# 0A3 and pentode questions (quickie)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

#### Simpleton

Gentleman, i bow before your knowledge and dare to ask: (ain't i a simpleton )

a) I read on and article from a 50's magazine that the gain of a pentode is G=gm x Ra
For a simple pentode stage, with gm=0.9mA/V and Ra=56KOhm, the gain goes to around 50X! Isn't this too much? (the pentode's internal anode resistence is around 2MOhm)

b) How can i use a nice 0A3, which i happen to have 2, as a power on indicator light? I know i could use a simple neon bulb, etc, but the 0A3 is sooo cute

Thank you all so much.

Cheers!

Francisco

#### Pentium100

I suppose 0A3 is voltage regulator tube with cold cathode. If so, you can use it as neon lamp (it may require lower resistanc).

#### Sch3mat1c

Simpleton said:
a) I read on and article from a 50's magazine that the gain of a pentode is G=gm x Ra

Approximately. For RL << Rp, it's pretty close.

For a simple pentode stage, with gm=0.9mA/V and Ra=56KOhm, the gain goes to around 50X! Isn't this too much? (the pentode's internal anode resistence is around 2MOhm)

Nope- works just right! Consider what Gm is to begin with: a change in plate current for a change in grid voltage. (Sidenote: ohms are volts divided by amps (R = V/I); conductance is the reciprocal of ohms, or I/V. Gain is rated in terms of output figure over input figure, in the case of a voltage-controlled device like a pentode, you have a change in output current divided by a change in grid voltage - because they aren't directly connected, we tack on a "trans" and name it "trans-conductance". )
Now, considering the plate resistance is up at 2 megs, we can ignore it altogether. That means a change in grid voltage of maybe 0.1V will cause a change in plate current of 0.1 * 900 x 10^-6 = 90 x 10^-6, 90 microamperes. Across a 220k resistor (current must be quite low for a mere 900umhos, so I'm assuming a large RL) this causes a change of V = IR = 19.8 volts, for a total gain of almost 200!
Indeed, into a constant current source, where Rp dominates, rather than RL, mu (amplification factor) is equal to Gm * Rp, or for your example, µ = 1800. Compare that to a single triode with 100 tops!

Typical gain stages (look at the GE datasheet for 6AU6 and other popular pentodes) quote gain in the range of 100 to 500.

b) How can i use a nice 0A3, which i happen to have 2, as a power on indicator light? I know i could use a simple neon bulb, etc, but the 0A3 is sooo cute

Hehe. One, you can always put a regulated supply in your amp - at 90V (75? I forget), it would work for screen supply for some late-era sweep tubes (6LQ6, et al), two in series (uh, 150/180V) for powering a preamp or regular (6L6, EL34, etc.) screens, or use one as voltage reference for a pass regulator of any voltage. If you can't squeeze that in, you can simply add a series resistor and connect it to any power supply, AC or DC. You'll need about 1-1.5k (2-5W) per tube to run from 120VAC, or 5-10k from +300V. Don't parallel them; like LEDs and zeners, one will hog all the current.

Tim

#### Simpleton

That was one great explanation Sch3mat1c! Really really good!
It's just because i made a pre-amp cicrcuit with a small russian tube (in the photo section) and the author claims a gainof only 5.5X!
This tube has a bigger transconductance and Ra than this particular tube i presented here.

My thing with the 0A3 is:

using it before the power tranny. Should i use something:

mains ---> ------->transfomer
|
R
|
0A3
|
mains---------> transformer

or in series

mains----- R---0A3 ---tranny

mains -------------------tranny?

my doubt with // conection as above is if it will regulate the mains or 75V

Thanks!

#### Tekko

Connect it in paralell with the power transformers mains winding with a current limiting resistor and it will glow like a big neon bulb.

#### PRR

Paid Member
The Pentode equation should mention the input resistance of the following stage. For Rp=56K, Rg=1Meg, this makes little difference.

Gains of over 400 are possible, and are done in the Dynacos. Too much? The input sensitivity would be ~0.1V, but then we put negative feedback around the whole thing to get input sensitivity about 1V, 10:1 given-over to negative feedback.

Don't run the OA3 on AC. I suppose it will work(*), but it will cause electric buzz like a Neon lamp, except 10 times worse because an OA3 needs about 10 times the current of a NE2 to get lit properly (and more for a fat glow). If fed AC, this large current will switch on and off 60 or 120 times a second: buzz-city.

"A series resistor must always be used... The resistance value must be chosen so the maximum current is not exceeded at the highest anode-supply voltage and so that the minimum current rating is always exceeded at the lowest anode supply voltage."

Wire it exactly like you would for a voltage regulator, just don't put any load on it if you don't have an appropriate load. You need over 125V DC, with more than 5mA of spare current. If you have a 250V supply that can stand an extra 5mA load, that will work fine. After starting, the OA3 will drop 75V. The resistor would then drop 250V-75V= 175V. The OA3 will not run well with less than 5mA. So the maximum resistor value is 175V/5mA= 35K. The resistor heat is 175V*5mA= 0.875 Watts, use a 2 Watt part.

If you do have a load that likes a steady 75V, you can tap it from the OA3. Reduce the value of the series resistor to pass the added current while leaving at least 5mA, never more than 40mA, through the OA3.

Do NOT put a capacitor across the gas-tube, or you create an Oscillator. The datasheets say that 0.1uFd is stable. Try a good 10uFd and you have a blinker (possibly with very short life).

Self-noise of a gas-tube can be as much as 5mV, so we don't power sensitive circuits directly. A capacitor can mute most of the noise but can not be placed directly across the gas tube. I believe 500 ohms between tube and cap is enough to supress oscillation.

(*) According to Philips: "The tube should be operated only with the cathode negative and the anode positive". I suspect it will work backward but perhaps at different voltage and much less power rating. STC says "Operation with reversed polarity will damage this valve."

#### Sch3mat1c

PRR said:
(*) According to Philips: "The tube should be operated only with the cathode negative and the anode positive". I suspect it will work backward but perhaps at different voltage and much less power rating. STC says "Operation with reversed polarity will damage this valve."

Pffbt. They rated it for regulating voltage, not merely producing light.

If you would like to use it on AC anyway, you'd use the parallel arrangement. The resistor prevents it from trying to regulate the mains and blowing the tube and a circuit breaker! (In that order: at least with small neons. That I tried once, years ago. I have never tried, nor do I wish to try, a 0A3 as so. )
If noise does prove to be an issue, you can try adding a capacitor across the mains lines. If it's still too noisy, you'll have to go with DC. The disadvantage is the resistor drops most of the voltage (75V tube, maybe 300V B+, resistor takes 225V!) and thus burns a whole lot of power. 30mA is also nothing to cough at; that's what a 6BM8 takes for 2W output!

Tim

#### Simpleton

I'm still waiting fot the chassis, a metal working friend of mine ordered one for me at the machining shop.

One thing i would like to ask (and i've been forgetting about it since i first posted ):

Should i move the B+ from one filament leg of a 5U4G ( the same PSU Sherman made for mikael's KT88 amp - my turn now) to the other from time to time to avoid depletion of the oxi layer?

#### Sch3mat1c

Huh? How? No. If nothing else, the filament resistance is utterly invisible compared to the voltage and current of B+ (though the 5VAC present *can* introduce a few volts of harder-to-filter 50/60Hz ripple if the filter is connected to the wrong cathode!).

Tim

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.