If you look at this design, you will see that it uses a cathode follower at the output to lower its source impedance. Also, it is designed for use as a line-level input to a standard line stage. At the time I started this project, I wasn't really aware that a phono preamp is perhaps the most difficult vacuum-tube audio device to build successfully. I was aware I wanted more gain than this design seemed to provide, and that I wanted to build an all-in-one unit that would serve as a line stage for other line-level sources as well as an RIAA phono preamp. So I kludged it up something fierce: I installed a selector switch with several decks that took the phono EQ components out of the circuit when it was switched to a line source, and converted the cathode follower to a plate loaded gain stage for more gain overall. I did this despite the pleas of the designer, Bob Danielak, that I try building it as drawn first to see how it went.

It didn't quite work. It hummed very badly, despite a DC filament supply, and I never was able to fix it.

Flash forward a couple of years. I now had a few successful, working projects under my belt, and one of the things I learned in troubleshooting these devices was that a high gain circuit is susceptible to oscillation at high frequencies above the audio band, and that this fault could sometimes manifest itself as excess hum. (The parasitic oscillations ride on the small residual hum signal and make it much, much more audible.) I thought to myself, well, that must be what's wrong with my phono stage! The leads that ran up and back to my kludge switch were longer than optimal and perhaps served as antennae for the radio frequency oscillations. I resolved to cut the switching stuff out, wire it up more neatly, and install grid stopper resistors to see what that would do to the oscillations.

This worked well - I now had a working phono preamp and it sounded pretty good.

I noticed that there was a poor match between channel EQs (I listen to a LOT of mono records, and their image was skewed towards one speaker or another in different frequency ranges) and not enough gain to match a CD player's output. Since I only drive a very short cable to the line stage, I decided I could live without the cathode follower and I used this half of the second tube (6SN7) as another gain stage, as I had tried to do with my first attempt. I also made a couple of other changes, and when they were made, they affected the accuracy of the RIAA EQ components.

I was fortunate in at least two respects: Bob is a patient guy who likes to think about changing his designs around, and Morgan Jones wrote a tremendous book, "Valve Amplifiers," which explains what is needed to calculate the RIAA components in some detail. I worked these out so I could eliminate all the cathode bypass capacitors and so I could use an AA Ni-Cd battery to bias the input stage. The result is the circuit you see here.

It sounds great and has been very easy to implement, once I got my goals clear.

Here is a photo of the breadboarded prototype, which is usably quiet for a MM cartridge despite having NO external shielding whatsoever: