Is coax 75ohm stupid legacy for home use?

[DonDish]

IM building a BO-box for ADI2-pro. It utilizes the DB-9 contact with AES and coax Spdif I/O. Researching this and with the help of others I concluded:

To maintain a 75ohm output from device and through the box would just drain the budget and be totally superfluos.

These are about €8. Perfect pin and coax arrangement for my need. I would buy with solder cups, pricing is same. They also need specially 2.85mm dimensioned 75omh cable which is mostly sold by the roll at €3k for 100m . Buying a few metre is NP.

Now for the issue. Ive been told for a shorther lentgh. Like the usual 1- 2m spdif cable. It doesnt even matter 75ohm or not. So why specify a product for home use when nobody needs it, more than to connect old gear?

A cheap USB cable can send 768khz. If you wire the house with audio cabling you could use network cable and so forth. Coax is just stiff and unagreeable to work with

I really cant see why I would need any coax in my house except for my old DVD player. TV and internet is over optical fiber and I dont have a parabola dish on my roof. Why would someone have RF coax cables exept for legacy? Time to take technology forward?

 

[voodooless]

These are about €8. Perfect pin and coax arrangement for my need. I would buy with solder cups, pricing is same. They also need specially 2.85mm dimensioned 75omh cable which is mostly sold by the roll at €3k for 100m . Buying a few metre is NP.

Why do you think it only fits this exact cable? Why not just buy the adapter from RME (RME BO968)? Also, that part you list doesn't seem to fit the ADI-2:

Now for the issue. Ive been told for a shorther lentgh. Like the usual 1- 2m spdif cable. It doesnt even matter 75ohm or not. So why specify a product for home use when nobody needs it, more than to connect old gear?

Yes, for these short distances, cable impedance is not that critical.

A cheap USB cable can send 768khz. If you wire the house with audio cabling you could use network cable and so forth. Coax is just stiff and unagreeable to work with

USB is a differential connection, making it more reliable, but USB also needs a proper impedance (of 95R) for reliable data transfer. Ethernet is no different, it is differential signaling on 100R impedance.

 

[DonDish]

Why do you think it only fits this exact cable? Why not just buy the adapter from RME? Also, that part you list doesn't seem to fit the ADI-2:

Sure to fit that contact into ADI-2pro you would need a hammer Did you read the post?

Ethernet is no different, it is differential signaling on 100R impedance.

Ethernet cables are cheap, so are switch boxes and contacts. ,and you can sure bend them to lesser than a 50cm radius Read the post again!

Why not just buy the adapter from RME?

The BO-968 is included. I want a weighted box with chassi contacts.

The real question is: If digital audio standard was established today, would it include coaxial transmission?

 

[voodooless]

Did you read the post?

I did, but I really can't see what it exactly is that you want to accomplish here, and why specifically you want to use this particular connector.

The real question is: If digital audio standard was established today, would it include coaxial transmission?

SPDIF, either via Toslink, Coaxial or AES is still the defacto standard for synchronous audio data transfer, and all the derivatives basically use the same infrastructure (like ADAT). So it's going to be around for a while. For asynchronous transfer, things like USB do an excellent job for the low bandwidth connections needed for audio. Then again, a coaxial cable isn't more expensive than any of the other cables. So I'm not sure what this price difference is about. Sure, if you insist on using some unicorn cable that you can only get at 100m lengths... That sounds more like a you problem, rather than a structural one

you can sure bend them to lesser than a 50cm radius

I'm pretty sure you can bend most coaxial cables with a way smaller radius: like 5 to 10 cm should be no issue.

 

[olieb]

I'm pretty sure you can bend most coaxial cables with a way smaller radius: like 5 to 10 cm should be no issue.

Of course, RG179 (75Ω) for example is available with min bending radius of 15mm.
I don't understand the problems of the OP either.

 

[MaxwellsEq]

The real question is: If digital audio standard was established today, would it include coaxial transmission

It might not. Before domestic use of coax S/PDIF, the professional world was using balanced microphone-like cable for AES/EBU (the pro version of S/PDIF) which is obviously not coaxial! I suspect when decisions were being made about what domestic interface to use, the S/PDIF designers picked RCA and coaxial cables because they were the dominant cable you could buy in a HiFi store. Back in the 80s virtually nothing domestic shipped with XLRs, only RCAs.

 

[DonDish]

Of course, RG179 (75Ω) for example is available with min bending radius of 15mm.
I don't understand the problems of the OP either.

Maybe its just a ruse. Anyway if you dont need/utilize the specification for a 2 meter homefi setup, whats the point? Even you can get a coaxial you could twist around a pencil, its certainely not the general coax. There is typically a thick solid isolator between core and shield, and with a foil shield it complicates even more. So from a DIY, installator or a home appliance perspective. This tech is outdated IMHO. Its not a problem in my life like inflation...

I did, but I really can't see what it exactly is that you want to accomplish here, and why specifically you want to use this particular connector.

This part: To maintain a 75ohm output from device and through the box. Its all in the text.

I found the connector while browsing for parts. It looks cool and sophistcated and more importantly maintain 75ohm impedance. So one coaxial line would go: ADI standard db9 -> coax cable -> db9 with 2 coax and 5 pin, male to female in breakout box -> internal coax cable -> Neutrik chassi RCA contact -> output.

The 5 pins are for AES I/O and will be drawn with 110ohm cable.

To this I want to add that I will NOT go further with this design. I will use standard db9 as in ADI2 pro and 75ohm coax cable. The more sane choice so to say.

This is for ADI2 pro and UCXII, base backplate for a stereo bench. Im also doing a four contact, 9.5 inch ADI2pro sized one. Wood is persimmon, rock hard. The bench will have cables drawn through a rubber gasket straight to device. The smaller box will have standard DB9 I/O like ADI2pro.

 

[AnalogSteph]

I suspect when decisions were being made about what domestic interface to use, the S/PDIF designers picked RCA and coaxial cables because they were the dominant cable you could buy in a HiFi store.

Don't think audio, think composite video, which happens to be quite similar in its bandwidth needs, uses 75 ohm coax and would have been super common (VCRs, home computers, ...).

 

[DonDish]

It might not. Before domestic use of coax S/PDIF, the professional world was using balanced microphone-like cable for AES/EBU (the pro version of S/PDIF) which is obviously not coaxial! I suspect when decisions were being made about what domestic interface to use, the S/PDIF designers picked RCA and coaxial cables because they were the dominant cable you could buy in a HiFi store. Back in the 80s virtually nothing domestic shipped with XLRs, only RCAs.

Thank you! You can see the woods even there are trees in the way. This was my suspision to somewhat

 

[DonDish]

I can get nitpicky about details. I had an idea to press the plastic out of the Neutrik, and put in SPDIF spec orange plastic instead. REFUSED! This partly explain why I woud use those 10x the price contacts. Stupid yes! Cool absolutely!

 

[olieb]

Even you can get a coaxial you could twist around a pencil, its certainely not the general coax.

The general audio coax cable connection would be this:

These are sold and provided in the millions.
Should be simple and easy to wrap around a pencil. Maybe they are not exactly 75Ω impedance but for someone who does not care for specifications and is not interested in the performance that comes with them (max length of working connection i.e.) it's probably close enough.
For an electrical two-pole signal cable a coaxial is a/the natural choice and I do not see what you are loosing by having a specification, you can still ignore it if you think it is bogus.

 

[sergeauckland]

Don't think audio, think composite video, which happens to be quite similar in its bandwidth needs, uses 75 ohm coax and would have been super common (VCRs, home computers, ...).

AES-EBU digital audio normally uses 110ohm twisted-pair cable, for the reason that this cable can be used equally well for balanced analogue audio, so when wiring a studio facility, only one type of cable needs to be installed, which can then be used for both digital and analogue audio.

In TV, or mixed audio-video facilities, AES-3 is more commonly used for digital audio, as this is the unbalanced 75 ohm version of AES-EBU. It has the same data as S-PDIF, but works at higher voltages, although is interchangeable with most S-PDIF ins and outs. This is done so that audio and video cabling can used the same 75 ohm cable, interchangeably.

Over the sort of cable lengths used at home, any old bit of cable can be used for digital audio, it's very rugged. Some of you may have seen a demonstration given by Canford Audio when they passed digital audio reliably over wet string. It was wet because it passed through a fish tank with fish swimming about.

I've passed digital audio through my body, holding on to the two ends of a cable with wet hands. It needed somewhat higher voltages to get over the losses, but worked fine for the short time I held on.

S.

 

[voodooless]

This part: To maintain a 75ohm output from device and through the box. Its all in the text.

I found the connector while browsing for parts. It looks cool and sophistcated and more importantly maintain 75ohm impedance. So one coaxial line would go: ADI standard db9 -> coax cable -> db9 with 2 coax and 5 pin, male to female in breakout box -> internal coax cable -> Neutrik chassi RCA contact -> output.

Okay, so you want to make some box to expose the IOs better, and preferably connect it with a single cable, much clearer now!

Remember that AES also needs impedance-matched cabled, 110 Ohm!

Given that all of these have quite a high tolerance, I would suggest using some 100 Ohm shielded ethernet cable for all IOs. Given that it's just a few meters, you can see how well this works. If not, you'll need to add some baluns on either side for the SPDIF IOs. The advantage of using a CAT cable is that you can also use a shielded RJ45 connector. Tiny chip baluns should not cost much and you can hide them in the box or connector easily.

 

[MCH]

Impedance calculators for both PCB traces and cables are available online and the information needed is normally available in the datasheet of the cable. I believe it is mostly a non issue even when one wants to do it right.

 

[DonDish]

Okay, so you want to make some box to expose the IOs better, and preferably connect it with a single cable, much clearer now!

Remember that AES also needs impedance-matched cabled, 110 Ohm!

Given that all of these have quite a high tolerance, I would suggest using some 100 Ohm shielded ethernet cable for all IOs. Given that it's just a few meters, you can see how well this works. If not, you'll need to add some baluns on either side for the SPDIF IOs. The advantage of using a CAT cable is that you can also use a shielded RJ45 connector. Tiny chip baluns should not cost mutch and you can hide them in the box or connector easily.

I really like the idea of going DB9 -> RJ45 with ethernet cable all the way. My solution was to bundle AES 110 ohm and coax from ADI to box. I have been twisting my mind to get from bratwurst to wiener sized cabling.

The thing with impedance transformers and baluns. It would be a leap with my feeble electronics skills. Im taking baby steps into that world so in the future, maybe. Im thinking of more active devices. Blinking leds to show activity and stuff like that. Im just digging my way really slowly with that.

 

[voodooless]

The thing with impedance transformers and baluns. It would be a leap with my feeble electronics skills. Im taking baby steps into that world so in the future, maybe. Im thinking of more active devices. Blinking leds to show activity and stuff like that. Im just digging my way really slowly with that.

Yeah, that's fine. The good thing is that it will probably work without it, so you can ease your way in.

I believe it is mostly a non issue even when one wants to do it right.

For these kinds of frequencies and distances, it mostly is. If you're dealing with radio or very high-speed digital signals, it certainly is not.

 

[DonDish]

Impedance calculators for both PCB traces and cables are available online and the information needed is normally available in the datasheet of the cable. I believe it is mostly a non issue even when one wants to do it right.

I wouldnt know how to take advantage of that. So doing it right for me is also not to mess it up. Setting up a string of 75ohm contacts and cables would be whitin, but applying the math to it would be more than a reach for me.

The bench will have cables drawn through a rubber gasket straight to device. The smaller box will have standard DB9 I/O like ADI2pro.

What do you think of this. 1. straight cable 2. chopped up with a DB9. Both will have correct cable 75 and 110 ohm.

 

[MCH]

I wouldnt know how to take advantage of that. So doing it right for me is also not to mess it up. Setting up a string of 75ohm contacts and cables would be whitin, but applying the math to it would be more than a reach for me.

What I mean is that many cables, even if not advertised as 75ohm impedance, are not very far away. You can use these calculators to select a cable you like or is available to you making sure you are not far off.

 

[sergeauckland]

What I mean is that many cables, even if not advertised as 75ohm impedance, are not very far away. You can use these calculators to select a cable you like or is available to you making sure you are not far off.

Indeed. Any normal coax will be near enough 75 ohms for it not to matter, albeit cables used for RF transmission are commonly 50 ohms, but still near enough not to matter for digital audio domestically.

Similarly, any balanced twisted-pair cable is close enough to 110 ohms for it not to matter. That's why UTP (unscreened twisted pair) or STP (screened twisted pair) cables work fine for digital and analogue line level audio, as does any balanced microphone cable. STP cable works fine even for microphone level signals.

S.

 

[Speedskater]

The higher the frequencies (digital or analog or TV) and the longer the cable, the more important the Radio Frequency Characteristic Impedance of the cable becomes.
For a 2 meter SP/DIF cable, it doesn't matter much.