AudioCircle
Industry Circles => Bryston Limited => Topic started by: caleb on 30 Jul 2006, 09:31 am
-
I have used my &B SST now for about a year,and thought that I would try the Balanced input rather than the RCA to RCA.
I previously used a TAG McLaren 250x3 power amp and had a spare set of RCA tXLR connectors which worked perfectly.
However, they don't work on my Bryston.
Can anyone tell me what may be the problem??
I tought that all connectors were wired the same.
-
To eliminate the possibilities...
Is there a switch on the SST series to go between balanced and unbalanced ?
-
Some equipment provide only balanced inputs with a convenience capability of accepted an unbalanced signal if the correct pins are shorted in the XLR connector. The cable required for this connection is an RCA to XLR (with pin-3 shorted to pin-1). When used in this fashion all the advantages of a balanced input are lost, but it allows the connection for hookup from unbalanced sources.
Bryston on the other hand provides both balanced and unbalanced inputs. The trick described above is not neccessary or allowed. The correct cables are required. Using the cable you have stated is shorting the input to the balanced connector. Remove it and use the RCA connection.
brucek
-
I have used my &B SST now for about a year,and thought that I would try the Balanced input rather than the RCA to RCA.
I previously used a TAG McLaren 250x3 power amp and had a spare set of RCA tXLR connectors which worked perfectly.
However, they don't work on my Bryston.
Can anyone tell me what may be the problem??
I tought that all connectors were wired the same.
There is no accepted standard of how to map the 2 conductors (gnd, +) of an RCA connector into the 3 conductors (gnd, +, -) of an XLR connector.
Assuming you have the switch on the back of the Bryston amp set to balanced, then the conductor mapping in your cable is incorrect for the Bryston.
You can convert the unbalanced output of your preamp to balanced by using a Jensen output transformer and then use a standard balanced cable into your amp.
-
Bob,
Where can you find a Jensen Transformer to convert the RCA to XLR.
Thank's
Lee
-
I guess this is what you are looking for?
http://www.jensentransformers.com/
-
Bob,
Where can you find a Jensen Transformer to convert the RCA to XLR.
Thank's
Lee
Hi Lee,
This is the link to the audio transformers:
http://www.jensen-transformers.com/iso_aud.html (http://www.jensen-transformers.com/iso_aud.html)
The output transformer is model DM2-2XX. You can order that unit to have RCA inputs and XLR outputs, DM2-2RX. I use that model on my Arcam CD player. Give them a call, they're good folks.
--Bob
-
Maybe I'm missing something, but why can't you just run RCA cables to the amps? Bryston amps have both types of inputs. Are they really long runs and you're worried about interferance?
bluejeanscable.com sells high quality RCA cables for very little money. Might be cheaper than buying a transformer.
-
RCA cables from Blue Jeans are definitely less expensive than the $210 Jensen transformer. In my system I prefer the sound of the balanced cabling, though I admit that I did not attempt any sort of blind testing. The Blued Jeans balanced cables are very economical as well and the potential advantages of balanced lines help me to sleep at night. :-)
-- Bob
-
Bob, I agree about preferring XLR. It sounds that the TAG doesn't have XLR outputs, though, so he'd need long XLR cables, short RCA cables, and the Jensen. Starts to add up. Ah well, what doesn't start to add up? :)
-
Caleb,
Yes--use that balanced input!
There actually is a more or less 'correct' way to connect your unbalanced source to the balanced amp--and you were on the right track.
No real need for a transformer.
You can use the balanced input to create a balanced connection.
Have Blue Jeans Cable (or whoever) make up a shielded twisted pair (e.g., Belden 1800F), wired correctly for the XLR end (double check the pin configuration with your manual or Bryston), and on the RCA end tie the cold wire and shield to ground.
That makes a balanced connection (do NOT use an adapter as it just wastes the differential input)--and will reject common mode ground noise.
Mark
PS: IF you were having problems with noise in the RCA to RCA configuration (which I did not see you mention) AND the added rejection of the balanced connection described above does not solve it....THEN a transformer would be a good idea. Otherwise it adds expense and electronics to the signal chain without need.
-
That makes a balanced connection (do NOT use an adapter as it just wastes the differential input)--and will reject common mode ground noise.
I'm afraid your method simply maintains the (+) signal source reference to ground. It remains a single ended unbalanced connection. There can be no common mode rejection at the balanced end when the (-) input has been shorted to ground.
A differencial input requires a (+) signal and (-) signal that are 180 degrees out of phase to provide its benefit. The differential amplifier allows the signal to pass (because it's 'different'), but rejects the noise since it is common to both lines. This is only possible from a balanced output at the source equipment or from a transformer. The advantage is lost if the (-) signal is at ground potential.
brucek
-
Thanks everyone for your very interesting posts.
I solved the problem - when I looked at the TAG wiring and the Bryston wiring diagrams, I discovered that if I swapped over the red & white wires at the XLR connector, leaving the shieldsing braid alone then it works fine.
Now dont go asking any deep technical questions - I just know that what I have done makes these interconnects work! !
-
I'm afraid your method simply maintains the (+) signal source reference to ground. It remains a single ended unbalanced connection. There can be no common mode rejection at the balanced end when the (-) input has been shorted to ground.
A differencial input requires a (+) signal and (-) signal that are 180 degrees out of phase to provide its benefit. The differential amplifier allows the signal to pass (because it's 'different'), but rejects the noise since it is common to both lines. This is only possible from a balanced output at the source equipment or from a transformer. The advantage is lost if the (-) signal is at ground potential.
brucek
Brucek,
Yours is a very commonly held view of what 'balanced' means--but it is incorrect. Balanced connections often do have signal on both lines but symetrical signal output has nothing to do with 'balanced', or noise rejection.
Balanced requires only identical impedances on both lines....and a differential input.
I'll dig up a link or two for you.
But....the main point is that a balanced connection (with improved noise rejection over unbalanced or an 'adaptor') can be made from an unbalanced output--if there is a well designed differential input. No need for a transformer (although a transformer will likely provide even better CMRR--if that is needed).
Mark
-
Balanced connections often do have signal on both lines but symetrical signal output has nothing to do with 'balanced', or noise rejection.
I guess we'll just have to agree to disagree on this one - no problem. :)
Yes, I've read the Jensens white papers you reference on this subject and I'm afraid I may not agree with your conclusions. Disregarding that Jensen has an interest in transformer sales, I have no real dispute with what they've said. It's rather well written.
Their heavily weighted references to the importance of impedances works toward the resulting use of transformers. Yes, it is important to ensure that you have matched output impedances on your source device or you can upset the level of the (+)(-) conductors. This level matching is what determines the final CMR ratio.
Some of the white paper deals with various methods of overcoming the poor practice of connecting balance to unbalanced and vise-versa.
The practice of using convertor cables that go from unbalanced RCA to balanced XLR usually increases noise in a system by injecting noise into one of the differential receivers from the XLR lead tied to ground.
They show a typical passive method to overcome this problem, but of course you then have to deal with its associated loss of signal or gain reach.
I do have a working knowledge of differential connections and CMRR principles, so believe me when I say that the basic principle of balanced interconnection is exactly as I stated in my upper post. This is a standard technique to cancel induced noise on a line using common mode rejection with differential circuits. It's normally done with operational amplifiers and precision resistors for matching the two channels.
The balanced connection doesn't use a current carrying shield used in single ended connections. It has a standard differential output on the source side feeding a twisted pair of wires that connect to an input differential receiver. That's it. Obviously this sort of arrangement has to be well done with matching input impedances and high quality differential amplifiers and precision resistors for matching the two channels because there is a lot of chance for distortion from a difference amp. If one channel handles the signal differently than the other, distortion will be introduced into the resultant output signal. But, this is a fairly well understood method of transmission.
The reason this principle works is because there is a mirror image on the two (+/-) current carrying conductors that will pass through the receiving ends difference receiver because of the fact that they are completely different (180 degrees out of phase). Any noise riding on the two wires will be the same and won't pass through the difference receiver. This is the principle of high common mode rejection. If the signal on both wires is common to both wires it is rejected.
brucek
-
Balanced requires only identical impedances on both lines....and a differential input.
I'll dig up a link or two for you.
Mark is correct. CMRR does not require signal symmetry. It works based on the impedances between the two phases. Bill Whitlock, president of Jensen, gives workshops at the AES cnferences and stresses this point. Google "Bill Whitlock Jensen AES" and you should find one of his presentations. It is this very point that makes the transformer implementation so effective at CMR. I can't say how effective the CMR will be with the cable that Mark describes.
-
Bob,
Whitlock estimates the cable configuration will provide c. 30dB better rejection than an unbalanced connection (or using an adaptor).
The actual CMRR will be determined (as pointed out in a paper by Benchmark Media--they call this method of connecting unbalanced to balanced 'forward referencing') by the ratio of input impedance of amp to output impedance of source (and the quality of the differential circuit).
So, not as good rejection as a transformer....but much better than leaving it unbalanced.
Mark
-
It works based on the impedances between the two phases.
And is only part of the story as I've described above.
I can't say how effective the CMR will be with the cable that Mark describes
I can. it will be seriously degraded since there is an impedance imbalance. The (+) input to the differential amp is sourced from the output impedance of the driving device of ~100 ohms (as a typical example) and the (-) is tied to ground with an impedance of zero ohms. The imbalance causes a poor CMR ratio.
brucek
-
I said I'd provide a link....and then forgot!
This is not one of Whitlocks AES presentations or Jensen White papers--but a bit more recent and succinct "Mix and Match Interfaces" article:
http://www.svconline.com/mag/avinstall_mixandmatch_interfaces/
He covers several methods of connecting unbalanced to balanced and in the paragraph starting "Figure 1A" he describes the 'psuedo-balanced' connection I related above.
Whitlock: for all intents and purposes it is "the correct way to connect an unbalanced component with a balanced one".
brucek, per your comment above....Bob is correct, impedance is the whole story for noise rejection. (one example, Whitlock: "The notion that signal symmetry has anything to do with noise rejection is simply wrong")
-
I always thought that balanced connections were usually used on long interconnects where
external noise could be induced into the cable by its surroundings after leaving the
pre-amp. Since the same noise is introduced into the + and - of the cable, it gets zeroed out
by the amp when it restores the signal as the difference of the + and -, hence the 6 dB gain.
In other non-audio fields, I have had people try to explain impedance matching using
transformers etc. Frankly it's all way above my knowledge, but I find it interesting
just the same.
Maybe we are discussing two different uses of balanced connections ?
-
Maybe we are discussing two different uses of balanced connections?
Nope, we're all discussing the same thing jethro, and your explaination is the accepted and correct interpretation.
There are a few that disagree though. That's OK, everyone is allowed their own opinion.
The pseudo-balanced connection technique described in the article is a poor way to connect an unbalanced signal to a balanced input. The small CMR advantage of unbalanced to balanced connections through the simple cable wiring technique compared to a proper unbalanced to unbalanced RCA connection is theoretical and quickly lost in practice. Disregarding the fact that there is an additional differential amplifier added to the circuit with its associated added noise, the reason is that the level out of the differential amplifier is at a 6dB (half the voltage) disadvantage to a proper symmetrical signal. The answer to obtaining the same level is to increase the gain from the source with commensurate raised noise floor. It's a losing battle.
If you have balanced outputs and balanced inputs, then they are certainly the best way to go. If you have only unbalanced outputs and a choice of unbalanced or balanced to connect them to, choose the unbalanced.
brucek
-
Jethro,
Practically speaking I would note that balanced interconnection is of utility in situations without long distances involved--the emphasis now is often that external noise (RFI, etc) is much less of a problem (for most) than simple ground potential differences between adjacent components (ground loops, etc), which creates noise in an unbalanced connection. Bruce Hofer (Audio Precision), Whitlock, Benchmark Media, etc, all have published recently on this point and interconnection in general (and all suggest the 'pseudo-balanced' solution I mentioned above--or varients therof).
Regarding 'how balanced works' and the disagreement regarding signal symmetry being involved in noise rejection--
There are not "a few who disagree", nor is it a matter of opinion.
However, debating on forums does not usually have a happy result--and Caleb seems to have resolved his question.
I will quote from IEC standard on amplifiers (IEC 602689-3:2001):
"The purpose of a balanced interface is to transfer a desired signal as a differential voltage on two signal lines"
So far so good, then:
"Only the common-mode impedance balance of the driver, line, and receiver play a role in noise or interference rejection. This noise or interference rejection property is independent of the presence of a desired differential signal. Therefore, it can make no difference whether the desired differential signal exists entirely on one line, as a greater voltage on one line than the other, or as equal voltages on both of them. Symmetry of the desired signal has advantages, but they concern headroom and crosstalk, not noise or interference rejection"
Hope that helps!
Mark