I'll do a SE vs. Balanced comparison after I get SE interconnects, leaving the right bank open. If I prefer SE then I'll grind off and silicone over the P8A side contacts of a failed pair of SP17c tubes and plop them into an extra set of TP adapters - so I can "call" your OCD and "raise" it to the level of mine.
On the issue of balanced vs single ended for the Lampi, here's my experience, FWIW.
As you know the debate between balanced vs single ended permeates almost every aspect of high-end audio. It made no sense to take a side without data, so I thought, what the heck, I'll do the experiment myself. What I was surprised to learn is that the absolute best sound
for me in my system is to run the Lampi single ended, but with a twist. I still use a balanced XLR cable to connect the Lampi H360 to my Soulution 727 preamp. Why? Well, the logic for running single ended has been advocated by many, including Dan D'Agostino who said that getting output devices to work equally well in a balanced configuration is nearly impossible due to differences in the individual gain and amplification differences in the transistors themselves. These minute differences matter so much to Dan that for his top-of-the-line Relentless monoblocks, he spends countless hours selecting and balancing the transistors for each leg of the XLR outputs, and this is what contributes in part to their extreme cost. Keep in mind this is for modern day transistors, which are made within a spec framework that many manufacturers consider good enough that they think the efforts that D'Agostino makes are not even necessary to get good balanced performance. Now, let; move to old NOS tubes, whose availability are nowhere close to those of a modern transistor. It would be essentially impossible to believe that the V vs I curves of such tubes could even come close to being perfectly identical from one tube to another right out of the box.
The bottom line is that in my H360, when I play balanced, the sound is ever so slightly less pure, less clean, and ultimately not as musically compelling as when running single ended. It's a case of something added in balanced mode that is ever-so-slightly pernicious
but audible, rather than subtractive defined as having something missing that is normally present when the comparison was made. The downside is that it's often hard for an audiophile with any component of OCD (who, me? Yes, sir!) to accept that their DAC sounds best with 2 tubes missing and that half of their expensive balanced cable is not being used at all! But it's such an easy experiment to do. Use simple material (girl with guitar, or a single piano recording). Just play something, remove the 2 tubes from the right bank, turn up your volume to compensate voltage 6dB and listen again. (Since power is proportional to the square of the voltage, a 6 dB increase in voltage results in a 3 dB increase in power if the load impedance remains the same). The balanced output always resulted in a slightly less musically pleasing sound
for me. The reason is presumably that its basically impossible to get both the Vhot and Vcold legs of the balanced cable to sound identical unless you have hundreds of output devices to use for precise output matching, as D'Agostino does with transistors. However, doing that with 75-year-old tubes is a non-starter. IMHO, balanced output should mainly be used for long runs of 25 ft or more to take advantage of the reduction of common mode noise rejection which is often useful in recording studios and other venues (i.e. sound reinforcement for large amphitheaters). I think the preference for single ended amplification is preferred by several major recording engineers (i.e. Keith Johnson at Spectral/Reference Recordings) for similar reasons.
When I look at my half naked H360 running single ended, I grimace. But when I listen to it running single ended, I just smile.
BTW, another way to make the comparison is the following. You need 2 identical cables, one being XLR terminated and the other RCA. Run both single ended and balanced out from the Lampi to your preamp (different inputs of course). Then you can just flip between preamp inputs to make the comparison. You just need to adjust the preamp by ~6 clicks (or whatever is necessary on your preamp) on the volume control to get a volume match between the two every time you switch inputs. This has the advantage of not having to get up to add or take away 2 tubes in the Lampi when making the comparison. While this method is very useful, there is still the vagaries of using 2 different IC cables each with different terminations and therefore very minute impedance differences. A less "pure" method perhaps, but it's still very useful and typically easier than getting up and adding and removing tubes each time you want to make a comparison. In addition, It is said that critical audio memory lasts about 7 seconds. I can't even get up from the chair and walk to my gear in 7 seconds, not to mention the time it would take to add or remove tubes and then go back to listen. Using this method therefore allows a more instantaneous comparison that might be very helpful as switching preamp inputs and adjusting the volume by 6 clicks remotely can easily be done in that time frame.