Non-Inverting Op-Amp Level Shifter & aplify

Thread Starter

CVMichael

Joined Aug 3, 2007
419
I have to amplify a video signal coming from a CCD array, and also to shift/offset the signal down by about 0.5V.

I found this site with a schematic: http://www.daycounter.com/Circuits/OpAmp-Level-Shifter/OpAmp-Level-Shifter.phtml

The problem is that I can't amplify with it... If I shift down, I loose signal strength, then I amplify it, and I'm back to square one, because if amplifies the "shift" also...

I am using an AD8041 op-amp for this.

I would like to use a potentiometer to manually adjust the shift, and also the amplification.

Can anyone help me with this please?

I attached a screenshot of the signal I get.
 

Attachments

Ron H

Joined Apr 14, 2005
7,063
1. How much gain do you need?
2. What is the required bandwidth?
3. What is your source impedance?
4. What power supply voltage(s) do you have available?
 

Thread Starter

CVMichael

Joined Aug 3, 2007
419
1. How much gain do you need?
2. What is the required bandwidth?
3. What is your source impedance?
4. What power supply voltage(s) do you have available?
This just shows how much electronics I know... (not enough)

1. I would like to amplify about 10 times the original signal. (is that a 10 gain ? :D)
2). Bandwidth? is that how fast i'm reading the pixels? right now about 800KHz...
3) Impedance? I have no idea... how do I find that out?
4) I use 5V power supply. But the ADC that is reading this signal goes up to 2.5V.
 

Ron H

Joined Apr 14, 2005
7,063
It looks like that positive pulse has a width of about 40usec. Do you also have to digitize that really short negative pulse that appears at about T+240usec?
 

Thread Starter

CVMichael

Joined Aug 3, 2007
419
Just to give you an idea of the kind of signal(s) I could get, I attached another pic where the camera is pointing at a source of light (at the ceiling). (Ignore the red circle for now).

And by the way, in this frame, there are 724 pixels (the CCD is actually 1024 pixels, but I don't need the beginning (100px) and end (200px) of it...)

[Edit]
This pic is taken with a long exposure (the shutter was open for a long time). What I need is to have the shutter open for a short time, and then amplify the signal... this is where I need your help.
 

Attachments

WBahn

Joined Mar 31, 2012
30,045
Your description is unclear because you say you want to do two things, amplify and shift, but you don't indicate what order you want to do them in. So do you want:

Vout = 10(Vin-0.5V) ?

Or do you want:

Vout = 10Vin - 0.5V ?

From the problem you said you had with the circuit you found amplifying the shift, I'm assuming you want the latter? Correct? But that is at odds with what people usually need, which is to remove a DC component in order to get the signal into the input range of their amplifier. So we need a bit of clarification.

You can make a single stage amplifier that does both without too much difficulty, or you might try taking a two stage approach and shift with one and amplify with the other. Ideally, you can do it in either order, but in order to stay within the linear range of your amp you will probably need to shift first.

The relation if you apply the gain and then the offset is basically:

Vout = GVin+Voffset

If you apply the offset first, then the you would need:

Vout = G(Vin+Voffset/G)

In other words, if you know you are going to be passing the thing onto an amp with a gain of 10, then only shift by -50mV and let the amplifier turn that into a shift of -0.5V.
 

Thread Starter

CVMichael

Joined Aug 3, 2007
419
The CCD array i'm using is powered at 5V, so it's output is up to 5V...
I would like to shift the signal by -0.5, and then amplify. But I don't mind if this can be done with one op-amp...

The problem is that I found that if I shift the signal down by half (let's say), and then I amplify by 2, then I get exactly the same ouput as if I don't do any shift and amplification.

What I need is to shift down, but when I amplify, I need the shift/offset to stay there!

On second thought, it looks like what I actually did was not shift, I actually divided the signal by 2 (or so), because that also got the signal peak to go down by half.
Then when I amplified, it just got it back the way it was before...

I edited the image in the first post to show you what the end result that I want. See attached.
 

Attachments

WBahn

Joined Mar 31, 2012
30,045
The only way your shift amount is going to remain unchanged after amplification is if it is shifted to exactly zero.

The figures aren't very helpful because there is no scaling information on them, so we don't know where the zero is, what the average value is, or what the min/max voltages are.
 
Top