Just wondering if this could work well.
Assuming I have a DIY DAC with a balanced output, and I wired it to send the output to 2 x RCA sockets. At the other end is a DIY amplifier with the same sockets and wiring.
GND to sleeve of both RCA's
+V to tip of one RCA
-V to tip of other RCA
[x2 - so there would be 4 RCA sockets for balanced stereo]
Any good?
Assuming I have a DIY DAC with a balanced output, and I wired it to send the output to 2 x RCA sockets. At the other end is a DIY amplifier with the same sockets and wiring.
GND to sleeve of both RCA's
+V to tip of one RCA
-V to tip of other RCA
[x2 - so there would be 4 RCA sockets for balanced stereo]
Any good?
Last edited:
Unconventional, but should not cause any particular issues, just make sure all RCA jacks are grounded to the same ground point.
I was thinking of leaving the output end of the shield disconnected. Both devices would share the same mains ground.
If that would help, I'm just guessing here.
You would then need excellent CMRR. Why are you using RCA with balanced connections? And why a pair of coax instead of a twisted pair?
Hopefully this won't go OT as I'm sure there are others with differing opinions on this but the main reasons are:
1. I cannot find any solid siver conductor XLR plugs or sockets for less than thousands of $'s. There are solid silver RCA plugs and sockets available for a lot less.
2. I cannot find a solid silver-cored shielded twisted pair available on the market for less than about $100/m. I am currently using a solid silver-cored coax for single ended audio that costs a lot less than that (and sounds great I might add).
Here's the idea:
I figure there might need to be some combination of resistors and/or capacitors between A, B and GND on the amplifier side? I'm no good at design, only theory and implementation so thanks for any help 🙂
perhaps you can try this out with two pairs of cheap rca - cables (equal lenght, sitting tightly together...... with short pieces of heatshrink or so ).
FYI, James Bongiorno used the two RCA sockets in a balanced cable connection in his Sumo line of amplifiers (circa. 1980's), preamp to power amp. This was before consumer electronics has balanced XLR connectors and associated circuitry.
The shield or in this case the two shields are not audio ground. They should only be connected to the chassis's. Not being part of the audio circuit, ground loops are not a mater of concern.
As "mjf" symmetry is necessary for a good balanced analog interconnect. So both cables (equal lenght, sitting tightly together...).
As "mjf" symmetry is necessary for a good balanced analog interconnect. So both cables (equal lenght, sitting tightly together...).
XLR to RCA Balanced cables
You are indeed correct the signal+, and signal - are the center conductors of each individual coax, and the shields are chassis ground. I tend to agree with Jan Didden, that the use of a shielded twisted pair with Cannon XLR connectors work best for balanced connections. I was just pointing out that the use of two coaxial cables with RCA connectors was used in a production amplifier back in the '70's.
A good reference for connector wiring can be found on Rane's pro-sound website under tech notes section. They also explain pin 1 grounding in a balanced interconnect.
You are indeed correct the signal+, and signal - are the center conductors of each individual coax, and the shields are chassis ground. I tend to agree with Jan Didden, that the use of a shielded twisted pair with Cannon XLR connectors work best for balanced connections. I was just pointing out that the use of two coaxial cables with RCA connectors was used in a production amplifier back in the '70's.
A good reference for connector wiring can be found on Rane's pro-sound website under tech notes section. They also explain pin 1 grounding in a balanced interconnect.
Good to know it was done before, on consumer equipment. It can't be that bad then!
Rane's wiring notes - Sound System Interconnection
Very interesting and useful.
So refering to my diagram, A to B to GND, and GND's within each unit is both the chassis ground and the audio ground.
I guess one option would be to remove the polymer shield from each coax, then twist them together so each shield is touching throughout the length of the cable. Then heatshrink the lot and seal the ends. Probably pointless though if ground loops aren't an issue.
Rane's wiring notes - Sound System Interconnection
Very interesting and useful.
So refering to my diagram, A to B to GND, and GND's within each unit is both the chassis ground and the audio ground.
I guess one option would be to remove the polymer shield from each coax, then twist them together so each shield is touching throughout the length of the cable. Then heatshrink the lot and seal the ends. Probably pointless though if ground loops aren't an issue.
Just a question, but would these signal still be trully balanced? as with this configeration both signals are referenced to the GND for thier return currents more than to each other, as they would in a twisted pair cable?
AFAIK a balanced signal is irrespective of ground. Only the differential between SIG+ and SIG- is referenced.
I've since found some affordable solid silver twisted pair shielded. However still no affordable solid silver XLR's. Maybe there's another solid silver 3 pin connector I could use. If I could I'd do away with them altogether - after all, for signal quality, the best connector is no connector at all.
I've since found some affordable solid silver twisted pair shielded. However still no affordable solid silver XLR's. Maybe there's another solid silver 3 pin connector I could use. If I could I'd do away with them altogether - after all, for signal quality, the best connector is no connector at all.
For digital LVDS the definition of a differential pair, is having the same impedance to ground, but are alwqays routed as a tightly coupled pair, or as twisted pair in cables such as ethernet.
With both your cables being shielded, the lack of twisted pair and common mode rejection wont be as critical. But I am curious about the return currents for the analogue signals. With LVDS the return currents are catered for by the close coupling of the signal traces or wires when going off board. My thought is would the seperated signals use the shield of the co-ax for the return currents.
With both your cables being shielded, the lack of twisted pair and common mode rejection wont be as critical. But I am curious about the return currents for the analogue signals. With LVDS the return currents are catered for by the close coupling of the signal traces or wires when going off board. My thought is would the seperated signals use the shield of the co-ax for the return currents.
Sounds like you know much more about this than I do, but yes I would assume so. Both chassis and signal grounds are connected in the same way normal XLR's, it's just in my case there are 2 parallel or twisted screens making the connection, rather than just one screen, just one core, or just one screen and core (depending how the XLR is wired).
Only digital, and as I learned recently some assumptions about low frequency analogue over interconnects are not as straight forward, and things that are
I am interested as I use both single ended and balanced (Behringer cross over), with some Ob speakers. I want to go all balanced, but my midrange amps are mono block class A valve (with solid state for the bass a Audiolab 8000A).
I do think with the distances involved etc that it wont make a bit of difference. Also as I do PCb design as a job, I have an interest in how signals get from a to b, and even though I have learned quite a lot over the years, the more I look into the subject, the less I find I understand, and assumtions and basic theory that is taught dosn't cover the full story.
courious, wont silver contacts tarnish easily, or is it not a problem?
I am interested as I use both single ended and balanced (Behringer cross over), with some Ob speakers. I want to go all balanced, but my midrange amps are mono block class A valve (with solid state for the bass a Audiolab 8000A).
I do think with the distances involved etc that it wont make a bit of difference. Also as I do PCb design as a job, I have an interest in how signals get from a to b, and even though I have learned quite a lot over the years, the more I look into the subject, the less I find I understand, and assumtions and basic theory that is taught dosn't cover the full story.
courious, wont silver contacts tarnish easily, or is it not a problem?
If any part of your system isn't balanced then going balanced won't make a difference, except for the fact you'll be using a different cable so that might sound different.
I think we're all learning about signal transfer, even those who know everything are finding out new things. There are just so many ways to transmit and recieve data it's incredible.
Funny you should ask about tarnished silver. Apparently even black-as-ink silver oxide is more conductive than copper!
Silver is 5-6% more conductive (less resistance) than pure OFC copper.
I think we're all learning about signal transfer, even those who know everything are finding out new things. There are just so many ways to transmit and recieve data it's incredible.
Funny you should ask about tarnished silver. Apparently even black-as-ink silver oxide is more conductive than copper!
Silver is 5-6% more conductive (less resistance) than pure OFC copper.
Yup...
Silver - Wikipedia, the free encyclopedia
Electrical resistivity and conductivity - Wikipedia, the free encyclopedia
And a subjective reference... Read from 'About Audio Note™ Silver Wires & the use of Silver in Audio Products'
Audio Note
Silver - Wikipedia, the free encyclopedia
Electrical resistivity and conductivity - Wikipedia, the free encyclopedia
And a subjective reference... Read from 'About Audio Note™ Silver Wires & the use of Silver in Audio Products'
Audio Note
- Status
- Not open for further replies.