but on 6L6G datasheet (considered identical to 807 tube, only Top Cap difference) is specified 2.5mA with zero signal and 7mA at max signal (18V AF peak)



as you say I must use the 7mA doubled and consider 14mA for math

so ... 250V  x 0.014A = 3.5 W, at least a .... 4 VA 230V/18V transformer if exist (the PT I've has 14V for Heaters and so I need a 230V/18V transformer to obtain 250V DC rectified)

Please, spent a bit of your precious time and write which will be your math

--

... Also, you may want to think about more filtering for the g2 supply ...

The PS I posted was only an example, but ... yes, my fault, I didn't considered a better filter

Many Thanks

Franco

Allowing for 2x current headroom on the 185VAC ‘secondary’*, that’s 185 x 0.014 = 2.59V.A (2.6W)

Current rating on the 12.6VAC ‘Primary’* = 2.59/12.6 = 0.2A


*going by your schematic

[kagliostro]

Thanks Friends

@ Sluckey

.... In my circuit the screen will always be 100V less than the plate.

That is appreciable, that way there is a more evident SAG effect

--

@ Tubeswell

Allowing for 2x current headroom on the 185VAC ‘secondary’*, that’s 185 x 0.014 = 2.59V.A (2.6W)

Current rating on the 12.6VAC ‘Primary’* = 2.59/12.6 = 0.2A

*going by your schematic

My thinking was "slightly" different, thanks to correct me

Again Many Thanks Friends

Franco

[pdf64]

Thanks Friends

@ Sluckey

.... In my circuit the screen will always be 100V less than the plate.

That is appreciable, that way there is a more evident SAG effect
…

I don’t see that, just a different lower g2 curve on the anode chart.
ie the g2 supply should track the main HT, minus 100V.
No additional sag, at least not by Aiken’s definition. That’s the whole point of the zener or OC regulator method, compared to a resistive dropper.  https://www.aikenamps.com/index.php/what-is-sag

[sluckey]

You could use suitable zener(s) rather than the 0B2 for a smaller footprint. Zeners don't look as cool though. 

[mresistor]

There are even SS equivalents of the tubes   http://www.solidstatetubes.com/0a2-and-0b2.html

[sluckey]

Hmm. A $29 SS replacement for a $5 tube? No thanks.

[kagliostro]

.... Zeners don't look as cool though.

I agree 

0B2 will give a charmant looking effect for sure if used

.... No additional sag

I was thinking to the way in some circuits VVR are connected (to G2 instead to feed plate) and so I was thinking a variable G2 voltage, instead of stabilized one, will allow an easyer SAG effect .... but if you disagree .... You are the expert

There are even SS equivalents of the tubes

That is interesting, never heard about it, thanks for the info, but at this point, on my case, a VVR feeding G2 will give the needed voltage on G2 and also allow to further decrease it to tame the power if desired

Many Thanks

Franco

[tubeswell]

variable G2 voltage

Cool idea methinks, but easier/more practical to implement on a cathode bias stage. Tube current (and by implication, screen current) and bias are interrelated. If you change Ig2 (by changing Vg2), you change Ik, and hence also the tube bias/operating point.  With fixed bias, a VVR for Vg2 would probably also need to change Vg1 somewhat, to keep the tube within happy operating parameters. Depending on bias point, if Vg2 is too low, tube current stops, and vice versa if Vg2 is too high. (Whereas cathode bias is 'auto biasing', which counteracts both of these tendencies)

Variable Vg2 on an output stage would be cool to experiment with to see how much effect it has on toan. Changing Vg2 changes input sensitivity, gain and Gm (without having to resort to a 'triode switch'). But I think you would want to keep the output tubes within a reasonable loadline ballpark (assuming you wouldn't be changing the load), and so probably wouldn't want to make drastic changes to Vg2. YMMV

[tubeswell]

Another way of regulating Vg2 (if you have enough heater juice in your PT) is to pick a 'conventional' tube that makes a good regulator 'diode' (it has to be able to comfortably supply enough total current for the screens without exceeding Pmax). The regulator tube's output voltage is modulated by the bias voltage on the grid, which is fixed w.r.t ground. (By the 1M/2M voltage divider in attached schematic. You could make this into VVR by putting a pot at the junction of the KT61 grid and the voltage divider to dial-in the regulator tube's current - and resulting cathode voltage). Changes in output tube screen voltage are resisted by the regulator tube’s bias point.

The regulator tube's heaters should be elevated sufficiently to stay within H-K insulation parameters (and it goes without saying that the bias adjustment shouldn't let the regulator tube exceed Pmax)

[kagliostro]

Thanks Tubeswell

... but easier/more practical to implement on a cathode bias stage.

Yes, I know, I agree with you

... Another way of regulating Vg2 (if you have enough heater juice in your PT) is to pick a 'conventional' tube .....

Seems interesting, but .... a bit more complicated to be realized in practice

Thanks again

Franco