Phono Preamp Tube Rush

I far prefer tube-based circuits to solid state, for sound quality, but all tubes do have a finite life span, and as the end draws near, they become noisier, until they reach a final common pathway of being noisy.  This malady will affect SLN tubes as well as non-select samples.  The variable is time from new to noisy, which is impossible to predict for a given sample of a new tube, but I would agree that for a tube that starts out SLN, perhaps (on average, but there are no guarantees) the time to its becoming noisy will be longer than for a non-select tube.  But the point I was trying to make earlier is that circuit design also has a lot to do with "noise" one perceives over the speaker.  For one example, ARC stuff was always known for driving tubes very hard, near their recommended limits for voltage and current.  In the old days, this led to short lifespan. Circuit layout also has an effect.  What tubes are chosen for a given task also has an effect.  Someone wondered why Herron(?) didn't choose to use a 6DJ8 instead of a 12AX7 in its phono stage; the simple reason is you could never get enough phono gain out of a 6DJ8, used as Herron uses the 12AX7.  The 12AX7 is a standard choice used in common cathode mode for the input of an MM phono stage; all the gain is done at the input before RIAA, as Hagtech mentioned.

I had brought up the question earlier in the thread about why not use a 6922 instead of the 12AX7.  You definitely could get enough phono gain with the 6922 and this is commonly used in the Manley phono pre's (for example).  Now would the overall sound quality be as good in the overall circuit design compared to what Herron achieves in the VTPH-2A?...I have no idea.  I use the Herron VTPH-2A and am happy with it.  I may try a coupe of Roger's SLN tubes in the 1st gain stage soon and see if it makes a notable difference but either way I'm perfectly good with the Herron. ;-)

three-easy, Manley, in the Chinook and in my Steelhead, use a hybrid cascode at the phono input.  "Hybrid", because the cascode is built out of a transistor (probably a JFET) and a tube (the 6922/6DJ8).  That is quite a bit different from using a 12AX7 all by itself to supply all needed gain.  A cascode per se will give you more gain and a wide bandwidth, plus shielding from input capacitance (because the JFET drives the 6922 via its cathode), compared to any single tube used in common cathode mode.  The trade-off is high output impedance of a cascode. The 6922 is well suited to this topology, because it has high transconductance, needed to take advantage of the drive from the JFET, rather than a high gain factor, in contrast to the 12AX7. 12AX7 has relatively low transconductance and very high gain.  You'd not use a 12AX7 in a cascode. Nor would you use a 6DJ8 by itself for the high gain needed at a phono input.

@lewm 

I appreciate the detailed explanation!  I was unaware of the Manley hybrid design.

Interestingly, changing from my Herron VTSP-1/166 to a Lamm LL2.1 made a big difference, and not because the Lamm itself is quieter because it isn't. Could it be something with input impedance?