PPA L-Cancelling setup in the C2/C5 positions
Jun 13, 2004 at 9:01 PM Thread Starter Post #1 of 15

CBMC

100+ Head-Fier
Joined
Jul 15, 2003
Posts
365
Likes
0
Can someone explain to me how and why this is done. I was looking too put some Black Gates in this spot in my ppa. Could I just solder the Black gates N-series(non-polarized) to the c2/c3/c5 positions. Are there any sonic benefits to l-cancelling? Does it cause stability issues in the ppa? Would you suggest using the black gates in the c2/c3/c5 positions? And finally if anyone has done this, has there been any noticable improvement? Hope someone can help with all of these questions.
 
Jun 13, 2004 at 9:24 PM Post #2 of 15
the L-cancelling pair works, as you ascertained, for identical non-polar caps. You connect the parallel pair in the opposite direction, and this reduces the inductance of the network considerably. Do a search for this, it's been well covered already.

The only position where you can effectively implement the pairs is C3, as they are both rail to rail. For C2 there are 3 positions, so no pair there. C5 is half supply, look at the low power section in the schematic, you can't parallel those unless you put non-polars into C4 as well, and you probably won't get enough capacitance in that position with an L-cancelling pair

g
 
Jun 14, 2004 at 3:36 AM Post #3 of 15
OK, thanks. One more question. Going through the forums I noticed that VodooChile had built a PPA with all black gate capacitors. My question is: Why use electrolytic capacitors instead of film? I always thought that film caps where supposed to sound better?
 
Jun 14, 2004 at 5:20 AM Post #4 of 15
Personally, I think this is a waste of money, but what do I know?
rolleyes.gif
 
Jun 14, 2004 at 5:46 AM Post #5 of 15
Yeah, I was kind of thinking that the rewards would be minimal. But I guess with audio you always hit a bar and it takes more and more money to move the bar even slight amounts as you go up. Morsel, what caps do you recommend?
 
Jun 14, 2004 at 5:47 AM Post #6 of 15
Quote:

Originally Posted by CBMC
Why use electrolytic capacitors instead of film? I always thought that film caps where supposed to sound better?


I've measured capacitors (various polypropylene, silver mica, high quality ceramic, Oscon organic electrolytics, and basic Black Gates) at 100, 1k, 10k, and 100khz. Polypropylene capacitors are much more uniform (near ruler flat) in capacitance values over the frequency range.


JF
 
Jun 14, 2004 at 6:18 AM Post #7 of 15
i have blackgates N in C2/C5

in C3 it didnt work too well because of the tight space and it was shorting on the overly big tin pads, so it was back to box caps
 
Jun 14, 2004 at 8:05 AM Post #10 of 15
Quote:

Originally Posted by CBMC
What type of black gates do you have the c2/c5 spot?


i have blackgates N 1.0uF caps in those positions, not even sure where misterx found them though
 
Jun 14, 2004 at 9:17 AM Post #11 of 15
Quote:

For C2 there are 3 positions


In PPA v1.0, yes, but in PPA v1.1 there are only 2. Because of the way they're connected in v1.1, though, I doubt any of this L-cancelling stuff would matter with them.
 
Jun 16, 2004 at 1:05 PM Post #13 of 15
Due to the Addition of R8 The So Called Supper E configuation will not work because of the added resistors. This is actualy better technicaly since the added resistors Kill the "Q" formed by the inductance of the PC foil trace and C4 and C5, in addition R8 also addes to RFI filtering and makes for a much smaller area for the High frequency decoupling loop.
see illistration at
http://www.elecdesign.com/Files/29/1478/Figure_04.gif
and Documentation at
http://www.elecdesign.com/Articles/I...leID=1478&pg=3


The Jfets and C4 provide first order decoupling and C5 provided Bypass in version 1.0 however with the addition of R8 C5 also in tandum with R8 also provide second order High frequency decoupling in addition to it's primarry task of Bypassing. The term Bypass and decouple are related but not the same as bypassing refers to diverting the noise to ground and decoupling refer to isolating the noise.
See Electronic gotchas at http://www.glencoe.com/ps/ee/william...g.html#defined
 
Jun 16, 2004 at 1:54 PM Post #14 of 15
Quote:

Originally Posted by CBMC
OK, thanks. One more question. Going through the forums I noticed that VodooChile had built a PPA with all black gate capacitors. My question is: Why use electrolytic capacitors instead of film? I always thought that film caps where supposed to sound better?


I do agree that ordinary electrolytics alone would make lousy bypass caps as compared to film, but I can't deny that the NX caps in l-cancelling pairs is more than adequate to quell oscillation (at least in this particular app, YMMV), and doesn't just 'do the job', the amp sounds better to me than it does with film caps. I don't believe it is the placebo effect either, as I have done cased side-by side comparisons with many people, and every single listener has chosen the blackgate amp.

I did it as much for as an experiment as anything else. The amp does not oscillate, and is extremely fast, and smooth. The bypass caps at the op-amps are the 0.1uF NX Hi-Q non-polars in l-cancelling pairs. In this configuration, the BG bypass caps are not much more than Wimas, although more than Phillips/BC. As for the main power rail caps, yes, the blackgates cost quite a bit. But the whole amp is still a tremendous value, even with the added expense of these caps, relative to buying a commercially manufactured product. So why not, I figured. I love the results, and still saved a good amount of money as compared to buying a commercially available product of similar performance. I cannot say if having l-cancelling blackgate bypasses coupled with Panasonics or Muse or Cerafine would sound better/as well/worse, as I have not tried that configuration.

So, I don't really want to go down this road about what is better, as it's all a matter of preference. For me, if it sounds good- it is good, and it sounds good to me. Everything I have heard or been told about them is excellent, and I would have to concur, having done my own experimentation. To each his own. I'd encourage anyone to try them, rather than disourage them. The economic difference is pretty insignificant in the scope of what you spend on this pastime as a whole.

Edit: WRT what PPL is saying about R8- R8 was not the same on my amp's board as it is now. In otherwords, things could well be different with the current crop of PPA boards. I have not built an all-blackgate amp with the newest version of the board.
 

Users who are viewing this thread

Back
Top