Measuring Sine wave accurately 500KHz

[adnan012]

hi,

I have source of fixed 500KHz sine wave. It has a constant amplitude of 300mV PP.I want to make a circuit to monitor the amplitude precisely and to detect any change in the amplitude if occurred. The change might be in few tens of micr-ovolt. What is the best possible solution.

Regards

 

[KlausST]

Re: Masuring Sine wave accurately 500KHz

Hi,

You are not the first one with this problem, therefore there are a lot of discussions about this topic: precise AC measurement.
--> read thorough them. The forum search is for free. ;-)

The first you have to decide:
Which value are you interested in?
Just the peak values, or the RMS value or just the fluctuations.
I assume no one can answer this for you. Its your application ... and every application has it´s own demands.

Klaus

 

[adnan012]

Re: Masuring Sine wave accurately 500KHz

Thanks for reply

The expected change in amplitude is no more than 5mv. And it is a fluctuation for 1 to 2 second . I am interested in peak value only.

- - - Updated - - -

If i can detect a change then it is also sufficient.

 

[KlausST]

Re: Masuring Sine wave accurately 500KHz

Hi,

then I recommend to use peak detector circuits.
This makes the ADC part simple, because you may attach a suitable low pass filter between peak detector and ADC.
ADC reading can be slow.

Mind that every noise at your signal will cause the same amount of error at the output.
Maybe consider to use low pass filters before the peak detectors to focus on the 500kHz signal.

Peak detectors give an absolute output value, thus you are free to use "the absolute value" or the "fluctuations" .. it´s just a (simle) question of software.

Klaus

 

[BigBoss]

Re: Masuring Sine wave accurately 500KHz

Since the frequency is fixed, the peak amplitude can be synchronized with a ADC to measure the peak values only in each cycle.
Similarly, if the derivative is taken of this signal, observing the zero crossing value will give the same effect.

 

[KlausST]

Re: Masuring Sine wave accurately 500KHz

Hi,

Similarly, if the derivative is taken of this signal, observing the zero crossing value will give the same effect.

True only for a noise free undistorted sine signal.

Klaus

 

[adnan012]

Re: Masuring Sine wave accurately 500KHz

Thanks for reply.

In spectrum analyzer we can change scale according to the expected input level. I my case i have a very small change in amplitude, is there any possibility to use some reference and cancel out large amplitude?

 

[FvM]

Re: Masuring Sine wave accurately 500KHz

I don't expect that direct peak detector is a good solution, too high noise bandwidth. Better use small band level measurement (e.g. lock-in amplifier or down-mixer and demodulator) and peak detector in the base band.

A spectrum analyzer in zero sweep mode with small filter bandwidth can be considered as model for optimal signal processing. Lock-in amplifier or similar method with I/Q processing is even better because it only sees the in-phase components.

 

[KlausST]

Re: Masuring Sine wave accurately 500KHz

Hi,

you talk about amplitide of 300mV pp and an amplitude change of 5mV.

even with a 10bit ADC - without additional signal conditioning circuit - you should be able to detect down to 0.3mV.

For sure any simple OPAMP with offset correction and suitable gain may improve resolution.

I don't expect that direct peak detector is a good solution, too high noise bandwidth.

I agree that the "peak" detector solution is the one solution which is prone to errors.
But the "lock-in-amplifier" solution involves the complete waveform and not only the peak, as the OP requested.

Klaus

 

[adnan012]

Re: Masuring Sine wave accurately 500KHz

Thanks for reply.

I am looking for look-in amplifier. Is it necessary to have reference and received signal in-phase?

 

[KlausST]

Re: Masuring Sine wave accurately 500KHz

Hi,

Is it necessary to have reference and received signal in-phase?

Yes, this is the working principle of a Lock In Amplifier.

Klaus

 

[adnan012]

Re: Masuring Sine wave accurately 500KHz

But how can i generate in phase reference generator of same frequency?

 

[FvM]

Re: Masuring Sine wave accurately 500KHz

You'll preferably use a PLL. The other option, small band demodulation without phase selectivity corresponds to a spectrum analyzer, gives 3 dB higher noise but should be sufficient to determine µV variations.

 

[adnan012]

Thanks for reply

I need more explanation for "small band demodulation without phase selectivity". Can i sue ZN414 type IC?