[zhoufang]

I'm designing a headamp similiar to MINT
As I'm considering how to bias the op-amps to class A, the idea come to my mind:
why not use a simple resistor for biasing ground channel op-amp to class A?
it can save me a little money.

since the output level from the ground channel op-amp is always constant at ground level, the voltage difference between it and the V- should be constant (ignoring to load modulated ripple on power rails), and by ohm's law so should the bias current through the resistor.

So my question is: is there any disadvantages by doing so compared to using a JFET or CRD solution.

 

[djwkjp]

http://www.tangentsoft.net/audio/opamp-bias.html

tangent does a good job of explaining things here.

 

[zhoufang]

deleted due to double posting.

 

[zhoufang]

Thanks for the info, but I have already read though that article.

What make me wonder is that if a resistor works just as well in ground channel, why do we see designs that use CRD/JFET based solutions which cost more and/or take up more PCB space.

The 1st problem of unconstant current stated inside that article does not apply to the groud channel op-amp.

I'm not quite sure about the second problem of op-amp's load impedance. If a CRD does the same thing as a resistor (i.e. drawing a constant current), to the op-amp shouldn't it looks to be of the same impedance as that resistor?

 

[amb]

Quote:

Originally Posted by zhoufang If a CRD does the same thing as a resistor (i.e. drawing a constant current), to the op-amp shouldn't it looks to be of the same impedance as that resistor?

The CRD is a constant current source that presents a very high impedance as seen from the output of the opamp. It is not the same as a resistor. The resistor is a passive device and the current flowing through it can be described by simple Ohm's Law. The CCS is an active device and the simple Ohm's Law does not apply.

 

[zhoufang]

Quote:

Originally Posted by amb The CRD is a constant current source that presents a very high impedance as seen from the output of the opamp. It is not the same as a resistor. The resistor is a passive device and the current flowing through it can be described by simple Ohm's Law. The CCS is an active device and the simple Ohm's Law does not apply.

Thanks!
I think I can get what you means.
I'm not sure how much difference a 10K ohm load impedance will make for a op-amp?
Can I say that the groud channel op-amp don't need to be of very high performance anyway since it only act as a buffer between TLE2426 and BUF634?

 

[amb]

Quote:

Originally Posted by zhoufang I'm not sure how much difference a 10K ohm load impedance will make for a op-amp? Can I say that the groud channel op-amp don't need to be of very high performance anyway since it only act as a buffer between TLE2426 and BUF634?

While the voltage output of the ground channel is fixed to zero at all times, it is not a dead circuit. It sources and sinks the return current from the two other channels. For best performance the ground channel opamp and its buffer should be just as high performance as the L and R channels.

 

[zhoufang]

Quote:

Originally Posted by amb While the voltage output of the ground channel is fixed to zero at all times, it is not a dead circuit. It sources and sinks the return current from the two other channels. For best performance the ground channel opamp and its buffer should be just as high performance as the L and R channels.

This makes sense to me.
Thanks for the explanation.