Why does LM358 gives output when input voltage is Zero?

[Kkein]

Gents,
I'm using LM358 to amplify 50mV DC voltage, connected to 9V supply and non inverting mode. I have set my gain to 100 as I'm using the opam in two stage mode - the gain for the first stage is 10 and the second stage gain is variable upto 10. The problem is the opam always gives voltage even when the input voltage is zero. What is the problem? can somebody help out as I have thoroughly checked my circuit.?
Kind regards.

 

[ericgibbs]

Gents,
I'm using LM358 to amplify 50mV DC voltage, connected to 9V supply and non inverting mode. I have set my gain to 100 as I'm using the opam in two stage mode - the gain for the first stage is 10 and the second stage gain is variable upto 10. The problem is the opam always gives voltage even when the input voltage is zero. What is the problem? can somebody help out as I have thoroughly checked my circuit.?
Kind regards.

hi,
The LM358 output will not goto zero volts output with a single supply voltage, what do you measure as the lowest output???

Its possible by adding a resistor from the output to 0V to pull the output closer to zero volts.

Look here:
https://www.electro-tech-online.com/threads/lm358-significant-data-sheet-errors.125929/

 

[atferrari]

Gents,
I'm using LM358 to amplify 50mV DC voltage, connected to 9V supply and non inverting mode.

Please post the circuit.

 

[MrAl]

Hello there,

Kkein:
As the link Eric posted should show, a resistor to ground helps in most cases, although in some cases a diode is used also to drop the voltage as well as the pull down resistor. This leads to a little upper range loss but may not matter if you dont need the full top end output.

Eric:
Thanks for the link, but i noticed we have a real problem in that thread with missing attachments. That makes the whole thread almost worthless
Is that going to be fixed or is that a permanent problem now or should i bring this up in "Forum Issues" or whatever?

 

[ericgibbs]

Eric:
Thanks for the link, but i noticed we have a real problem in that thread with missing attachments. That makes the whole thread almost worthless
Is that going to be fixed or is that a permanent problem now or should i bring this up in "Forum Issues" or whatever?

hi Al,
I would suggest its highlighted in Site Issues.

E.

 

[Mike odom]

Also, with dual supplies, even with the two inputs tied to ground for a differential of zero volts, with a gain of 100 you could see as much as +/-700mV on the output, as the input offset voltage is 7mV (max) for a generic LM358.

 

[MrAl]

Hi,

Yes good point. If there is even a small input offset (and there always is) it could cause a large output differential even with 0v input. This is especially true with high gain like 100. 2mv (typical for this op amp) gets amplified to either plus or minus 0.2 volts. Usually it's positive too with a positive supply voltage.

 

[MikeMl]

...The problem is the opam always gives voltage even when the input voltage is zero. What is the problem? ....

There is no problem; it is not an OpAmp. You are expecting it to work like one...

By design, when both inputs are grounded, the output is supposed to sit at Vcc/2.

 

[MrAl]

There is no problem; it is not an OpAmp. You are expecting it to work like one...

By design, when both inputs are grounded, the output is supposed to sit at Vcc/2.

Hi Mike,

I think i understand how you can come to that conclusion, but i dont think i can agree with that. If we call vn the inverting terminal and vp the non inverting terminal for ease of notation, when vp=vn we have a 'no change' condition, where the output does not change one way or the other. So if it was 2v before than it is 2v after. We can have three conditions:
vp>vn, output ramps positively.
vp<vn, output ramps negatively.
vp=vn, output stays the same.

So in other words, when vp=vn we can show just about any output voltage is obtainable depending on the feedback and input to a standard amplifier circuit using an op amp for example.

Alternately, we might specify that when vp=vn we have 0v output, but then there is nothing to force that condition just like there is nothing to force the output to go to Vcc/2 either.

In the real life op amp however, when vp=vn externally, internally either vp>vn or vp<vn because of the small but significant input offset. In that case the output ramps according to
vp+voffset=vn

where voffset can be plus or minus. So the output either ramps up or down when externally vp=vn.

But a lot of times when someone says the input is at zero, they dont really mean the input to the op amp itself is zero but rather the input to the entire circuit is zero, such as when we have say an inverting amplifier with two resistors and the input is applied to the first resistor and the second resistor is the feedback. But yeah, a circuit schematic would help to clarify this which we'll have to wait for i guess

 

[MikeMl]

Hi Mike,

I think i understand how you can come to that conclusion, but i dont think i can agree with that.

Then you are disagreeing with what it says on the LM386 data sheet (Page 1 second paragraph):
...
The inputs are ground referenced while the output automatically
biases to one-half the supply voltage.

 

[ChrisP58]

Then you are disagreeing with what it says on the LM386 data sheet (Page 1 second paragraph):
...
The inputs are ground referenced while the output automatically
biases to one-half the supply voltage.

The part being discussed in this thread is the LM358, not the LM386.

 

[MikeMl]

The part being discussed in this thread is the LM358, not the LM386.

No point in being dyslexic unless you show it...

 

[MrAl]

No point in being dyslexic unless you show it...

Hi Mike,


And partially increnumericalexic too
I've done worse in the past

Yeah i think we are talking about the more general purpose op amps rather than an application specific one with special features.

 

[RogerAF]

Then you are disagreeing with what it says on the LM386 data sheet (Page 1 second paragraph):

You are confusing the LM386 Dual Preamp with the LM358 of this thread's subject. According to TI's copy of the original National datasheet the LM358 is part of a general group of dual Op Amps including: LM158, 258, 358, & 2904. All devices include an A suffix series in addition to the no suffix devices. They are in deed a dual op amp device.

 

[Mike odom]

that's why I'm a member of DAM... Mothers Against Dyslexia... (DAD just doesn't work the same)

 

[MrAl]

Hi,

The LM386 is an audio power amplifier isnt it?

Anyway, as most of you know this site had a crash awhile back which took out some images. EM is busy trying to restore those images, and i was able to manually restore my own images in the thread Eric originally linked to in this thread:
https://www.electro-tech-online.com/threads/lm358-significant-data-sheet-errors.125929/

which contains a more detailed explanation of the problems using the LM358 and how to effectively deal with them. He also linked to another note which helps too. This is a must read thread for using the LM358 and other op amps like the LM324. Just for reference because it has been also talked about, the problems associated with the LM386 are entirely different however.

 

[RogerAF]

You are correct, my mistake.

Something for reference notes:

LM38x series Audio Devices:
LM380 = 2.5W Pwr Amp = 14-pin DIP
LM381= Low Noise Dual Preamp = 14-pin DIP
LM382= Low Noise Dual Preamp = 14-pin DIP (Gain pins)
LM383 = 7W, Pwr Amp = T05B = TO220 5-pin package
LM384 = 5W, Pwr Amp = 14-pim DIP
LM385 = Adjustable Micro Power Voltage Reference TO92
LM386 = 0.5W Low Voltage Audio Pwr Amp = 8-pin DIP
LM387 = Low Noise Dual Preamplifier = 8-pin DIP
LM388 = 1.5W Audio Pwr Amp = 14-pin DIP
LM389 = 0.325W Low Voltage Audio Pwr Amp = 16-pin DIP

Wattages into 8 ohms.

A few of these are obsolete. check TI data; however, Jameco still carries many of them.

We won't get fooled again!