Cables... advice on what digital cables to use

[Barwell1992]

Hi everyone,

Looking for some advice on digital audio cables as there’s quite a lot of conflicting info.

So what is the consensus on Coaxial vs Toslink, specifically QED performance Coax vs either Qed performance optical or Fisual Hollywood Glass optical.

Or would I be better of getting a better Coaxial cable?

Source: Tidal upto 24/48Khz over UPnP or CD
Renderer: Cambridge Audio CXUHD
DAC: Beresford Caiman SEG 30 Inc Dorado

2nd question

Toslink to mini optical adapters, are they a bad idea (for my CCA)?

Thanks

[Light Dependant Resistor]

Hi Dan

Rather than the brand of coax, the important parameter is the cables stated impedance.
The characteristic impedance of the cable is created by having uniformity from inner conductor to outer shield
with then inner conductor at exact spacing from the outer shield. We can see straight away a RCA at either end is
not obeying this essential requirement. Rather digital links must be used with a BNC connector. But as usual
the marketplace wins by offering basically silly RCA's for this purpose.
.
Coax is in the case of digital links is forced to usually 75 ohms by a fixed resistor, so the cable appears
as having the cables characteristics impedance. If the resistor was not there the characteristic impedance
is then tied to the length of cable requiring knowledge of wavelength. Whilst 75 ohms is common there
is good reason to rebuild your equipment to use 50 ohm which is the broadcast standard. The lowering
of impedance also lowers propensity for noise intrusion.

In this regard coax wins hands down over toslink as it is being forced to be a transmission line
( although toslinks works and is convenient ) and obeys well known principles https://en.wikipedia.org/wiki/Transmission_line

In the circumstance of a digital link it is usually best to make your own, using high quality coax cable.
which leads to converting your equipment ,or indeed buying equipment with a BNC already fitted.
Which will be as rare as the proverbial you name it... so out with the soldering iron if you want it done properly.

[Discopants]

First i would ask which input gives the best SQ on the DAC, coax or toslink?

Then chose a cable , looks like Beresford make a coaxial cable, this might be a good option for synergy.


Sent from my iPad using Tapatalk

[Pharos]

"there’s quite a lot of conflicting info"

What, in Hi-Fi choices, surely not so, I though it was all resolved with no conflicting views.

[struth]

Hi everyone,

Looking for some advice on digital audio cables as there’s quite a lot of conflicting info.

So what is the consensus on Coaxial vs Toslink, specifically QED performance Coax vs either Qed performance optical or Fisual Hollywood Glass optical.

Or would I be better of getting a better Coaxial cable?

Source: Tidal upto 24/48Khz over UPnP or CD
Renderer: Cambridge Audio CXUHD
DAC: Beresford Caiman SEG 30 Inc Dorado

2nd question

Toslink to mini optical adapters, are they a bad idea (for my CCA)?

Thanks

if done well i doubt you will notice much difference. i use all 3 (usb too) and dont notice a lot of difference, if any.

[Barwell1992]

Thanks for the reply’s so far !




In the circumstance of a digital link it is usually best to make your own, using high quality coax cable.
which leads to converting your equipment ,or indeed buying equipment with a BNC already fitted.
Which will be as rare as the proverbial you name it... so out with the soldering iron if you want it done properly.

Very interesting, if my equipment were older it would defiantly be on the cards to crack out the iron, however not sure I fancy doing it in a 2 month old £700 Bit of kit.

If it had to be Toslink VS RCA would you sway towards Toslink ?


First i would ask which input gives the best SQ on the DAC, coax or toslink?

Then chose a cable , looks like Beresford make a coaxial cable, this might be a good option for synergy.


Sent from my iPad using Tapatalk

I’ve messaged Mr Beresford him self about something else so I’ll also ask these questions.


"there’s quite a lot of conflicting info"

What, in Hi-Fi choices, surely not so, I though it was all resolved with no conflicting views.

:rolleyes: maybe I should just buy what the magazines say I should buy :eek:

[Barwell1992]

if done well i doubt you will notice much difference. i use all 3 (usb too) and dont notice a lot of difference, if any.

To be honest that’s what I’ve always thought. Just sanity checking really, I know swapping the plastic Toslink cable for glass is an easy win as it seems clear cut that glass is better.

[struth]

To be honest that’s what I’ve always thought. Just sanity checking really, I know swapping the plastic Toslink cable for glass is an easy win as it seems clear cut that glass is better.

yup, i used a fisual glass one for a while. currently not tho as ive misplaced it :eyebrows: a well made one of any type is fine. its as much to do with how well the ends are cut n polished

[Light Dependant Resistor]

Yes given the industries wrong choice with using RCA connectors vs BNC the correct choice to terminate digital cables, Toslink
has much appeal. which begs the question is there anything that can be done to improve it. Possibly ensuring no radiated red frequency light
escapes at the junction points. The junction points too might be improved by focussing the light beam - to a degree this is already done,
via the fittings, but I think could be improved upon with encapsulation. Also the receiving and sending conversion I do not think is optimized
for light transfer,rather just borrowed from the coax way of doing digital transfer. Might be a future topic in the DIY section. to discuss.

[Discopants]

Yes given the industries wrong choice with using RCA connectors vs BNC the correct choice to terminate digital cables, Toslink
has much appeal. which begs the question is there anything that can be done to improve it. Possibly ensuring no radiated red frequency light
escapes at the junction points. The junction points too might be improved by focussing the light beam - to a degree this is already done,
via the fittings, but I think could be improved upon with encapsulation. Also the receiving and sending conversion I do not think is optimized
for light transfer,rather just borrowed from the coax way of doing digital transfer. Might be a future topic in the DIY section. to discuss.

Ive got a chord mojo , generally accepted sound quality of inputs in the mojo owning community is USB > coax> optical and this is backed up by a hierarchy of sampling rates supported. So I don’t think a single rule can be applied to all circumstances.


Sent from my iPad using Tapatalk

[Ianmac]

Chris
75 Ohm RCA connectors and sockets have been available in the UK for some years. I bought mine from Canford Audio and my standard BNC crimp worked with them. I think they are offered on Ebay.
I agree 75 Ohm BNC is the best solution if properly engineered.
Most of the issues come from the various manufacturers not adhering to an audio industry standard- none exists- and therefore a mixed bag of RCA sockets appearing for SPDIF inputs.
Making it a field day for the cable vendors !

[Barwell1992]

Just as a bit of an update to this..

I decided to make a punt on the coveted Fisual Hollywood optical cable in 0.5m long, linked it up with my Cambridge Audio CXUHD (using it as UPnP renderer) and Beresford SEG 30 DAC and gave it a run.... If I'm honest wasn't sold by the sound.. in fact my chrome cast audio over a QED performance optical with adapter sounded better and it defiantly sounded better previously with my QED performance coax, so i had a good look and it seems that the poor sound quality is most likely down to a fairly crap Toslink connector on the back on the Cambridge, its just not a good fit and allow's quite a lot of movement, not helped by the stiff nature of the fisual, bit of a shame given the Cambridge its a £700 unit.

So that leads me to the next cable, a blue jeans coax cable, this cable uses Canare RCAP crimped on plugs,these are a true 75 Ohm rca plug mated to a Belden 1694A 75ohm cable. So knowing exactly what i was getting 12 hours and £20 later its running in my system, well so far all i can say is its a great cable for the money all the detail is back that was missing from the Fisual and it feels very well made, its not the best looking cable but seems like it will do exactly what its supposed to, the fisual has been moved over to chrome cast duty and sounds good but to me it sounds no better than my plastic QED cable..

I'll update again once i've got some more hours on the cable.

Blue jeans
https://www.bluejeanscable.com/store/digital-audio/index.htm

Connectors
https://www.bluejeanscable.com/pages/technicaldocs/canarercap.pdf

cable
https://www.bluejeanscable.com/pages/technicaldocs/1694tech.htm

25903

[NRG]

The dimensions of an RCA plug mean it can’t meet 75 ohm impedance. Somebody measured the canare plugs on diy audio ages ago and they didn’t meet 75ohms...also if it’s plugged into a standard RCA socket then theres no chance of it being 75ohm. Then there’s the consideration of the pcb traces and termination of the DAC itself, does that meet 75ohm. It’s a folly to chase this requirement.

[Barwell1992]

The dimensions of an RCA plug mean it can’t meet 75 ohm impedance. Somebody measured the canare plugs on diy audio ages ago and they didn’t meet 75ohms...also if it’s plugged into a standard RCA socket then theres no chance of it being 75ohm. Then there’s the consideration of the pcb traces and termination of the DAC itself, does that meet 75ohm. It’s a folly to chase this requirement.

Very interesting, I guess the closer to 75ohm the better. Beresford recommends 75ohm for the DAC but I can’t be sure that it meets 75ohm’s.

[Barry]

The dimensions of an RCA plug mean it can’t meet 75 ohm impedance. Somebody measured the canare plugs on diy audio ages ago and they didn’t meet 75ohms...also if it’s plugged into a standard RCA socket then theres no chance of it being 75ohm. Then there’s the consideration of the pcb traces and termination of the DAC itself, does that meet 75ohm. It’s a folly to chase this requirement.

Agreed, the RCA phono connector cannot have a characteristic impedance of any more than 57 Ohm (assuming the two conductors of the coaxial pair are air spaced). With a typical dielectric support for the centre conductor, the characteristic impedance will in practice lay somewhere between 30 - 40 Ohm.

It has to be remembered the RCA phono was never intended to be a precision connector, unlike BNC, CAMAC, SMA, SMB, etc. connectors. So one wonders why Sony/Philips chose the RCA connector for their SPDIF. Well the reason was the RCA phono is a cheap connector that has unfortunately been adopted as the hi-fi industry standard. The saving grace is that the connection is electrically 'short'. So the electrical mismatch between 75 Ohm and the ~ 35 Ohm of the RCA connection will have a voltage reflection coefficient of -0.36, whereas the mismatch between ~35 Ohm and 75 Ohm will have a voltage reflection coefficient of +0.36. Since these two mismatches are electrically close to one another (~1.5cm at the most), the two mismatches largely cancel one another.

Let us assume the rise time for the sampling rate is 25ns, that is a signal of 10 MHz, the overall mismatch reflection coefficient can be shown to be 0.015 (or -36dB return loss). Thus despite the large impedance mismatch, the use of an RCA 'phono' connector is not quite as bad as it might appear.

Yes, a true 75 Ohm BNC would be ideal, but the 75 Ohm version is easily confused with the 50 Ohm version, and mating a 50 Ohm BNC plug to a 75 Ohm BNC socket will damage the latter. And regarding those RCA phono connectors sold as being designed to have a 75Ohm impedance, once they are mated with a standard RCA phono connector, the impedance mismatch is just the same (and just as bad) as mentioned above.

However the most important thing is to use coaxial cable that has a 75 Ohm characteristic impedance. Regardless of connectors used, this cable will 'sound' better than say 50Ohm impedance cable or any other audio interconnect cable. There are several 75 Ohm cables that one can use: RG59/CU, URM70, Trompeter TC75-2 etc.

[Light Dependant Resistor]

The dimensions of an RCA plug mean it can’t meet 75 ohm impedance. Somebody measured the canare plugs on diy audio ages ago and they didn’t meet 75ohms...also if it’s plugged into a standard RCA socket then theres no chance of it being 75ohm. Then there’s the consideration of the pcb traces and termination of the DAC itself, does that meet 75ohm. It’s a folly to chase this requirement.

No its not a folly at all, rather it is a folly NOT to. What you get by not abiding by the characteristic impedance of the cable including the circuit board traces is reflections of the required signal interfering with the same signal
known as standing wave ratio (SWR). Whilst some manufacturers cannot be bothered by providing RCA terminations, rather than BNC, this does not mean the problem magically disappears.rather it just remains
to detract from ideal transfer.

[Light Dependant Resistor]

Reflections on cables caused by not adhering strictly to 75 ohms then causes the reflection voltage to superimpose itself on the needed
transition, in turn causing jitter.

[Ianmac]

The dimensions of an RCA plug mean it can’t meet 75 ohm impedance. Somebody measured the canare plugs on diy audio ages ago and they didn’t meet 75ohms...also if it’s plugged into a standard RCA socket then theres no chance of it being 75ohm. Then there’s the consideration of the pcb traces and termination of the DAC itself, does that meet 75ohm. It’s a folly to chase this requirement.

Im surprised to read that a company with Canare's reputation are not making a product that meets their published specification.
I use Benchmark Dacs and their specification for input impedance on the RCA type connectors used on the DAC2 is 75 Ohms. From past experience, I have a great respect for Benchmarks engineering standards and would consider it unlikely that they would publish 75 Ohm input Z if it wasnt.
Now retired and no longer have access to the correct equipment needed to accurately measure these Zs but I suggest caution re believing everything one reads on DIY audio.

It might be interesting to look at the various quality DACs in the Hi Fi market and see how many actually use BNC.

Dan
So that leads me to the next cable, a blue jeans coax cable, this cable uses Canare RCAP crimped on plugs,these are a true 75 Ohm rca plug mated to a Belden 1694A 75ohm cable. So knowing exactly what i was getting 12 hours and £20 later its running in my system, well so far all i can say is its a great cable for the money all the detail is back that was missing from the Fisual and it feels very well made, its not the best looking cable but seems like it will do exactly what its supposed to, the fisual has been moved over to chrome cast duty and sounds good but to me it sounds no better than my plastic QED cable..

Good to read that progress is being made

[Pharos]

These posts are very interesting, and largely right in principle, (it was my specialist subject), but is this a significant factor in these contexts of audio?

There have been numerous articles on Z matches over the years, and in particular, although considering audio frequencies, ones on power amp into speakers comes to mind, most of which state that it is not significant.

[struth]

Ive got a Mogami 2964 Digital Coaxial Cable, 75 Ohm S/PDIF | Canare Gold RCA version.

[NRG]

I refer anybody disagreeing with me to Barry’s detailed post above. Chris, yes it’s not a folly for the designer but it is for the consumer who has no control over such design aspects. The measurements on DIY audio where done by an established RF engineer along with the TDR results. Also, there was something about the canare plugs having to use the matching RCA socket to meet 75ohm but if that is really necessary then you may as well use the cheaper and proven BNC connectors instead.

[struth]

This is the one I got.. Nicely made etc. https://uploads.tapatalk-cdn.com/20190427/adf824f129b10c568d94e8cf322f1b04.jpg

[Barry]

I like the connectors - they appear to have been crimped onto the outer braiding/shielding. This is important, as the interconnect has to be treated as being an RF interconnect. So, no place for crappy 'bullet' plugs here.

[struth]

I like the connectors - they appear to have been crimped onto the outer braiding/shielding. This is important, as the interconnect has to be treated as being an RF interconnect. So, no place for crappy 'bullet' plugs here.Yup they are crimped on. Quite thin cable but it is nice

[Barry]

It doesn't matter they are thin. True, thicker cables will have a lower RF loss, but as we are only talking a metre or two here, loss is largely unimportant.

[struth]

A better look at size of gauge and of the plug.
Pi to vena and sounds better than last one. Much smoother imo. Timing flows betterhttps://uploads.tapatalk-cdn.com/20190428/17440035b3a29d561f23ab2d1de16547.jpg

decay is better too..

[goraman]

If this wont do it , nothing will. Read the specs in description.
https://www.ebay.com/itm/CANARE-L-5CFB-Coaxial-Digital-Audio-Interconnect-Cable-1m-Free-Ship/401759560137?_trkparms=aid%3D111001%26algo%3DREC.S EED%26ao%3D1%26asc%3D20160908105057%26meid%3D380de 21c45e341409abef4958622608b%26pid%3D100675%26rk%3D 3%26rkt%3D15%26sd%3D362625703394%26itm%3D401759560 137&_trksid=p2481888.c100675.m4236&_trkparms=pageci%3A3683962c-6ee5-11e9-a789-74dbd1809929%7Cparentrq%3A8604fd9e16a0ac1c51a7e759 ffed8e74%7Ciid%3A1

[Barry]

If this wont do it , nothing will. Read the specs in description.
https://www.ebay.com/itm/CANARE-L-5CFB-Coaxial-Digital-Audio-Interconnect-Cable-1m-Free-Ship/401759560137?_trkparms=aid%3D111001%26algo%3DREC.S EED%26ao%3D1%26asc%3D20160908105057%26meid%3D380de 21c45e341409abef4958622608b%26pid%3D100675%26rk%3D 3%26rkt%3D15%26sd%3D362625703394%26itm%3D401759560 137&_trksid=p2481888.c100675.m4236&_trkparms=pageci%3A3683962c-6ee5-11e9-a789-74dbd1809929%7Cparentrq%3A8604fd9e16a0ac1c51a7e759 ffed8e74%7Ciid%3A1

It's physically impossible for an RCA 'phono' plug to have a characteristic impedance of 75Ohm. Disappointing to read a company such as Canare propagating the myth of 75Ohm RCAs; crimped or soldered, it makes no difference. But as I commented on above, the effect of the mismatch is relatively small (hence Canare's claim of a mismatch VSWR of < 1.1).

[goraman]

I was pointing to the cable specs as far as bandwith.
But it is true all round connectors will be under 75 ohms.
Amphonal makes some of the best commercial connectors for coax , I wish they did more for audio.
There BNC stuff is really good.

[Fourlegsgood]

My endeavours in this make me think that all digital cables get the signal to the other end completely unaltered as otherwise we would hear clicks or pops. This applies to usb, bnc or optical. The difference is in whether they also transmit unwanted RF noise overlaid on the digital signal and which will eventually get into the dac analogue stagecand and cause intermodulation distortion. To avoid this, optical is best in theory because the optical only transmits the digital signal and not any overlaid RF noise.

On this basis, the difference in sound of different digital cables comes down to how they handle RF noise. More noise equals a bright or false sense of detail. Less RF noise is indicated by a softer or darker sound.

Cables selected because they have more detail are almost certainly transmitting more RF noise.

So, my suggestion is to select whichever digital cable has the darkest, least bright presentation. That one is then transmitting nearest to the bare digital signal and therefore is closest to the original musical file.

[Stratmangler]

I was pointing to the cable specs as far as bandwith.
But it is true all round connectors will be under 75 ohms.
Amphonal makes some of the best commercial connectors for coax , I wish they did more for audio.
There BNC stuff is really good.

There's nothing to stop you using BNC coax connectors in place of RCA connectors.
The oft quoted 75 ohms is of no relevance - it's not DC resistance. It's 75 ohms impedance, so only of relevance when you're dealing with frequencies far above what's considered to be those of human hearing range.
Radio frequencies.

BNC connectors would give a good, tight connection on interconnects.
A pain in the backside to fit, but a decent audio connector nonetheless.

[JohnNB]

I've been following this interesting thread, but, as usual, got lost in the technical stuff a few posts in!
It did get me thinking about optical vs coax though. I use two digital cables into my Roksan DAC; a Atlas coax from the cdp (£40) and a cheap optical (£5)from the DAB tuner. I decided on this set up after listening and swapping cables around and felt that the coax was - as mentioned in this discussion - darker sounding, so chose it for the cpd, which I listen to most. I didn't even consider buying a quality optical thinking they're all much of a muchness, but after reading the discussion maybe I should. Atlas do a toslink for about £40. Getting one of these would also make all my inter-connects the same model of the same brand. My questions are; is it worth spending the cash on presumably a better quality optical lead, and does it benefit overall SQ of the set up if all interconnects are by the same manufacturer, and the same model?

[Light Dependant Resistor]

There's nothing to stop you using BNC coax connectors in place of RCA connectors.
The oft quoted 75 ohms is of no relevance - it's not DC resistance. It's 75 ohms impedance, so only of relevance when you're dealing with frequencies far above what's considered to be those of human hearing range.
Radio frequencies.

BNC connectors would give a good, tight connection on interconnects.
A pain in the backside to fit, but a decent audio connector nonetheless.

As I understand the question is " Advice on what digital cable to use" Coax cable or optical cable remain the correct choices for consumers using digital cable
http://audioworkshop.org/downloads/AES_EBU_SPDIF_DIGITAL_INTERFACEaes93.pdf

Deviation from impedance of 75 ohms including cable interfaces simply invites error known as jitter, as you are then moving away from the needed properties of coax
BNC you are right happens to be a well designed coax cable interface.