Petp Capacitors-one Of The Best?

Status
Not open for further replies.
Well now that you mention it.....,

I did a well matched B1 buffer to go between the Lightspeed and the F3 and it never did come good - I couldn't bakance the current to the caps (sounds a bit ridiculous, seeing that written like that, but that's how it seemed at the time!)

It needs about 10uf as o/p cap and with the 10 volts across it, one of the 10uF/63volt K73-16 might just do the job - will try it again and see how it goes.
 
Yes, you can just add it in line with the source and have a listen and you'll get a rough idea about what sort of sound it has.

Unfortunately, this same device will probably sound rather different in a different situation like with high impedance and low bias voltages, for example, or perhaps, a series cap in a Sallen Key filter, or a feedback cap in an amp - you get the idea.

When people get different/poor result from an "unsuitable" application, an otherwise good device may be written off altogether.
For quite a few years, all poly/mylar caps have had a pretty average reputation - with me, too!

Now, we're seeing a general production device in a new light - it would be even more useful if we knew where they didn't function well, and why not.

From what's written on the thread above, most of the ccts that these caps have shown exceptional performance have been valve ccts with quite a bit of voltage across them - with the much lower voltages that I'm using them for, we're looking to confirm that we also get the same exceptional results. Hence the idea with a current common project, the B1 with 9 volts on it.

Pity they're so damn big tho! And perhaps a heatshrink sleeve/jacket on the metal case. The wires need rubbing clean, too.

Sorry for the mumbo jumbo and vague terminology - not intentional.
 
Havoc,
This is a subject that's available on quite a few different threads here - with those that agree and those that think it's nonsense - as usual, it's somewhere in between.

Most big electro caps aren't as good at higher frequencies as smaller, "better quality" caps, so if you connect a smaller cap across the big cap terminals (as a "bipass capacitors"), this generally gives better sound.
There are sometimes problems - sometimes "ringing" occurs and resonance between the 2 caps, so it's not as simple as it first appears.

For example, if you add a 0.1uF Polypropylene bipass across a 220uF Elna Silmic, you generally get more top end and a "faster" sound (sorry for the vague) in some circuits like I use, but often it also introduces a ringing of the top freqs (eg: cymbols sound unnatural, that some people quite like). If you change the type of bipass to say a good quality polystyrene, you get a quite different top end sound altogether that you may not like at all - it get's a bit trickie (sometimes produces "fatigue").

And to add to the "fog of confusion", often when you use a 400volt bipass and only apply 10 volts across it, the caps doesn't function anywhere near it's potential. When you use these bipass caps in the signal path with low voltage but high impedance (like an amp's input cap) they will behave quite differently to when you put them into a high current use like the output caps in my set amp, power supply cap bipasses, etc.

What the guys here have found is that the humble, inexpensive PETP cap from Russia, seem to do it all!

There has been an enormous amount written about bipass caps over the years, so you'll just have to start reading. Perhaps someone can recommend a good starting point
 
Thank you james for the comprehensive explanation.

So you can actually use bypass caps both on signal level and on output?
You just parallel the bypass cap with the cap already there?

It sounds like it could be fun to try and at the price og the russian caps it's possible even for a student :)



@ analog_sa
lol.... I'm set to go then ;)
 
Yeah, you can have quite a bit of fun with this bipass thing.

Most people recon that the caps are the weak link in the signal chain and are the major source of the loss of fidelity - well, i disagree with this, as fidelity of the signal is dependent on everything starting from the power transformer type, the supply diodes, the connection wire, those caps, etc ,etc - everything makes a difference and you can use this in the final sound of your system.

A general rule of thumb to start this bipass thing is to choose the electro, lets say it's a 10,000uF/50 volter in the supply cap (the 2nd one in a C-R-C system) and let's say you're using basic Phillips caps and you want to make them "faster,brighter, etc" - I would begin with a 10uF/63v Propylene like a Wima, Solen, etc and listen to the effect. If that doesn't lift the sound enough, I'd try a good quality 220uF electro across that and see.
It's quite straight forward and teaches your ears "how to listen" to a sound difference.

Another example, in the power supply for say Rock/Roll and heavy classical, for a classA amp, you would look towards a "heavier sounding" supply, and it starts with a bigger transformer than required, slightly slower diodes like RURP860s rather than MUR820/ETH06/BYW29 and use caps like the Roe/Mundorf/F&T rather than brighter ones like Rifa, BHC, etc.
The choice of the components directly effects the sound, and this is just the power supply! Then comes the choice of rail regulation, if any, and so on.....

Mind you, there will be a lot of folks that think this is all a lot of rubbish and that adherance to the theoretical design will guarantee good results.
Well, I can't dissagree with that, but there is so much more to be had for a little bit of stuffing about and patient listening, and an open mind - the cost is generally minimal if your sensible and swap parts with others in your area.

I hold the view that the "amp + speakers + room" is an interdependent system and the source, preamp, crossover, etc are a seperate section and have their own particular compromises - this is the area that you can save an enormous amount of money with some smart shopping and get very good quality.

For now .....
 
Yes I have noticed from alot of reading here that for every argument or oppinion there are always others opposite or pointing in a 3rd direction :)

I'll definately try some of your suggestions.

Any recommendations for vocal lovers? :) (in general I listen to almost all genres but high res, well recorded vocals and acoustic music really makes my hair rise on my extremities ;) )
 
Well, just off the top of my head before "noddy land" I would look towards the more efficient drivers and lower power amp combination - maybe this is wrong, but I think many listening rooms in Northern Europe are smaller, so looking towards smaller boxes too, that will tolerate being located near side/rear wall.

Unfortunately, I'm not conversant about the wide range, high efficiency, reasonable priced drivers over there - I do know about the AERs, Lowthers, PHLs, etc, but for the lower cost units, I look more towards the Fostex, and such. I'm pretty sure that you have quite a few, but a lot aren't available for diy anymore - Audax, Focal, etc.

Curiously, I've just got a pair of the Fostex 167 drivers from Planet10 in Canada, that have been "EnABLed" and also have the little phase plugs, etc - quite remarkable and can be fitted to good boxes to reach a well controlled 60+ Hz (actual efficiency now about 93dB as no more high peaks and such) - exceptional for voices and acoustic instruments.
If you were to cross these over around the 100Hz area, could use seperate base drivers that can be located away from the "full ranger" with far better base driver/room response results - they call this "distributed sub-base" rather than just subwoofers banged into corners - there is a bunch of rules about all this too that's quite different to the usual way of looking at base/mid integrated speakers.

Open baffle can give really good results, too - these have their own rules, and most of the problems are about matching up the base drivers - definitely worth looking at for your use, not too good for R & R!

Sorry I can't be more specific or more help.

My current system is a bit unusual - a Nos dac (1541A) CD player into a Passlabs X2.5 pre, into FirstWatt F3 into 96dB 2way (modded Coral betta8s and Lorrantz 307 base), so you can see where I'm "at", so to speak.
 
i've tried the K73-16 ( 4.7uF/63V ) , both as bypass in powersupply and in the signal path
after few seconds , i already knew that it's my new favorite cap
the sound is similar to Blackgate BGN in super-cap configuration , but not grainy at all , and with more dynmic and resolution
the cap seems to have no parasitic resonances
i use them // with 0.1uF K73-16 or // with 0.1uF FT-2

thank you , Audiojoy , for the tip
 
Status
Not open for further replies.