Emergency help needed!!!

shawn said:
It's the only solution. I didn't know twisted pair RCA existed until five minutes ago. I can't think of a good application for them, but this definitely isn't it.
Click to expand...
Twisting the pair will help cancel out EMI-conducted eddy currents. We do it in the lab w/ improving SNR w/ EEG and other really small amplitude near-DC AC signals when shielded isn't available, its too small, or its a COTS part that didn't use it. Same logic as why Cat5 etc is twisted.
But an actual shield is better. Twisting is just the cost-effective half-assed solution that costs $0 extra.
 
RatLabGuy said:
Twisting the pair will help cancel out EMI-conducted eddy currents.
Click to expand...

"Will help"

Cat5 isn't a good metaphor, because Ethernet has error correction in the protocol, it's a balanced signal, and there's a shield (granted, an ungrounded one) wrapping all four pair.

Coaxial RCA is better and appropriate for the application. Twisted pair will pick up noise.
 
Valid point. This is an analog signal, where EMI is a much bigger problem.
Twisting is better than nothing, but definitely beat by proper shielding. Maybe even janky foil coating.

Actually, the balancing is an interesting point. Ideally you'd want the cables to be something like XLR that has an inherent separate ground, which would also kill this problem.

Interesting fact: Edison had the first patent on twisted pair wire, where he explained the phenomenon of how it reduces inducted signals.

EDIT - Graham Bell... b/c I'm an idiot
 
There are balanced-signal audio components that would be used with those twisted pair RCA cables. Most audio components have unbalanced single-ended inputs and will have better noise rejection with coaxial RCA cables, yes. It's not a matter of one type of cable being better than the other when used in the wrong application, it's a matter of using the proper type of cable for the application.

RatLabGuy said:
Actually, the balancing is an interesting point. Ideally you'd want the cables to be something like XLR that has an inherent separate ground, which would also kill this problem.
Click to expand...

Lots of audio components (not usually for automotive) use XLR cables for things like balanced line drivers and preamps. They are shielded twisted pair, not the same type of cable that XLR microphones use. Back in the '90s when I was dabbling in signal processor design for car audio, I used Lemo connectors instead of XLR for differential inputs (balanced) because they were a lot smaller and easier to run. Easy when someone else is paying for Lemo connectors.
 
Last edited:
Fabrik8 said:
There are balanced-signal audio components that would be used with those twisted pair RCA cables.
Click to expand...

How would that work? A balanced signal would require three conductors, but the RCA only has two in the connector. If you're sending sig+ on one cable and sig- on the other with grounds on both, the twist isn't bringing much to the table as compared to a pair of coax.
 
shawn said:
How would that work? A balanced signal would require three conductors, but the RCA only has two in the connector. If you're sending sig+ on one cable and sig- on the other with grounds on both, the twist isn't bringing much to the table as compared to a pair of coax.
Click to expand...

That's a good point, I was thinking of the differential RCA cables. They are single-pair twisted over the entire length, shielded overall, broken out to individual RCA connectors (very small length compared to the overall cable length) and the ground/shield is just split for each jack. Not coax. You need twice the RCA connectors obviously, as each channel needs two connectors instead of one. It's really the same concept as any other shielded twisted pair cable, but with dual RCA connectors instead of a single XLR connector or whatever. Not sure if anyone is actually doing the differential RCA inputs like that anymore, it's been probably 20+ years since I've played with that stuff.

I just poked around a little bit, and those unbalanced twisted pair RCAs are all over the place. I agree with you after realizing what they actually are: Pointless. Someone must be doing a lot of marketing bullshit. Totally different animal than the balanced RCA cables.
 
Last edited:
Fabrik8 said:
They are twisted over the entire length, shielded overall, broken out to individual RCA connectors (very small length compared to the overall cable length) and the ground/shield is just split for each jack. Not coax.
Click to expand...

Yeah, that's basically the same as you'd get with an XLR on the end, but sometimes you'll see it split out into a pair of RCA connectors instead of an XLR. I'm sure there's a reason for it... maybe the ability to take a balanced or unbalanced signal through the same jack. But functionally it's no different.
 
shawn said:
Yeah, that's basically the same as you'd get with an XLR on the end, but sometimes you'll see it split out into a pair of RCA connectors instead of an XLR. I'm sure there's a reason for it... maybe the ability to take a balanced or unbalanced signal through the same jack. But functionally it's no different.
Click to expand...

I think it's either for form factor/packaging or because RCA connectors are very cheap compared to XLR.
You theoretically could design it so that unbalanced and balanced could be through the same input connector, but there would have to be configuration switches or whatever. There would have to be some way to reconfigure the differential input network to ground one side of the opamp input and change the input impedance, input gain, etc. (differential setups usually use twice the voltage per signal and the differential input stage sums it back to half of that). Back when XTANT (I think it was them) made their sweet high-end amps, you could get different circuit board plug-in modules for things like that, so if you wanted a differential input instead of a single-ended input then you would swap out the plug-in board. That way seems to make more sense.
 
Last edited:
Is it a new radio? Some are more sensitive to rfi/emi than others - and that's part of why they're cheap. Check your alternator if it only does it while running - a bad diode will cause a whine.
 
Not a feed back issue ..... that would be squealing through the speakers.
Ignition noise. Might want to check your alternator and make sure it's not about to crap out on you!
Twisted pair wires of anything is elcheapo coax. It works. kinda.
Clamp on ferrites on the speaker and power wires might work, but i wouldn't
bet on it.

If your alternator OK, start going over any ground wire you may have touched and make sure it has a good connection.
One alternative is to go straight to battery for all power and ground connections. ( but might not work)
 
RatLabGuy said:
Actually, the balancing is an interesting point. Ideally you'd want the cables to be something like XLR that has an inherent separate ground, which would also kill this problem
Click to expand...

It's not the ground that's magic - it's that you're sending the signal on one leg and the inverse of the signal on the other. Any noise that gets picked up along the way is on both legs (call it noise+). But the inverse signal is inverted again (de-inverted??) by the receiver and combined with sig+. You end up with noise+ on one leg, noise- on the other, and they cancel one another out.
 
Fabrik8 said:
I'm talking about AC and DC crossing. Well, DC that apparently has AC noise riding it from something.
Click to expand...
RatLabGuy said:
yep.
Sound = AC signal ;-)
Click to expand...

Just checking in.
Certainly agree with you guys here. I was skimming the thread while distracted by a clown at lunch and a waitress with a tiny waist and an impossible amount of just jiggly enough it has to be real but not so jiggly to be flabby cleavage....thats my story and Im sticking to it.


I thought we were talking about his radio power wire crossing the ground was the reason I made the ac/dc reference. Naturally speaker/sound signals are wave based.
Carry on.
 
shawn said:
It's not the ground that's magic - it's that you're sending the signal on one leg and the inverse of the signal on the other. Any noise that gets picked up along the way is on both legs (call it noise+). But the inverse signal is inverted again (de-inverted??) by the receiver and combined with sig+. You end up with noise+ on one leg, noise- on the other, and they cancel one another out.
Click to expand...

The inverted signal doesn't get inverted again at the receive end, it gets summed with the other one (subtracted, technically). The noise is the same amplitude and same phase on both signals, so it gets summed out to zero. It's conceptually the same thing that you're saying, just a little more concise.
 
Last edited:
Fabrik8 said:
The inverted signal doesn't get inverted again at the receive end, it gets summed with the other one (subtracted, technically). The noise is the same amplitude and same phase on both signals, so it gets summed out to zero. It's conceptually the same thing that you're saying, just a little more concise.
Click to expand...

The noise starts out on the same phase on both legs. You have to invert before you sum, or else the noise gets doubled and the signal gets cancelled. ;)
 
shawn said:
The noise starts out on the same phase on both legs. You have to invert before you sum, or else the noise gets doubled and the signal gets cancelled. ;)
Click to expand...

You're saying the same thing as I am, and you quoted me saying it. ;)

The differential amplifier does a subtraction (which is functionally the same result as an invert and sum, if you want to think of it that way). Standard notation for a differential receiver is a subtraction node, as you're already dealing with a positive signal and a negative (inverted) signal.
Subtracting a negative (the inverted signal) is operationally the same as inverting the signal and adding it. Noise on both signals is the same amplitude and phase, so subtracting a signal of the same sign and same magnitude from another signal with the same sign and magnitude equals zero, which is the entire point of differential signals.

You're arguing "No, you don't subtract a negative from a positive, you multiply the negative signal by -1, and then add them". Think about that.
 
Last edited:
obullfish said:
@McCracken looks like I’ve either misplaced or am out of the RCA connectors. I’ve got plenty of RG6 coax and all the tools if you want to order some compression connectors.
Click to expand...
I appreciate it but don't worry. I'll Amazon some shit and get it here. That way you don't have to bother.
 
a_kelley said:
Is it a new radio? Some are more sensitive to rfi/emi than others - and that's part of why they're cheap. Check your alternator if it only does it while running - a bad diode will cause a whine.
Click to expand...
New radio? Old radio, new wiring? What have you changed if it is the old radio? If it's a new radio, try the old radio. If you have access to a scope, look at the power supply at the radio, and the output when making the noise..
 
98_tj said:
But it's a really good band-aid that doesn't come off the first time you wash your hands.

Sent from my SM-G965U using Tapatalk
Click to expand...

If it's electrical noise from a ground loop problem, extra shielding on an already shielded cable is not going to do anything useful.
 
Back
Top