6H6pi curves available?

[disco]

Doing the math with the formulas presented by M. Blencowe, I adjusted the load resistance to the (for practical reasons) chosen 690 ohm kathode resistors.

Rload = (mu*Rk1,2 - ra) / 2 = (22*690 -1800) / 2 = 6690 ohm
Quiescent current Iq = HT / (2*ra + 2*mu*Rk) = 285 / 2*1800 + 2*22*690 = 8.4mA
Max undist voltage swing = 4* 8.4mA = 33.6 in 6690 ohm = 225Vpp or 80Vrms
Output impedance ca 600 ohm.

Even when optimising was an theoretical effort, the amplifier sounds finer with the lowered grid leak. Worth the trouble I'd say.

[revintage]

With Blencowe THD gets lower but the harmonic pattern gets worse. 2H lower but 3H higher.

[disco]

It's in anti phase to the output tube, wasn't it? That way lowering 3H would not harm.

I've just begon voicing but it has potential. First a smaller coupling cap as output impedance of the SRPP is that low. Bold bass response from that 150VA output, quite different to the AD1 sound I'd say

[mogliaa]

To my surprise as I was looking to use this valve in an OTL headphone amp, found it was not very linear:

http://www.bartola.co.uk/valves/2012...thd-benchmark/

Close to the top chart of the worst ones am afraid, however this is just a THD test...
Cheers,
Ale

[disco]

Well, the charm is in the topology as SRPP operates in class A only. Under condition that both triodes contribute equally to the load, all even harmonics (not only 2H) cancel, resulting in minimum THD. This deep null is obtainable for a fixed load only and it does not clip pleasantly.

[vinylkid58]

Originally Posted by mogliaa

To my surprise as I was looking to use this valve in an OTL headphone amp, found it was not very linear:

So was I.

What circuit were you considering?

jeff

[waltube]

In my opinion the best sonic results are when the idle current is about 20mA /section with a Vdc at 200-240 Vdc; -Vg at 2,5-3 volt; also the test on Thd are quite good.
I built a Gomes preamp with this specification and the results is fine.

Ciao

Walter

[revintage]

It's in anti phase to the output tube, wasn't it?

That is correct. But this mostly affect 2H. Without distortion cancellation this circuit and many other would have gross distortion.

Well, the charm is in the topology as SRPP operates in class A only. Under condition that both triodes contribute equally to the load, all even harmonics (not only 2H) cancel, resulting in minimum THD. This deep null is obtainable for a fixed load only and it does not clip pleasantly.

The lack of charm is that the less pleasant uneven harmonics are worse than for other topologies. This can be bettered going mu-follower instead where gain will also be higher. You unfortunately need somewhat higher B+. If using sand as upper mu-element or a "iron-follower", B+ is OK.

[disco]

Originally Posted by revintage

That is correct. But this mostly affect 2H. Without distortion cancellation this circuit and many other would have gross distortion.

It has, the loudness is still on
The 300V HT is too low to drive the SRPP with more current and the available power tranny was a design parameter, unfortunately.

I thought about feedback for cancellation. If my numbers are right there's a 50Vpp margin for driving the power triode to full output. That would make NFB over the OPT to the driver input possible. Things seem to get too complicated to feed the anode of the power tube back to another node of the SRPP and still maintain some sort of even push pull operation..

Yet another possibility to lower or substantially diminish uneven harmonics would be a like-drive-like situation with a second SRPP. Anyone tried this? That would leave us with plenty of gain for overall feedback

The lack of charm is that the less pleasant uneven harmonics are worse than for other topologies. This can be bettered going mu-follower instead where gain will also be higher. You unfortunately need somewhat higher B+. If using sand as upper mu-element or a "iron-follower", B+ is OK.

I was thinking about a Beta follower with a BC560, is that what you meant by iron follower, or just an interstage?
Perhaps I'll have another look at a penthode driver and see with what I can come away with driving that miller capacitance. I married a C3G happily to a 2A3 but that's an easy load

[disco]

Originally Posted by spswmsw

Ea is NOT Ua
Ea is a power supply voltage
Ua is a plate voltage

Under 420 V power supply and 12 k anode load the plate dissipation is less than maximum threshold for this tube (dashed curve).
Just such a mode is preferable to drive 'heavy" output triodes (no impact of their input impedance due to Miller’s capacity ).
Taking into account the outstanding linearity of 6N6P one can chose the bias point here in a rather wide range.
Sincerely yours.

Has anybody tested the harmonic spectrum of 6H6pi in SRPP?