precise peak detector

[nima_1981]

Hi friends ,
I have 100 mV to 3 V Ac signal with 700 Hz frq and i should to measure the rms value with Mirocontroller (Avr -BasCom) then i use this circuit but that is very big error that you can see in attached picture , please help me for find best method to convert Peak Voltage to Dc value .
Thanks

 

[Warpspeed]

Your problem may be due to the very heavy loading on the output (10uF + 10K). The op amp will struggle to keep that load charged right up at the actual peak.

Try it with a much smaller capacitor, say 100nF and a 1 meg load, the time constant will be the same.

If the real output load must be 10K, then fit a unity gain buffer on the output of the above circuit.

 

[nima_1981]

Thanks Warpspeed
I try your suggestion Cap & Res but i have not Dc now ? see this picture

 

[Warpspeed]

You are using a x1 probe which is also loading the circuit.
Try using a x10 probe and remove the 1 meg load resistor from the circuit.

 

[Audioguru]

Remove R1 that is quickly discharging the 100nF capacitor C1. Then the 1M for RL will slowly discharge the 100nF capacitor.

 

[nima_1981]

You are using a x1 probe which is also loading the circuit.
Try using a x10 probe and remove the 1 meg load resistor from the circuit.

with X 10 probe and without 1MG Ress i have this

 

[Audioguru]

Did you remove the useless 10k resistor R1?

 

[nima_1981]

Did you remove the useless 10k resistor R1?

Yes i remove all resistor

 

[KlausST]

Hi,

Use the scope and measure directely the output of the opamp. Then you see what's happening.
I recommend to use a second diode from output to inverting input. Cathode to the output.

This should prevent the opamp output from saturation.

Maybe this can solve the problem. But I'm not sure.

Klaus

 

[nima_1981]

Hi,

Use the scope and measure directely the output of the opamp. Then you see what's happening.
I recommend to use a second diode from output to inverting input. Cathode to the output.

This should prevent the opamp output from saturation.

Maybe this can solve the problem. But I'm not sure.

Klaus

Dear i use another diode and use oscope for directly measurement , but as you see in picture they are not equal , did you have another method for convert sin wave to dc for measurement?

 

[crutschow]

It's not desirable to use peak detection to determine the RMS value of a sinewave, due to its sensitivity to small amounts of noise and waveform distortion.
Here's a precision full-wave rectifier circuit with an output 2nd order filter that has its gain selected to give a DC output equal to the RMS value of the waveform average value.

(But if you need the true RMS value then you could take many samples of the waveform (20 or more per cycle) using the processor A/D converter, and then calculate the root-mean-square value of the samples.)