Hey, just a thought - oversized PS with free 2x300V winding, big chasis - why not put RIAA preamp inside, selectable to the input (100k pot) of the line preamp?
I tinkered for a few hours about this idea today - I tried to be as simple as possible ("all in one go" EQ, just one double triode), and without "exotic", hard to find values for the RIAA EQ - R1, R2, C1, C2. The little "trick" is to find the C1 cap with little lower value then standard 10nF, about 9,76 nF (not a big problem actually, most of the standard 10 & 20 % caps I tried are little lower in value). Then, with our RIAA equations: R1C1=2187uS, R1C2=750uS, R2C1=318uS, C1/C2=2,916 we can find:
R1=224 kOhms - our Rout from the first stage is about 3k, and we can use standard E96 value, 221k/1% (even E24 value 220k with small selection)
R2=32,58 kOhms - little tricky, we can select between some 33k resistors, or use some cobination, or just use E96 value 32,4k
C2=3,347 nF - "input" capacitance of the second stage is about 47pF, and just use standard 3,3nF cap of your choice (2,5% tol. if possible)
If we want to use this preamp to drive more then few cm of cable, then we can use active DN2540 load for R7=15k, and use low imp. out.
Theoretical amplification is about 80 times/1k, more then enough for "normal" MM cartridges + our line stage.

Well, this is just some thoughts/computing/simulations (almost perfect, of course:-)). Is not a problem to build it, problem is that I don`t have the quality measuring devices to actually try RIAA accuracy:-(. Maybe...hm...Doug?
And, BTW - some other tubes besides E188/7308 can be implemented, probably E180CC, maybe 6N1P or so... Both triodes work with rel. high current, Ia1=10mA, Ia2=9,5mA.
Thoughts?

Second, I'd suggest the 6BQ7A/6BK7B valve. It is inexpensivew and a little higher gain. Cg-a is comaprable to any of the mentioned alternatives.
regards,
Douglas

the troubles I can see from here are mainly the high output z of the cascode stage and the correspondingly high z RIAA EQ. The Guinevere-style plate load will do a good job of driving cables.

I may decide to put in a variable capacitance in the front, as the cascode will be nearly free of Miller effects.

Still looking at it.
regards,
Douglas

That and noise....heh-heh-heh!

Do you know of a sheet with its triode curves? With that I could generate the cascode plate curves and see what gain I could realize. mu-squared is unbelievable ( 3600 ). If I could get 10% of that I'd be in business.
regards,
Douglas

• http://www.tubes.mynetcologne.de/roehren/roehren.html