45mVpp 50hz AC to 0-5V DC how?

Im looking at building an op amp circuit which takes a small AC voltage and outputs a DC voltage for an AD Convertor. I've done a search but I don't really understand the theory that well and how to pick the correct op amp.

Input:
45mVpp
50Hz

Output:
0-4.096v (4-5v will do) DC

I've designed the following circuit in multisim using filter pro for the circuit, however when I do it in altium with real components it doesn't seem to work to well.
**broken link removed**

So I guess Im asking what the best way to create this circuit is, and how do I pick the correct op amp for the specification? we haven't picked a power supply yet so anything from +-5 - 15v should be ok.

You need two parts to the circuit, first an amplifier (which you already have a poor confusing example of), follow that with a precision rectifier circuit.

As you're amplifying AC only you need an AC amplifier not the DC one you've shown, check the sticky at the top of one of the forums for correct examples of how to connect an opamp amplifier.

I would suggest an MC33202 opamp for that one. Output swings rail-to-rail, input near ground sensing, and you can drive it off a single +5v supply after you find Audioguru's schematic.

hi rolls,
The type circuit you require will be a version/adaptation of this image.

Why use a precision rectifier instead of just adding a dc offset? wouldn't that be more accurate? Also should I be using a filter for any noise or that a waste of time?

edit: Whilst I have one recommendation for the op amp I'd really like to know how to pick them myself, what sort of specs should I be using to search for them?

rolls said:

edit: Whilst I have one recommendation for the op amp I'd really like to know how to pick them myself, what sort of specs should I be using to search for them?

Click to expand...

Some pertinent op amp characteristics:

Gain-bandwith product (GBW) - This determines the gain and bandwidth the circuit can have. If you divide GBW by the circuit gain, you get the maximum bandwidth the circuit can have.

Input bias current - This determines the amount of output DC offset you will get with the input and feedback resistors you use. You multiply the input bias current by the input resistor and the circuit gain to get the output offset voltage with zero input.

Input offset voltage - You multiply this offset by the DC circuit gain to determine the amount of DC output voltage you will have with zero input. You add this to the offset from the input bias current to get the worst-case total output offset.

Operating voltage - Determines the power supply voltage(s) you need.

Operating current - Determines how much power the device uses.

Slew Rate - This is how fast the op amp can swing the output voltage. It often limits the voltage swing at higher frequencies. For example the slew rate of a sine-wave is 2Pi x Vp x f where Vp is the peak voltage and f is the frequency of the sine wave. When this value equals the op amp slew rate you can go no higher in frequency or amplitude without distortion at the output.

Input/Output Type - Rail-to-rail can have inputs and outputs at the supply rails. They are generally what you want if you only have one supply voltage. Non rail-to-rail usually can only operate within 2 or 3 volts of the supply rails. They are more typically used with dual power supplies.

There can be other characteristics the may be important in some applications, such as the input noise, but the above covers the basics. Often you have to do a trade-off between characteristics for a particular application. That's why there's no "one size fits all" in op amps.

rolls said:

Why use a precision rectifier instead of just adding a dc offset? wouldn't that be more accurate? Also should I be using a filter for any noise or that a waste of time?

edit: Whilst I have one recommendation for the op amp I'd really like to know how to pick them myself, what sort of specs should I be using to search for them?

Click to expand...

hi,
If you use a dual supply of say +/-9Vdc to the OPA, with a 50Hz signal input of 45mV and output of +4.096Vdc
any general purpose amplifier would get the job done. I use the CA3240.

What do you mean by just adding a dc offset?.
If its to remove the forward voltage drop of the half wave rectifier diode, won't that drop change with forward diode current.???

a precision rectifier is just what it says, precision.

ericgibbs said:

What do you mean by just adding a dc offset?.
If its to remove the forward voltage drop of the half wave rectifier diode, won't that drop change with forward diode current.???

Click to expand...

I mean using a differential amplifier and instead of hooking one terminal to ground you use a voltage divider for wherever you want the output to swing around. Eg this: **broken link removed**

I guess what Im asking is what will be more accurate, using a precision rectifier and plugging that straight into and ADC, or using the differential amplifier I've drawn in the above picture and plugging it into an ADC and averaging the voltage over say 5 cycles via software.

rolls said:

I mean using a differential amplifier and instead of hooking one terminal to ground you use a voltage divider for wherever you want the output to swing around. Eg this: **broken link removed**

I guess what Im asking is what will be more accurate, using a precision rectifier and plugging that straight into and ADC, or using the differential amplifier I've drawn in the above picture and plugging it into an ADC and averaging the voltage over say 5 cycles via software.

Click to expand...

You need to rectify it - if you don't you can't easily measure the level.

You would also do better to post your pictures here, rather than on a third party site where they will disappear.

rolls said:

I mean using a differential amplifier and instead of hooking one terminal to ground you use a voltage divider for wherever you want the output to swing around.
I guess what Im asking is what will be more accurate, using a precision rectifier and plugging that straight into and ADC, or using the differential amplifier I've drawn in the above picture and plugging it into an ADC and averaging the voltage over say 5 cycles via software.

Click to expand...

How will you time the precise sampling point over the 5 cycles.?
Or
do you plan to sample at high speed testing for the peak of the 50Hz sinewave.?

ericgibbs said:

How will you time the precise sampling point over the 5 cycles.?
Or
do you plan to sample at high speed testing for the peak of the 50Hz sinewave.?

Click to expand...

That was the plan, however if the signal is changing fast then it might be too difficult to do reliably, I'd really like to test out both designs though, see what gives us better accuracy.

Could someone point me in the right direction for designing a precision full-wave rectifier and how to choose the capacitor values for it?

edit: I gave this a shot and am using 10k for R, what should I use for the cap?

rolls said:

That was the plan, however if the signal is changing fast then it might be too difficult to do reliably, I'd really like to test out both designs though, see what gives us better accuracy.

Could someone point me in the right direction for designing a precision full-wave rectifier and how to choose the capacitor values for it?

edit: I gave this a shot and am using 10k for R, what should I use for the cap?

Click to expand...

hi,
I dont see a R or C which components are you referring too.?

ericgibbs said:

[/B]

hi,
I dont see a R or C which components are you referring too.?

Click to expand...

Sorry the image didnt upload properly, here it is now.
https://www.electro-tech-online.com/attachments/esp01-jul-31-gif.21299/
This pic is what I based it on.

rolls said:

Sorry the image didnt upload properly, here it is now.
This pic is what I based it on.

Click to expand...

hi,
Look at these two images for ideas and component values.

Thanks for that ericgibbs you have been very helpful, I will have a play over the next few days and report back if I have any issues.