Wanted->Discrete balanced to unbalanced CMR converter

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Here is something I came up with:
 

Attachments

  • cmr1.png
    cmr1.png
    16.5 KB · Views: 789
Yes, I thought of that as well. My goal here was to use the least amount of silicon in the circuit. Also, aditional buffers may add DC drift to the outputs. As long as the VGS of Q1 & Q2 are well matched & the +/- supply is balanced, output should be biased very close to 0v.

BTW, the 180 ohm resistors should be changed to 470 ohm, otherwise their load will exceede 1/2 watt.
 
Brian Guralnick said:
Yes, I thought of that as well. My goal here was to use the least amount of silicon in the circuit. Also, aditional buffers may add DC drift to the outputs. As long as the VGS of Q1 & Q2 are well matched & the +/- supply is balanced, output should be biased very close to 0v.

Have you considered Nelson's Balanced Zen Preamplifier?

Look, Jam! No transformers! :D
 

Attachments

  • balzenpre.jpg
    balzenpre.jpg
    28.1 KB · Views: 715
Brian Guralnick said:
I've seen OP-Amp designs. I've also seen audio transformer designs.

I'm looking for the bare minimum discrete CMR circuit to convert balanced to unbalanced & new balanced signals.

--------------------------------------------------

Any circuit will affect sound quality. Why not just take signal from the balanced output directly. Coupling capacitors particularly affect the sonics however good.
 
Originally posted by fmak
Any circuit will affect sound quality. Why not just take signal from the balanced output directly. Coupling capacitors particularly affect the sonics however good.


If you take only the (+) and GND line from a balanced signal to make it unbalanced (as seen in many amplifiers with RCA and XLR connectors wired in parallel), your CMR will be zero.
 
Another good way would be this approach as used in the Accuphase C-290V. This circuit gives you both balanced and unbalanced outputs (unbalanced after Amp.1) and can restore a balanced signal with unequal amplitudes on (+) and (-) lines to perfect symmetry again. Also, if you short one side of the balanced output to ground, the circuit will continue to operate and the output voltage will be the same.

The "Amp" modules are discrete unity-gain opamps.
 

Attachments

  • c-290v_fig4e.jpg
    c-290v_fig4e.jpg
    9.8 KB · Views: 638
Steve Eddy said:


Have you considered Nelson's Balanced Zen Preamplifier?

Look, Jam! No transformers! :D


I've seen it before. In visualizing this circuit, if I were to tie the +in & -in together & apply a test signal between the GND & ‘incorrectly’ tied inputs, I believe that the outs would still leak the test signal, instead of being canceled out..

Have I visualized the circuit incorrectly?
 
Brian Guralnick said:
I've seen it before. In visualizing this circuit, if I were to tie the +in & -in together & apply a test signal between the GND & ‘incorrectly’ tied inputs, I believe that the outs would still leak the test signal, instead of being canceled out..

Have I visualized the circuit incorrectly?

The circuit is just a classic differential input pair. It's primary function is to amplify only the difference between the two inputs and reject anything that's common to both inputs.

How much common-mode signal gets through it will depend on the particular circuit. Nelson measured its CMRR at "slightly greater than 80dB." That's using its output balanced. If you use the output unbalanced, it only does about 20dB.

I wouldn't expect to be able to do much better than that if you're planning to use just two active devices in the circuit.

se
 
Steve Eddy said:


Nelson measured its CMRR at "slightly greater than 80dB." That's using its output balanced. If you use the output unbalanced, it only does about 20dB.

se

20db...... Bingo..... I'm trying to create a set of unbalanced outputs which zero to the local ground. Since my powersupply is shared between the left & right, & I wish to feed a unbalanced stereo amp, which also shares a common L&R ground, just wireing the -out on both L&R preamp channels to the GND & +out as my new audio signal would kill the intended function of this amp.

Using this circuit's GND instead of -out will only give me that 20db CMRR.
 
Brian Guralnick said:
20db...... Bingo..... I'm trying to create a set of unbalanced outputs which zero to the local ground. Since my powersupply is shared between the left & right, & I wish to feed a unbalanced stereo amp, which also shares a common L&R ground, just wireing the -out on both L&R preamp channels to the GND & +out as my new audio signal would kill the intended function of this amp.

Using this circuit's GND instead of -out will only give me that 20db CMRR.

Awwww screw it. Just grab a transformer and be done with it. :)

Sorry, Jam. I held on for as long as I could.

se
 
Well, I prety much gave up. I guess I will use a transformer.

Q1: Which companies make these high performance line matching transformers?

Q2: With a transformer, should I buffer the output?

Q3: Using a transformer would make life extra easy for integration into the Balanced Zen Pre-amp. However, I do not need the signal gain, would a simple voltage follower buffer make things any better?

Q4: Are there any worthy pre-amp output buffer stages which can drive headphones?
 
li_gangyi said:
Ok...let's assume he's a little cash strapped right now...what would you guys give him?? A cheap0 trafo or a well-designed circuit that could possibly beat that trafo in terms of pricing and performance...I'm not saying that trafos are bad...but given the costs...I would think twice...especially for a kid like me...

Oh ye of little faith. I knew he'd crack sooner or later. :)

se
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.