Diff amp driver -> SE output

[Saurav]

Hi,

I have 2 ECC99s for 2 channels, which means I have a pair of triodes sitting around unused. My initial idea of paralleling them to increase gain was aborted quickly when I learnt that it didn't work that way. The triodes that are in use are CCS loaded, so I don't want to try SRPP or mu stage. I was thinking of things I could do with those triodes, and wondered if a diff amp would make sense? I'd lose gain, which won't be good, I don't have much gain to spare. But would I gain anything? Would my PSRR increase much over what the CCS load gives me right now? Is this worth pursuing?

Any other ideas for using the two spare triodes?

Thanks,
Saurav

[fdegrove]

Hi,

My initial idea of paralleling them to increase gain was aborted quickly when I learnt that it didn't work that way.

It doesn't work that way...
BUT it halves impedances, noise and doubles transconductance, the latter is a nice bonus but requires good matching so no current hogging occurs.

Cheers,

[universum]

Quote: Any other ideas for using the two spare triodes?

Well,what about using them to condition your supply up..,by making two (seperate for each channel)shunt-regulators?Good to do with those two *spares*.
It probably helps you to lower the capacity on the supply,thus reducing phaseshift.
Or u use them as kathode loads,when you can make a negative rail to it?
Just thinking out loud...

Have a nice day!

[Saurav]

Hmm... I hadn't made the connection between triodes and shunt regulators. That could be an interesting idea. This obviously wouldn't be as straightforward as a VR tube, right, I'd have to build a regulator circuit around the triodes.

I'm not sure I understand what you mean by a cathode load.

BUT it halves impedances, noise and doubles transconductance, the latter is a nice bonus but requires good matching so no current hogging occurs.

And also increases input capacitance, which given the fact that I'm driving this with a passive linestage, may be the overriding factor.

It did sound a little smoother/softer with the paralleled triodes though. Not sure what factors contributed to it. One possibility would be the increased heater current draw, which brought the voltage down closer to 6.3V (it was around 7.1V without the spare triodes being turned on).

[fdegrove]

Hi,

And also increases input capacitance, which given the fact that I'm driving this with a passive linestage, may be the overriding factor.

Certainly...
Add to that the cable capacitance and it sure will roll off the highs.

which brought the voltage down closer to 6.3V (it was around 7.1V without the spare triodes being turned on).

One way of burning tubes like candles it to run them at too high heather voltages...
It will sound great for a while but you're just exhausting the cathode at an accellerated rate.

Any tube tester will show you that...
In fact I bet those tubes already are at 50% as it is now...

Cheers,

[Saurav]

I have a resistor in there now to bring the voltage down. 50% already... that's not good news. How long does it take for the tube's life/emission to go down when operating at elevated heater voltages?

[fdegrove]

Hi,

How long does it take for the tube's life/emission to go down when operating at elevated heater voltages?

Once I did a test running a 6DJ8 at 7V heater supply...It took no more than a few minutes to bring cathode emission down to 80% of a new tube.

Once it's down, it wont come back up...that much I can tell you.

Sorry to bring the bad news...If you have, or have a friend with a tube tester you can have your tubes checked for emission.

Cheers,

[Saurav]

Wow. That's not good news. I don't know anyone locally with a tube tester, but I'll keep that in mind.

Maybe this means it's time to try the 6C45-pi My CCS boards are set for 20mA so that should be fine, just need cathode resistors for 9V and I already have the capacitor. Or try 9V batteries like the 4V batteries I'm using now. I'll get some extra gain too.

How does one hook those tubes up? 100 ohm grid stoppers close up against each grid pin, and then connect the other ends of those resistors together and that goes to the input RCA? I think that's what I remember reading. And what about the multiple cathode pins - connect them all, or use any one?

[universum]

[QUOTE]Originally posted by Saurav
[B]Hmm... I hadn't made the connection between triodes and shunt regulators. That could be an interesting idea. This obviously wouldn't be as straightforward as a VR tube, right, I'd have to build a regulator circuit around the triodes.

yep it's quite simple too...see attachment...look at Rk,and the two possible connections...see (hear) what's the differance...it works fine with any tube as long as you keep it within it's ratings ofcourse:-)it's not a really regulation,but keeps out any buzzzzzz,quite well,it's works by putting the buzz into the grid and it ""amplifies""the buzz and put it out of pahse at the anode,so killing the buzz...

and:[QUOTE]Originally posted by Saurav I'm not sure I understand what you mean by a cathode load.


Well instead of a kathode resistor you can use a triode at the bottom but you'll need an negative rail to do so...i would stick to the shunt (regulation) buzzkiller...

[fdegrove]

Hi,

How does one hook those tubes up? 100 ohm grid stoppers close up against each grid pin, and then connect the other ends of those resistors together and that goes to the input RCA?

Yes, that's one way of tackling it and you could apply the same method for the cathode connections.

If you have a scope you can keep the gridstopper's total value to a minimum...You want to listen to the tube, not the resistors.

Cheers,