Provide offset voltage for non-inverting amplifier

[miketan3904]

Hi guys! Does anyone has any idea to shift the output voltage of an non-inverting amplifier up by 2V? How can I introduce an offset voltage to the non-inverting amplifier? I do not want to use another op amp to shift the output voltage.

[Zero999]

http://earmark.net/gesr/opamp/gain_offset.htm

[OM222O]

there doesn't exist a such thing.

For an inverting amplifier, you can create a summing amplifier which allows you to add offsets, there exists a non inverting summing amplifier, BUT it has many issues, so nobody actually uses it.

The only other solution that allows you to do that is using a differential amplifier which has an offset pin, but that requires "a different op amp" (assuming you have one already).

You can also chain a second op amp but again, that's a new op amp added.

[Zero999]

there doesn't exist a such thing.

For an inverting amplifier, you can create a summing amplifier which allows you to add offsets, there exists a non inverting summing amplifier, BUT it has many issues, so nobody actually uses it.

The only other solution that allows you to do that is using a differential amplifier which has an offset pin, but that requires "a different op amp" (assuming you have one already).

You can also chain a second op amp but again, that's a new op amp added.

Nonsense. It's fairly easy to add a 2V offset to a signal. Look at the site I linked to in my previous post.

[OM222O]

Nonsense. It's fairly easy to add a 2V offset to a signal. Look at the site I linked to in my previous post.

The website you provided shows absolutely no circuit / actual values?

It's easy to remove DC offset (AC coupling) but not to add it.

[Zero999]

Nonsense. It's fairly easy to add a 2V offset to a signal. Look at the site I linked to in my previous post.

The website you provided shows absolutely no circuit / actual values?

It's easy to remove DC offset (AC coupling) but not to add it.

No, it's easy to add DC offset. Enter the numbers in the boxes, click on the links and it'll generate the circuit and component values.

I don't need that website. I can design it in my head.

[OM222O]

the only one which it suggests for non inverting, is the summing one, and like I mentioned, that has several design issues.

[Zero999]

the only one which it suggests for non inverting, is the summing one, and like I mentioned, that has several design issues.

Please enlighten me of the design issues.

EDIT:
Here it is. I thought about using a 2V reference, but that's too easy: just a differential amplifier.

R3 & R4 form a potential divider. The values are selected to give 2V out, with 12V in. Obviously, the 12V needs to be well-regulated.

R5 needs to be equal to R3 & R4 in parallel. Two resistors could be used or a single one, like here.

R1 & R2 should be approximately R5. It doesn't need to be exact. It's normally a good idea to keep the input impedances matched, to minimise the effect of bias currents.

[David Hess]

Add the offset at the bottom of the feedback divider.

[oschonrock]

Add the offset at the bottom of the feedback divider.

+1 to that. Was going to write the same thing and this is very common when using single ended op-amps

Zero999 was also correct that the non-inverting summing amp is totally possible. It's just more complex to work out the resistor values for non-trivial situations (this one is trivial)  so most people avoid it and are "cognitively biased against" it..

If you want to be clear on the maths this guy is always brilliant I reckon:


His whole playlist on op-amps (or filters etc etc) is actually 2nd to none if you want the full blow mathematical design approach. Need good maths skills though, complex numbers the lot..

[Zero999]

Add the offset at the bottom of the feedback divider.

+1 to that. Was going to write the same thing and this is very common when using single ended op-amps

Yes, that'll work, but it requires a negative supply and only works for gains >1.

[TimFox]

Non-inverting feedback amplifiers made from op-amps must have a gain of +1 or more.  You would have to add a voltage divider in front of the amplifier to reduce the gain.
Inverting amplifiers can have negative gains with magnitude greater than or less than 1, with appropriate input and feedback resistors.

[oschonrock]

Non-inverting feedback amplifiers made from op-amps must have a gain of +1 or more.  You would have to add a voltage divider in front of the amplifier to reduce the gain.
Inverting amplifiers can have negative gains with magnitude greater than or less than 1, with appropriate input and feedback resistors.

Correct, but non-inverting summing amplifiers need a gain of say 2 (1+1)  for 2 balanced inputs to results in a "gain of 1 sum". See zero999's circuit and that video.

[oschonrock]

Add the offset at the bottom of the feedback divider.

+1 to that. Was going to write the same thing and this is very common when using single ended op-amps

Yes, that'll work, but it requires a negative supply and only works for gains >1.

Agree with gain of >1. But the O/P was a little vague about what gain he actually wanted, and he is using a non-inverting amp, so this is already the case no? If non-inverting meets the requirements without the offset then it still should?

Not sure the negative supply point is correct. Are you sure about that? We are not proposing to put -2V there. It should be +2V where ground is normally in a non-inverting amp. You might run out of headroom at the top in all cases, but that's entirely separate and we just don't have enough information about what the OP actually needs.

We are in danger of over analysing this. OP isn't even around any more ;-)

[Zero999]

A negative power supply is required for David's suggestion, because the bottom of the feedback divider goes to the  inverting input.

Add the offset at the bottom of the feedback divider.

[oschonrock]

A negative power supply is required for David's suggestion, because the bottom of the feedback divider goes to the  inverting input.

Add the offset at the bottom of the feedback divider.

Sorry, you're right. It was late. 

That trick with the elevated ground reference only works with inverting amps unless you use a negative refrence (2 minuses make a plus) and THEN you need a negative rail obviously. 

Well I am glad that's fully solved now. At least your interpretation of it, with a gain=1, which the original guy probably didn't/couldn't have meant because he couldn't have got that with his existing non-inverting amp   

Where is he?