Diy spdif cable.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I can't find many diy designs on the net (many reference a certain brand/model of cable)...so last night I just took 1...tp from cat5 cable..and some 3m shielding tape around it soldered it to bnc on dac side and rca on transport side.

Good enough? Any suggestions for improvement?

I will listen to it soon! (Don't have coax dig out on my cd player)

Also ......brace yourself..this may sound very stupid.....how do you measure that it is 75 ohm?

Regards,
Bas
 
Bas Horneman said:
I can't find my diy designs on the net (many reference a certain brand/model of cable)...so last night I just took 1...tp from cat5 cable..and some 3m shielding tape around it soldered it to bnc on dac side and rca on transport side.

Good enough? Any suggestions for improvement?

I will listen to it soon! (Don't have coax dig out on my cd player)

Also ......brace yourself..this may sound very stupid.....how do you measure that it is 75 ohm?

Regards,
Bas
Hi Bas,
You can simply use double shielded TV/video/satellite 75 Ohm coax. But in order to avoid reflections you have to use 75 Ohm connectors and chassis parts. F-type connectors for satelliet TV would work but are totally hopeless in construction. The most readily available is 75 Ohm BNC.
Just ask for 75 Ohm coax, double shielded in the shop.

:cool:
 
Diy SPDIF Cable

Bas Horneman said:


Thanks Elso...I will...

Don't you think it is funny they use RCA though for co-ax out....when it is not a 75ohm part?

Still that leaves me with the question...how the hell do you measure it to be 75ohm? Either the cable or a RCA plug? (Or both)

Cheers,
Bas
Hi Bas,
A 75 Ohm RCA plug does not excist to my knowledge. And even if the plug is claimed to be 75 Ohm the chassis part is not.
For the cable you have to trust the vendor, but 75 Ohm TV coax is fairly common.
It is extremely curious that a RCA phono connector is the standard as described in table IV of IEC 268-11 for the consumer interface.
Ask Jocko or Fred if you are suspicious of my answer. They are the expert on this subject.

:angel:
 
Re: Diy SPDIF Cable

Elso Kwak said:

Hi Bas,
A 75 Ohm RCA plug does not excist to my knowledge. And even if the plug is claimed to be 75 Ohm the chassis part is not.
:angel:

These guys claims to have one ! ;)

http://www.bluejeanscable.com/articles/75ohmrca.htm
http://www.bluejeanscable.com/pages/technicaldocs/canarercap.pdf

vcomp255.jpg
 
AFAIK, the previous suggestions and considerations about RCA are absolutely correct.

I would just like to add a piece to the picture. The problem is that apparently it is the wrong piece.... it does not fit in the frame!!!!

Just to summarize: at tnt-audio we are testing a bunch of digital interconnects. We tested them under different aspects including audio jitter at the output of a DAC outout, smoothness of the transition in time ("eye" in the oscilloscope without neither too much rounding nor spikes), correct impedence, etc. and obviously sound.

What apparently results at present time is that the best cables (which are MUCH better then all the others) from the sound point of view are the ones with the most regular eye pattern, as expected.

The problem is that they are also the ones causing more jitter and with the impedence more different from 75ohm (one far above and one far below)... And this AFAIK is both in contrast with theory (and eye pattern results too).

The impedence test was performed with a MFJ-259, an antenna analyzer which is able to give the impedence module and phase of any complex system, and the measure sounds correct (the measures perfectly fit technical data for all the cables for which these are available...).

I could think there might be a problem in jitter measures, but even them have been performed so many times and sound so stables that I am not so sure about this explanation. The difefrences in jitter are anyway rather reduced.

Has anyone any idea ?
:xeye:

Next steps could be to measure the wire isolation from external disturbs, but the previous tests leavme me rather upset. Has anyone seen anything similar??

Thank you in advance

Giorgio
 
IT IS NOT 75 OHMS!!!!!!!!!!!!!!!!

How many times do I have to say that???

I don't want to have to 'splain this again.

So Bas.........what you wanted to know.....

2 ways:

Measure it on a TDR,

Or better yet.........

Pay someone like me who has one to do it for you.

Hint, hint....have TDR, will travel.

Jocko
 
Depends who asks!!!!!

Why 75 ohms..........?

Because that is the spec for SPDIF..........0.5 Vp-p, into a 75 ohm load.

Yes....if you are making your own.........you can make it anything that you want. As long as the TX, RXC, and cable impedances are all the same.

As for antennas.............don't get me off on that tangent.

Antennas are not designed to "look" like a specifed impedance over a few decades, as transmission lines are.

Jocko
 
Jocko,
You are right about antennas, I think that almost everyone knows that they are efficient only on a reduced band.

The MFJ-259, however, has a variable test oscillator and is able to measure complex impedence and SWR (and a lot of other related things) on a 1.7-170MHz range. Should be enough, in your view?

What I am not sure of, by the way, is if the SWR sensitivity is high enough. It is clear that 1 would be perfect, but what would be in your view an acceptable SWR, and on which range?

Many years ago, when I worked in telecom on baseband modems with biphase code the required bandwidth was considered twice the bitrate, but in my view this is ridiculously low, if we need precise edges. The eye pattern in fact was very rounded.

By the way, what TDR model have you? HP, isn't it?

Kind regards
Giorgio
 
Correct-a-mundo......

As we say in Texas.

Yes, you would use an impeance bridge when measuring an antenna. And yes......I am old enough to remember the first MJF product..........along with the Omeag-T noise bridge. Almost went to work there after I moved to Texas.

However.........transmission lines are measured with TDRs, which are little more than a glorified step generator built into a very high speed 'scope. If you were stuck using an impedance bridge.......I would look to at least the seventh harmonic to see how low the rho (reflection coefficient) is.

On my fancy-schmancy SPDIF TX and RX boards.......the rho is around 0.02. Return loss greater than 30 dB, in other terms.

My TDR is an H-P......made in '63.....still working........bought military surplus......where it spent the first 30 years of its life at "Dickey Goober AFB", in Missouri.

Jocko
 
F-Type connector

lucpes said:
Short question: anything wrong with F-type connectors/RG59 or 'sattelite' 75 ohm coax besides the not-so-reliable appearance if it's not going to be much fiddling around with it?
Hi lucpes,
Yes they will work as my satelliete receiver is proving everyday.
The construction is sloppy as the center pin is the core of the cable and a pull on the cable will slip it out of the connector.
:bawling:
 
Bas,

Elso's suggestion on TV/SAT cable works for me. But as our dear Jocko said, impedances must be matched all along the way, for start to end, including connectors. Making one's own SPDIF cable to achieve good performance doesn't make sense if you still use RCA connectors. IMHO, just begin to use 75 Ohms BNC connectors at both ends, and connect a 75 Ohms SAT cable between (Jocko prefers lengths > 1m, if I remember...) And make sure your TX and RX are 75 Ohms matched too (check the components on the PCBs). This includes replacing the existing input/output connectors on the equipments by 75 R BNC sockets. Once you have the BNC connectors, feel free to test any make of cable, provided its impedance is 75R... And keep the one you like the most.
If you want to be anal, get your hand on a 75R wideband reference load, calibrate your BNC-cable-BNC setup with the load, and TDR it...
But IMHO, your ears are your best measurement devices, and just try different existing cables. Of course you might want to diy one, but just keep in mind that you'll have to achieve a 75R impedance on a wide frequency range. You can find formulas on the web for various configurations of transmission lines, so take the configuration you find the most suitable for DIYing, and compute parametres to get the right impedance.

Just let your ears speak ;)
 
CheffDeGaar,

My ears don't say anything but if they start I’ am going I'll have to see the doc about some better drugs.... :D

Well, I do listen to Jocko when he talking about the SPDIF interface since he know what he's talking about. I know that difference his attention to detail has made in the sound of my systems. And round two is coming up....Right

Man, you don't have to be anal to test your designs with TDR or a spectrum analyzer. He know how to get the best results because he not guessing. Too much of this golden ear crap leads to poorly design equipment. Don't get me wrong, listening is the final proof. Having said that, if you don't know where to start, then you don’t know where you’re going, then your lost.
:)
 
In theory.....

If all your impedances are amtched, then every cable should sound alike.

Ah....but we don't live in a perfect world.

As for the ears.....they tend to tell the truth, but we don't always know how to understand what it is telling us.

I think that I told this one before.....so I'll ty to be brief.

(Dragon Master ought to find this especially amusing.)

A big name audio guru from the West Coast made one of the silliest SPDIF cables that I have ever seen.

He could not understand how to make a proper build out resistor on th TX end, so he found out......by using his ears...that a 93 ohm cable sounded best. Lo and behold.......his ears then told him that it sounded better still with some goofy ham radio connector......also close to 93 ohms.

Because he was off by about 20 ohms too high on his TX impedance.

Too bad when you used in any other piece of gear that it sounded like dreck.

Only he wouldn't accept that, because:

It worked so well in his system.

True story.

Jocko
 
World is far from perfect...

In theory I pefectly agre, I have been up to few days ago a strong supporter of perfect 75 ohm alignment... Now I doubt of anything...

As per previois e-mail in same thread, at tnt-audio we have been testing a bunch (8) of digital cables, and the acustic tests results were perfectly aligned on various different systems and with different reviewers; by the way, I am told they are aligned with world consideration (I have stopped reading reviews.... unless I am very interested in the unit ;-))). The only problem is that if we assume that MFJ impedance measures are vaguely correct, the best wires from the sound point of view are at FAR, FAR away from 75ohm...

And I have definitely not only no explanation for this, but looking to the live SPDIF signal on my scope, the best cables seem to have the best signa form, which again maks sense but does not match with theory.

One explanation, in my view, could be in fact be that the terminations are not correct, but I have not checked yet...


br
Giorgio
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.