help me to understand: bal to unbal, 6 bd loss?

An audio transformer is a wonderful thing.

It lets us hook things up flexibly, forgetfully, foolishly, badly, and even drunkenly...to almost anything and still it works pretty well, and has a tendency to remain safe and quiet.

Sadly they are heavy and expensive, but other than that....

Another subtle unintended consequence of shorting one output line, is that when the opamp loses linear negative feedback, this clipping/saturation can reflect back through the input resistance and introduce distortion into the good output. 

Click to expand...

  JR, that is exactly what I had when shorting the negative line of the Apogee AD8000 analog outputs.

JohnRoberts said:

http://www.douglas-self.com/ampins/balanced/balfig5b.gif

Yes, that simple dual polarity output is OK for dedicated applications where you are the only person using the gear, but I found when designing products for the mass market you must make the designs customer-proof.  Driving one of two output opamps into ground will not make for very happy opamps, and all that signal current in ground could cause crosstalk issues. Not a good design practice.

Click to expand...

JR, I rather thought you would be more critical than that.  It's all that heavily distorted signal current in ground could cause crosstalk issues.(?!?) 

But what the heck?  Loadsa pro-sumer gear, eg M-Audio & Behringer use this evil scheme.  What's a few extra % of THD to their customers?  As long as all the lights flash in time to the music (music?)  ;D

ricardo said:

JohnRoberts said:

http://www.douglas-self.com/ampins/balanced/balfig5b.gif

Yes, that simple dual polarity output is OK for dedicated applications where you are the only person using the gear, but I found when designing products for the mass market you must make the designs customer-proof.  Driving one of two output opamps into ground will not make for very happy opamps, and all that signal current in ground could cause crosstalk issues. Not a good design practice.

Click to expand...

JR, I rather thought you would be more critical than that.  It's all that heavily distorted signal current in ground could cause crosstalk issues.(?!?) 

But what the heck?  Loadsa pro-sumer gear, eg M-Audio & Behringer use this evil scheme.  What's a few extra % of THD to their customers?  As long as all the lights flash in time to the music (music?)  ;D

Click to expand...

Perhaps I was a little soft on Mr Self... It's better when the readers of the thread can see the flaws and decide for themselves how bad a given approach is.

JR

Peterson's "Explain like I'm 5" has a podcast with some basic information on this subject:
http://www.diyrecordingequipment.com/5459/podcast-8-explain-like-im-5-balanced-vs-unbalanced-connections/

gunpoint recording said:

Peterson's "Explain like I'm 5" has a podcast with some basic information on this subject:
http://www.diyrecordingequipment.com/5459/podcast-8-explain-like-im-5-balanced-vs-unbalanced-connections/

Click to expand...

Unfortunately, most of this is too basic. The subject is complex; balanced is not a binary thing. A connection can be balanced regarding amplitude and badly balanced regarding impedances, or vice-versa. Some admittedly unbalanced connections will provide excellent interference rejection by being perfectly impedance-balanced. And some balanced outputs are better balanced than others.
Saying that "3-pin is balanced" is very misleading, and reducing interference rejection to signal polarity is crude.
This kind of "information" is not up to the general level of this thread.

JohnRoberts said:

Perhaps I was a little soft on Mr Self... It's better when the readers of the thread can see the flaws and decide for themselves how bad a given approach is.

JR

Click to expand...

Just to note:  the page from which that image originated does state that grounding pin 3 will cause problems.

Oddly enough, the (EXPENSIVE!!) Avid/DigiDesign 192HD boxes (and, probably other models from those jerk-offs...) use the "brain dead"  "balanced" outs.  I found that out years ago when TRYING to make a pro studio function regardless of what was patched.

Bri

Brian Roth said:

Oddly enough, the (EXPENSIVE!!) Avid/DigiDesign 192HD boxes (and, probably other models from those jerk-offs...) use the "brain dead"  "balanced" outs.  I found that out years ago when TRYING to make a pro studio function regardless of what was patched.

Click to expand...

There's little difference apart from the initial $$$ you fork out, between expensive sh*t (Avid/DigiDesign) and cheap sh*t (M-Audio/Behringer)

ricardo said:

Brian Roth said:

Oddly enough, the (EXPENSIVE!!) Avid/DigiDesign 192HD boxes (and, probably other models from those jerk-offs...) use the "brain dead"  "balanced" outs.  I found that out years ago when TRYING to make a pro studio function regardless of what was patched.

Click to expand...

There's little difference apart from the initial $$$ you fork out, between expensive sh*t (Avid/DigiDesign) and cheap sh*t (M-Audio/Behringer)

Click to expand...

Ouch! But manufacturing cost and what is MSRP are not always related. Yes I too feel that with the increase in smaller and less comercial spaces and roving recording equipment, compounded by the anybody can hook this stuff together and go to work world we live in has resulted in a very difficult and somewhat ill informed enduser community. 

This is the basic building blocks of good signal transmission and is obviously greatly misunderstood.

The best book on the subject of audio system integration I have ever read is this: http://www.amazon.com/Systems-Design-Installation-Philip-Giddings/dp/0240802861. Good luck trying to find it though. Every book I ever read regarding audio when I was a wee one referenced this one and a lot still do. It has always amazed me that one of the bibles of equipment interfacing has been so backgrounded by watered down versions.

Less voltage less signal.

Edit: CMRR is what tranformers do also the best the only thing I have ever seen that comes close to a Transformer for this is what THAT corporation calls there Ingenius Chip.

What about the Mytek 192 DAC converter, the conversion is good but it is again those uggly "balanced" outputs... expensive half-shit?

saint gillis said:

What about the Mytek 192 DAC converter, the conversion is good but it is again those uggly "balanced" outputs... expensive half-sh*t?

Click to expand...

The "uggly" hard differential output is only a problem when one signal leg is shorted, otherwise they will work OK.

A nice ancillary benefit of transformers is that they bandpass the signal scraping off DC and RF. It is good practice to bandpass solid state inputs to avoid undesirable effects related to out of band signals.

I do not know a good way to interface 2 wire I/O, only less bad. As long as all inputs are 3 wire balanced inputs, they will generally read almost anything you connect to them.

JR

The "uggly" hard differential output is only a problem when one signal leg is shorted

Click to expand...

That's the point isn't it?

JohnRoberts said:

saint gillis said:

What about the Mytek 192 DAC converter, the conversion is good but it is again those uggly "balanced" outputs... expensive half-sh*t?

Click to expand...

The "uggly" hard differential output is only a problem when one signal leg is shorted, otherwise they will work OK.

A nice ancillary benefit of transformers is that they bandpass the signal scraping off DC and RF. It is good practice to bandpass solid state inputs to avoid undesirable effects related to out of band signals.

I do not know a good way to interface 2 wire I/O, only less bad. As long as all inputs are 3 wire balanced inputs, they will generally read almost anything you connect to them.

JR

Click to expand...

OK! A truly balanced 600 OHM transmission line needs only 2 (two) wires to be balanced in a twisted pair configuration. The 3rd wire is the drain and is not necessary in a balanced system unless there is significant EMI and RFI to overcome. And then it need be only connected at one end usually the output side as you do not want to inject noise into an input (usually). Or Phantom power is in use and the drain then becomes the third leg or return leg in the Phantom Standard and needs to be connected in circuit and unimpeded by any system to the Phantom PSU. This is what a truly balanced system is all about! CMRR Common Mode Rejection Ratio!!! This is what the entire audio standard is based on that is why I chose the 600 OHM standard it is what Telco uses and has since day one, almost, and was the first impedance standard for audio thats why so many older pieces of gear are 600 OHM like the original 1176's. Thats why we use Long Frame and TT patchbays. Etc.....

The problems begin when:

A. We have impedance mismatches. Which is off topic and a big discussion. Or.

B. We are interfacing Balanced and Unbalanced gear of varying input design.

That is the subject and is why people like me get paid oodles of money to hook-up peoples gear. Sorry I vented there. It comes from hours spent trying to explain this to people. That maybe you don't want to buy that as you are going to have to put in matching transformers or special wiring schemes or that piece of kit sounds great BUT!

This is sadly a truly misunderstood concept. Also with the advent of fiber systems this is really getting lost in the shuffle but still rears its ugly head once the fiber lands at the racks.

Edit:
If you loose or tie one conductor to ground usually the low side or pin 3 you also loose between 3 and 6db of signal and unbalance the system at the same time. A nasty state of affairs.

Want to really crack your head look up Litz wire.

6db loss? It's so simple - Poles stole it 

Goodnight to all!

Pip said:

JohnRoberts said:

saint gillis said:

What about the Mytek 192 DAC converter, the conversion is good but it is again those uggly "balanced" outputs... expensive half-sh*t?

Click to expand...

The "uggly" hard differential output is only a problem when one signal leg is shorted, otherwise they will work OK.

A nice ancillary benefit of transformers is that they bandpass the signal scraping off DC and RF. It is good practice to bandpass solid state inputs to avoid undesirable effects related to out of band signals.

I do not know a good way to interface 2 wire I/O, only less bad. As long as all inputs are 3 wire balanced inputs, they will generally read almost anything you connect to them.

JR

Click to expand...

OK! A truly balanced 600 OHM transmission line needs only 2 (two) wires to be balanced in a twisted pair configuration. The 3rd wire is the drain and is not necessary in a balanced system unless there is significant EMI and RFI to overcome.

Click to expand...

I guess I wasn't clear... I have written about this so many times I do not repeat the entire story every time. The two wire i/o I was referring to was unbalanced consumer gear, not 600 ohm telco pairs. 

And then it need be only connected at one end usually the output side as you do not want to inject noise into an input (usually). Or Phantom power is in use and the drain then becomes the third leg or return leg in the Phantom Standard and needs to be connected in circuit and unimpeded by any system to the Phantom PSU. This is what a truly balanced system is all about! CMRR Common Mode Rejection Ratio!!! This is what the entire audio standard is based on that is why I chose the 600 OHM standard it is what Telco uses and has since day one, almost, and was the first impedance standard for audio thats why so many older pieces of gear are 600 OHM like the original 1176's. Thats why we use Long Frame and TT patchbays. Etc.....

The problems begin when:

A. We have impedance mismatches. Which is off topic and a big discussion. Or.

B. We are interfacing Balanced and Unbalanced gear of varying input design.

That is the subject and is why people like me get paid oodles of money to hook-up peoples gear. Sorry I vented there. It comes from hours spent trying to explain this to people. That maybe you don't want to buy that as you are going to have to put in matching transformers or special wiring schemes or that piece of kit sounds great BUT!

This is sadly a truly misunderstood concept. Also with the advent of fiber systems this is really getting lost in the shuffle but still rears its ugly head once the fiber lands at the racks.

Edit:
If you loose or tie one conductor to ground usually the low side or pin 3 you also loose between 3 and 6db of signal and unbalance the system at the same time. A nasty state of affairs.

Want to really crack your head look up Litz wire.

Click to expand...

This territory has already been covered while I don't think we need to veer even further off the reservation (skin effect?) for a discussion about to misconnect 3 wire interfaces to lose signal level or more. 

JR

I didn't think I strayed or misunderstood. But it wouldn't be the first time. I wish that "consumer" or "prosumer" or "professional" gear stuck to a two wire or three wire system. Sadly what market it is targeted at has little to do with the system implemented. Sometimes it is a design choice, sometimes a price choice and sometimes its just bad design.

I have a console that was made years ago. It was made by a company that is still around today. This company is making products that are being used all over the planet for a bigger company. They actually did this console with a hybrid system design of balanced ins and unbalanced outs. Not throughout but that was sort of the norm. The thought was you want your ins clean and your drive lines or outs are usually short so it doesn't matter. Ugh!!! Console sounds great and I have made a lot of records on it and others in its family tree. BUT IT WAS AN INTERFACING NIGHTMARE!!!!.

The point I was trying to make, and I am sorry if my self admitted vent was taken as a dig, it was not meant that way,  was that balanced systems of any impedance (again I quoted the 600 OHM standard as it was the first one and was not reserved for Telco alone) are sorely misunderstood by most people who use this gear everyday. 

Misconnecting is not what I was pointing to. That does not work. I was actually saying that you can interface balanced and unbalanced gear without much loss if you put the proper systems in the middle. Also that the loss of Voltage which is measured in this case as db is brought about by loosing part of the transmitted energy.

Lastly I meant if you wanted to read more on how the relationship of wires to one another in a system effect that system read about Litz wire. Something that is I think fascinating and actually was developed so that underground transmission lines could not be listened in on by inductive listening devices during the war.

Pip said:

...
Misconnecting is not what I was pointing to. That does not work. I was actually saying that you can interface balanced and unbalanced gear without much loss if you put the proper systems in the middle. Also that the loss of Voltage which is measured in this case as db is brought about by loosing part of the transmitted energy.
...

Click to expand...

+1

If you read my posts you will observe that I'm fond of hard-clamped, hard-driven (and if possible, impedance balance optimized), semi-permanent connection between the critical components in the signal path, and this sort-of implies that some sort of "inbetween" matching sub-systems will have to be put to interface "other stuff", where more flexibility is required.