GoodSoundClub - Romy the Cat's Audio Site

In the Forum: Audio Discussions
In the Thread: Additive capacitance: mixing oils and films?

Post Subject: Re: Additive capacitance: mixing oils and films?Posted by Axel on: 8/16/2009

well, to add my bit of mustard to this here oldie thread.

I HAVE mixed film and oils successfully in e.g. a tweeter x-over of a Burmester 961. The tweeter is an AMT (air-motion-transformer) and very revealing.

The original C's are two sets of 1uF 630V || 3.3uF 400V + 1uF || 3.3uF (same voltages i.e. physically close in size). In between the two parallel C's is the 'tap-off' for an RCL. All caps are Mundorf MCaps. I have now added to each parallel set 1 Mundorf silver-oil of 0.01uF and it is quite amazing how much it cleared-up the tweeter. (If I remove the by-pass it seems to haze-over the treble!)

So this worked fine, BUT having tried some other by-passes with a 5.6uF C, parallel to signal path in the mids x-over, using e.g. 0.1uF tin foil, or also 0.01uF silver oil all produced a very incoherent sound.

Lastly, by-passing big value electrolytic (220uF and 47uF) with a 1uF tin foil worked very well, -- using a 0.68uF tin foil didn't, causing incoherence also.

Long story short: it is pretty much a case by case trial and error situation and much depends on the particular circuit.

My own rules:
- Do not mix cap brands the chance for incoherence will be greater

- Stay below 1% of base value for 3rd (cascaded by-pass cap)

- Tin foil caps actually can sound tinny in tweeter circuit, avoid it.

- Tin foil in mid circuit are fine and seem to work OK to 'clean up' electrolytic caps both, smooth and rough caps.

- By passing electrolytic caps in woofer circuit will very likely produce some 'glare' in the upper mid. A 180Hz xo point will still have enough energy with 2nd order (4th acoustic) to mess noticeably in the upper midrange!

- Reduce electrolytic 'rough' value and make it up with electrolytic 'smooth' in woofer circuit. This can yield some unexpectely good results in an improved upper bass performance.

- Same applies to mids circuit but one added foil by-pass cap can further improve things (note < 1% of base value)

- One more mid alternative for ~> 10uF electrolytic is to replace 1 value of a 'rough' (etched) cap, with ~ 2 'smooth' smaller values and making up the final value with e.g. polyprop foil cap of not more then ~ 10% of total value.

- Do not mix caps to of very different physical dimensions (incoherence!), use lower voltage for base value and go up with voltage for smaller values.

Axel

Rerurn to Romy the Cat's Site