Hi Tom,
Very informative post. Thanks. I will be taking your hard earned lessons to heart.
A couple of thoughts. For smaller polystyrene caps, PCX has surplus caps available at very reasonable prices. The ones I have purchased all have copper leads and seem like nice quality. I will say that I purchased smaller value ones, 1 nF and lower. HiFiCollective seems to have similar parts available.
So far, I have only used them as high frequency compensation caps to reduce MHz gain on an op amp, so it is not a place where their sonic character shows through. That said, everything sounds good so far.
As for size matters, I agree. I think 100:1 could even turn out to be on the large bypass cap size in certain circuits. For example, my 1 uF C13 is bypassed with a 1 nF or 1000:1 and I can tell the difference to a 3 nF. If I were buying a cheaper cap with the intention of bypassing it, I would be tempted to buy a few sizes of bypass and experiment. Who knows, maybe 10,000:1 is perfect in some cases?
I think if I was going to bypass at 20:1, the only way it might work would be if the caps were a big and little version of the same cap, or at least something quite close to the same design. I do believe that you can get some of the benefit of bypassing when using big and small of the same cap. The smaller caps have a higher resonance, so they continue to act as capacitors to a higher frequency, even when they are the same design.
Jac
Last edited:
Interesting idea. Im using Sonicaps for c13 and have 2 pieces of those boxed orange Siemens polystyrenes caps, 15nf value. Havent check if they got copper leads but i will definately try them as a bypass to sonicaps
The guideline regarding ratios was simply wild guessing on my part. It could very well be that smaller ratios are effective, and 20:1 harmful. This entire business is so circuit and component dependent that no one can be any more specific or authoritative about anythig other than what they have experienced themselves.
The 15nF/Sonicap could be the "sweet spot" for an initial trial. Please give it a try and let us know.
I am not trying to hijack the thread, but I have a comment that pertains to an aspect of performance that is very important to me and probably others, and a metric for evaluating that aspect. If you're not interested in a personal recommendation, skip the rest of this post.
I believe a stereo can sound good without perfectly accurate frequency response. Some amps (notably, many tubed) have flabby bass and rolled off highs and still do a good job of playing something that sounds like music. Then there are plenty (too many, in fact) of amps with solid bass and extended highs that sound like perfect machines but nothing like music. In my experinece, a system that cannot reproduce at least a certain amount of spaciousness does not ever sound like real music. If instruments or the entire image is flat, I cannot fool myself into believing the illusion, so I spend a lot of time (and words) attempting to capture and describe that attribute. The MyRef amp is among the best SS designs, at any price, I have ever heard at projecting real space, and it can also excel at producing realistic tonality with accurate FR.
The most recent configuration of my electronics/speakers has afforded me very realistic representation of three dimensional music in my listening room. I have lots of CD's and SACD's that sound good, and a few that are exceptional, but one that is truly unique in a completely unexpected manner.
There is an obscure but excellent recording (how many dozens or possibly hundreds of CD's fit that description!) called "Trailer Park" by Beth Orton, which contains a song entitled "Galaxy of Emptiness" that presents the deepest image I have ever heard from a stereo. I admit that I have been entranced by this CD, and specifically this song, in various states of legally and illegally altered consciousness. But Friday night I was perfectly sober, listening with my audio buddy who has better HF hearing than I do and whose opinion I respect. I played this song, and we were both mezmerized by the effect of hearing sounds coming from around the room, filling the entire space within all four walls, from two speakers which had totally disappeared. After he left, I had to replay that song just to be sure that I had really heard what I thought I did. It was better the second time! If you need a test of whether your system is ideally equipped and situated, this song could serve as a standard. This is not an audiophile sanctioned disc, but it should be. The rest of the CD is a superb recording of very good music, which has been described as a marriage of folk and trip hop. Not as bizarre as it seems: Beth has an engaging voice, and her songs are poetic and melodic, some of them infused with a sort of Chemical Brothers flavor, others played simply on acoustic instruments, including violins and cellos. The production is flawlessly realistic, and in this one case, uniquely hypnotic. The bass should vibrate loose windows, and I swear the synthesizer will swim alongside your head while ethereal voices echo in the emptiness of the galaxy. Spacey, in every way that word can be used.
We now return you to our tediously earthbound programming...
Peace,
Tom E
The 15nF/Sonicap could be the "sweet spot" for an initial trial. Please give it a try and let us know.
I am not trying to hijack the thread, but I have a comment that pertains to an aspect of performance that is very important to me and probably others, and a metric for evaluating that aspect. If you're not interested in a personal recommendation, skip the rest of this post.
I believe a stereo can sound good without perfectly accurate frequency response. Some amps (notably, many tubed) have flabby bass and rolled off highs and still do a good job of playing something that sounds like music. Then there are plenty (too many, in fact) of amps with solid bass and extended highs that sound like perfect machines but nothing like music. In my experinece, a system that cannot reproduce at least a certain amount of spaciousness does not ever sound like real music. If instruments or the entire image is flat, I cannot fool myself into believing the illusion, so I spend a lot of time (and words) attempting to capture and describe that attribute. The MyRef amp is among the best SS designs, at any price, I have ever heard at projecting real space, and it can also excel at producing realistic tonality with accurate FR.
The most recent configuration of my electronics/speakers has afforded me very realistic representation of three dimensional music in my listening room. I have lots of CD's and SACD's that sound good, and a few that are exceptional, but one that is truly unique in a completely unexpected manner.
There is an obscure but excellent recording (how many dozens or possibly hundreds of CD's fit that description!) called "Trailer Park" by Beth Orton, which contains a song entitled "Galaxy of Emptiness" that presents the deepest image I have ever heard from a stereo. I admit that I have been entranced by this CD, and specifically this song, in various states of legally and illegally altered consciousness. But Friday night I was perfectly sober, listening with my audio buddy who has better HF hearing than I do and whose opinion I respect. I played this song, and we were both mezmerized by the effect of hearing sounds coming from around the room, filling the entire space within all four walls, from two speakers which had totally disappeared. After he left, I had to replay that song just to be sure that I had really heard what I thought I did. It was better the second time! If you need a test of whether your system is ideally equipped and situated, this song could serve as a standard. This is not an audiophile sanctioned disc, but it should be. The rest of the CD is a superb recording of very good music, which has been described as a marriage of folk and trip hop. Not as bizarre as it seems: Beth has an engaging voice, and her songs are poetic and melodic, some of them infused with a sort of Chemical Brothers flavor, others played simply on acoustic instruments, including violins and cellos. The production is flawlessly realistic, and in this one case, uniquely hypnotic. The bass should vibrate loose windows, and I swear the synthesizer will swim alongside your head while ethereal voices echo in the emptiness of the galaxy. Spacey, in every way that word can be used.
We now return you to our tediously earthbound programming...
Peace,
Tom E
Last edited:
"Trailer Park" by Beth Orton. is available in "Auto-Rip". First time I'v seen something like that at Amazon.
"Galaxy of Emptiness" for $0.99 at iTunes.
"Galaxy of Emptiness" for $0.99 at iTunes.
Last edited:
What the heck is auto-rip??
Tom,
Were you playing the CD or vinyl? If CD, the original 1997 recording or the remastered 2009 "Legacy Edition"? I always like a well recorded album and have enjoyed other Beth Orton songs in the past.
Jac
Thanks, Bob, for pointing that out. It would be interesting to know if a ripped version has the same 3D nuance as the standard CD. When I rip CD's at home, there is a definite loss of fidelity that sometimes I can tolerate, other times not, depending on the quality of the original.
Plenty of people downloading stuff these days, so it can't be all bad. I'm just not sure it will be as good.
The new CD can be bought for bargain alert a dollar, plus shipping. Worth the difference if you like the music?
I have the original CD. I dislike much remastered stuff due to loudness wars tinkering, and I've come to distrust them in general until proven otherwise. No black discs left in my house, but I get to hear vinyl on my buddy's expensive system. Sometimes it's better, other times not. So it goes with all things audio.
End of hijack.
Peace,
Tom E
Plenty of people downloading stuff these days, so it can't be all bad. I'm just not sure it will be as good.
The new CD can be bought for bargain alert a dollar, plus shipping. Worth the difference if you like the music?
I have the original CD. I dislike much remastered stuff due to loudness wars tinkering, and I've come to distrust them in general until proven otherwise. No black discs left in my house, but I get to hear vinyl on my buddy's expensive system. Sometimes it's better, other times not. So it goes with all things audio.
End of hijack.
Peace,
Tom E
Last edited:
This is the mastering I like : Jessie Ware
It makes speakers work
Whole albums is good, but I ll point out
Wildest Moments and
If you're never gonna move
It makes speakers work
Whole albums is good, but I ll point out
Wildest Moments and
If you're never gonna move
Oh No No, You don't get away that easily.
"I admit that I have been entranced by this CD, and specifically this song, in various states of.............."
After hearing the music we want a complete and detailed history of your "Music Listening" methods, surroundings and exhancment aids.
While we are discussing tracks, remember that I set up a folder, I believe back in the beta phase, for MyRef builders on Boxnet, . It is still there and I will hunt down the original post put it here and on the Final Version thread. I think it will be useful in the tuning/comparison process.
Boxnet is free but there are a few hoops to jump to avoid any copy-write problems.
Edit: Oops - that wasn't too difficult. FE Audio Tests.
Boxnet is free but there are a few hoops to jump to avoid any copy-write problems.
Edit: Oops - that wasn't too difficult. FE Audio Tests.
Last edited:
The bypass capacitor will be the smallest capacitor value. It is incapable of passing LF & MF due to high impedance at these frequencies.
That leaves it passing HF only.
To do this best the HF cap MUST use the shortest possible route with the LEAST inductance. I would suggest the ONLY place for the bypass cap is ON the PCB using the pads provided, or on new holes drilled into the traces, if this will achieve the shortest possible route (least inductance) for the the bypass cap.
The LF passing cap is predominantly passing the LF and MF signals. It is much more tolerant of inductance added to the signal route. This cap should be tacked on above, or below, or beside the board as best fit can be achieved.
If you use three paralleled caps then the highest value can use the longest route. The lowest value can use the shortest route and the intermediate value uses an intermediate length/inductance route.
I suspect a lot of you are putting the highest value cap direct to the PCB in the best fit achievable and then tacking on the HF/bypass cap in any location that then fits. This MUST lead to inductance effects on the HF signal.
That leaves it passing HF only.
To do this best the HF cap MUST use the shortest possible route with the LEAST inductance. I would suggest the ONLY place for the bypass cap is ON the PCB using the pads provided, or on new holes drilled into the traces, if this will achieve the shortest possible route (least inductance) for the the bypass cap.
The LF passing cap is predominantly passing the LF and MF signals. It is much more tolerant of inductance added to the signal route. This cap should be tacked on above, or below, or beside the board as best fit can be achieved.
If you use three paralleled caps then the highest value can use the longest route. The lowest value can use the shortest route and the intermediate value uses an intermediate length/inductance route.
I suspect a lot of you are putting the highest value cap direct to the PCB in the best fit achievable and then tacking on the HF/bypass cap in any location that then fits. This MUST lead to inductance effects on the HF signal.
Last edited:
Andrew,
Thank you for this. I had never thought of it this way, but it makes excellent sense.
Happily, by accident, I have been doing the right thing up to now.
Now the challenge is how to handle this best when the input cap is oversized. I had planned on making the LF cap the signal connection between the amp board and a balanced line receiver board. My choices appear to be either;
- run the HF bypass cap and return wire between the boards as a twisted pair and parallel to the LF cap
- run both the LF and HF (shortest path) caps on the board with the LF cap hanging off the side of the board, then run a twisted pair of wires from the other board
My first thought is that the second choice is better because it puts both caps closer to the ground plan. Any thoughts, anyone?
Jac
There is science behind every result we hear.
But if you guess you can never be sure you are reaching any of the correct conclusions.
I suspect much of the reporting of capacitor sound is flawed because none are asking themselves "why did I hear that" and come up with a method of testing that reveals why they heard the change.
Once you have a hypothesis you can start trying to prove it. From that science you can then make progress on determining what characteristics of the filter component are important.
But if you guess you can never be sure you are reaching any of the correct conclusions.
I suspect much of the reporting of capacitor sound is flawed because none are asking themselves "why did I hear that" and come up with a method of testing that reveals why they heard the change.
Once you have a hypothesis you can start trying to prove it. From that science you can then make progress on determining what characteristics of the filter component are important.
A further thought.
I posted an idea on running two return wires diametrically opposed around the filter capacitor.
Only one Member replied. Again I suspect that noise on the separated return route can confuse the "hearing" result.
Why has no one else reported on experiments for the two wire return?
I posted an idea on running two return wires diametrically opposed around the filter capacitor.
Only one Member replied. Again I suspect that noise on the separated return route can confuse the "hearing" result.
Why has no one else reported on experiments for the two wire return?
I have a slight problem.
My design requires me to put two tweeters in series. They have nominal impedance 4Ohms and are well behaved. Crossover is at 7kHz.
http://www.falconacoustics.co.uk/downloads/Seas/h1396_datasheet.pdf
What should I choose as the AC voltage for the toroid?
For the mids I have single 4Ohm elements and there I have chosen 22V AC as per earlier recommendations.
/Kazam
My design requires me to put two tweeters in series. They have nominal impedance 4Ohms and are well behaved. Crossover is at 7kHz.
http://www.falconacoustics.co.uk/downloads/Seas/h1396_datasheet.pdf
What should I choose as the AC voltage for the toroid?
For the mids I have single 4Ohm elements and there I have chosen 22V AC as per earlier recommendations.
/Kazam
find one that does 199.9mVdc and does 199.9mVac without any of the other functions.
These can be bought for <£5
If you can get the DC accuracy to <=+-0.2%+1digit, then even better.
If you want other functions then buy cheap and when accuracy becomes important you build or buy the specialist instrument for that function.
If you require better accuracy go for a secondhand bench meter.
These can be bought for <£5
If you can get the DC accuracy to <=+-0.2%+1digit, then even better.
If you want other functions then buy cheap and when accuracy becomes important you build or buy the specialist instrument for that function.
If you require better accuracy go for a secondhand bench meter.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.