Zen Mod I was thinking of doing the same Elna Silmic II bypassed by Wima MKS4 that I have laying around.
My question is how would you connect the Elnas? Would you use 2 back to back with their negatives connected? If so what value in order to achieve 10uf in AC? 2x 20uf? I.e. like that below?
---(+)||---||(+)---
Last edited:
I have a long story of measure once, cut once, then go to the store to buy more wood...Now I measure twice!
I know that lesson well!
Russellc
I know that lesson well!
Russellc
Hah! I measure 3 times and cut...only to find I measured incorrectly
all three times.
Dennis
The output capacitor pads are marked C3.
One is large enough to hold a big fancy film cap, the other pad, connected in parallel, is for a small 10uF can electrolytic. the electrolytic may be polar, and the small pad is marked for orientation.
The provision for an inexpensive (yet high performance) cap such as a Silmic or audio-grade Nichicon has already been placed on this PCB. You can bypass that with a small value film cap if you wish.
C1 and C2 are part of the bias voltage supply, and have nothing to do with the output signal.
Here’s my attempt at the BA-3 FE as a pre
[/URL][/IMG]
2x 20V 80VA transformers, Will’s 317/337 reg boards, Alex and Greg’s JLH Ripple eaters.
I’m using stacked relay IP switchers for IP and Ground switching.
BA-3 board uses matched LSK from DIYA store and matched Fairchild mosfets. Resistor value changes: R5 220R R6/7 50R R13 200R. OP cap are Mundorf MKP. All resistors are 1/2W PRP except R13 Mills 5w wire wound.
Volco is a 50K DACT switcher.
P3 is in the central position. I'll have a play with that later
Shamefully everything except the BA-3 is from my parts box hence the large transformers.
Looks so, so but sounds superb
2x 20V 80VA transformers, Will’s 317/337 reg boards, Alex and Greg’s JLH Ripple eaters.
I’m using stacked relay IP switchers for IP and Ground switching.
BA-3 board uses matched LSK from DIYA store and matched Fairchild mosfets. Resistor value changes: R5 220R R6/7 50R R13 200R. OP cap are Mundorf MKP. All resistors are 1/2W PRP except R13 Mills 5w wire wound.
Volco is a 50K DACT switcher.
P3 is in the central position. I'll have a play with that later
Shamefully everything except the BA-3 is from my parts box hence the large transformers.
Looks so, so but sounds superb
Two are better than one.
This from Bateman 2003 confirms that back to back Bi-polar when used as a filter have significantly less distortion than a single Bi-polar. Another para in the same report confirmed that back to back polar electrolytic are significantly lower distortion than single polar when used as a filter.
For coupling duty the distortions are even lower, probably unmeasurable.
Bateman tested this and pooh-poohed the added film cap
This is again based on the test method where the electrolytic is used as a filter.
Last edited:
Thank you Andrew, that's very interesting stuff.
It makes sense that a single polarised cap will have more distortion, however I'm not too hung up on distortion - as long as it's low enough.
I am more worried about damaging the resistor with reverse biasing, as well as the sonic qualities. For the latter, I guess there is only one way to find out!
It makes sense that a single polarised cap will have more distortion, however I'm not too hung up on distortion - as long as it's low enough.
I am more worried about damaging the resistor with reverse biasing, as well as the sonic qualities. For the latter, I guess there is only one way to find out!