Dear Muhammad,

The first step to solve your problem is to know well the strain signal characteristics. In general any signal has a specific waveform having certain range of amplitude and frequencies. Then to acquire any signal by the computer one has to convert it into digital form by sampling it and convert each sample into an n-bit code with n so high that the waveform can be reproduced accurately. The sampling frequency must be greater than 2x the maximum frequency in the signal of the sensor. The resulting bit rate from every sensor is nxf s. If you have M sensors multiplexed together in one stream the the total data rate will be mnxfs.

As an example assume fs= 1000 Hz which is very exaggerated, n=12 bits and M=30, the bit rate will be 360 kHz. As Zeashan pointed such rate can be handled by suitable Data Acquisition cards. However as the sampling speeds increases the card gets expensive.

Wish you success.

Dear Muhammad,

The first step to solve your problem is to know well the strain signal characteristics. In general any signal has a specific waveform having certain range of amplitude and frequencies. Then to acquire any signal by the computer one has to convert it into digital form by sampling it and convert each sample into an n-bit code with n so high that the waveform can be reproduced accurately. The sampling frequency must be greater than 2x the maximum frequency in the signal of the sensor. The resulting bit rate from every sensor is nxf s. If you have M sensors multiplexed together in one stream the the total data rate will be mnxfs.

As an example assume fs= 1000 Hz which is very exaggerated, n=12 bits and M=30, the bit rate will be 360 kHz. As Zeashan pointed such rate can be handled by suitable Data Acquisition cards. However as the sampling speeds increases the card gets expensive.

Wish you success.

Thank you dr Vasile for your amendment. My concentration was .to estimate the speed required to acquire many sensor signals simultaneously. I know that you always take care of the practical considerations. Really, one samples a signal with a rate greater than the nyquest rate for easier antialiasing filtering before sampling and for easier reconstruction of the signal after the sampling.Yes, 2xfs is a practical value.

My best regards.

I agree with the comments made by the teachers. With a multi-channel acquisition card data you could solve the problem. It only remains to provide that if you have to face an electronic design from scratch , I recommend using a single analog to digital converter and analog switches such as the ADG5401 . With such analog switches , digitally controlled , would select a sensor to measure at all times. You repeating the above cycle , activating and deactivating analog switches , 'd readings from all sensors with a time spacing will depend on the chosen AD converter and analog switches . In this scenario , it is important to count on a good designer of PCBs ; since we're talking about low-level analog signals , high speed and perhaps long distances from the sensor to your data acquisition system . It is therefore important that each analog channel is decoupled from the other channels, which can be achieved with analog switches plus multiplexing technique mentioned above.

Mucha Suerte!

A sample and hold on each channel is a classic solution for measuring fast signals with a slow DAQ. They will sample at the same trigger and hold their value until you complete your acquisition. A simple single shot monolithic setup that triggers at the start of each of your DAQ's MUX sweeps would remove some signal error.

But if your DAQ is not sampling fast enough to get at least two samples per significant transition you will be storing bad data. If you were to visualize your signal with a Fourier transform, under sampling is discarding all high frequency information.

At an extreme you can build a very fast ADC for each channel and queue through a two port FIFO to a DAC that will slowly feed your slow daq a MUCH faster signal. Or send from the FIFO to a capture application on a PC instead.