This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hi team,
We can see that there is a super balanced circuit built with ne5532 in professional audio equipment.
Best Regards
Wesley Huang
Wesley,
This does make the input appear the same impedance to common mode. Both IN- and IN+ see 20k to ground for common mode. As input common mode changes, the op amp VCM is 1/2 of the input VCM. The gain is 1/2 (not 1 from a conventional diff amp with all resistors the same); For a difference input voltage the resistance is still 20k and equal and opposite input current flows.
This will not reduce signal to noise ratio because gain is 1/2 (less signal) and second op amp will add noise too. Limiting the bandwidth with a matching caps across R2 and R4 with help with total noise
As an alternative, you could buffer both inputs with two amp amps (high impedance for both, add a real resistor to set the impedance) then combine them with a third op amp.
Here is a TI-Tina circuit of your circuit (with my caps)
Hi Ron,
Thanks for your reply.
Compared with the traditional differential amplifier structure with R4 grounded, what are the advantages of this circuit? Because it seems that the traditional differential amplifier structure can keep the same impedance of in - and in +.
Best Regards
Wesley Huang
Wesley,
For the single amp input (R4 ground) the input impedance for common mode is 20k just like the super bal. However, for an input difference. IN+ doesn't get the same termination voltage as the super bal. So current for IN+ won't be in scale with IN- That is the difference.
Another way is a 3 amp diff amp that has high impedance inputs that can be used as is or you could add resistors to ground to make any input impedance that is desired.
IN+ is top left , IN- goes to bottom left
Hi Ron,
Thanks for your reply.
Why the IN+ doesn't get the same termination voltage as the super bal ? Could you please kindly explain that?
How does this benefit the CMRR?
Best Regards
Wesley Huang
Wesley,
Moving R4 connection from ground to an inverted output does make a difference when the output is not zero.
Tomorrow , I'll sim all three circuits and see what the results say about CMRR
Hi Ron,
Thanks for your reply.
Looking forward to your simulation.
Appreciate that!
Best Regards
Wesley Huang
Wesley,
I simulated with a RC4558 because I don't have a NE5532 model.
Below are the results for CMRR and noise. Noise was totalled from 1Hz to 10kHz.
The last two columns compared the results to the forward gain.
Circuit | CMRR | Noise BW 10kHz | Signal Gain | CMRR S/N | Noise S/N |
SuperBal | 102dB | 2.2uV | -6dB | 96dB | 4.4uV |
Diff Amp | 96dB | 3.6uV | 0dB | 96dB | 3.6uV |
3 AMP | 102dB | 3.7uV | 0dB | 102dB | 3.7uV |