Balanced connections question #342

[Coytee]

Something dawned on me yesterday and I became puzzled.

If you run a Peach into a (insert brand) unbalanced/RCA amplifier, we have an unbalanced signal. I got that.

If you run a McIntosh (insert model) with BALANCED outputs, into a balanced amp, we have a balanced signal. I got that too

If I run my Peach into my EV Dx-38 with unbalanced cables (RCA--->XLR), the EV has balanced outputs. If I then take the XLR outputs and take them to my XLR McIntosh inputs, is that a TRUE balanced signal?

In other words, will the EV 'convert' it to a balanced signal? Since the originator of the signal (Peach) wasn't balanced, is this a balanced signal but one of lessor strength?

If I add the transformer into the circuit, does it "convert" to a balanced signal or does it simply step up the signal level?

I'm just trying to get my arms around the logic of "what must create" the balanced signal and if something doesn't create it per se', (Peach to EV unit, via RCA-->XLR cables), do you then have a "true" balanced signal further down the food chain??

My brain keeps going back to the logic of "garbage in, garbage out" and if I have unbalanced in, then it can't be a "true" balanced out.

[sfogg]

"In other words, will the EV 'convert' it to a balanced signal?"

Probably, it is not a big deal to do this.

Shawn

[Guest " "]

"In other words, will the EV 'convert' it to a balanced signal?  Since the originator of the signal (Peach) wasn't balanced, is this a balanced signal but one of lessor strength?"

You are seeing the light.

One can argue that optimal results would be balanced from end to end.

But if you apply any type of signal balancing in the connectivity path, the goal of noise reduction as a result of amplified inverted signals can be realized.  The only issue is the closer to get to the speakers, the more costly the implementation is.  You can run a balanced 40lb transformers that connect to the speakers directly or you can use balanced transformers in the low level siganl path that only weight a few grams.

In your senerio...the key feature you want to leverage is the MC-2102.  It's balanced inputs uses an active circut that routes inverted siganls thru two different tubes in parallel.  When this signal meets in your driver stage...noise is cancelled out.   

Bottom line is use balanced where ever possible to allow any added noise that is captured at that link to be cancelled out.

Also, I did some A/B test using my pre-amp and amp which have both XLR and RCA connections and found the XLR to perform the way you intially reported.  I'm not  using any black boxes.....just pure XLR cables.  I was able to switch back and fourth and my findings was the same as yours.  XLR rocks.

[sfogg]

"the goal of noise reduction as a result of amplified inverted signals can be realized. "

That is a common misconception. You do not need an amplified inverted signal on a balanced connection to still get the CMRR benefits. You just need the impedance in both legs to be matched/balanced as then noise couples equally into both legs and the receiving end of the balanced connection cancels the noise by how it works (assuming it is done properly, not always is).

"Also, I did some A/B test using my pre-amp and amp which have both XLR and RCA connections and found the XLR to perform the way you intially reported. I'm not using any black boxes.....just pure XLR cables. I was able to switch back and fourth and my findings was the same as yours. XLR rocks."

Did you account for the 6dB level difference? Most don't and therefor get tricked by the level difference so that they think balanced sounds so much better then single ended. It doesn't, they are just hearing the loudness difference. Match the levels and things chance. Balanced just has greater immunity to noise pickup in the lines and less chance of a ground loop.

Shawn

[Coytee]

Mike, I'm glad you've seen some of the results I thought I saw. I had Dale over last Friday to give a listen. Thursday (day prior), I pulled my RCA/XLR cables out and reinstalled the transformer and all the XLR cables. I had a hum that didn't exist before so not having the time to tinker with anything, I swapped it all back to the RCA/XLR setup and let it go. Ironically, when I first had the tranformer and XLR cables in...it was when I took them OUT that I had a hum. I didn't have a hum with it in. Now I'm reversed (sigh)

For the 10 minutes I DID play around (and thanks to MAS and his comments) I did turn the gain volume of the XLR transformer down from full gain to something like 1/3rd. I DO think that helped the noise issue but again, wires were flying around in there so I can't be conclusive that it was perhaps the culpret.

Unfortunately, Mark (Deneen) doesn't have the time/setup to convert the Peach to XLR outputs (or allow dual outputs). I really don't want that transformer in the signal path if I can get around it. I DO however, like the additional gain the transformer gives me (conflicting desires).

I know I'm making this into a mole hill...and will probably continue on to make it into a mountain. Here's an interesting happening to me...

I had a cd in. I KNOW good & well that not all cd's have the same signal levels on them but the truth is... I can say the following for "most" of my cd's (though not all)

I had a cd in... I was listening to it loud. I then PEGGED the volume on the Peach. I had the volume pegged as well as each channel control. I also had it on "lo" mode which I belive gives yet another small boost of some kind in the signal. In other words, the Peach was INCAPABLE of any more increases on any of its controls. They were all absolutely maxed out.

I was in the same room with it... yes it WAS loud, no question...but I remained intrigued. I looked over at the meters on my MC-2101 (which again runs ONLY the HF horn, my solid state running the LF horn has NO meters). The 2102 was PEAKING at about, or just by a hair, over 10 watts on the meter.

Got that?? Peach was PEGGED on volume and the 2102 was peaking at 10 out of 100 watts. I don't know what the solid state was using but even if it was 10x, that would only be 100 watts (and I doubt it was 10x)

With the transformer in the line, I once had the volume between 9:00 and 10:00 and had to stop as it was getting to these same levels on my meters.

The system simply has SO much more power available to it when I use the transformer...it's almost like the mute button is on when the transformer is out.

I don't want to use an outboard transformer, I really like the Peach... I've been wondering if I ought to look into a pre that comes with balanced outputs. I've decided to wait on that journey.

With all the above said, it might be good to mention that when I'm cranking it, I'm probably never using much more than 10 watts on the HF horn anyway so it's somewhat of a moot point...but just knowing there's more ability there, kind of makes me want to be able to have it on tap.

[Coytee]

What Shawn says makes some sense to me as it's the stronger signal that I think perked things up for me. Meaning, although I DO think the bass was a lot stronger when I had the transformer in, I can see where 'everything' was stronger and I simply focused on the bass. I don't think the sound quality (eq'ing of the speakers) was necessarily any different for me... I think that I simply had more power available to the speakers when I cranked it (from another room).

I've made comment before that I could get the same "flapping of your pants" at 30' with the Jubilees, that I had with the Khorns at 10'. Although that is true, I think it was more true when I had the transformer in than today with it out. If that's simply because I had 6 db hotter signal, then that is why (who am I to argue?)

I guess I need to get off my hiney and FOCUS on getting my basement done so I can move things down there and tinker with the room a bit. Everything I do in current room is temporary at best and as far as anything 'room treatment' related...forget it... it ain't gonna happen as long as SWMBO is around.

[Coytee]

That is a common misconception. You do not need an amplified inverted signal on a balanced connection to still get the CMRR benefits. You just need the impedance in both legs to be matched/balanced as then noise couples equally into both legs and the receiving end of the balanced connection cancels the noise by how it works (assuming it is done properly, not always is)

Shawn... is it feesible that I'm somehow doing it wrong? Is there more than one way for me to do it since I've got RCA-->XLR cables?

[sfogg]

"The system simply has SO much more power available to it when I use the
transformer...it's almost like the mute button is on when the
transformer is out."

Are you flipping the switch on the Mc to go from single ended to balanced inputs?

Shawn

[mas]

Richard,

You do not have to go to the trouble of using an expensive quality transformer to balance the inputs.

Enclosed is a manual and schematic. Its small but you can increase the size (and the resolution is great and very clear) and see exactly what they are doing. All of the similar units operate in the same manner.

My only concern in the process is that you employ as few additional gain stages as possible in adjusting the gain (level matching - as will will see is done via the op amps in the circuit - as well as phase inversion) . In other words, you do not want to amplify the signal at more than one point if at all possible, allowing any additional gain stages to be run at unity (0) in order to minimize contributions to the noise. The optimal place to do this would be on the output of the Preamp prior to the input of the EV crossover.

BTW, while you were out I spoke with Chloe [&] and she was fine with this. [Y]

Enjoy. [pi]

Rolls MB15 ProMatch Manual.pdf

[sfogg]

"I guess I need to get off my hiney and FOCUS on getting my basement done so I can move things down there and tinker with the room a bit. "

Do it!

Having a dedicated room is one of the best things you can possibly do to the sound quality of your system. The differences in pre-amps, amps, sources and such pales in comparison to the difference between a good room and a bad room.

"Is there more than one way for me to do it since I've got RCA-->XLR cables?"

If they are just plain RCA->XLR cables you won't get the noise reduction benefits with them because the impedance between the legs won't be balanced. Doesn't the Mc have regular single ended inputs? If you are just running singled ended out just run to the signal ended inputs. You potentially would have less circuitry in the signal that way depending upon what the Mc does with its balanced and single ended inputs.

Shawn

[Coytee]

Shawn: Are you flipping the switch on the Mc to go from single ended to balanced inputs?<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Answer is yes, it wont work if I dont.

Mas, the unit you showed looks a great deal like the one I have. Im ALL for not having too many gain stages. That is one reason I sold my MC-30s as they created yet another place for me to worry about level setting (with their input control on back). As for best location (between preamp & EV), that is EXACTLY where Ive put mine.

Shawn: Doesn't the Mc have regular single ended inputs?

If by that, you mean RCA inputs, then yes, the Mac has both, the dbx amp only has RCA. The fly in the ointment (for me anyway) is my EV unit ONLY has balanced inputs/outputs so Ive got to deal with them somewhere, somehow.

What I find so interesting is even with the volume on the Peach maxed, I at times, can still get no more than 10 watts out of the McIntosh meters. Yes there are other cds where I can PEG the McIntosh meters but Id say those cds are the exception.

[Guest " "]

"Did you account for the 6dB level difference? Most don't and therefor get tricked by the level difference so that they think balanced sounds so much better then single ended. It doesn't, they are just hearing the loudness difference. Match the levels and things chance. Balanced just has greater immunity to noise pickup in the lines and less chance of a ground loop."

a. The MC-2000 has a switch for XLR vs RCA and another for the level differences. It's covered in how to spell Kat with success.

b. Balanced is more than immunity to noise.  In the case of passive balanced via transformers, there's a change in the ground reference in a manner that creates an invert signal that once re-united, cancels out noise.  In the case of active balanced the signal is split and sent thru two different paths, one is inverted, when re-united prior to final output, noise gets cancelled.   

Think P-P circuts in amps and how they cancel noise.

[Guest " "]

"I've been wondering if I ought to look into a pre that comes with balanced outputs."

Thats what I did.....Mcintosh C-2200...it has both XLR and RCA inputs and outputs.

I can run directly to the MC-2000's XLR inputs.

Your findings will hold true, inspite of the level compensation switch.  

Level's are easy to calibrate using tone burst signals and the meters on both the C-2200 and MC-2000.

The difference in sound is not a level issue at all.   Your getting cleaner sound, reduced noise floor, better dynamics using XLR.

[DrWho]

For the longest time I thought the signal was split and one leg inverted when running balanced lines....but that couldn't be further from the truth. I saw the light after taking some signal processing courses - turns out they only send the signal down one leg, and as Shawn mentioned there only important thing is the differential impedance. And then it finally made sense about the whole "which pin is hot" debates that show up in the pro-audio world. Is pin 2 the accepted standard now? I know for a while there there was a lot of gear on both pins (my church being one such example). The polarity inversion everywhere gets annoying after a while.

That's not to say that you can't split and invert the signal...but that's not how / why it works.

To answer your question Richard, yes without a doubt the signal will be balanced on the output of the EV. It's interesting to note that almost all equipment is unbalanced on the sindie - mostly because it doubles the cost if you try to maintain a balanced topology throughout (it's really more than double because you gotta spec match every single part). It's just not worth the trouble.

[Guest " "]

DrWho/Sfogg

Don't know what you folks are thinking about....but I'm talking about the MC-2102 that Cyotee has and the MC-2000 that I have.  On the XLR side the signal is split and sent thru two different tubes in parallel, then the signal gets re-joined prior to the pre-driver stage.

Please stop telling me it does not get split and inverted....you can download the technical details on these two amps if you have questions on how XLR  is implemented on them.

The references to general concepts is confusing specific's about a particular equipment specific implementation.

Please review the technical material for the MC-2102 and the MC-2000 before responding further.....while you folks might have some insight and be able to add some value....let's start with the same baseline by  reviewing the tech data on the equipment in question.

[sfogg]

"b. Balanced is more than immunity to noise. In the case of passive balanced via transformers, there's a change in the ground reference in a manner that creates an invert signal that once re-united, cancels out noise. In the case of active balanced the signal is split and sent thru two different paths, one is inverted, when re-united prior to final output, noise gets cancelled. "

Both examples are just talking about immunity to noise picked up in the cables.

Think about it... if the noise is in the *source* guess what happens to it when it hits the transformers or active stage.... exactly the same thing that happens to the signal. As far as the transformer or active stage is concerned the noise is SIGNAL... so it gets transmitted to the other end and into the circuits there. It looks no different then signal.

The noise that can be canceled is noise that is coupled into the cables themselves, nothing more.

And the reason that gets canceled is because if the line is truely balanced (impedance matched) the noise couples equally in both legs. At the receiving end of the cable one leg is subtracted from the other leg. Since the noise is in phase in both legs (impedance matched) when you subtract one leg from the other the noise cancels itself out.

If the signal is on one leg or on both legs (out of phase) is irrelevant to the CMRR advantage.

If the signal is in one leg when it hits the receiving end one leg is still subtracted from the other. Since one leg has no signal nothing gets subtracted from the signal so the signal gets through. If both legs had signal (one phase inverted) the inverted (negative) phase gets subtracted from the positive phase. minus a minus is a plus.... you end up with a hotter signal... that is where the 6dB gain increase comes from.

Anyone that says otherwise doesn't understand how balanced connections actually work. Think about it.... does the noise level suddenly rise when you hit pause on your CD player and no signal is passing through the cables?

Of course not. Because what occurs with the signal is irrelevant to what happens to the noise.

Shawn

[sfogg]

"Please stop telling me it does not get split and inverted....you can download the technical details on these two amps if you have questions on how XLR is implemented on them."

It does not get split and inverted on the inputs to Coyotees amp.

Split and inverted occurs on the *source*/output side of a balanced connection.

That is not what occurs on the receiving/input end. On the receiving end the two legs are subtracted from each other to go back to single ended. This is how a balanced connection works.

Further on down the line because the amp is push-pull the signal will be split/inverted again but that has nothing to do with what is being discussed... balanced interconnects.

Shawn

[mas]

Fritz...

I can certainly appreciate your fascination with transformers...

And yes, they can do some very useful things.

But they are not a one trick pony for everything that ails ya...

[sfogg]

"That's not to say that you can't split and invert the signal...but that's not how / why it works."

Right, plenty of gear does that too (that is where the +6dB comes from) just that that is not a requirement for the CMRR advantage of balanced interconnects. Few actually understand that though....

Shawn

[Coytee]

Mark, to try to answer some of your questions to my ability

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

I do not have a lot of noise. Indeed, for the little bit of noise I DO hear, I wonder if that is simply to be expected with such efficient speakers? Now, with that said, when I very first installed the speakers and the EV unit, everything seemed to work fine. I REALLY wanted to simplify things and take anything not needed out of the system. I made new cables with RCA/XLR connections and swapped them in the evening before JC & Mark were due to arrive & listen to the Jubilees. I had some kind of noise that was not in the system PRIOR to yanking the transformer out. I started looking at each cable and it seemed some dufas (me) cross connected one of the wires. Thankfully, I only did one and it happened to be the first one I checked. Fixed that and the hum was gone (without the transformer in the foodchain). Later, I reflected on the difference in sound, you know, the old logic of listen to it for a month and THEN make a thought as to sound differences, rather than make a snap judgment every 10 seconds. I thought the system sounded stronger before (with the additional 6 dbs I guess) so was going to go back and try that setup and see what I thought.

I got the RCA/XLR wires out of there and reinstalled the transformer and its set of wires Upon turning it on, I immediately had SOME kind of hum. This wasnt there before and I didnt know what was causing it. I had Dale coming over the next day to give a listen so I yanked the transformer out and went back to the RCA/XLR cables which is where it is today. Sound is great, no extraneous noise. Unless perhaps volume is maxed and cdp is not playing.

Just tonight, I was listening to David Gilmour. Peach volume was set about 2:00 Id say. McIntosh meters (and I realize theyre not perfect) were again PEAKING at just under 10 watts. Volume was again, about 2:00, the individual channel controls were maxed and the other switch was set on lo Z. (just reporting the settings). Everything sounded VERY nice. I walked a distance away from the speakers again to get a feel for them a distance away (nearing 30) and again, there simply wasnt the snap that was there like when I had the transformer in the chain (and the additional 6dbs)

So, for ME, this is less about any noise, as contrasted with striking a balance within the signal chain of wanting enough OOMPH when Im not in the room and want to crank it.

I suspect this is my actual area of concern but out of ignorance didnt realize it. Meaning, I suspect part of me wants that additional 6dbs. Another part is fine without them as they do seem to raise the noise/hiss level a bit more. I think Im beginning to see that the additional hiss was probably from the transformer which I had on full gain.

3 & 4 are over my head in a technical sense although I get the gist of them so Ill just nod in agreement

Since this is my first experience ever with balanced stuff. I know as much about it as I do brain surgery. Actually, I DID once watch a show on brain surgery so I probably know more about it. []

very interesting comments. Unfortunately, I AM stuck with XLRs in the system somewhere as long as I plan on using the Dx-38. Given that, the anal person in me needs to understand them a bit better so Im comfortable with what Im doing instead of being totally in the dark.