Are op-amps really superior to transistors for an audio tone control?

Question

I'm now working on a new hobby project, an audio amplifier using TDA2050 ICs in bridge configuration, to be fed by phone or PC earphone output. I already made the amplifier, it sounds fantastic. Two of these per channel.

When I started looking for tone control circuits, I find those using op-amps like 4558, 741, etc. and those using general purpose transistors.

Beyond any underlaying economical or simplicity reasons, is there a real overwhelming benefit on choosing op-amps over transistors for tone controls for HIFI home audio amplifiers?

As examples.

This:

VS. this:

Answer 1

Yes, likely you can buy better op-amps for Hi-Fi than you can make yourself from multiple transistors. And likely op-amps are better than just a single transistor amplifier, but it might suffice if you don't require much performance.

Also, op-amps like 741 or 4558 are made of transistors, so in the end they are also just a bunch of transistors, better than a single transistor. Before op-amp ICs were available, op-amps were modules made of discrete transistors and other components. Before transistors were available, op-amps were even larger modules made with thermionic valves (electron tubes).

Although if you want Hi-Fi like you say, you might want to use other components in place of TDA2050 or ua741, better components exist.

Answer 2

Now that all of the (1968) 741 thumping is out of the way,,,,,

An opamp is superior because of how it interacts with the frequency-modifying networks. Compared to any single bipolar transistor circuit configuration, an opamp will have a nearly perfect zero-ohm output impedance (critical for driving feedback networks), and either a nearly-infinite impedance non-inverting input or a nearly perfect virtual-ground inverting input.

All of that means that the tone control networks will behave exactly as the math predicts. The driving-point impedance seen by the lower schematic R2 and C4 is a significant fraction of the impedances of those components, and will change as a function of signal amplitude. This will cause non-ideal frequency-shaping effects, and interaction between the two control settings.

Similarly, the upper schematic R3-C3 node is tied to a virtual GND, so there is no interaction between the bass and treble settings. The Q1 input impedance is not nearly low enough to prevent this.

Answer 3

A modern audio opamp is made for and used for hifi. It has many transistors for extremely high gain and bandwidth that creates very low distortion and full audio bandwidth.
Here is the awful distortion of a single transistor similar to the old tone controls circuit you found:

Answer 4

Either is, of course, the standard Baxandall tone control circuit that essentially zeroes a frequency-dependent adjustable bridge between signal and negative feedback from an inverting amplifier.

The single-transistor version has no dependable amplification factor, limited linearity, variable noise performance, considerable quiescent current that drives noise via heating components.

The opamp version, in contrast, has quite constant design parameters. It will inherently have crossover distortion but will go to considerable lengths to obviate its effects, and in connection with bypass capacitors there are well-documented ways to get rid of it.

Noise performance will be good, quiescent current for low-noise low-distortion operation will usually be considerably less than the single-transistor version.

You wouldn't use transistors from the 70s, and of course you wouldn't use the 741. You can pick opamps that fit your use regarding the specs. Inverting configuration means that you need not worry about common-mode distortion, so you can pick a JFET opamp like the OPA2134 that has no specific common-mode distortion compensation (like OPA1642 has) that will have very high input impedance (and thus not sap the Baxandall network operation), no relevant input current noise given the variable Baxandall network impedances, and input voltage noise of 8nV/√Hz which corresponds more or less to the noise of a 3.2kΩ resistor which is small in relation to the other impedances (one might lower all impedances for less overall noise).

You can also pick other opamps with different noise figures, input impedances, slew rates and so on and match the circuit to it (with regard to a tone network, it would be customary to add a small capacitor as direct negative feedback for high frequencies that cuts off amplification outside of the audible range).

In short, using opamps allows you to design your circuit in terms of directly audio-relevant parameters and design constraints with high precision and dependable results.

And on-chip compensation networks and laser-trimmed components allow for a precision and functional simplicity that you just don't get with isolated components.

Answer 5

"Are op-amps really superior to transistors for an audio tone control?"

Well, the answer depends on how you define the "superiority". As tone controls are there to modify the signal, or to put it differently to wilfully "distort" the signal, it depends on if you like the resulting effect better or not.

In hifi circuits (the people, not the electronics) intentional distortion is sometimes used to say that, as example, a tube amplifier makes the sound warmer. In studio recording "intentional distortion" is used a lot, all from tone controls and various compressors and echos and so on, in order to "heighten the experience".

It all comes back to what you want to achieve by "distorting the signal". If the purpose is to "heighten your experience" use whatever sounds best to your ears (never mind other peoples ears). If the purpose is to "somewhat mitigate errors in other parts of the sound chain" you are probably better off using op-amps than using a few transistors, or even better might by using a signal processor using FFT or other algorithms to modify the sound (so called room correction as available in some speaker systems).