?#%&* lm358

[Vizier87]

Hi guys, the title says it all... my projects with LM358s as the chosen op-amp is always frustrating.
Now, my network will contain motors and some noisy stuff, so which part ID is the most suitable as an alternative to LM358s?

 

[Speakerguy]

There are literally tons. What are your supply voltages? What is your price limit per dual op amp?

 

[audioguru]

The lousy old LM358 is one of the first low power dual opamps ever made. Its opamps are the same as the ones in a lousy old LM324 (it has four).
They are bad for audio because they have crossover distortion (3%) and are noisy. Their max frequency response is about only 2kHz at high output levels.

The MC33171 single, MC33172 dual and MC33174 quad opamps have exactly the same low power, operate from a supply as low as only 3V and have inputs that work at the negative supply which can be ground. The output also goes to ground. But they have no crossover distortion and have a bandwidth to 35kHz.

The MC34071 single, MC34072 dual and MC34074 quad opamps have the features of the MC3317x opamos but have normal supply current and a response to 100kHz.

I don't know what goes wrong with your network with motors and stuff.

 

[bountyhunter]

Hi guys, the title says it all... my projects with LM358s as the chosen op-amp is always frustrating.
Now, my network will contain motors and some noisy stuff, so which part ID is the most suitable as an alternative to LM358s?

That's kind of like asking which wine tastes best.

You have to define the operating conditions and what you are trying to do to select an op-amp.

 

[Vizier87]

welll... simple operations regarding amplification didn't work, and I did a little bit fancier stuff with my thermal anemometer using TO-92 transistors as sensors... they didn't work too...
Point is that I've heard some grumbles regarding 358s and want to confirm it. Well I got my answer.
Thanks audioguru.

 

[Vizier87]

Anyway here's some layout: 5V reference input and ground fed by the LM7805 to the op-amp, amplifying signals from multiple sensors (7 to be exact) which are analog outputs fed into a PIC18F4455, and digital outputs sent via RC modules, with 4 motors for mobility of the proto. And oh, there is a possibility a metal detector will be in proximity too.

I'd like extra suggestions, thank you.

 

[audioguru]

Post your schematic for us to see why it does not work.

 

[picbits]

I have over 2000 LM358 opamps here and use them for the majority of my project.

Not had a problem with one yet.

 

[Speakerguy]

We need a schematic. They might work crappy but they should be working, unless something is terribly wrong.

 

[Vizier87]

Wait for a few days then..... I'll need to REALLY test the 358s again... picbits might be right..... I might have made a mistake.

 

[bountyhunter]

welll... simple operations regarding amplification didn't work, and I did a little bit fancier stuff with my thermal anemometer using TO-92 transistors as sensors... they didn't work too...
Point is that I've heard some grumbles regarding 358s and want to confirm it. Well I got my answer.
Thanks audioguru.

The truth is the LM358 and quad version LM324 have probably shipped a billion pieces over the years. They are a good choice for an op amp if you need one that allows the input voltages to go all the way down to ground (or the ngative rail). Their output also swings all thw way down to ground which is not a common feature. They are unity gain stable and pretty tolerant of capacitive loading. They are not a good choice for audio amps or wide band signal processing but they were never intended for that. I have used them dozens of times for various projects.

 

[bountyhunter]

I have over 2000 LM358 opamps here and use them for the majority of my project.

Not had a problem with one yet.

I would wager they are the most commony used op-amp in electronics.

 

[audioguru]

The LM358 and LM324 opamps have a low bandwidth of only 2kHz and 3% of crossover distortion.
The Motorola (ON Semi now) MC33171 single, MC33172 dual and MC33174 quad opamps have exactly the same low supply current, exactly the same minimum supply of 3V, exactly the same inputs that work at the negative supply voltage that can be ground and exactly the same output that goes very close to the negative supply voltage that can be ground.

But they are much better because they do not have any crossover distortion and have a bandwidth to 35kHz. They also have a max supply voltage of 44V which is much higher than the max of only 32V for the National Semi opamps.

 

[bountyhunter]

The LM358 and LM324 opamps have a low bandwidth of only 2kHz and 3% of crossover distortion.
The Motorola (ON Semi now) MC33171 single, MC33172 dual and MC33174 quad opamps have exactly the same low supply current, exactly the same minimum supply of 3V, exactly the same inputs that work at the negative supply voltage that can be ground and exactly the same output that goes very close to the negative supply voltage that can be ground.

But they are much better because they do not have any crossover distortion and have a bandwidth to 35kHz. They also have a max supply voltage of 44V which is much higher than the max of only 32V for the National Semi opamps.

Like I said, the LM358 was never intended to be an audio amp and it's also a design that is probably 30 years old. There are now much better op amps available, but they cost more as well. For general purpose work, the 358 has been a pretty solid workhorse.

The fact that general purpose op amps don't do audio well is one of the reasons Signetics came out with the NE5532 in the early 80's. Now there are lots of good audio grade amps.

Interestingly, the first CD players marketed way back when sounded like crap because they were using 741 type op amps in the audio signal path. The moral is never let digital designers do analog circuitry....