Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

filtering and amplifying a sensor signal

Status
Not open for further replies.

aruna1

Member
Hi all,

I have some questions about filtering and amplifying a sensor signal.

lets say my temperature sensor outputs about 100mv analog output and is combined with noise.

so i need to remove the noise from the signal then amplify it so it can be read using microcontroller's A/D converter.

So i guess i need a one opamp configured as an active low pass filter (what topology should i use, salen key, RC low pass with buffer amp etc) and then another opamp configured as an amplifier with about gain of 25.

My questions are;

1.) how what is the most appropriate way to cascade the filter and the amplifier

sensor ----> low pass filter ----> amplifier

or

sensor ----> amplifier ----> low pass filter


or is there any method that combines thee filter and amp stage together so that single op amp can be used.



2.) what type of low pass filter should i use?
salen-key or RC filter with buffer amplifier


thank you
 
Last edited:

duffy

Well-Known Member
I would just slap a low-pass RC on the output and send that to an op-amp with a gain of about 20. Mainly because that's quick and easy, and will get you up and running with tests to see what other issues there might be.

Temperature sensors don't usually generate appreciable amounts of noise. A smarter approach would be to identify the source of the noise, because that sounds like the real issue - what you may actually need is a power supply filter on the temperature sensor, or to re-route the signal lead, or look for a ground loop.
 
Last edited:

aruna1

Member
View attachment 61236
I would just slap a low-pass RC on the output and send that to an op-amp with a gain of about 20. Mainly because that's quick and easy, and will get you up and running with tests to see what other issues there might be.

Temperature sensors don't usually generate appreciable amounts of noise. A smarter approach would be to identify the source of the noise, because that sounds like the real issue - what you may actually need is a power supply filter on the temperature sensor, or to re-route the signal lead, or look for a ground loop.

hi
thank you for the reply.
sensor is not actually a temperature sensor. I need solution for a general case which is applicable to any kind of sensors (temperature, current etc) I used "temperature sensor" just to name that it is a sensor.


My question is if i first filter the signal before amplifying it will it remove necesary details of the signal along with the noise?
I feel like if i first amplify the signal then send it through a filter it will remove the noise and will retain the details of the signal.


and another thing.
i have attached a circuit diagrams of two low pass filter types with 1kHz cutoff frequency.
first one is simple 2nd order rc filter with buffer amp and second one is 2nd order sallen key filter.

I could n't find any explanation for the difference between them. can you please explain the advantages and disadvantages each of this filter types and which one is better than the other

thank you
 

duffy

Well-Known Member
I see. So this is a homework problem?

I need solution for a general case which is applicable to any kind of sensors

This is like asking for a general-purpose glue that will stick to anything, or a universal linkage that will work as any kind of lever. A real temperature sensor (even one with a very small thermal mass) will have a very s-l-o-w response time in electronic terms, and even a simple 1st-order filter would be enough to get rid of any real-world high frequency noise I can think of, and still retain the details of the signal. A current sensor is very different, the signal you are filtering out may be exactly what you need to be reading. You need to look at it. There is no general case.

For a low-pass filter like your example, there isn't much advantage to one over the other. The active filter will have a somewhat sharper cutoff (owing to a higher Q), and exhibits slightly less noise from the amp (owing to a higher signal at the input). The real advantages are in things like bandpass filters, where you can jack the Q up until it rings (if you want, I made a cool electronic drum set this way), give it a much higher input impedance, and you don't need inductors - which can easily cost more than the rest of the circuit combined.
 
Last edited:

aruna1

Member
I see. So this is a homework problem?



This is like asking for a general-purpose glue that will stick to anything, or a universal linkage that will work as any kind of lever. A real temperature sensor (even one with a very small thermal mass) will have a very s-l-o-w response time in electronic terms, and even a simple 1st-order filter would be enough to get rid of any real-world high frequency noise I can think of, and still retain the details of the signal. A current sensor is very different, the signal you are filtering out may be exactly what you need to be reading. You need to look at it. There is no general case.

For a low-pass filter like your example, there isn't much advantage to one over the other. The active filter will have a somewhat sharper cutoff (owing to a higher Q), and exhibits slightly less noise from the amp (owing to a higher signal at the input). The real advantages are in things like bandpass filters, where you can jack the Q up until it rings (if you want, I made a cool electronic drum set this way), give it a much higher input impedance, and you don't need inductors - which can easily cost more than the rest of the circuit combined.

LOLz this is not a home work problem :) . I just passed out from the university. I'm trying to learn filtering thing so i can apply it to my future circuits :)

can you point me to a good tutorial on filters? some thing that teaches from basics.

thank you
 

duffy

Well-Known Member
I learned from a book by Don Lancaster called the "Active Filter Cookbook". If it's still in print I would recommend it.
 
Status
Not open for further replies.

EE World Online Articles

Loading
Top