I stand corrected.
I always thought that ferrites should be placed very near the cable connecters.
But i quickly checked into the papers of the ferrite go-to expert Jim Brown (also the EMI/RFI interference expert). So with short cables (such as yours) that are generating noise (such as yours) you can place the errite near the middle of the cable.
Now to chose the ferrite core material. With 4 common types to chose from:
Fair-Rite #31, #43, #61 & #78 it would take some more serious reading.
"RFI, Ferrites, and Common Mode Chokes For Hams"
http://www.audiosystemsgroup.com/RFI-Ham.pdf
"Measured Data For HF Ferrite Chokes"
http://www.audiosystemsgroup.com/FerriteDataHF.pdf
************************************************
Do we need chokes on both ends of that cable? The answer is, it depends. The Ethernet devices
on each end of the line are both potential generators of RF trash (because they both include digital
electronics). We also need to look at the length of the series circuit that includes the Ethernet cable.
If the cable is shorter than about λ/10 at the highest interfering frequency, a choke roughly
near the center of the cable may be entirely sufficient. A cable that is electrically longer than λ/10
(or is radiating VHF trash) is far more likely to need chokes at both ends.
Last edited:
I agree and are often used inside equipment where cables connect to a board and between stages of circuitry, they are extremely useful.
When you are using ferrites to STOP emissions you generally look to the system clock and its harmonics to determine a value to use, as the system clock tends to be one of the peaks on the EMC graph of generated noise. For input protection it can vary. For serious protection both ways a barrier across the PCB at the connector position can be used with a Pi filter formed by two low value caps and a ferrite, with a moat formed across the board for any supplies and ground. The ground connects across its own ferrite only, no caps.
Even though I would not recommend a cable geometry such as you have done, because of the braiding you signal and return are at least in close proximity. The lack of shielding is still my major concern.
Thanks for posting the pics.
🙂
Do you have a set of stock answers you put on every thread.
For the record, it is harder to remove low frequency noise (especially in the audio band or very near) than it is to remove any digital noise that is high frequency in nature.
Consumer audio is probably often better at removing this noise than some high end stuff because of things like CE and FCC, but neither noise is a s irritating at some pointless posts.........
While equipment in general suffers from these "maladies" it will be relevant to keep mentioning the issue - if one is satisfied with sub-standard sound then there is no problem, but there are a few who are interested in advancing "the art". The ability to see the trees, rather than continuing to blast holes in leaves with high powered laser vision helps here ...
Killmister, now I have seen the pictures I understand what you did; I was under the impression that you had a ferrite introduced into your signal wire, which would not be appropriate. The sort of clamp-on ferrite you use is ok, since it only suppresses common mode signals (with a good cable, not completely in the case here).
I share Marce's question about the braiding. It is not just a matter of shielding. Although digital signals are fairly robust for what gets into the cable, you need shielding to prevent the cable to radiate. Another reason why a braided cable ain't that good is because of cable inductance, which is bound to be higher than a twisted pair and much higher than coax. Because the current centroids of signal and return are not coincident in a braided cable, currents produce magnetic fields = the origin of inductance. It is also the reason why a comon mode ferrite will not function optimally.
I share Marce's question about the braiding. It is not just a matter of shielding. Although digital signals are fairly robust for what gets into the cable, you need shielding to prevent the cable to radiate. Another reason why a braided cable ain't that good is because of cable inductance, which is bound to be higher than a twisted pair and much higher than coax. Because the current centroids of signal and return are not coincident in a braided cable, currents produce magnetic fields = the origin of inductance. It is also the reason why a comon mode ferrite will not function optimally.
It is NOT art it is engineering design, first point.
Instead of harping on about sub-standard provide hard evidence, instead of generically dissing all equipment, show where the problem is.
I have had a lot of involvement with audio and listening tests in the past, and it is surprising how little is different to most of the participants when there is no other reference than listening. But in this instance listening alone is NOT enough, hard evidence must be produced and that means measurements and often EMC style evaluation to determine the effects of external noise etc. on a systems abilities.....
I just find the constant engineer bashing irritating, it gives the impression of a load of half assed guys hanging around throwing designs together with not a care in the world...I find in all industries though most if not all take pride in their work and want to produce the best they can within any confines the project requires.
Fortunately, digital audio is sufficiently robust that for short distances any old rubbish cable will do. At longer distances you need the correct characteristic impedance and reasonably good screening - just buy any decent commercial quality 75ohm coax.killmister said:
No. In most cases NOS = no serious attempt to reproduce the original band-limited signal.
It isn't parasitic capacitance. It is distributed capacitance, and it is a vital part of the way the cable works (in conjunction with the distributed inductance). I will hazard a guess that you have never studied electromagnetism, and have no idea how a TEM transmission line works?
The only way a degraded cable can be audible is by being sufficiently bad that it affects signal integrity. You need to be absolutely clear on this: no cable can improve sound, all competent cables do not affect sound, so therefore any cable which genuinely sounds different must be so bad that it is degrading sound.
A ferrite does no harm provided that all the cable conductors go through it. It then acts as a crude common-mode choke, and may reduce EMC issues.vacuphile said:
I am aiming very specifically at system engineering, just having a particular piece of gear up to some nominal standard is not sufficient to ensure overall capability. So, the problems are, firstly that the means to measure that system competence are very poor, and secondly that measuring such is not considered very important ... you don't measure the links of a chain individually, you pull on the ends until something snaps; that tells how strong the assembly is.
Listening tests would typically be of little value, because they don't stress test: the latter is what I do, which makes it trivially easy to identify flaws, weak links in the chain.
The characteristics of the ferrite i used:
65 ohm at 25MHz / 95 ohm at 100MHz.
This cable does not seem at all conventional and not optimized, i agree .
But why it works good? I don't know exactly ...
I specify that it has been tested on different hardware of mine with the same conclusion : it exceeds in subjective quality some cables which are expensive !
Here is the history of this cable :
I read somewhere in a French forum audiophile that a guy was making SPDIF braided cables from silver copper single wires or litz wire but no more precision. But he said its cables worked better than cables costing a fortune !
I had two coaxial cables made with Belden 1694 and generic RG59 . Both worked well but a friend had brought a Wireworld cable that worked much better than my cables . I do not want to spend $ 200 or more .
So I thought about what I had read in the forum and as I had several types of cables in my workshop , i decided to make two test braided cables of 30 cm with MGTF wire and the other with some litz and some 10 cents plastic RCA .
When I tried these two little shabby cables, I was very surprised !My two coaxial cables were far exceeded at the musical rendering and $ 200 Wireworld too..
and i was not the only one to hear : my wife, my friend have confirmed.
Then I made a better finish and longer cable (90 cm) as you see. I thought the treble was a little worse than my two little test cables of 30 cm.
I had the idea of adding the ferrite and everything is back to normal , treble and focus was back ...
That's all the interest to share this experience on this forum, I read lots of interesting things, and I appreciate reactions and debates to try to understand.
I hope that someone will have probably an explanation. I invite you to try this cable , all my audiophile friends have one and i spent several long evenings to braid.
thank you to all !!
65 ohm at 25MHz / 95 ohm at 100MHz.
This cable does not seem at all conventional and not optimized, i agree .
But why it works good? I don't know exactly ...
I specify that it has been tested on different hardware of mine with the same conclusion : it exceeds in subjective quality some cables which are expensive !
Here is the history of this cable :
I read somewhere in a French forum audiophile that a guy was making SPDIF braided cables from silver copper single wires or litz wire but no more precision. But he said its cables worked better than cables costing a fortune !
I had two coaxial cables made with Belden 1694 and generic RG59 . Both worked well but a friend had brought a Wireworld cable that worked much better than my cables . I do not want to spend $ 200 or more .
So I thought about what I had read in the forum and as I had several types of cables in my workshop , i decided to make two test braided cables of 30 cm with MGTF wire and the other with some litz and some 10 cents plastic RCA .
When I tried these two little shabby cables, I was very surprised !My two coaxial cables were far exceeded at the musical rendering and $ 200 Wireworld too..
and i was not the only one to hear : my wife, my friend have confirmed.
Then I made a better finish and longer cable (90 cm) as you see. I thought the treble was a little worse than my two little test cables of 30 cm.
I had the idea of adding the ferrite and everything is back to normal , treble and focus was back ...
That's all the interest to share this experience on this forum, I read lots of interesting things, and I appreciate reactions and debates to try to understand.
I hope that someone will have probably an explanation. I invite you to try this cable , all my audiophile friends have one and i spent several long evenings to braid.
thank you to all !!
Last edited:
I wouldn't make that conclusion.. I only trust long term listening evaluation.🙂
I have a few hundred LP to CD's transfers.. I use a Sony CDR-W33..
I assure you that the burned CD's sound different than the LP and there is losses.. Is there something wrong with my Burner or the way I'm operating it. Maybe I need higher quality burner..
If I was to compare the two discs under ABX conditions it's possible I might have trouble distinguishing, But then I don't trust ABX for proper evaluation🙁
Defects are everywhere, Part of choosing audio equipment is about picking the defects that bother you the least amount..
For you, yes, excellent reasoning..
In OP, should have been " CD Standard digital source"
The higher rez transfer I heard of an old analogue master tape was amazing and transfers like this could possibly make me sell my LP's.. LP nostalgia wouldn't make me hold on to the LP collection, cost of replacing them would..
Last edited:
Because the digital system is very robust, and it takes some real effort to screw things up. Even an incompetent wire like this, or some of the others you mentioned, won't usually cause major issues as long as it's kept relatively short.
Typical objectivists have a lot of trouble comprehending that it's a system that produces sound, rather than a DUT - and in the case of a ABX which has a A/D/A conversion vs. a straight LP, the system includes all of the machinery that is operating at the time of listening. Just saying that there should be no cross-interference doesn't make it so, and this flawed aspect to the testing means that little trust can be placed in what comes out of it ...
I can't stay with the reason "the digital system is very robust" and ".... kept relatively short" whereas cables of the same length were compared ?
Wireworld cable at 200$ and Fadel Art at 300$ are beat by this cable for the same length of wire.!!
I find your conclusions are sloppy and precipitate, and I remain convinced that, beyond a certain length, it is best to avoid coaxial cable if possible .
Last edited:
I would suggest that it is the recording chain (ADC) where the difference is occurring, writing to a CD is just slapping data onto it.... To check the CD burner though you can compare files on your hard disk if you are storing them there first with the CD data.
Perhaps they did - in audio there is very weak correlation between cable cost and cable quality, and the curve may have a negative slope at the really expensive end.killmister said:
I have good news for you: you don't need to. 75ohm coax is fairly cheap, widely available, and for the fairly short runs typical of domestic audio its doesn't even need to be particularly good quality coax.
That suggests that your cable was picking up RF, and the ferrite reduced this. Coax probably would not need the ferrite.
Yes. Your cable, although poorly constructed for the task, is sufficiently short that it does little damage to the signal. It may pick up some RF, which will add to jitter and so add to distortion which you and others then assume is an improvement in sound.
Thanks for providing exactly the right words to express what I think of your conclusions. All an SPDIF link has to do is get some pulses from here to there with the minimum of distortion. As the source and receiver both expect to see and drive a 75ohm unbalanced link there can be no better way to do this than a 75ohm coaxial cable. Anything which sounds different from this must necessarily be worse.
Last edited:
- Status
- Not open for further replies.