• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Single Ended, Balanced, Differential

Jdunk54nl

Addicted to Fun and Learning
Joined
Aug 5, 2020
Messages
968
Likes
1,048
Location
Arizona
Hello all.

I was asking this question over on a car audio forum as I see this more in car audio, but wasn't getting very many responses that were actually useful and not really backed by great data.

Here is the question: Why do amps that have ONLY rca inputs have balanced/differential inputs? What are their benefits?

Example:
https://jlaudio.zendesk.com/hc/en-us/articles/209357588-Differential-Balanced-Inputs

Balanced can be done technically with 2 conductors as long as both carry a signal and same impedance to ground, so I suppose you can have a balanced signal on RCA's (I couldn't find any "in the wild" examples though).

I have read both of these by Bill Whitlock. AMAZING reads, well worth it. They both have basically the same info.
https://centralindianaaes.files.word...notes-v1-0.pdf

https://www.jensen-transformers.com/...ic-seminar.pdf


Pg 16 from the Jensen Article: From this, it seems that unbalanced, single ended sources, doesn't benefit from differential amplifiers




Screen Shot 2021-05-26 at 17.54.40.png


Pg 22 from the Jensen Article: This again says single ended sources, since they have different impedances to ground, won't benefit from an amp input that is "balanced and differential". Nothing really comes from the differential part it appears.



Screen Shot 2021-05-26 at 17.53.21.png


Pg 29 from the Jensen linked article:
This is where I get confused, in the second "correct" example, wouldn't the shield being connected to ground and to the sleeve of the RCA, which is connected to pin 3, cause the lo and hi conductor to have unequal impedances to ground? So how do we get noise rejection here? The signal can no longer be balanced with unequal impedances to ground!



Screen Shot 2021-05-26 at 17.55.09.png
 

Attachments

  • 1622076652678.png
    1622076652678.png
    234.6 KB · Views: 94
OP
Jdunk54nl

Jdunk54nl

Addicted to Fun and Learning
Joined
Aug 5, 2020
Messages
968
Likes
1,048
Location
Arizona
If I am understanding correctly, this seems to imply the following.
1) electricity is lazy so it takes the path of least resistance
2) if you have single ended source grounded and a direct to ground line, chassis voltage differences should flow through that ground wire as path of least resistance
3) this means it won’t flow through the lo side in my picture above as it is a harder path (well not much anyway.
4) this should be similar to the signal flowing through the lo side and not through the shield

Is that correct?
This would no longer technically be balanced though right? Even if the single ended source wasn’t grounded, but since it is on the receiver end, impedance to ground can’t be the same anymore?

Also forgot to mention, someone on the car audio forum also said that a balanced differential input (and jl also hinted at this) has the benefit of really accepting any signal from singled ended to fully balanced and differential.
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,334
Likes
3,278
Location
.de
Car audio is hit by the exact same problem as analog audio passed around inside a computer (*), or an unbalanced connection between a PC and typical studio monitors - there is an already existing common circuit ground. Given the massive and dirty currents going around in a car chassis (notably alternator charge current along with fuel injection and ignition systems), you can probably imagine that insisting on 100% unbalanced connections would be pure ground loop hell. They cannot differentiate a voltage differential between output ground and input ground from an actual signal, and no chassis has a resistance of exactly zero.

*) Look up the Crystal CS4297A datasheet, notably the CD input - it's a pseudo-differential affair.

No, usually these pseudo-differential jobs do not have the same amount of CMRR as a proper balanced input, but they do tend to get the job done. Remember we are talking the relatively modest dynamic range requirements of car audio here, and 20-30 dB is an awful lot better than nothing. There probably is too much inertia and too little incentive to change to proper balanced wiring at this point.
 
Top Bottom