intermitant phono tube cut out?

I think the discussion in an audiocircle forum by Dan was pretty straight forward. However, as you suggest ,there may be a better way to skin the cat. How would the 12au7 negatively effect the sonics? enhancing or suppressing high frequency for instance? Do you have a recommendation on a 5751 type that is a quiet tube on the slightly organic side of the spectrum?

I collect 5751s and have some experience subbing them in for 12AX7s. If you want a softer edged fuller sound then go against the wave of opinion and get an older grey plated example. If you crave detail space and definition then get black plated tubes. The best tube were the triple mica supported late 1950s to early 1960s, however it is not absolutley necesarry for a creamy musical tube. Probably the best depending on your budget are the Sylvania Gold Brand which were double mica in the grey plated versions. If you just want to experiment get the ubiquitous GEs of that era which are very underrated, they deliver volumes of good bass, and are cheaper that most other brands.
If you want to hear what a blackplate sounds like you can get RCA white lable triple micas or the later "Command" series.
If you really just want a sense of what a 5751 sounds like get fairly current (approx. 1980+) JAN tubes or the current Sovtek butI warn you that Sovtek just doesn't sound right to my ears. There is a lot of 5751s out there espcially GEs so finding a decent example shouldn't prove hard or very expensive. If confused write me an email and I will guide you.

Mechans apparently knows much more about 5751 variants than I do. I would take his advice.

Do I understand correctly now that Modwright is not the source of your idea to sub a 12AU7 for the 12AX7? It makes sense that this is so, because the idea of doing that sub makes no sense.

I would not be able to predict how a AU7 for AX7 would affect the sonics of a phono stage where the 12AX7 is used for phono gain, without seeing the schematic, but I can tell you that the RIAA equalization curve depends on very accurate filtering at very specific frequencies. The output impedance of the gain stage forms one of these filters in conjunction with a precise parallel capacitance and usually a series resistance. Since the AU7 has a VERY different output impedance from an AX7, the substitution would have a very large affect to screw up the RIAA equalization. However, nothing will blow up or anything of that sort.