Mike,
I have the opposite experience of yours; my bookshelves are -3dB at around 60Hz too and I run them full range with no problem at all.
However, let's explore the possibilities.
If you use a simple 1st order high pass - viz a small cap and a R, best achieved by removing the top of R17 on the GK-1 pcb and moving it to the top of C19) - it will pass a gently falling slope to the tube stage so that all frequencies below a certain point - adjust to taste - will begin to tail off. But it will be very gentle, perhaps not enough for your purposes.
I'd suggest a 3rd order, which can be done with three RC networks in front of a quality opamp, or even a simple emitter follower design. Then you can select a reasonable corner frequency, say 3dB down at 58Hz and 21dB down at 29Hz.
You could also buy an off-the-shelf solution for this, but likely will have to tolerate lesser quality.
Once again, this comes back to the strong desire many people have for active crossovers. Once you enter this galaxy, you confront many nasty demons, however.
Phase shift. Dysfunctional, even absent, imaging. Loss of vitality, much more complexity, issues with choosing and dimensioning the frequency determining components. A big one is the notion people have that driving a speaker direct, with no passive components anywhere near it, is a good thing. It's not. It's a bad thing, because the highly reactive load of a driver voice coil needs compensation to keep it's impedance changes in check so that the amp can do its job properly. It is a hugely complex task, and let me tell you plainly, I have not heard a marvellous, 3D image from an active crossover system yet. I'm sure it can be done, but it's MUCH more difficult is commonly held, and most of the ultra-fi systems, in the $US50K plus bracket, still haven't got it right........ A simple, passive system is still superior in this regard. If you like imaging, think hard on this one........
I perceive that these comments will get me into more hot water than I care for, but perhaps it's hubris speaking. AndyR, one of our subscribers, has perhaps the best active system I've ever heard, but it's not perfect, and it's complicated. Ask him here - he'll be reading this with a smile - we enjoyed coffee together yesterday.
In closing, I should explain I'm not so much opposed to a simple high pass or low pass. The ones I really bristle at are the bandpass filters. To me this is like trying to compete in a hill climb in a Cadillac - the complexity is just too much, and the nimble musical signal is badly damaged.
I'd be quite happy to give you start out figures for R17, so email me privately if you want to take this further. I can even give you a trial circuit for a 3rd order high pass you can insert between the output of the SS section and the input of the tube section on the GK-1.......
Have a great weekend
Cheers,
Hugh