Digital Drive

Upsamplers, DACs, jitter, shakes and analogue withdrawals, this is it.

Return to Digital Drive

Do Good Coax Cables Make a Diff?

71.209.163.147
Posted on October 22, 2009 at 17:01:46
Lewis Moon
Audiophile

Posts: 156
Location: Arizona
Joined: November 14, 2008
 I just switched out a low/medium quality digital coax cable with a nice Blue Jeans Coax and, counter to my intuition, I could hear a difference. A noticeable change for the better with more detail, space and tighter, deeper bass. I really expected nothing....really. The only reason I went for the BJC was it was cheap (used) and was the right length. I always just assumed that digital signals were just ones and zeros. Am I drinking something?

RE: Do Good Coax Cables Make a Diff?, posted on October 31, 2009 at 22:13:29
akltam
Audiophile

Posts: 688
Location: Hong Kong
Joined: September 29, 2002
 According to this guy, not all cd player has good transmitter for these SPDIF signal and bad cables will make things worst. And he also inspected the spdif signal under the scope and some looked horrible. I did built his jfet version of the spdif buffer. I think it does improved the quality of the final performance.







RE: Do Good Coax Cables Make a Diff?, posted on October 25, 2009 at 11:34:55
Bob Neill
Dealer

Posts: 1927
Location: New England
Joined: October 1, 1999
 Can make a huge one.

RE: Do Good Coax Cables Make a Diff?, posted on October 24, 2009 at 11:26:59
cfraser
Audiophile

Posts: 2070
Location: Pickering, Ontario
Joined: April 30, 2000
 They do. And I'd say a length change can sometimes make more of a sonic improvement than a cable-type change, or than a termination change. IOW a "good" cable doesn't have to be an expensive one.

Besides the cable, but related, is what's just before and after the cable jacks. I have found that isolating the coax, (very) preferably at both ends, makes a big diff. And drastically reduces the perceived diffs between cables.

RE: Isolation., posted on October 24, 2009 at 11:44:01
rick_m
Audiophile

Posts: 2539
Location: Oregon
Joined: August 11, 2005
 Yes, even a paltry 1/2" of isolation can reduce distortion to absolutely inaudible levels. Or were you thinking of some other sort of isolation?

Rick

RE: Isolation., posted on October 24, 2009 at 12:50:19
cfraser
Audiophile

Posts: 2070
Location: Pickering, Ontario
Joined: April 30, 2000
 I was thinking of electrical isolation of the coax cable, by I/O transformers specifically.

If you think "I tried isolated with Toslink and didn't like it", bear in mind there's a lot more to the Toslink interface, isolation just happens to be a natural part of it (and a wasted part, considering what it could have been and so uncommonly is).

RE: Isolation., posted on October 24, 2009 at 14:20:43
rick_m
Audiophile

Posts: 2539
Location: Oregon
Joined: August 11, 2005
 Cool, for some reason it didn't occur to me that you were actually doing something, well, normal. Guess I've been spending to much time on AA, I visualized something along the line of fishing weights suspended from the cable by cat gut...

I've been thinking of rigging up galvanic isolation and isolated power for a USB A/D to do my records, but it never crossed my mind to isolate my DAC inputs on the regular system. But, now that you mention it, it surely makes sense, I'll try it. Did you keep using coax or go to a balanced line?

Thanks for the info.

Rick

RE: Isolation., posted on October 24, 2009 at 14:44:47
cfraser
Audiophile

Posts: 2070
Location: Pickering, Ontario
Joined: April 30, 2000
 I still use the coax. I have never owned or borrowed a DAC with an AES/EBU input, though one of my players has the output, so never tried it.

I have found this mod very effective for minimising most cable effects. It's not a magic bullet. I'd see what others say. Compare with what Al said below, another method along the path to achieving a similar end.

The input/output isolation is quite cheap to do. Usually easy in the transport as you can install right on the output jack in many cases. A little trickier at the DAC input end usually, strictly a space issue not a "technical" difficulty. Warranties are out the window.

Edit: I over-simplified what needs to be done for installation. You really need schematics or to trace the coax I/O circuits to do it properly.

RE: Isolation., posted on October 24, 2009 at 17:24:40
rick_m
Audiophile

Posts: 2539
Location: Oregon
Joined: August 11, 2005
 If I put it internally I'll put them in the DAC as it's roomy. I haven't looked but unless they already use floating RCA's I'd have to float them. I was just thinking to build a little thingy that plugs between the cable and source/sink. It would make A/B'ing easy and I could just build a couple and try them in various places and see what sort of effect they have.

Seems worth looking into: little cost, risk or time and yet, maybe better sound... Another good thing about having it stick out the back is I can easily eyeball the final signals with everything in place and see how it's going, maybe even do a little compensation. Nothing like having a new thing to tweak!

Rick

RE: Isolation., posted on October 25, 2009 at 13:27:47
cfraser
Audiophile

Posts: 2070
Location: Pickering, Ontario
Joined: April 30, 2000
 The thing about putting the isolation externally is it may just not work at all. As you said the I/O jacks must be isolated for a start. They may or may not be in the stock config. At the CDP end, you have to make sure the coax output circuitry can drive the transformer; it should be able to, but some cannot very well. In one case I saw lots of waveform distortion even though the signal was getting through to the DAC; we don't want that, so a scope is almost essential to verify. At the DAC end, chances are it's a single-ended input config even if the jack is isolated, so a slight mod is needed there to completely isolate the cable. Because you are using a transformer input you can toss most of the DC filtering input stuff as there won't be any DC getting into the DAC chips (unless things have gone VERY wrong).

RE: Isolation., posted on October 25, 2009 at 15:22:40
rick_m
Audiophile

Posts: 2539
Location: Oregon
Joined: August 11, 2005
 Hmmm, maybe I'm completely missing the point. I thought the idea was to galvanically isolate the source and DAC with pulse transformers at the cable ends to prevent ground loops and perhaps provide some rejection of HF longitudinal noise and allow a balanced cable to reduce coupling to nearby cables.

Are you also trying to improve something within the gear? I'm with you on the output drive, it would be a PITA to put an active buffer in am external blob. But I do have a scope and can keep an eye on signal integrity.

What transformers have you been using?

Thanks, Rick

RE: Isolation., posted on October 25, 2009 at 15:43:00
cfraser
Audiophile

Posts: 2070
Location: Pickering, Ontario
Joined: April 30, 2000
 It sounds like you have the right idea, I'm not sure what I said is so different... The secondary of the driving transformer will be isolated, and the primary of the receiving transformer too, that's pretty obvious. Thus the cable is now isolated. I guess what you're asking about is the secondary of the receiving transformer? There are options there depending on the DAC's receiver chip. It is preferable to not have the secondary of the transformer ground-referenced IME, though you can if you want; we are dealing with an AC signal and an external DC reference on this input isn't needed (helps keep any local ground noise/crap out of the signal). It is removing this ground-referencing stock "stuff" at the DAC end that is a bit beyond "just isolation".

I have always used Schott, but there may be better. They are small so fit in fairly easily. At the DAC end I'd typically use mini Teflon 75 ohm coax from the input jack to the xfmr, so that I could put the output of the xfmr right near the receiver chip (sometimes mounted the xfmr upside down on top of the rcvr chip). At the CDP end I mount the xfmr on a small piece of PCB and try to install it right at the output jack.

RE: Isolation., posted on October 26, 2009 at 06:34:35
rick_m
Audiophile

Posts: 2539
Location: Oregon
Joined: August 11, 2005
 I forgot to mention that the Schott website is dead. I wonder if they are still in business. Do you recall what disty you got yours from?

TNX Rick

RE: Isolation., posted on October 26, 2009 at 12:08:25
cfraser
Audiophile

Posts: 2070
Location: Pickering, Ontario
Joined: April 30, 2000
 Not a good sign. A few months ago it came up that somebody wanted this mod. I had no transformers left in "stock" so checked digi-key, and was a bit surprised they no longer sold the required model, or any Schott (I forget which). So I said if the person got the transformers themselves, I'd put them in. The Schott site was still up then. There are probably suitable units available from other manufacturers at digi-key.

Edit: I see Schott was acquired by TACNA Magnetics "recently" (this year??). I can't seem to access any Tacna web page either. You'd think there'd be some news if one or the other disappeared, unless it just happened LOL. Maybe it's just a major site update...

RE: Isolation., posted on October 26, 2009 at 06:28:40
rick_m
Audiophile

Posts: 2539
Location: Oregon
Joined: August 11, 2005
 Thanks for the info.

If I understand you correctly it's better to establish the reference close to the receiver rather than at the chassis and that there are usually some passives at the input to "block DC" and they are best removed as they aren't needed with the transformer.

That makes sense to me. I'll take a good hard look at the existing implementation before doing anything. The DAC has multiple inputs selected (by the sound of things) by relays and I don't know where they are in the grand scheme of things, probably at the inputs.

I've got to add a more modern DAC very soon and may wait to see how that goes, probably it makes the most sense to gussy that one up since I may end up not using the older one. One of the nice things about hanging it on the input is it's transferable but I can see that it may not be optimum.

Rick

In my system they make a more significant differance than any other cable, posted on October 23, 2009 at 22:31:56
Julien43
Audiophile

Posts: 2572
Location: Western NC
Joined: May 5, 2001
Contributor
  Since: March 7, 2002
 Speaker and/or IC
Julien
"There's someone in my head, but it's not me"
This is the first cable I could really hear, posted on October 24, 2009 at 07:02:24
Lewis Moon
Audiophile

Posts: 156
Location: Arizona
Joined: November 14, 2008
 ...All my other cables are BJCs but this was the one that really changed things.

I have persuasive (at least, to me) experience with this., posted on October 23, 2009 at 14:50:48
Al Sekela
Audiophile

Posts: 8281
Location: Northern California
Joined: February 18, 2002
 A friend has a system with a separate transport and DAC. He is also a cable designer, and uses the digital coax cable between the transport and DAC as his most sensitive test location.

My pass-time is experimenting with electrical noise filters for the AC. His system exhibited sensitivity to details in my filter recipes UNTIL he fixed something in his cable design. Once that happened, his system basically ignored my filters.

From this I conclude that the digital cable affects the overall performance through its participation in the RF noise sensitivity of the system. The ones and zeroes may get through intact, but the cable may inject noise into other parts of the system, or make it more sensitive to external noise.

Yes, they can., posted on October 22, 2009 at 17:49:37
rick_m
Audiophile

Posts: 2539
Location: Oregon
Joined: August 11, 2005
 If you heard a difference they must be making one. Most everyone else also hears differences if that's any comfort.

The punch line in my words is that 1's and 0's are an abstraction, the data are actually transmitted by time and voltage encoded analog symbols. The combination of a finite bandwidth and variations in cable characteristics, especially dispersion, impedance and dielectric absorption can affect the amount and style of jitter at the receiver's output.

There's a lot of technical information available if you wish to know more. I don't know if you are drinking anything, but that is a rather common practice amongst those that actually work with this stuff...

Rick

RE: Yes, they can., posted on October 23, 2009 at 03:13:53
ramw
Audiophile

Posts: 315
Joined: August 31, 2003
 I still don't get it. If at the output of different digital cables the BER (bit error rate) measures zero, why the audible outcome is different ?

RE: Yes, they can., posted on October 23, 2009 at 05:42:47
rick_m
Audiophile

Posts: 2539
Location: Oregon
Joined: August 11, 2005
 "If at the output of different digital cables the BER (bit error rate) measures zero, why the audible outcome is different?"

Well, in a word: jitter. Do a search on it on this asylum and at Stereophile and you will be inundated with information, views and arguments. Take a look at the link below.

The deal is that SPDIF whether coax or optical is especially sensitive to cable characteristics because the clock is embedded with the data and the finite bandwidth of the cable, transmitter and receiver can make it hard to extract it cleanly. On top of that reflections from small impedance mismatches, especially at the receiver can further muddy the picture.

At the receiver end various techniques are used to extract the clock and clean it up, some more successful than others. In general few manage to truly get it below the amazingly low level of audibility. When any jitter left in the recovered clock hits the D/A it causes IFM (incidental frequency modulation) which even in very small doses causes unnatural sounds that our ears are extremely sensitive to. So the data may be intact, but the clock isn't.

This very issue has probably received more development effort and hair pulling than anything else among manufacturers and users and yet still raises it's ugly head. You are very much in the mainstream.

Regards, Rick

Page processed in 0.183 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2009, Audio Asylum, all rights reserved ] [ Go Green ] [ Take5 ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]