Balancing card (2x unbal-to-bal & 2x bal-to-unbal) R vs G graph added

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

audiox

Well-known member
Joined
Feb 25, 2007
Messages
610
Balancing amplifier with 2 balanced-to-unbalanced converters and 2 unbalanced-to-balanced converters on one 70x50 mm card. Only standard components used (no special ICs).

Documents:

http://audioxxx.googlepages.com/bal_card_block_diagram.pdf
http://audioxxx.googlepages.com/bal_card_schematic.pdf
http://audioxxx.googlepages.com/bal_card_layout_designations.pdf
http://audioxxx.googlepages.com/bal_card_layout_values.pdf
http://audioxxx.googlepages.com/bal_card_copper_mirror.pdf

Gain resistor selection:
http://docs.google.com/fileview?id=0B_XBYPfmG8pVNWQyYzgxOGEtMWNjYy00ZDJmLWI3YzMtOWRkZWIwZGZkZjRk&hl=en


Specifications are very good:

Bal-to-unbal:
THD 0.002%
Output noise -102 dBu

Unbal-to-bal:
THD 0.002%
Output noise -101 dBu

Gain/attennuation can be freely selected by changing only one resistor. Gain equations can be found on the block diagram sheet.

**********

My other projects:

Studer 169 EQ
http://www.groupdiy.com/index.php?topic=33996.0

Neumann W492 EQ
http://www.groupdiy.com/index.php?topic=28332.0

Neumann OA10 discrete op-amp
http://www.groupdiy.com/index.php?topic=34105.0

Balanced 12 input summing amplifier
http://www.groupdiy.com/index.php?topic=29082.0
 
very low noise? distortion? any numbers? I would like to know. (If you dont have an analyzer me or someone else could accept a mail parcel and test and return it for you)

I too am building a console but wont be balancing too many things. obviously mic inputs.

I would be interested in the project if there are some real good numbers on the noise floor and THD.

But correct me if Im wrong, isnt the only reason to balance is to CM reject RF noise found in an existing environment? IE: used to fix rather than used as a matter of course? hence "unbalanced insert" in that your rack shouldnt be far enough away to warrent rf removal and you shouldnt put your board in a high rf environ in the first place.?

But I
 
I might be interested. Are you going to use dedicated chips SSM or THAT etc. or opamp based?
 
cool stuff  ;D

this is a great project for a beginner like me...

hopefully this forum stay alive for the next couple years,
because i always find new projects wich i would like to build  ;D
 
I don't think you will achieve a CMRR anywhere close to that of the SSM or THAT balancing devices. This is a very important parameter when doing balancing/unbalancing in order to achieve the desired effect of balancing.
 
helterbelter said:
I'd definately buy a small stack of these if you (or Gustav ?) is going to produce these.

As usual I don't sell anything related to my projects. If someone wants to manufacture the boards, no problem. I noticed that I haven't published copper layer pdf for home etchers (I just forgot to do that). I try to post it later today. The board is single sided.

gswan said:
I don't think you will achieve a CMRR anywhere close to that of the SSM or THAT balancing devices. This is a very important parameter when doing balancing/unbalancing in order to achieve the desired effect of balancing.

You can achieve those datasheet CMRR figures in lab environment only. Something like 10 dB can be lost just by adding connectors in the signal chain. If you need high common mode impedance, use transformers. In practice there is not much difference between active balancing circuits.

It is important to understand that the balanced input stage does not determine the CMRR alone. Think the whole system as a bridge (and remember to include cables, connectors, RF-filtering etc). You will soon find out that 40 dB "real" CMRR is very good. If you build a balanced input stage using 1% resistors the "lab" CMRR is typically close to 60dB up to 15kHz (worst case is 40dB but that is not likely if you use resistors from the same batch).

The circuits used in this project are the same as in most pro audio equipment. So I wouldn't worry too much.
 
helterbelter said:
Nice !

I'd definately buy a small stack of these if you (or Gustav ?) is going to produce these.

yes PCB`s wouldnt be a bad idea... but the circuit is so simple that you probably will
be able to transfer the curcuit on to those grid boards
 
Advantgage of these drv, that, etc chips is that they're semi floating.

I've always wondered how stressful it is for an opamp when connecting differential outputs to unbalanced in's like you would with transfo's
 
Copper layer pdf uploaded:
http://audioxxx.googlepages.com/bal_card_copper_mirror.pdf
 
An advantage of this design is that both legs of the balanced input have an equal impedance.
Most of the time you will see a balanced input with one opamp stage, using 4 resistors with the same value.
This means (IMHO) that the impedance of the input connected to the non-inverting input of the opamp is twice the impedance of the input connected to the inverting input. (Virtual ground.) I even found this in designs made by well known manufacturers!
 
I know this thread is old but...

If looking at channel 1 for example, if I want to attenuate signal and add gain can sub in a 20k pot as a variable resistor for R5 and be rocking?

Also... does the absence of a differential amplifier between the two signals still result in noise reduction or is this more of a conversion to unbalanced without the potential cancellation of noise?

CC
 
conleycd said:
If looking at channel 1 for example, if I want to attenuate signal and add gain can sub in a 20k pot as a variable resistor for R5 and be rocking?

The idea of this circuit was to make gain adjustable by changing just one resistor. You can replace R5/R11 and R14/R20 by a log potentiometer for variable gain/attenuation. Equations can be found in the block diagram.

conleycd said:
Also... does the absence of a differential amplifier between the two signals still result in noise reduction or is this more of a conversion to unbalanced without the potential cancellation of noise?

It is a differential amplifier (made of two op-amps).
 
Audiox, could you please explain the equations in the block diagram, I'm embarrassed to say not only do I not get them but I don't get them so thoroughly I don't even know what to google to figure it out for myself. 

For instance, on channel 3 I get that changing the value of r14 would change the gain, and I get the "x" divided by 10k part but it's been a long time since high school math, what's the story with 20 log?

thanks!
 

Latest posts

Back
Top