I definitely got the sense that a desire to "personalise" your new headphone system was a significant motivator behind your desire to replace that "big", fat'n'juicy wall wart.
I too, am not the biggest fan of its form factor. It's prone to sitting awkwardly in the receptacle, particularly horizontally on my power strips.
OTOH, I am even more grateful that I got a quieter & cleaner linear adapter than some dinky & noisy SMPS wall-wart...
You appear to have gone to a not-insignificant amount of trouble to get a PCB made (?) & sourced for a better transformer than what Koss gives us. But where is the safety improvement in a homemade, non-CE approved power supply project?
I highly commend your creative spirit, & I am sure that SOME sort of sonic "difference" will result. I would even grant you that there probably WILL be a sonic improvement, HOWEVER SMALL due to the ACTUAL power supply being the DC-DC converter(s) inside the E90. But ONLY if you don't get too fancy & say, upgrade the amount of filter capacitance (or not!) but DON'T implement an "active" regulator circuit into your "pre"-PSU. That will add its own sound, & the result may not to your liking. Personal experience is against it!
There are three issues here:
1) Whilst I am utterly & completely convinced that lowering power supply source impedance is only a good thing ("high current", which is the part non-audio electronic engineers don't understand; the high current capacity is merely a byproduct of the solutions used in LINEAR power supply design to improve sonic & electrical performance), I am rather more skeptical about taking such measures when dealing with switching, regulated psu's "upstream" of a linear one like we have here.
Now, without inspecting the insides of the E90 to see how exactly Koss is stepping up the DC input voltage 75-odd times, I could be wrong about the degree of possible sonic improvement. I say this because the E90 is so small & the entire system so..."inexpensive", the implementation of the DC-DC conversion may well be rather basic. There isn't any strict requirement that the high voltages be regulated, especially the 600VDC polarising voltage.
OTOH, regulating the polarising voltage is kind of impossible as a high source impedance here is necessary for lowest distortion from a push-pull electrostatic diaphragm. I wouldn't be surprised to see AT LEAST 1,000,000 ohms of added resistance in series with the polarising part of the PSU, as per Stax & others!
2) Because the diaphragms have to charge through such a large amount of series resistance, it takes a long time for Electrostatics to charge optimally. If you keep turning off the E90, your headphones will NEVER sound their best. You won't be getting the sound you paid for, the amazing performance it's capable of. You get lower efficiency (subtle, but noticeable), less treble extension (VERY noticeable), & less detail overall plus a longer wait for the E90 amp section to warm up optimally (QUITE noticeable, too!).
Many audiophiles swear by leaving their preamps & power amps on continuously; For example, Bryston recommends it & they offer a mere 20 year warranty as well!
So for someone chasing sonic improvements, why modify your system, potentially void your warranty by using an unapproved power supply?
Wouldn't THE BEST SONIC UPGRADE be merely to figure out how to set up your E95X system optimally, & just use it accordingly? This is my primary argument!
Figure out how to optimise the performance of the "stock" product first, & live with that for a while first.
THEN maybe consider safe & thoughtful upgrades instead of just IMMEDIATELY ASSUMING YOU KNOW BETTER!
Heck, even a Dekoni or whomever earpad "upgrade" would be a far wiser & more meaningful performance improvement...
2) As for safety, this is a "red herring". When a product has an extremely low maximum power consumption (the supplied wall-wart, as beefy as it is & rated to supply a full ampere, is STILL only rated for 9 watts!), fuses are mostly more trouble than they are worth. The incredibly fine wire on the primary winding of that wall-wart are hardly more tiny than the wire in say, a 1/10th Ampere fuse. Which would burn first? It doesn't matter!
Yes, I've seen the rare DC wall-wart develop internal parts failure, like a shorted dinky little electrolytic filter and/or one of 1A 1N4002/3/4 diodes, resulting in enough overheating to partially melt the case. But it is far, far rarer (& has never knowingly happened to me or anyone I personally know in my 55 years on this planet) to have the wall wart catch fire as opposed to merely quitting after that bit of overheating.
Most (98%+) of the time wall warts just safely & silently quit.
My Stax SRD-7 power supply unit for my SRX-III headphones use only 160VDC polarising voltage. There is no power transformer, just a small 1A diode (half-wave rectifier) feeding a tiny electrolytic (1uF max!) & a 1M resistor straight to the front panel connector. Protection is via an 0.3A fuse mounted in an (unlabeled for amperage) chassis-mounted fuseholder. A foil label affixed to the underside of the unit by the then-Canadian distributor (TAE Electronics of Montreal) shows proof of CSA "special" approval by them (the Canadian version of UL, which was more stringent than Underwriter's Labs as well). Pretty Mickey Mouse, & entirely safe. 40+ years of flawless & near-continuous use!
The E95X system meets all relevant CE electrical safety approvals; why are you so unconvinced of its safety?