[SOLVED] how to remove 400mHz envelop signal?

[D.A.(Tony)Stewart]

Re: how to remove 400 microhertz envelope signal?

I changed frequency by 10Hz but I get the same wave form as previous. Period is 40min. It did not change.

Is this some magneto wobble of the earth? Ha
Perhaps your ADC clock is a harmonic on 50Hz off by this amount. Can you change the ADC clock with finger capacitance a few PPM.

 

[janaranjanahl]

yes actually we tested in different place in the same building. we noticed that even due to metal( steel) magnetic field changes around the sensor. Also we noticed that wave like previous figure on that setup too. but we don't have the frequency /period information since we didn't measure it that long.

- - - Updated - - -

Can you change the ADC clock with finger capacitance a few PPM.

how can I do that? It is an lock in amplifier.

 

[janaranjanahl]

Could a magnetic field be formed by your circuit

our circuit is a small op- amp circuit and we tested with vero board version and printed circuit. We saw same out put in both cases.
we applied high pass filter between circuit and the lock in amp but we could not remove the that envelop.
then we applied high pass filter (RC) between lock in and digital multi meter following is the figure (fig.2) for that

we can see the same effect from 400uHz.
also we checked with the oscilloscope wave shape is same. I dont think it is a problem with ADC..

 

[BradtheRad]

we suspect that it comes from sun's radiation.

Here is an article telling how measurements detect periodic gravity fluctuations emanating from the sun. These include a 40 minute cycle and 160 minute cycle.

"The Sun as a Pulsar"



Gravity is not directly related to a magnetic force, but perhaps indirectly, in your particular setup.

- - - Updated - - -

Example, sun's fluctuations impact earth's magnetic field, which influences your sensor.

 

[D.A.(Tony)Stewart]

Re: how to remove 400 microhertz envelope signal?

Is this some magneto wobble of the earth? Ha
Perhaps your ADC clock is a harmonic on 50Hz off by this amount. Can you change the ADC clock with finger capacitance a few PPM.

Congrats on your measurement of the modulating 40 minute Solar magneto pulses.

I wonder if the daily orientation of Earth is alternating the polarity of the pulses in a 24h cycle.

Your first photo seemed to show 31 cycles with a linear ramp with much noise and the last one had very sharp but clean decaying impulses.

So you still wish to remove the pulses?
I know how to filter it from a spreadsheet with the data.

 

[janaranjanahl]

Re: how to remove 400uHz envelop signal?

hello, We were trying to figure out what causes the problem. we did several things and tested with another sensor has different frequency in our set up. Also we change the lock in amplifier and to different one. but output was the same. now we are using MODEL SR830 DSP Lock-In Amplifier. Also set up running using UPS . (tested with UPS connected with power and and only battery incidents).
I don't have a data set of fig.2 but I have fig.1 row data set.

now we are using only one signal generator for sensor. for lock in we use internal frequency generator.

thank you all for different thoughts.

- - - Updated - - -

View attachment sample_data_set.rarView attachment sample_data_set.rar
fig. 1 Data set attached.

 

[D.A.(Tony)Stewart]

This chart of 31 cycles is only 10 minutes of data ???

 

[janaranjanahl]

any way, problem is we need to know why it is giving the envelop. we can remove the noise by signal processing. I'm looking for "hardware solution" for that. because it takes 40 min for a correct data sample. thanks.

 

[BradtheRad]

I'm looking for "hardware solution" for that.

A typical method is to take data from two sensors. The first sensor is the one in your experimental apparatus.

The second sensor is located at a distance, far enough from your apparatus so it detects only ambient noise (sun's fluctuations). Subtract its readings from your first sensor's readings. The result is your desired data.

 

[D.A.(Tony)Stewart]

This chart of 31 cycles is only 10 minutes of data ???

Your xls date:time data indicates the frequency is a 20 second cycle, not 40 minutes. Care to explain?

 

[janaranjanahl]

Hello, yes 20 s cycle is AM modulation frequency.
we identified the problem. It is due to two different beat frequencies of two signal generators. And phase has not locked in the lock in amplifier. So now we are giving same frequency by one signal generator (to sensor and lock in ).
We are getting good signal.
Thank you everyone for supporting me.

 

[BrunoARG]

SO to summarise the envelop signal was formed by a phase between both generators, why didn't you use the sync output? Most signal generators have that synchronization BNC connector.

 

[D.A.(Tony)Stewart]

Most Magnetometers only have differential excitation coils with one common differential driver at one frequency.

1st 400mHz then microHertz then ~20s or 50millihertz all caused by obvious aliasing but took 32 attempts to resolve with sufficient detail.

Congrats , but too bad it wasn't real magneto-wobble that Carrington predicted in the 19th century... Which caused telegraph lines to arc and other solar effects.....