The plan is to order a lot of different types and give them all s listen. Gotta design adapter board to accommodate them all.The ones I mentioned were just a few examples.
There are/were much more comparable types from other manufacturers.
As e.g. RiFa (PFE-216) in Sweden, Suflex (SUF677, SUF710) in England, Mial in France.
Some switched hat over the years, e.g. the boxed Siemens KS, became S+M KS manufacture (Siemens + Matsushita).
At a later stage, the technology plus the machinery was sold off to EMZ, who manufactured the exact same KS parts with an EMZ logo for a number of years.
What all these types and brands differ in, is not just the technology, but the manufacturing tools.
The German cap winding machinery is/was SOTA, and as you heard from Stuart, tightness of winding the foils (and variations) is half the story.
Not merely the choice of dielectric.
Main and only reason why the boxed Siemens polystyrenes could also be ordered in 0.1% and 0.05% accuracy, straight from the production line (see the link I posted for you).
Difference between a Chevy small block, and one by AMG (with an individual engine computer data file).
Doesn't mean you shouldn't try newer foil cap types, it's the continuous change in technology that makes audio so fascinating.
(and sometimes, a step back can become a step forward)
Ever seen a 0.5uF polystyrene ?
Would you post Foolish Diyer version of SCH so even I can have a go and try please
I am working on a balanced update for the Paradise.
I am planning to put together a kit with everything necessary, including noise screened and selected transistors. I really would like to avoid the trouble we had with Hfe matching and wrong components and delay in the group buy.
The implementation is pretty easy.
The biggest problem is how to setup the servo. We can not use the same method then in the R3 paradise because the biasing is so different.
I am planning to put together a kit with everything necessary, including noise screened and selected transistors. I really would like to avoid the trouble we had with Hfe matching and wrong components and delay in the group buy.
The implementation is pretty easy.
The biggest problem is how to setup the servo. We can not use the same method then in the R3 paradise because the biasing is so different.
I can not help, i will post an even better circuit where everything is cascoded.
There is still an unnecessary high impedance point between the top transdiodes and the output transistors. That is the beauty ones you understand topology.
That topology can even be improved by earth referenced or floating current sources
Without the current sources the helpers can only drive 1V, no matter how i set the idle.
It is fascinating, like looking into a kaleidoscope.
There is still an unnecessary high impedance point between the top transdiodes and the output transistors. That is the beauty ones you understand topology.
That topology can even be improved by earth referenced or floating current sources
Without the current sources the helpers can only drive 1V, no matter how i set the idle.
It is fascinating, like looking into a kaleidoscope.
This is totally off topic but nice reading nevertheless. It also contains a primer on how audio came about :
http://www.peavey.com/support/technotes/hartley/chapter_1.pdf
http://www.peavey.com/support/technotes/hartley/chapter_1.pdf