Hello,
I need to multiplex 32 channels to 1. The signals come from 32 piezoelectric transducers (transducer datasheet: https://acs-international.com/downloads/S1803_Data_Sheet.pdf). The goal is to acquire the signals , regardless of whether there are delays between the different signals. It seems to me that using a 32:1 MUX is a good choice. I plan to drive the MUX selects (S0-S4) with a micro controller ( eg: Arduino). I will use one amplifier at the output of the MUX. The problem is that , the transducers (A0-A31) output low voltages of the order of (µV-mV). So I don't know which multiplexer to choose. Will any MUX be sensitive to low voltages and not distort the signal ? What are the important parameters I should take into account when selecting the multiplexer?
Hi Rafik Moulouel ,
you might try a tree containing 5 pieces of TMUX1308:
https://www.ti.com/lit/ds/symlink/tmux1308-q1.pdf
Advantage: low channel capacitance (typ. 11 pF, max 40 pF = ca. 1/20 of the capacitance of the transducer)
There are 32:1 multiplexers but with much higher channel capacitance (300 pF = 1/3 of transducer capacitance), resulting in considerably attenuated signals.
In your application, I guess noise is critical but it is unfortunately not documented. Since these devices are not expensive the best way might be an experimental setup with one or two sensors. You could place a low pass filter, e. g. with fg = 200 kHz, between the output of the tree and the amplifier in order to reduce the noise.
Dear Josef Punčochář
Use pre-amplifier before MUX so that any normal MUX will full fill the requirement
with regrds
vasanth
Dear G.s. Vasantha Kumar ,
your solution provides certainly the best result but I understood that Rafik Moulouel 's intention is to reduce the number of amplifiers from 32 to 1.
Dear Rafik Moulouel ,
You can use analog multiplexers based on CMOS transmission gates. They have linear transfer characteristics. So, thy will convey your small signal to your instrumentation amplifier.
For selecting the most suitable CMOS analog multiplexer please follow the link:https://www.maximintegrated.com/en/design/technical-documents/app-notes/5/5299.html
Best wishes
Update (solved):
Actually, an analog CMOS multiplexer, with low Ron and Con worked just fine. It introcuded quite a significant attenuation though. I've used an excellent amplifier, with a very low power noise.
You need to pay attention if you plan to use these CMOS multiplexers for higher frequencies applications ( in the order of MHz-GHz). You'll be limited by their cut off frenquency.
Thanks for your replies.
RM
- Badges
- Science topic
- Similar topics
- Engineering
- Petroleum Engineering
- Oil