LM358 output voltage non-zero even though input is zero [closed]

I tried this experiment with LM358 and LM741. I used the circuit mentioned from this question:

I have not connected any input source. But I am getting an output of 7.5V. I used R1 as 130 Ohm and R2 as 1K. VCC is 9V battery. I was expecting the output as zero. Connecting the input to a source (another 9V or 1.5V battery) did not make any difference in case of LM358. With LM741 I get the output as 1.5V, when my input is connected to 1.5V source.

Can anybody help me?

I am experimenting this for a simple voltage amplifier where the input source is around 100mV max. I needed output as 0-12V.

• What happens when the input is grounded? – markrages Apr 2 '14 at 18:06
• Your schematic does not even match your own description. -1 – Olin Lathrop Apr 2 '14 at 18:10
• Olin... Its the same except that values are different. – user1827678 Apr 3 '14 at 5:23
• @user: Exactly, which makes it different. To decipher the circuit, one has to read your text and then mentally replace values clearly marked on the schematic with different ones. That's a hassle. There are many ways to post a schematic here, including the built-in schematic editor. There is simply no excuse for posting the wrong schematic. – Olin Lathrop Apr 3 '14 at 12:22

Without an input, pin 3 will find its own level and the output could be anything between 5mV and about 7V from a 9V Vcc. OK that's based on the spec and you say +7.5V - I'm not surprised.

When you connected the input to 1.5V, the gain of the op-amp circuit being about 8.7 will still mean the output voltage is end-stopped at 7.5V.

With a 741 - you can't use inputs close to the negative or positive rail so this would be your problem here. The spec implies you need rail clearance of at least 2V. It's the same with the output of a 741 - depending on the load (and yours is actually the feedback resistor) the output might get to within 1 or 2 volts of the power rails but no better.

As a footnote for 741s - don't bother using these any more - get your self a stock of decent rail-to-rail op-amps and, if you also need to run stuff in a quad package on a split supply (such as +/-12V) and you don't need MHz performance you'll not find a better device than the OP4177. It doesn't do rail to rail but for low frequency stuff it beats 741s and stuff like that into oblivion (not cheap though!!).

• Thank you very much. I understood my mistake and I solved it. – user1827678 Apr 5 '14 at 15:56

There is no guarantee what a opamp will do with one of its inputs floating.

Your schematic shows a gain of 2, but your description implies a gain of 8.7. If the gain really is 8.7 (this kind of ambiguity is annoying in a question since we have to waste time dealing with it), then putting in 1.5 V should result in 1.5 V x 8.7 = 13 V. The opamp can't produce that with a 9 V supply, so the output goes as high as it can make it.

Try a reasonable input voltage that should result in a output within range and see what you get. For example, let's aim for 3 V out. 3 V / 8.7 = 345 mV. Try putting that in and see if you get about 3 V out.

• Hi.. I used the schematic mentioned in the link as an example. Actual values are different and that is what I mentioned in the text. With this experiment, I am just tryig to find if I can make an op-amp circuit myself. I will use right values of R1 and R2 when for my actual input source. I will try your suggestion. Thank you very much for the answer. – user1827678 Apr 3 '14 at 5:26

Well, you have a gain of 1+1000/130 = 8.7. So connecting the input to either 9V or 1.5V will rail the output. The LM358 allows single supply operation and sensing near ground, but if you leave the input floating you are also likely to have the output go to the rail. (You might be better off with dual supplies.) Did you tie the non-inverting input to ground and still observe 7.5V on the output? If so, you likely have a wiring error. Do you have bypass caps on the power supply close to the IC?

With a gain of 8.7 and an input of 100mV you will get 870mV out, not 0-12V. You will need a lot more gain, meaning offsets and bandwidth limitations should be checked.