MF A1, Kudos to Tim de Paravicini, Ode to my Youth, Babelfishing

I might actually understand why harmonic distortion 'profile' changes with load

Youngsters are used to play with LTSpice, while even substantial part of us Geezers abandoned looking at U-I graphs of parts used

if you invest some time in reading Toob books from Yore - there you'll see everything based on silly graphs and graphically based calculus

most recent musings here, I can remember, are connected to important contribution of Mr. MR - his ( evervanishing on his numerous also evervanishing web-sites) SIT Curve making apps

that's exactly base to understand why same part with one load is giving XXX Watts and YYY THD, while same part with other load is giving XX Watts and Y THD


completely naturally that you can expect change in THD SPectra too

now - remember how we all are used to take Papa's words as 15 10 Commandments ........ he explained more than once effect of SIT-1 "Happy Dial Button" - changing THD Spectra simply by turning Iq biasing pot at front plate; that's practically same mechanismus

in general - say that gadgets with greater amount of OLG (thus NFB substantially higher) are more immune to that sort of changes ( having more stable THD Spectra) while gadgets with lesser amount of OLG (with or without NFB) are more prone to Chameleonning ..........

ZM Omnichameleon :clown:
 
  • Thank You
Reactions: Hotep
^ Thank you! 🙂

I still look in amazement (meaning I have no clue, but it sounds nicer to say I'm amazed) at the source resistors carefully chosen on the F6 and scratch my head.

Making transistors behave to the master's will is ... magic.
 
well, in some cases luck is what helps

practically - entire range of similar small amps I made in recent time, was practically investigation what are effects of using "same", "similar" and "not-so-similar" parts in OS

all that being possible when I stumbled (reinventing the wheel) on whatiscalled "level shifting follower stage", thus drive-biasing stage always seeing benign load

in case of Babelfish A1, difference made with use of mosfet followers in place of double bjt one, and choosing modern complementary BJT pair instead of beancounter's choice

old bjts shown in pics (post #13, 2SA1106/2SC2581) are no slouch, but tons better than TIP35/36 (shown in #7)

while TTA/TTC modern Toshiba Biguns are bingo comparing to pretty much everything else sanely priced

now, mentioning F6 - part of my fixations of late is SL OS arrangement, avoiding degeneration wherever possible

that dictates other techniques to alter THD spectra where needed ...... just observe different values of (OS) base resistors in B. A1; difference is subtle, but there is a difference; who knew that yesterday?

I certainly didn't :rofl:

and that's main reason why all this is fun to me ........ luckily I've just scratched what's possible to learn , so it'll stay as that at least 100 more yrs of my life


ZM Omnicunning


Baldrick.jpg
 
Last edited:
  • Like
Reactions: Vix
I like the story very much, reminds me on my youth some 170 years ago :rofl:
My dream amp was the Quad 405, built it several times. I was living behind the iron curtain at the time, so the options were very limited. After the third copy I've built, I thought - I could ask. Letter sent to England... and a few weeks later I received two original Quad 405 pcb's! But that's another story.
 
Is there a point in time when Tim stopped developing complementary transistors push-pull amplifiers? It may well be that at some point he too became aware of the physical, "objective" "reality", the audible difference between identical and, above all, non-identical components. Unfortunately, we can no longer ask him about this)-;