some slight help needed

[narsil]

If you don't have an oscilloscope, so how would you know if your amplifier has crossover distortion?

Hey, any idea of how to answer this question? any way of calculations or any formula i can make use of?

 

[Nigel Goodwin]

If you don't have an oscilloscope, so how would you know if your amplifier has crossover distortion?

You listen to it!, at very LOW volume - crossover distortion only occurs at low levels, and is EXTREMELY distasteful!.

 

[audioguru]

Feed about a 400 Hz pure sine wave to an audio amplifier and you might hear the 3rd harmonic at 1.2kHz in addition to the 400Hz. If crossover distortion is there then it sounds like a high pitched buzz noise.
Then turn up the treble and cut the bass and the distortions will be much louder, and the level of the 400Hz will be reduced.
In my workshop I can position myself so that sound reflections from nearby walls cancel the 400Hz, leaving only the distortions to be heard.

A lousy old LM324 or LM358 opamp has tonnes of crossover distortion that is very obvious. They designed them to be low-power so starved their output transistors of sufficient bias current. Their narrow bandwidth helps attenuate the distortion. :lol:

 

[Nigel Goodwin]

A lousy old LM324 or LM358 opamp has tonnes of crossover distortion that is very obvious. They designed them to be low-power so starved their output transistors of sufficient bias current. Their narrow bandwidth helps attenuate the distortion. :lol:

Not to mention that the feedback applied around the chip drastically improves the situation! - both devices have been commonly used in audio equipment with no ill effects, nor excessive distortion.

 

[audioguru]

A lousy old LM324 or LM358 opamp has tonnes of crossover distortion that is very obvious. They designed them to be low-power so starved their output transistors of sufficient bias current. Their narrow bandwidth helps attenuate the distortion. :lol:

Not to mention that the feedback applied around the chip drastically improves the situation! - both devices have been commonly used in audio equipment with no ill effects, nor excessive distortion.

Correct, but only when the output is biased with a DC load resistor to operate in class-A, to eliminate the crossover distortion in the first place as recommended in the datasheets.
Single supply circuits usually have a DC load resistor.

I installed Goldstar (now called LG Electronics) made in Korea telephone systems in 1980. The telephone speakers were driven with a pair of complementary emitter-followers with their bases tied together, something like the output of an LM324 or LM358. They were driven by a pretty good opamp with overall negative feedback. The crossover distortion caused many complaints. :cry:
I added a resistor from the output of the opamp to the speaker so that the opamp drove the speaker directly at low levels and the transistors began working for higher levels. Luckily, the speakers were 63 ohms so the opamp could drive them to about 0.7V pretty well.
When users heard the big improvement I couldn't modify them all fast enough. :lol:

 

[Nigel Goodwin]

Interestingly enough, the datasheet says:

To reduce the power supply drain, the amplifiers have a
class A output stage for small signal levels which converts to
class B in a large signal mode.

And:

For ac applications, where the load is capacitively coupled to
the output of the amplifier, a resistor should be used, from
the output of the amplifier to ground to increase the class A
bias current and prevent crossover distortion.

Where the load is directly coupled, as in dc applications,
there is no crossover distortion.

Although I'm rather dubious about no crossover distortion!.

Certainly you see the LM324 and LM358 used in huge amounts of audio equipment, with no crossover problems.

 

[Optikon]

A lousy old LM324 or LM358 opamp has tonnes of crossover distortion that is very obvious. They designed them to be low-power so starved their output transistors of sufficient bias current. Their narrow bandwidth helps attenuate the distortion. :lol:

Not to mention that the feedback applied around the chip drastically improves the situation! - both devices have been commonly used in audio equipment with no ill effects, nor excessive distortion.

Correct, but only when the output is biased with a DC load resistor to operate in class-A, to eliminate the crossover distortion in the first place as recommended in the datasheets.
Single supply circuits usually have a DC load resistor.

I installed Goldstar (now called LG Electronics) made in Korea telephone systems in 1980. The telephone speakers were driven with a pair of complementary emitter-followers with their bases tied together, something like the output of an LM324 or LM358. They were driven by a pretty good opamp with overall negative feedback. The crossover distortion caused many complaints. :cry:
I added a resistor from the output of the opamp to the speaker so that the opamp drove the speaker directly at low levels and the transistors began working for higher levels. Luckily, the speakers were 63 ohms so the opamp could drive them to about 0.7V pretty well.
When users heard the big improvement I couldn't modify them all fast enough. :lol:

The resitor idea has been around for a long time.. I believe it's called "current dumping" in audio literature. In the situation where the opamp cannot drive the load even at low levels (like say for a 4 Ohm or lower load), I've used constant current diodes there to drive as hard as it can while not short ckt current limiting the opamp. That alone is usually enough to make any distortion much improved.

 

[ThermalRunaway]

As I understood it, if you were to use 100% negative feedback this would give a closed loop gain of 1 for the Op-Amp, which would significantly reduce the cross-over distortion, pretty much to a negligable level. There would still be some cross-over distortion present due to the small time it would take for the op-amp to move it's output from +0.7V to -0.7V. The faster the Op-Amp that you use, the further the cross-over distortion is reduced.

Brian

 

[audioguru]

On an LM324 or LM358 circuit, a load from the output to the negative supply bypasses the output PNP transistor, causing the output NPN darlington transistor to do all the work in class-A with no crossover distortion.

There is a 50uA current sink from the output to the negative supply. Therefore the output NPN transistor does all the work in class-A when the load is direct coupled and connected to the negative supply, or when the load is capacitively coupled to a very high resistance load. The 50uA sink must be increased by adding a resistor from the output to the negative supply so that the output NPN transistor can continue operating in class-A with a lower resistance load.

The crossover distortion is caused by an NPN darlington transistor pulling-up the output and a PNP transistor pulling down, with their bases tied together for low idle current. Therefore the output has about a 1.8Vp-p to 3.0Vp-p dead zone.
Negative feedback doesn't help reduce the crossover distortion very much because the IC has a narrow bandwidth and an open-loop gain of only about 60 at 10kHz. If your circuit has a gain of 200 like an LM386 power amp with its gain-boosting capacitor, then there isn't any negative feedback in an LM324 or LM358 above about only 2kHz. :lol: