2018-10-16 05:05 AM
Hello all
I am new at this forum and have almost zero knowledge in electronics but look at acceleration data on a daily basis (in the geophysics field).
My experience is that MEMS accelero are quite noisy, especially at "low" frequencies (<~1 Hz).
Just wondering if multiple MEMS can be connected together in order for the individual signals to be summed giving a sqrt(N) pure random noise attenuation.
I am thinking here about a hardware solution (e.g. N analog output sensors connected in serie, but is that that simple?), not a stack after individual sensor data digitalisation (something I could not afford for a number of reasons)
Your comments and advices will be much appreciated!
Regards
Pascal
2018-11-01 03:38 AM
Hello
thanks for the feedback - I agree it may be challenging for me alone but the idea is to figure out how complex this could be in order to discuss it with a more competent person who could help.
Coming back to the initial idea of summing the analog outputs, I found the following circuit schematic to stack multiple MEMS microphones:
https://www.analog.com/media/en/technical-documentation/application-notes/AN-1328.pdf
Could you explain why MEMS accelerometers (analog and single ended - I guess) could not be used instead of those MEMS microphones?
Best regards
Pascal
2018-11-01 06:17 AM
The analog sensor can be connected as indicated on the schematics in the application note.
The sensors are not connected in series they are connected to summing operational amplifier.
2018-11-01 08:26 AM
any kind of analog MEMS acc?
Is there a limited number of MEMS that can be stacked with a single op-amp?