Making my own RCA -> XLR cables

[Jdunk54nl]

A couple of other notes: The capacitance of the Mogami is far lower than the Canare (even Mogami star-quad vs Canare star-quad, but more so on the 2549 single conductors). Star quad isn't going to have an advantage for your use, and it's easy to solder connectors with the two-conductor plus shield than the quad. (XLRs aren't so bad with quad, but with phone and RCA plugs quad is tight to work with.)

PS—I like the idea of the adapters, to a degree, but they most often create a bad leverage point for stressing the connector/cable. So it depends on what's going on behind the unit, whether it's acceptable.

I will just go with the mogami 2549. Seems like it will work for what I want and we are talking like a $5 difference in cost.

I will also order the Rean RCA's

Are those neutrik XLR's the "right ones" They have a bunch of different 3 pin XLR. I thought these were but I could be wrong.

 

[Wayne A. Pflughaupt]

Surprising...it's been decades since I've used Canare, and it was in pre-built cables, but Mogami is known for being supple—if you search on that word with Mogami, it pops up quite a bit. And with the Canare having a braided cable, I'd be surprised if it was more so than the Mogami. But maybe in a different way...(besides the star quad Mogami, I have a half-spool of their 8-cable snake bought at the same time and used for a few things that's amazingly supple for 8 x two-conductor plus shield in a jacket.

I'm sure Canare is fine, not disagreeing, just registering my surprise about the flexibility issue.

I suspect that it’s mainly an “issue” (not that it’s really that big a deal!) with the 2549 specifically. The insulation on the individual conductors is thick and stiff, and the shield wire is pretty coarse, not finely stranded like Mogami’s star-quad braided shield. So the combination of those two things stiffens up the overall package.

Not saying it's like RG-59 or anything, just not as supple as the comparable Canare offerings.

Regards,
Wayne A. Pflughaupt

 

[Wayne A. Pflughaupt]

I will just go with the mogami 2549. Seems like it will work for what I want and we are talking like a $5 difference in cost.

I will also order the Rean RCA's

Are those neutrik XLR's the "right ones" They have a bunch of different 3 pin XLR. I thought these were but I could be wrong.

As long is its 3-pin, take your pick.

Regards,
Wayne A. Pflughaupt

 

[earlevel]

I suspect that it’s mainly an “issue” (not that it’s really that big a deal!) with the 2549 specifically. The insulation on the individual conductors is thick and stiff, and the shield wire is pretty coarse, not finely stranded like Mogami’s star-quad braided shield. So the combination of those two things stiffens up the overall package.

And that totally makes sense.

I do recall liking Canare, back in the '70s, for musical instrument cables, I believe it held up well for playing on stage. I imagine that's still true—Mogami would probably not be my choice for something that's going to get stepped on a lot, but it's great for the studio.

 

[solderdude]

There is a difference between leaving the RC out and shorting it.
When you short it 3 and 1 are connected via the RCA shield.

Pin 1 (when shorted RC) is connected to shield ... shield is connected to RCA shield.... RCA shield is connected to 1 wire... that wire is connected to pin3. Ergo pin 1 and 3 are connected as they should be.
When RC is there pin 1 and 3 are also connected but for low frequencies (groundloops) there is 100 Ohm between - input (3) and common and for high frequencies (RF) there is a short and pin 1 and 3 are tied together.

 

[Jdunk54nl]

Ended up going with the Canare L-2T2S at $0.49/ft. At half the cost I couldn't justify buying the mogami. Also ordered the Rean RCA and Neutrik RCA connectors and solder with 3% Silver.

Should be fun to make my own. I may even make them prettier with some leftover tech flex that I have.

 

[L5730]

To solve a ground loop problem I had going from (USB DAC) RCA to XLR (Powered Monitor).
I was using RCA-RCA cable with Neutrik RCA-female to XLR-male adaptors, but had a ground loop.

Neutrik adaptors are NA2MPMF, which have RCA tip to XLR pin1 (+ve), and ties RCA sleeve to XLR pin2 (-ve) and pin3 (GND)

Ended up modifying the existing RCA-RCA cable, cutting off the RCA at one end and replacing with an XLR.

RCA tip (signal)     > XLR pin 2 (hot/+ve)
RCA sleeve (ground) > XLR pin 3 (cold/-ve)
RCA sleeve (ground) > cable shield
                      XLR pin 1 (ground) unconnected.

I have this cable coiled (yes, naughty) sitting on top of a PC tower, running alongside a mains multi-gang extension socket, next to a router and modem. No noises or funny business.

I can see why putting a resistor on the XLR pin1 side of RCA GND / cable shield / XLR pin1 GND. The resistor would lower the current (V=I x R) along the ground connection and reduce the impact of a voltage difference causing an annoying ground loop.
With RCA GND connected also to XLR pin3 (-ve), the balanced input at the XLR end (+ve / -ve opposite phase) would give some noise rejection. Any noise that is the SAME between +ve and-ve would be cancelled through phase nulling. That's how balanced connections work.

I should really re-do my cable with a 100 Ohm resistor on pin1, but meh. It's labelled and I know not to use it for other purposes.

 

[Jdunk54nl]

That was one of my motivations for making my own cables. To test out some different configurations of the cables to see what works.

I know I have a noise (ground loop most likely but nothing solved it yet) issue with my AVR to crown amps. I know it is caused by the Cable Coax on TV and sent through HDMI to AVR and AVR pre-outs to Crown AMP. It is fine with sub woofer duty, but if I connect it to my speakers, the noise is quite clear. I want some separate amps and the issue will be in them as well so I am attempting to solve it. But I also want to make the cables correctly and not mess anything up.

 

[Rock Rabbit]

The short between GND and COLD must be done at the RCA side to use the differential input amp (XLR side) to reduce induced noise, but then a shield must be grounded only on the XLR side.

3 wires + shield, star ground between equipment helps too.

A balanced input can't resolve a ground loop.

 

[Speedskater]

A balanced input can't resolve a ground loop.

A well designed balanced input stage will not be impacted by a ground loop in the Safety Ground system.

 

[Rock Rabbit]

A well designed balanced input stage will not be impacted by a ground loop in the Safety Ground system.

A "well designed" balanced input even with matched resistors is limited by CMMR of the op amps. With an RCA input (with finite output resistance Rs) you destroy the matching resistance of the balanced circuit adding Rs to pin 2 input resistance, with 100 ohm typical for Rs the CMRR of the "balanced" input is limited to only 46 dB!

 

[peng]

I guess I am missing where it shows them connected at the XLR side. I also assumed that shorting the RC network just meant to eliminate that, not to connect pin 1 and 3 together.

So eliminate what I did in red and just make it a wire, but it doesn't appear 1 and 3 are connected here. At least I would think they would make that clear if it was...they did on the RCA side at least. Does this even make a difference on the XLR side.
View attachment 97878
Again, this was my issue with the monoprice cables because they did have the XLR pin 1 and 3 physically soldered together and the RCA shell soldered to shield and -input.. Does this soldering at the XLR side and RCA side make a difference to just having them soldered at the RCA side?


In all of the threads here that I have read, the hypex document kept coming up (at least the most recent ones) as the way to make your cables. It is what benchmark, march audio, and a few other companies do. I just am not 100% sure if they are actually soldering pin 1 and 3 and if it makes a difference. Why wouldn't hypex have that connection shown in their diagram if it was supposed to be?

I plan to update to hypex or purifi amps so I want to do these right.

I know this is old, but just want to ask how did it go in the end, it did the job without any ground loop resulted hum right?

My 2nd point is, I don't know why even after solderdude' explanation you still think the monoprice connector "have the XLR pin 1 and 3 physically soldered together and the RCA shell soldered to shield and -input.."

I have that Monoprice cable and they are in fact connected as shown in your diagram, so NO, pin 1 and 3 are not soldered together directly, but via the shield and the shell casing of the XLR connector only. Pin1 is not "solidly" soldered to pin 1 directly. So it will work the same as if you make your own following Hypex diagrams, it is the same!!

 

[Cbdb2]

A "well designed" balanced input even with matched resistors is limited by CMMR of the op amps. With an RCA input (with finite output resistance Rs) you destroy the matching resistance of the balanced circuit adding Rs to pin 2 input resistance, with 100 ohm typical for Rs the CMRR of the "balanced" input is limited to only 46 dB!

CMRR is for rejection of airborne EM that the conductors pick up, it does nothing for ground loops. Ground loops cause noise in unbalanced circuits because the signal is between the positive and GROUND and any noise on the ground becomes part of the signal. In balanced signals the signal is between positive and negative and ground doesn't matter. However not all balanced gear is designed properly and there are other ways for ground loops to interfere in balanced equipment.

 

[Speedskater]

However not all balanced gear is designed properly and there are other ways for ground loops to interfere in balanced equipment.

How very true!
Some new audiophile equipment still has the infamous "Pin 1 Problem". First written about in 1995 in the AES Journal.
* * * * * * * * * * *
and lots of new RCA unbalanced input equipment have the RCA equivalent of the "Pin 1 Problem".

 

[Jdunk54nl]

I know this is old, but just want to ask how did it go in the end, it did the job without any ground loop resulted hum right?

My 2nd point is, I don't know why even after solderdude' explanation you still think the monoprice connector "have the XLR pin 1 and 3 physically soldered together and the RCA shell soldered to shield and -input.."

I have that Monoprice cable and they are in fact connected as shown in your diagram, so NO, pin 1 and 3 are not soldered together directly, but via the shield and the shell casing of the XLR connector only. Pin1 is not "solidly" soldered to pin 1 directly. So it will work the same as if you make your own following Hypex diagrams, it is the same!!

I ended up needing to tie chassis’s together to eliminate the noise by running a separate wire between chassis bolts on all equipment. That was what ended up solving my ground loop. Now I don’t even have any of that equipment anymore.

As far as the monoprice, I knew it did because I had that cable before and could physically view it, and it was physically soldered. I know there have been changes in that monoprice wiring over the years too, so you may have a different version.

 

[peng]

I ended up needing to tie chassis’s together to eliminate the noise by running a separate wire between chassis bolts on all equipment. That was what ended up solving my ground loop. Now I don’t even have any of that equipment anymore.

As far as the monoprice, I knew it did because I had that cable before and could physically view it, and it was physically soldered. I know there have been changes in that monoprice wiring over the years too, so you may have a different version.

Thanks, as I said, I have those cables, and they are wired like the way they show on their website. If yours have pin1 and 3 soldered together, that's wrong. May be yours were from a bad batch that was made incorrectly. Anyway, if tying the two chassis worked for you, that's great, I would think in most cases that wouldn't work.

 

[Jdunk54nl]

Thanks, as I said, I have those cables, and they are wired like the way they show on their website. If yours have pin1 and 3 soldered together, that's wrong. May be yours were from a bad batch that was made incorrectly. Anyway, if tying the two chassis worked for you, that's great, I would think in most cases that wouldn't work.

Nearly every time I’ve tried tying cases together it has lowered noise. Usually some stuff is using a two prong plug and some is using a 3 prong with earth ground. That’s usually when I see ground loops like this and when it has worked.

 

[peng]

Nearly every time I’ve tried tying cases together it has lowered noise. Usually some stuff is using a two prong plug and some is using a 3 prong with earth ground. That’s usually when I see ground loops like this and when it has worked.

Yes, it will work in some cases, just too hit and miss for me to recommend it as a fix, other than no harm trying first. The better way is to try and identfy the ground loop current path, and then break the path, such as using a good rca to xlr cable, Monoprice has the cheap and dirty one, Benchmark's probably is among the most effective ones, but $100 for a pair of 3 footers.

 

[Jdunk54nl]

Yes, it will work in some cases, just too hit and miss for me to recommend it as a fix, other than no harm trying first. The better way is to try and identfy the ground loop current path, and then break the path, such as using a good rca to xlr cable, Monoprice has the cheap and dirty one, Benchmark's probably is among the most effective ones, but $100 for a pair of 3 footers.

My ground loop path was coming from the cable coax and being transferred via the hdmi cable. I could never find the cause of the cable coax other than disconnecting it all the way at where the cable was coming into the house.

Agree that grounding chassis isn’t always going to fix it, but definitely worth a shot. All you need is a good spot on each device and a piece of wire stripped to test it out.

But having a common ground for every device is always a good idea, so if your devices do not, doing that is always recommended.

 

[earlevel]

Nearly every time I’ve tried tying cases together it has lowered noise. Usually some stuff is using a two prong plug and some is using a 3 prong with earth ground. That’s usually when I see ground loops like this and when it has worked.

Yes, this means the various grounds are at different potentials (to state the obvious). The best cure is to improve the house/room wiring, but another possibly cure is to take the problem equipment off the same outlet, where practical.

My house had a standalone building in back, apparently a "recreation" room (sliding glass door, lots of window area, had the outline of what might have been a bar bolted to the floor, a concrete pad just outside that was likely for a ham radio antenna). Really bad ground loops (all 2-prong outlets). It was uninsulated and cheap wood paneling, so it was an easy job to take it down to the studs. My electrician friend ran new mains wiring from the main panel at the house, put in a subpanel, conduit, ran proper wiring, grounding rod. Never a ground loop since.

But yeah, if you're stuck with poor wiring, you do whatever works.