As I mentioned elsewhere, I have added new plate chokes to my Kaiju rebuild. The purpose of the rebuild was to convert the amp to a dedicated headphone amp for planar magnetics -- presently my LCD4 headphones which I like very much -- and just to have fun tweaking stuff. Changes made prior to new plate chokes:
(1) A CLCLC power supply with (a) doubled capacitance of the voltage doubler (PSUD indicates that this cuts ripple in half) and (b) polypro filter caps including a 1500 uF as "reservoir" cap (it stores a lot of juice - if the power switch is turned off while music is still playing, the amp will continue playing normally for another 3 seconds!)
(2) 100,000 uF bypass cap across the TL431 regulated driver bias (I still want to try replacing the reg with a high quality resistor as one other forum member has done - just to hear)
(3) A 330 uF polypro 300B Kthode resistor bypass cap in place of the stock electrolytic.
(4) Nano-C core output transformers with nominal 3.3K primary and switchable 16/32/64 ohm secondaries, plus parallel load resistors across the outputs that can be toggled on/off to hit the target nominal secondary impedances in parallel with the 200 ohm impedance of my headphones.
(5) 7000H "supermalloy" grid chokes that replace the grid resistor that loads the driver tube (I still need to do some experiments with different interstage coupling cap values w/ and w/o damping resistor across the choke -- as suggested by PJ)
Plus a new bigger chassis and changing the interstage and parafeed film caps to my preferred flavor. I've posted before about my purely subjective impressions of all of the above.
The newest change was to replace the Lundahl 53H Si-Fe plate chokes with 69H Nano-C plate chokes from Monolith Magnetics. The new chokes only have about 10-15 hrs on them - but I am liking what I am hearing. Nothing totally game changing - it doesn't sound like a totally different amp. Instead there is even more and tighter low end bass (likely attributable at least as much to the increase in inductance as to the change in core material) and even more clarity and reduction in "glare" (which I attribute to the new core material). Acoustic instruments in particular sound cleaner and more textured and full.
Here's the speculation part. I'm fascinated by the relationship between external stimulus (music sound waves) and internal ear-brain reaction/processing (what I hear). Some maintain that much of the burn/break-in experience is really the brain "rewiring" itself to better process new stimuli. (Still others believe that much or all of it is just a placebo effect -- sort of a matrix "there is no spoon" kind of thing -- I think this is possible but doesn't explain everything).
If we admit or at least hypothesize that the ear-brain "constructs" the experience of music on the basis of sound wave "inputs" then it follows that there are 2 sides to the equation: input and construction. Is it all just "construction"? I don't think it can be. Music played through your car's am radio sounds crappy compared to the same music played through, say, a BH Crack with Sennheisers. You can spend 20 years listening to that am car radio and your brain is not going to turn it into a "HiFi" experience. So the input matters a lot.
So is it really all just about the input - with the ear-brain, if functioning properly, doing nothing more than translating eardrum enervations into the experience of sound according to a fixed, automatic process -- i.e., by a non-adaptive process? I don't think that can be right either. Consider hearing aids - I don't yet need them, but they strike me as providing a good example of how the brain "constructs" the experience of sound through some sort of iterative learning process. Do a search for "getting used to hearing aids" and you'll see a ton of info online to the effect that, initially, using a hearing aid can make some sounds incredibly loud, difficult to locate in space (where it's coming from), and even leave one feeling disoriented. But with repeated use, those problems resolve. The inputs don't change (a bird chirp is no less loud one month later); the ear-brain "construction" operations do.
Here's another potential example of the same adaptive learning phenomenon that's probably a little more relatable to BH forum members. Have you ever changed something in a BH amp -- e.g., volume pot, film cap for electrolytic, fancy "low noise" bias resistor, etc. -- and experienced a huge change in sound between how the amp sounded before turning it off on Day 1 and after warming up on Day 2? Now, I think that in some cases some of the change may be due to an actual physical change -- e.g, in the case of caps, "cap forming" or some such. I've "broken-in" teflon caps over a period of 100+ hours using dummy load resistors during which time I only briefly listened (matter of minutes or even seconds) once a day or so to see how the amp sounds. My experience with teflon caps is that they go through an extended period of 50+ hours where they don't sound so good, and then gradually start sounding better and better. This gradual change coupled with very little listening suggests to me that the sonic behavior of the cap (the "input") is indeed changing.
But what about a volume pot, a resistor, or even some new iron? [Edit: In cases where it is not simply the placebo effect or "confirmation bias"], I suspect that any experience of additional "detail" is in fact the result of a physical change (change to the " input")-- but that at least some, and perhaps all, of the subsequent experience of the sound becoming more "coherent", "even", "separated in space", etc. is ear-brain learning. I'm no neurologist or psychologist, but I've seen and read a number of docs/articles exploring the important role of sleep in learning. I wonder whether much of the "next day improvement" is result of brain adaptations during the intervening sleep?
Potential case in point: When I turned off the Kaiju last night I was hearing lots of new detail and clarity, but the "imaging" and "soundstage" was really off -- sometimes instruments seemed superimposed in the same space and others were just difficult to "locate". And some of the highs even sounded a bit shrill. Complete change this morning: sounds really good, soundstage and separation are "returning" and no shrillness.
All of this is just a long-winded way of saying that my working guess is that physical soundwaves (inputs) provide the raw data and basic skeleton upon which the ear-brain constructs the experience of music -- fleshing it out, so to speak. The ear-brain can't create detail without a stimulus, but it has some play in how it can arrange that detail into a more or less coherent soundscape.
A neat way of excluding (controlling for) the effects of any physical changes due to turn-off/cool-down would be to: (a) make a change (e.g., new choke or output trafo); (b) listen for a number of hours; (c) take a nap while the amp keeps playing (eg., with dummy loads in place of speakers, or just in another room if using headphones); and then (d) relisten. If there's a big improvement, my money would be on it being the result of brain adaptation.
Any tweakers out there that are also fans of midday naps? If I remember, I will try this the next time I make a significant change or build something completely new.
cheers, Derek