These alternatives have different applications.
Transformers are the traditional favorites when it is key to avoid a galvanic or metallic connection between the devices.
Transformers can add noise pickup, distortion, frequency response, and phase shift problems. They can also be costly - top quality audio transformers can run a $Benjamin ($50) or more.
Balanced inputs can have benefits when used with unbalanced outputs.
The Babyface Pro has XLR outputs.
The Original Babyface has TRS outputs that may behave advantageously with balanced inputs and a suitable adapter.
IIRC the TRS was also balanced, just a mechanically different implementation from the XLR.
Jan
Truly ac does not have a polarity, And shouldn't. So, every time we force it to have a polarity, that goes badly.
Truly, that's just about the most nonsensical statement I've heard on this forum in a long time.
Dave.
AC certainly has a polarity. Take a standard XLR/XLR interconnect and switch pins 2/3 at one end. You'll find yourself with a problem immediately. Left/right will be out of phase, or a mid/tweeter are not summing correctly in a bi-amp system, or maybe no signal at all if pin 3 was grounded at the source end, or a variety of other possible issues.
In this particular case queried by the OP, we're not "forcing AC to have a polarity" by re-configuring a balanced output into an unbalanced interface. I think Daniel understands that, but it's just an extremely poor choice of words that clouds the issue even further.
This thread is typical of many on DIYaudio.com. I believe most people are indeed on the same page, but semantics and labels and various other language gets used that causes confusion.
Cheers,
Dave.
In this particular case queried by the OP, we're not "forcing AC to have a polarity" by re-configuring a balanced output into an unbalanced interface. I think Daniel understands that, but it's just an extremely poor choice of words that clouds the issue even further.
This thread is typical of many on DIYaudio.com. I believe most people are indeed on the same page, but semantics and labels and various other language gets used that causes confusion.
Cheers,
Dave.
Last edited:
FWIW I must assume that unless stated otherwise which is not the case, in any modern electronics device, specially a DAC, for God's sake !!!!! , "XLR outputs" means:
* 2 out of phase signals,
* connected respectively to pins 2 and 3 ,
* each of them referred to ground because it comes from an Op Amp or equivalent,
* no "galvanic isolation" implied unless specifically stated (not that it's indispensable in any case) ,
so the basic answer for the OP would be:
connect XLR pin 1 to RCA ground, pin 2 or 3 (your choice but do the same both Left and Right)to RCA Hot/center pin and call it a day.
And do NOT ground the unused "balanced" pin.
* 2 out of phase signals,
* connected respectively to pins 2 and 3 ,
* each of them referred to ground because it comes from an Op Amp or equivalent,
* no "galvanic isolation" implied unless specifically stated (not that it's indispensable in any case) ,
so the basic answer for the OP would be:
connect XLR pin 1 to RCA ground, pin 2 or 3 (your choice but do the same both Left and Right)to RCA Hot/center pin and call it a day.
And do NOT ground the unused "balanced" pin.
It might help to provide some caveats:
(1) A high proportion of so-called XLR or TRS balanced outputs put the signal on the equivalent of pin 2, and either ground pin 3 directly or ground it through an impedance matching resistor. The pin 1 equivalent is grounded and should be attached to signal ground on the equipment receiving the signal.
(2) If the device has a transformer output there are three common possibilities:
(a) The transformer winding is floating completely and both ends are not connected to any part of the piece of equipment. They are typically connected to XLR pin 2 and pin 3 or their equivalent on the source equipment. Hook one to RCA ground and the other to RCA signal connections.
(b) The transformer winding has a center tap that is grounded. Hook RCA ground to the same grounding point (as is possible), and hook RCA signal to one of the ungrounded terminals.
(c) The transformer winding has one end connected to ground and the other has the signal. Hook RCA ground to the same grounding point (as is possible), and hook RCA signal to the ungrounded terminal.
Consumer devices? This is the rare exception in the pro and semi-pro world in my experience, usually reserved for lowest end 'prosumer' gear.
Not all active input/output stages are referenced to ground in the classic sense suggested here. When I asked Bruno he confirmed Hypex, for example, uses an instrumentation type which comes close to emulating a transformer. The test is trivial; if grounding either pin 2 or 3 reduces the signal between pin 2 and 3 by 6 dB, the i/o is referenced to ground.
Sorry to disagree. The best way to get a clean signal is to connect the unused phase of the XLR also to gnd. Remember that either phase can have appreciable hum/noise - normally in a balanced system not a problem because a balanced receiver only takes the difference between the phase so cancels hum/noise on each phase (which should be the same on each in a well-balanced system).
If you take only one phase and let the other dangle, that one used phase most probably carries some hum/noise. By connecting the unused pahse to ground you force the unbalanced input to take the difference between the phases and help cancel hum/noise.
Jan
Jan,
I certainly understand where you're coming from on this.
There are two (maybe more) schools of thought on this particular bal/unbal interconnect scheme. I've seen it recommended in a few different ways.
Here's one that might be of interest:
Sound System Interconnection
Take note of example 4 about two-thirds of the way down the page. This refers to the XLR/RCA example in the first line of the chart above.
I have absolutely no problem pulling an active pin 3 XLR output to ground.......with the proviso that said output won't be damaged by doing so. As I said earlier, I've seen that happen. It certainly wasn't the best engineered piece of kit, but there it was.
Cheers,
Dave.
I certainly understand where you're coming from on this.
There are two (maybe more) schools of thought on this particular bal/unbal interconnect scheme. I've seen it recommended in a few different ways.
Here's one that might be of interest:
Sound System Interconnection
Take note of example 4 about two-thirds of the way down the page. This refers to the XLR/RCA example in the first line of the chart above.
I have absolutely no problem pulling an active pin 3 XLR output to ground.......with the proviso that said output won't be damaged by doing so. As I said earlier, I've seen that happen. It certainly wasn't the best engineered piece of kit, but there it was.
Cheers,
Dave.
Well I think that is very suboptimal, I can't figure out why anyone would do that. Your S/N is at the mercy of whatever is inside the source. That source may have an astronomical S/N in bal-bal mode but in the example may be horrible with that conversion. Just stupid!
The ONLY way to keep a good S/N when going from bal to se is to wire it such a way that the se side also 'looks' at the difference between the two bal signals. And that only happens when one of the signals is referred to the se reference which is its signal ground. Surely this is clear?
Now, I understand that if your source is not a xformer and not a x-coupled output, you may get into trouble. But hey, there's so much competently designed equipment around, why settle for 2nd rate ;-)
Jan
Last edited:
Likewise, does it make sense to upgrade the preamp by adding THAT1200 to get the XLR jacks at the preamp?
Sorry to disagree. The best way to get a clean signal is to connect the unused phase of the XLR also to gnd.
If you take only one phase and let the other dangle, that one used phase most probably carries some hum/noise. By connecting the unused pahse to ground you force the unbalanced input to take the difference between the phases and help cancel hum/noise. (how does this reach the other end through a single RCA connector?)
Jan
Sorry to disagree on your disagreement
but what you say applies IF there is also a "cheating" "fake" XLR input at the receiving end, which if well made, might benefit from what you say and in which case I would agree.BUT
The OP is going to a simple unbalanced input through an RCA connector, for God's sake, so all elaborate thoughts don't apply for such a crude connection.
To boot, I very much doubt both devices are separated by a couple hundred feet of unbalanced cable passing through an electrically dirty environment, I rather imagine a couple feet, and themain problem being differing *connector* matching, rather than anything else.
In this case, I fail to understand what benefit might the OP receive by grounding an active output which sends nothing to the other device
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.