Length of digital cable?


Picking up where I left off a year ago, somewhere on the net I ran into a recommended length of 18 ft of digital cable between transport and D/A converter. Various lengths were discussed and one source was very confident of the 18 ft length to successfully resolve jitter. And the Beldon 1694A was recommended. For $30+ or whatever I gave it a shot and was surprised by the quality of sound, and I swear on a stack of owners' manuals that this length with this brand of cable resolves jitter. The sound is rock solid gorgeous.

The Beldon lies behind the system mostly in a coil, and I think coils are probably not ideal. I'm wondering if anyone has found a shorter length that works. Al suggested 8", but I was unable to make it work because the components aren't close enough. I know that 1 ft" doesn't work and that 1 meter doesn't work.

Part 2: Does length of ethernet from computer to avr also affect jitter? My sister who is a computer person and not an audio person set up my avr with computer using cat 5. The sound quality is bad. I'd like to get good sound quality with computer as source down the road but need advice on how to get it.
arnettpartners
Arnett - 18 feet is totally unnecessary and you can do a LOT better than Belden 1694A. Many years ago when I still had a set of inexpensive cable from 1964A. I have a giant spool of it. I finally concluded that it cannot be done.

If you want a fantastic sounding cable, get the 4 foot BNC-BNC cable with RCA adapters from my website for $250. It is literally unbeaten. Use the pull-down on this page:
http://www.empiricalaudio.com/products/off-ramp-converter

Here are some reviews:
http://www.audiocircle.com/index.php?topic=115008.msg1208227#msg1208227

http://www.audiocircle.com/index.php?topic=115008.msg1208373#msg1208373

The optimum length of a digital cable is entirely a function of the risetime of the signal on that cable due to the driver in the source device. The analysis is in my white-paper here:

http://www.positive-feedback.com/Issue14/spdif.htm

Steve N.
Empirical Audio
Hi Steve, You were around last year when Al and you and others were pounding into my head what a digital signal is. I thought I could do better than the Beldon but thought I had to be on the right track because suddenly the signal sounds right. Will check out your cable although haven't found it on your website yet. I also noted your suggestions for using computer as source that are listed on your website. Lynne
The BNC-BNC cable is on the pulldown menu at the top of the Off-Ramp page. It is generally sold as an accessory to the Off-Ramp, but can be purchased alone. You can also email me and I can send you a paypal invoice.

Steve N.
Empirical Audio
We found that the shorter the better if the cable is well designed. I use 40cm, it is astoundingly detailed and tonally very accurate..
Ps, you use 40cm of what cable?
Hi Lynne,

I agree with Steve's comments about length, and I don't doubt that his cable is an excellent one. As he indicated above and in his Positive Feedback paper, the optimal length is dependent on signal risetimes and falltimes, which will vary from transport to transport and are usually unspecified. But as indicated in his paper, what probably stands the best chance of being optimal in any given setup is a length of around 1.5 meters (about 5 feet), assuming that a very short length is not practicable.

In addition to signal risetimes and falltimes (the amount of time it takes for the signal to change from its low voltage state to its high voltage state, and vice versa), what length is optimal in any given setup can also be affected by the propagation velocity of the particular cable, by the accuracy of the impedances of the cable and the components it is connecting, and by the amount of electrical noise that may be riding on the signal that is provided by the source, or that may occur as a result of ground loop effects or RFI pickup. So it's all somewhat unpredictable.

Regarding your second question, most likely the length of the ethernet cable is unrelated to the poor sound. The timing of D/A conversion in the AVR (which is where jitter matters) and the timing of data transmission via ethernet are inherently unrelated. It's conceivable that the cable type and/or its length could have a slight effect on sonics by affecting the amount of digital noise that could radiate or couple from the cable to circuit points within the AVR, but I'd expect that kind of effect to be subtle at most.

Best,
-- Al
Hi Al,

I need to order a second cable for a second system, so I will try a 1.5m length. With this 18 ft Beldon, I'm getting a smooth high end like analogue. I can live without stonishing detail if I have to but not the "digital-sounding" high end.

Further, I will work on RFI with tweeks--noise caps etc. Al, I have the strangest phenomenon. My hk990 was upstairs. When I would change sources, the middle LED's on the read-out would go out. When I powered off and on again, the lights were on again. I tried a different circuit but the issue was the same. My contact at hk said he thought it was really strange and had never heard of it but sent me a new amp. De javu all over again. I recently moved that system to the basement. Issue is gone.I haven't tried computer source in the basement yet. Might be improved.

I exagerate a little in regard to the sound quality of the computer source. It's not horrible. My sister doesn't notice the difference. A friend who is somewhat an audiophile does. But I take note of your comments.

Thanks.

Best,

Lynne
Hi Lynne,

I have no thoughts about the HK990 issue.

Regarding the ethernet cable, I can recall a couple of members here whom I consider to be particularly credible reporting that upgrading their ethernet cables to SHIELDED CAT6 resulted in significant improvements. It wouldn't cost much to give it a try. Newegg.com is a good source.

The likelihood of that providing a benefit may depend, though, on the shielding (if any) and other characteristics of the mating jacks.

Best,
-- Al
OK, Al. I'll try it. Thanks. Best, Lynne
HK990 issue was likely a ground-loop issue.

One must take care to use the optimum AC outlets and circuits for digital, particularly when the devices don't have galvanic isolation from each other. Never use two different phases from a panel and best to use the same circuit for all devices.

Steve N.
Empirical Audio
I was using separate circuits for digital and analogue. That is, transport was on separate circuit but was using the amp's onboard D/A converter. I do have all components on the same circuit in the basement.

I would like to try your cable. Will get in touch.

Thanks.

Lynne
"I was using separate circuits for digital and analogue. That is, transport was on separate circuit but was using the amp's onboard D/A converter."

That could easily do it.
ok. This site is screwing up. It won't let me respond.
Update: I bought the 4-ft digital cable with RCA adaptors from Empirical Audio that Steve recommended. Break-in is typical with a rough patch after about 24 hrs. It emerges hrs later with a very broad soundstage and becomes light and airy, sounding more like live music. It's probably 75% broken in now, and the highs are quite controlled; guitars sound steely; sound in general sounds cleansed--really not what I expected. Transients are delightful.

I don't have another high end cable with which to compare; it costs less than the Cardas Lightning 15 and the better Audioquest. No question that it is extremely cost effective in the SQ it produces and strikingly illustrates that cable should be treated equally with other components.
Steve has recommended other products that will get me world class sound, but I probably can't handle world class sound at the moment. I'm too overwhelmed by the SQ I have now.

Thanks to Al and Steve for catching me up. Audiogon is a valuable source thanks to these gentlemen and many more members.
Steve, I understand one can't make a true 75 ohm digital cables with RCA connectors. That's the advantage of using BNC connectors, right - that is the BNC connectors allow for a true 75 ohm cable? So if you need to use RCA adaptors on the BNC cable, doesn't that negate the benefits of a true 75 ohm BNC to BNC cable?
Camb - Its true that RCA connectors are more like 45 ohms, but if you use good 75 ohm BNC adapters like I do, the effect is negligible. If you have BNC on either the output or input device, you can use the native BNC which is better.

The benefit is really the cable itself and how it is terminated to the BNC and shielded. It cannot be just any 75 ohm BNC. The BNC must be designed specifically for that particular cable to avoid an impedance change at the cable termination.

If you terminate to an RCA, it is very difficult to avoid a big impedance discontinuity because it is never a good match for the cable.

Steve N.
Empirical Audio
Thanks Steve,

I wish more manufacturers offered a BNC option. RCA still seems to be the norm, unfortunately.

Cam
Although BNC would theoretically be preferable to RCA connection, there is actually next to no audible difference between an spdif using either. Steve mentions that the BNC must be designed for a particular cable, which frankly is a dubious statement. The connectors 'are what they are', and the cable that exists to connect them together should offer as close to the 75ohm Zo as is possible, which is not difficult. If using RCA connectors, the shorter the better, i.e. Furutech FP126, and designing the cable to terminate at 75ohms exactly can make a perfect match. Using adaptors also makes little difference, both on a scope and audibly. Saying that, I do agree that more equipment manufacturers could do with offering them.
Cable length should not be an issue if the said cable is an exact impedance match for the output and input, as refections should not occur. Cable length is an issue in mismatched impedances though. As I mentioned earlier, I use 40cm of my own design, which offers one of the cleanest, detailed and most spacious and unfatiguing sounds we have heard from many digital cables, which are in fact just high speed analog signals really anyway. It is most likely that there may be system limitations that would mask the subtleties of the true sonic differences between connection options.

Best Rgds,
Paul.
Thanks Paul!

I've read that 1.5 meters is the optimal length to minimize the impact of reflections in RCA terminated digital coax cables. Is this true? Or is shorter always better?
"Although BNC would theoretically be preferable to RCA connection, there is actually next to no audible difference between an spdif using either. Steve mentions that the BNC must be designed for a particular cable, which frankly is a dubious statement. The connectors 'are what they are', and the cable that exists to connect them together should offer as close to the 75ohm Zo as is possible, which is not difficult."

Whether you hear differences will depend on your source and your system. If the source is truly low-jitter and has fast risetimes to insure that the low-jitter is maintained through the receiver, the BNC will make a BIG difference.

Termination of the coax to the RIGHT BNC connector is critical. Even a millimeter difference in diameter inside the connector will change the impedance. Many so-called 75 ohm connectors are also not 75 ohms exactly.

"If using RCA connectors, the shorter the better, i.e. Furutech FP126, and designing the cable to terminate at 75ohms exactly can make a perfect match. Using adaptors also makes little difference, both on a scope and audibly. Saying that, I do agree that more equipment manufacturers could do with offering them.
Cable length should not be an issue if the said cable is an exact impedance match for the output and input, as refections should not occur."

Also untrue. There will ALWAYS be reflections, even in the best of cables and connectors. The reason to have a 1.5m minimum length as recommended in my paper is to insure that when those reflections get back to the receiver, that it is not when the receiver is detecting the edge. Its just common-sense.

You obviously did not read my white-paper, which has been the standard for S/DIF cable length for many years now.

Steve N.
Empirical Audio
Hi Camb,

Thank you, it all depends on the cable design, termination, and transmission and receiver circuits. In our tests, the audible results are superior with a specific design over short lengths, which also minimises potential interference from all the mass of power cables etc usually found behind a system. What would be the shortest you could use?

Rgds,
Paul.
Hi Paul,

I would need 1.5 to 2 feet to connect my Meridian streaming gear to my DAC.

I've got a 1.5 meter MIT Oracle digital coax RCA cable now.

Cheers,

Cam
The 1.5m cable length is a good guideline. Shorter cables can be problematic depending on the signal risetime. The 1.5m length insures that any jitter causing reflections occur too late to have any impact. Really short, say 6 inches can also be good, but between 1 foot and 1.5m, you are taking a risk IMO. Dont spare the change.

Adapters do make a difference, but again depending on the signal risetime. The faster the risetime, the better the jitter result. Here are some plots showing jitter with different cables and adapters:

http://www.audioasylum.com/forums/pcaudio/messages/13/132743.html

I have been doing this a LONG time and I know what I'm talking about.

Steve N.
Empirical Audio
Cam,

It would be an interesting shoot out, the IC you have has a reputation for excellence!

Rgds
Paul.
The best performing and also the best price coax cable is my 4-foot BNC-BNC cable with RCA adapters. Beats everything in customer shootouts, at any price. Silver with expanded Teflon dielectric and terminated to low-loss true 75 ohm BNC connectors. A steal at $250

Steve N.
Empirical Audio