Tweakers' Asylum

Tweaks for systems, rooms and Do It Yourself (DIY) help. FAQ.

Return to Tweakers' Asylum

Message Sort: Post Order or Asylum Reverse Threaded

RC AC FILTER: 100 OHMS TOO LITTLE?

71.92.218.178
Posted on May 28, 2010 at 19:02:39
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009
 I have made my first attempt at the much-covered AC filter (Sekela, et al. version) using .47, .047, .0047 polyprop caps and 100 ohm resistors.

Had it in for a couple hours. What I hear is more detail and clarity but also the high end is a little tizzy. Is this part of burn-in or sign that I need to alter the resistance?

 

Hide full thread outline!
    ...
Both, but ask Al first, posted on May 28, 2010 at 21:12:37
bartc
Audiophile

Posts: 4424
Location: SF Bay Area
Joined: November 25, 2002
 I recall a different value for R, something like either 60 or 120 Ohms, probably the latter. Go back to his posts or ask him directly.

And there is burn in, but not much. The results should be LESS tizz, not more!

Also, there is polarity to this. Again, go back to Al's posts or ask him.

 

Much trial-and-error..., posted on May 29, 2010 at 15:11:46
Al Sekela
Audiophile

Posts: 9169
Location: Northern California
Joined: February 18, 2002
 I use Wima metallized paper caps, as none of the synthetic polymer caps I've tried give acceptable performance.

Each separate size of capacitor needs a different amount of series resistance to sound its best. Too little resistance can indeed lead to treble enhancements, as the caps are resonant objects themselves and the resistors have to damp them as well as the AC line.

 

RE: Much trial-and-error..., posted on May 29, 2010 at 16:31:55
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009
 Thanks Al, I added another set of 100 ohms to the caps that I intended for a second unit, creating an RCR of 200 ohms per cap. The treble was reduced. I then switched polarity and the sound became kind of closed-in though not unpleasant.

I just happened to find several .01 and .001 wima mp3 caps I had stowed away so maybe I buy some .1 wimas and give another try.

Is there a burn in time for this filter? I didn't see any mention of that in my searches.

Hate to ask these questions when they've probably been asked before. I spent many hours searching but there're only so many permutations of 'ac rc filter' I can think of to enter. Wish this forum was the kind where you could put stickies...

thanks much

 

Reduced treble means you are on the right track., posted on May 30, 2010 at 10:17:13
Al Sekela
Audiophile

Posts: 9169
Location: Northern California
Joined: February 18, 2002
 One aspect of RF noise in an audio system that makes tweaking difficult is that a little of it sounds like enhanced detail and spaciousness. This is because the spurious audio artifacts are most noticeable in the treble. It took me a long time to figure this out, and to learn how to listen for improvements.

If you are used to a bright treble, immediate reduction of the RF noise sounds like something went wrong. A little extended listening reveals that the true musical details are presented more honestly, however. There should be a more relaxed feel and the sibilants and cymbals should be more in balance with the rest of the music. It does not take long for the improved presentation to become your new standard.

Yes, there is a burn-in time of at least a few days, IME.

Try splitting the resistance by putting resistors of half the total on either side of the cap.

 

RE: Reduced treble means you are on the right track., posted on May 30, 2010 at 10:55:12
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009
 Thanks Al, yes I put 100 ohm on either side of each cap the second time around in stead of just one and the sound was improved. The polarity difference I'll have to ponder as to which is the most effective, as one way seems almost *too* relaxed in presentation, but I'll figure it out. I have 68 ohm PRPs on the way to make R-C-R versions. Hopefully that will be the sweet spot.

Last 2 questions:

1) The .1, .01, .001 should be fine, yes? I have not read anyone else trying this cascade.

2) Is there no benefit to using higher cap values with higher wattage *non* inductive wire wound resistors?


 

Any suitable cascade should work., posted on May 31, 2010 at 14:27:59
Al Sekela
Audiophile

Posts: 9169
Location: Northern California
Joined: February 18, 2002
 I only used 0.47 microfarads as the starting value because it limits the power dissipation in the 0.5 watt resistors to below half their rated value (this a conservative design principle).

Resistor type is as important to the performance as capacitor type. I chose not to use 1.0 microfarads as the largest value because I did not want to spend the time experimenting with power resistors. The PRP half-watt resistors give the best sonic performance within their power limitations, and it is not clear to me that there is a power resistor with equivalent sonic performance. Caddock are good but require heat sinks.

Also, there is no a priori requirement that all the cap values be spaced by decades. We know that all caps have upper frequency limits imposed by self-resonance, but the self-resonant frequencies should be measured to determine the cap value spreads. I have a neighbor who works for Agilent and who has helped me in the past with cap measurements, but it is a major imposition on him to set up the work. If I ever get the chance, I will test the caps I use and see if another set of values would give more efficient coverage.

Alan Maher's Circuit Breaker Filters seem to enhance the performance of the R-C filters nicely. They are non-contact and are not subject to the same sort of self-resonance limitations that caps with resistors are.

 

RE: Any suitable cascade should work., posted on May 31, 2010 at 15:46:20
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009
 Got it. I'm a big fan of PRPs in general. Thank you for the explanation. Appreciate it.

 

RE: Any suitable cascade should work., posted on June 7, 2010 at 23:03:18
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009



Since I have yet to see any pics of people's builds I've uploaded some photos...

 

RE: Any suitable cascade should work., posted on June 7, 2010 at 23:06:02
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009



Another pic. Looks kind of like a bug...

 

RE: Any suitable cascade should work., posted on June 7, 2010 at 23:06:50
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009



Another angle.

 

RE: Any suitable cascade should work., posted on June 7, 2010 at 23:08:06
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009



And all done. Hot glue secures the .1 to the enclosure and blue tack secures the caps to each other...

 

Thanks for sharing your photos., posted on June 9, 2010 at 11:53:13
Al Sekela
Audiophile

Posts: 9169
Location: Northern California
Joined: February 18, 2002
 It would be better to put the smallest cap closest to the plug. This network affects the highest frequencies.

 

RE: Thanks for sharing your photos., posted on June 9, 2010 at 12:03:55
wushuliu
Audiophile

Posts: 117
Joined: March 19, 2009
 *groan*

Thanks Al.

If it's ok, I think I will write up a brief post on putting these together at AudioCircle w/ pics so that there will be some kind of single reference point for people who want to put these together without asking the same question over and over and spending thousands of hours using the search function to glean different options/versions.

 

OK so consider these extra steps, posted on June 9, 2010 at 17:43:42
bartc
Audiophile

Posts: 4424
Location: SF Bay Area
Joined: November 25, 2002
 Al uses Teflon tubing over the exposed leads and the Rs, while I do the same with Teflon plumber's tape wrapping. Al showed me how to Kynar tape the whole bundle, which I also do with all of them now.

We both use TI-shielding, and mine is grounded too (12 AWG copper wire here). So that's under the Kynar wrap as well. The cotton damped Ti surround, taped with Kynar, makes a tidy package that is electrically and mechanically well damped.

CDC suggested replacing as much of the native leads as possible with silver 20 AWG wire for the speaker cable filters, which I did to greater advantage as he stated.

Al suggested changing the AC filtering R-C-Rs (make sure you use safety rated caps!!!!) native leads to Teflon 24 AWG copper (CAT 5 for instance). I had been using 18 AWG copper hookup wire for this purpose until then. Al told me the slimmer wire would give better sonics (more detailed without loss of true LF info), and I tried it and found his theory to be correct in my installations as well. I also wrap all this in Teflon tape to give some measure of greater safety (real or imagined).

As you appear to be doing, Al recommended directly soldering the leads to the tines of the plug, which I'm also doing and it helps sonically.

 

Page processed in 0.024 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]