|
|
Post by scanspeak8620 on Oct 27, 2010 12:52:55 GMT -5
I own two XPA-1 amps and am looking to purchase an XDA-1. My understanding is with this setup there is a balanced connection all the way from the DAC chip to the outputs on the XPA-1 amps without ever converting to unbalanced. (Which sounds awesome!! If that is the case…?)
If that is the case, do I have to worry about LONG balanced cables between the XDA-1 and my amps?
My previous knowledge of Balanced cables is when the signal gets to the amp, the positive and negative signals are added together and whatever remains is noise and is removed from the signal. This makes sense in an unbalanced amp because you would want to get back to an unbalanced signal. But in this FULLY Balanced scenario, does some type of Noise cancelation occur? Or should I be using short balanced cables to avoid noise?
Thanks
|
|
|
|
Post by Chuck Elliot on Oct 28, 2010 17:54:57 GMT -5
....when the signal gets to the amp, the positive and negative signals are added together and whatever remains is noise and is removed from the signal.... Thanks Actually, subtraction is what removes the noise. The noise is induced in phase in both lines and therefore cancels during combination: ((+Signal+Noise) - (-Signal+Noise)) / 2 = Signal |
|
|
|
Post by ajani on Oct 28, 2010 19:35:11 GMT -5
Balanced Cables are generally used on the pro audio side, as they are better for long runs than RCA... So you have NOTHING to worry about in terms of cable length...
|
|
MikeWI
Emo VIPs 
DC-1, ERC-1, USP-1, UPA-2, Sub 10, Emotiva 4S
Posts: 346
|
Post by MikeWI on Oct 29, 2010 5:31:46 GMT -5
Balanced Cables are generally used on the pro audio side, as they are better for long runs than RCA... So you have NOTHING to worry about in terms of cable length... Can someone confirm this: XLR cable does not necessarily mean balanced So, "pros" may use an XLR cable, but that doesn't necessarily mean that the system is balanced (depends on all of the components)? Mike |
|
spork
Emo VIPs
Posts: 152
|
Post by spork on Oct 29, 2010 9:10:30 GMT -5
Short answer:
You need not be afraid of using a long XLR cable.
More information:
A balanced signal sent over an XLR cable means that for each channel you have:
1. Ground (shield)
2. Hot (signal)
3. Cold (inverse of pin 2 or -signal)
The major benefit of such a cable is that you have a grounded shield surrounding a twisted pair carrying inverse signals which makes for a cable that is very robust towards electromagnetic interference from outside sources.
Also, since the twisted pair is made up of two inverse signals, the electromagnetic fields surrounding both the hot and cold cables in the twisted pair will cancel each other out, preventing the cable from being a source of noise to other cables it may cross.
Now, for an amplifier that is not fully balanced, aka the +/- signals are "combined" as input and fed to a single amplifier circuit, the signals from above will be "combined" to remove noise as cfelliot posted.
For a fully differential amplifier such as the XPA-1, the +signal and -signal are sent to two distinct amplifier circuits where the +signal becomes the positive speaker terminal and the -signal becomes the negative speaker terminal ("push-pull") such that each post is carrying an amplified signal. (In a traditional amplifier the positive terminal will be the signal and the negative terminal will be ground.)
In the fully differential amplifier's case, noise is effectively removed since, if the same noise is added to the output of each terminal, the differential between each terminal remains the same.
Thus the same equation that cfelliot posted still applies (simplified), the only difference is that the signal subtraction is the differential between the terminals and the signal is not divided in two at the end, so you effectively have:
(+Signal+Noise) - (-Signal+Noise) = 2 x Signal (which in this case is a differential taken at the speaker terminals and not at the input to the amplifier)
|
|
rakash
Seeker Of Truth
Posts: 2
|
Post by rakash on Oct 29, 2010 10:19:24 GMT -5
....when the signal gets to the amp, the positive and negative signals are added together and whatever remains is noise and is removed from the signal.... Thanks Actually, subtraction is what removes the noise. The noise is induced in phase in both lines and therefore cancels during combination: ((+Signal+Noise) - (-Signal+Noise)) / 2 = Signal Forgive me if I'm wrong though I do believe you're a bit unacurate in your description of how balanced cables work. In electronics, subtracting a signal is rather hard and is not by any normal means done straight forward. Adding them together, is however rather simple.
How one solves this is simply by adding signals that are inverse of each other to make them cancel each other out. One may look up superposition of waves if anyone's interested. How this is used in an XLR cable to get reduced noise is as following: At first the signal is split and one of them also inverseInput1: Signal -> Conductor 1 Input2: Inverse(Signal) -> Conductor 2 Through the cable both signal experience approximately the same noiseConductor1: Signal + Noise Conductor2: Inverse(Signal) + Noise When the two signals reach their destination, the one that was originally inversed is once more inversed making the Signal once more equal to the original, though the Noise inverse in this, nr 2 in my example, conductor inverse of what it is in nr1.Output1: Signal + Noise Output2: Inverse(Inverse(Signal)+Noise = Signal + Inverse(Noise) Finally both Signals are added to each other and the noise that now happens to be inverse of each other completely cancel each other out(in theory):Final output: Signal + Noise + Signal + Inverse(Noise) = 2 * Signal + Noise + Inverse(Noise) = 2 * Signal Note: The factor of 2 is simply an amplification factor of 2 on the signal. It does nothing other than increase the amplitude of the signal and one does not have to "divide" the final output by 2 to make a proper signal.
Chuck Elliot: I am aware that you might understand the concept very well, thus this might seem like me repeating what you said, though your explanation you made was a bit unclear and unaccurat as to what exatcly happens. One signal may not be "subtracted" from another like simple mathemathics. Both signals are added to each other, though when the signals are of different polarity what happens is equivalent to subtracting. The more accurate thing to say is that the signals cancel out. The way XLR cables are implemented practiacally is however not my strongest field, thus for other reasons there might be a reason for doing certain things differently from what I explained. This does however not change the concept of the XLR cable =) Cheers from Norway! |
|
|
|
Post by Chuck Elliot on Oct 29, 2010 11:24:55 GMT -5
…. although when the signals are of different polarity what happens is equivalent to subtracting…. In my book a difference that makes no difference IS no difference! My equation was just a simplification to explain the noise cancellation properties of balanced signals being converted to unbalanced output.  |
|
|
|
Post by oppman99 on Oct 29, 2010 11:41:36 GMT -5
Balanced Cables are generally used on the pro audio side, as they are better for long runs than RCA... So you have NOTHING to worry about in terms of cable length... Can someone confirm this: XLR cable does not necessarily mean balanced So, "pros" may use an XLR cable, but that doesn't necessarily mean that the system is balanced (depends on all of the components)? Mike You are correct Mike. As an example, take the XPA-5. It has XLR connections, but the amp is not fully balanced. Think of it as the same thing as putting a converter on a single ended cable so the opposite end has an XLR connecter. It doesn't mean you have balanced circuitry inside. I believe only the XPA-1 is fully balanced as far as the Emotiva amps go. In my experience so far, you most likely will only notice a sonic difference unless all the gear in the chain is fully balanced (not just having XLR connections). I've recently switched to all fully balanced gear and definately notice an improvement in clarity and detail. |
|
rakash
Seeker Of Truth
Posts: 2
|
Post by rakash on Oct 29, 2010 12:11:26 GMT -5
…. although when the signals are of different polarity what happens is equivalent to subtracting…. In my book a difference that makes no difference IS no difference! My equation was just a simplification to explain the noise cancellation properties of balanced signals being converted to unbalanced output. First off I do not like the use of caps nor bold to emphasize a point as a good argument should be enough to prove it. What is what in your book really isn't the subject. And I do understand that you have more than grasped the concept of super-positioning two signals, so to repeat myself I'm not claiming your "wrong" in the sense of what works. And yes, I understood that your post was to explain this in a simple manner. I just felt that the simplicity of your way of presenting it made it unclear what is actually going on. Actually, what makes no difference in one scenario may make a huge difference in another. If A+B=C, and A = B = -1, then it is mathemathically equivialent to calulate -1-1=C, though logically correct to calculate (-1)+(-1)=C. The missunderstanding comes when one thinks A-B, if that is the most common scenario, and one suddenly has a rear scenario where B is +1 (A and B being nothing in particular). Cheers |
|
|
|
Post by Chuck Elliot on Oct 29, 2010 12:57:16 GMT -5
The emphasis was within the statement and not directed at you!
I find the words “inaccurate” and “wrong” pretty much synonymous, but I stand by what I originally posted and hope the full circuit diagram and formula above clarify what’s going on.
Understanding basic algebra and signs is paramount to understanding basic electronic theory.
If this is not desired, then the conversation just stops at “It works” or “balanced cables and equipment have more noise immunity”.
|
|
|
|
Post by CrazyBlue on Nov 1, 2010 3:01:44 GMT -5
For us dummies, is there still an advantage to the balanced, dual differential output stage of the XDA-1 into an unbalanced amp like the XPA-5 or UPA-1? Do these amps realize the benefits of the output stage or does the "night and day" difference Lonnie heard in testing apply only into a fully balanced amp like the XPA-1?
|
|
|
|
Post by rockguitar on Nov 1, 2010 8:54:54 GMT -5
For us dummies, is there still an advantage to the balanced, dual differential output stage of the XDA-1 into an unbalanced amp like the XPA-5 or UPA-1? Do these amps realize the benefits of the output stage or does the "night and day" difference Lonnie heard in testing apply only into a fully balanced amp like the XPA-1? I would believe that the majority of the "night and day" difference referred to by Lonnie would be because the preamp is not in the circuit and the XDA was corrected directly to the amps. I have heard similar improvements with such a connection before using DAC --> AMP using the RCA connections. 2nd, regardless of when the amp is fully differentially balanced, i.e, any kind of balanced connection will have the noise cancellation. Lastly, I've thought a bit more about the fully differentially balanced circuit. This circuit would have to have the EXACT same signal on the two signal paths (except for polarity), however due to the tolerances of the components, there are going to be small differences in the signal and therefore when the subtraction of the signals occurs, you are actually going to lose a little of the original signal because of those slight variations. However, most of what I've read in the audio press touts its audible benefit, so I'm presuming that the gains (less noise, more drive) outweigh the losses. Most pro components are not fully balanced, they would be twice as complicated, heavier, and more expensive and they use the balanced connections primarily for the higher signal to noise ratio. |
|
|
|
Post by Chuck Elliot on Nov 1, 2010 9:08:09 GMT -5
I would believe that the majority of the "night and day" difference referred to by Lonnie would be because the preamp is not in the circuit and the XDA was corrected directly to the amps. I have heard similar improvements with such a connection before using DAC --> AMP using the RCA connections. 2nd, regardless of when the amp is fully differentially balanced, i.e, any kind of balanced connection will have the noise cancellation. Lastly, I've thought a bit more about the fully differentially balanced circuit. This circuit would have to have the EXACT same signal on the two signal paths (except for polarity), however due to the tolerances of the components, there are going to be small differences in the signal and therefore when the subtraction of the signals occurs, you are actually going to lose a little of the original signal because of those slight variations. However, most of what I've read in the audio press touts its audible benefit, so I'm presuming that the gains (less noise, more drive) outweigh the losses. Most pro components are not fully balanced, they would be twice as complicated, heavier, and more expensive and they use the balanced connections primarily for the higher signal to noise ratio. Exactly! The primary use of balanced lines is to get a signal from point A to Point B with less noise. Most equipment converts it back to unbalanced immediately after input. |
|
