What makes a great concertina triode?

[leadbelly]

Question as above. What is most important: high transconductance, low plate resistance, or something else? What about less important factors such as mu?

[vinylkid58]

I may be wrong, but I read a post where Eli mentioned the 12AU7 as having the best characteristics for concertina duty. In some applications, maybe you could use other tubes, 12AT7's, 12AX7's etc.

My PP EL84 amp was designed with 12AX7's in the concertina, which I quickly removed in favour of 12AT7's that sound better to me.

Jeff

[aardvarkash10]

kinda depends on whether it is just splitting or if it is also carrying out driver duties as well. Also what it is driving into...

[Wavebourn]

The best phase splitter is when Concertina drives LTP. A sweet pair combination.

[SY]

The best concertinas have matched loads under ALL conditions. That means interposing something between them and any stage that will transition out of class A. The LTP that Wavebourn mentioned is a good one- also see the very clever Williamson circuit.

High gain, high transconductance, and low distortion, that's what you want. 12AU7 is not good in any of these categories. 12At7 is better but not ideal. Something like a triode-connected D3a or 7788 is near-ideal. A 6DJ8 is a good compromise practical choice.

[leadbelly]

Thanks SY!

[Eli Duttman]

Originally Posted by vinylkid58

I may be wrong, but I read a post where Eli mentioned the 12AU7 as having the best characteristics for concertina duty. In some applications, maybe you could use other tubes, 12AT7's, 12AX7's etc.

My PP EL84 amp was designed with 12AX7's in the concertina, which I quickly removed in favour of 12AT7's that sound better to me.

Jeff

Jeff,

You misconstrue my remarks. I stated that cathode followers and "concertina" phase splitters were the only jobs the 12AU7 is suited to. As SY stated, there are any number of types better than the 'U7 in those roles. I abominate 'U7 section voltage amplifiers.

A "concertina" phase splitter is a tricked out cathode follower. As such, high gm rates (IMO) 1st. In spite of its μ of 22, the ECC99 is good for the job, thanks to high gm and low R_P.

[Wavebourn]

Originally Posted by SY

The LTP that Wavebourn mentioned is a good one- also see the very clever Williamson circuit.

Right; in both cases common mode errors are minimal due to Concertina usage. Common mode errors cause syn-phase currents in output tubes that waste power. But in Williamson case 2 coupling caps are needed; in mine only one. It gives a possibility to stabilize working point using output from the long tail to the screen grid of the 1'st pentode; also my pentode LTP is convenient to drive linearized by parallel feedback by voltage output tubes that working as pentodes with low Cmiller have lower output resistance than triodes may have. Such a way wider frequency band may be obtained from output transformers, also phase intermodulations are minimal.

However, in my case Concertina + LTP form an APF, so big time constant is needed for stability on low frequency end.

An advantage of Williamson is possibility of better balance of amplification factors between 1'st tube and a driver, so choice of the 1'st tube and it's regime is less critical. Common triode may be used, while in my amp a very linear pentode stage is required.

[vinylkid58]

Originally Posted by Eli Duttman

You misconstrue my remarks. I stated that cathode followers and "concertina" phase splitters were the only jobs the 12AU7 is suited to.

Sorry Eli, I am the misconstruer. After two days of struggling to rid this computer of malware & trojans, my head is spinning, therefore I shouldn't be quoting anyone.

Originally Posted by Eli Duttman

A "concertina" phase splitter is a tricked out cathode follower. As such, high gm rates (IMO) 1st. In spite of its μ of 22, the ECC99 is good for the job, thanks to high gm and low R_P.

Is there any way of implementing CCS in the concertina of my HK A224, or is this just a waste of time?

Thanks,
Jeff

[Wavebourn]

Uhm.... Jeff, who made such a thingy with feedback from OPT to Concertina?

It needs to be thoroughly discussed...