Hi All!
I am trying to build a modular addition to a spectrophotometer (it needs to be mobile and easily removed) thus I have decided to base my design around an Arduino.

My application requires that I power/control a High Power (1W) Infrared LED, and read off temperature measurements from an NTC thermistor.
Currently I am primarily having difficulty with the Thermistor part of the circuit (left side for all schematics)

My original design (See Image):

Used a voltage divider to measure the resistance of the thermistor. This worked fine, however due to the limitations of my thermistor and the small range of temperatures I am interested in, the voltage changes in that setup were minimal (on the order of a few tenths of a volt maximum). Thus I decided to add an OP-AMP in differential arrangement to amplify (and give a desired DC offset) my signal.

This led me to the following (current) design:

The OP-AMP that I am using called for a dual polarity power source. As such I put together the arrangement at the top left of the above schematic. My understanding is it produces a virtual ground at 8V.

The Issue I am having is tying this ground into the Arduino ground, so that the signal entering the Arduino at A0 (an analog input port) is between 0-5V

I really am stumped as to what I should connect my "Virtual Ground" up to.

Currently I keep reading non-trivial bias voltages between the input to A0 and the Arduino ground.

I have been beating my head against this for a few days now I would most appreciate some help.

Thank you so much,
Justin

 

it's never going to work... read the data sheet...

if you are trying to use R3 as constant current source, it should be at least 330k (better 390..470k).

any particular reason to use stone age opamp in a sensitive measuring instrument? 741 was a very nice chip - about half a century ago (eons when considering how electronics develops).

and why are you trying to measure voltage drop across R3? if you measure across NTC, then one of opamp inputs and the virtual ground connects to negative rail of the 5V (USB) supply. can't get any simpler...

all you need is amplifier that will produce reasonable output swing when NTC is subjected to various temperatures. T room temperature and 470k for R3, means that voltage across NTC would be about 0.1V. you need opamp with gain sufficient to bring this to say 3-5V (gain 30..50). as temperature increases, NTC resistance drops and so does the voltage across it. perhaps having R8 as trimmer wouldn't hurt.

 

it's never going to work... read the data sheet...

To be perfectly honest I have scoured the data sheet and am not sure what I am looking for. This is my first op-amp project (really first electronic project ever) and would be very grateful if you could point me to the part of the data sheet you are referring to.

if you are trying to use R3 as constant current source, it should be at least 330k (better 390..470k).

Do you mean constant voltage source?

any particular reason to use stone age opamp in a sensitive measuring instrument? 741 was a very nice chip - about half a century ago (eons when considering how electronics develops).

Because I had it lying around and would not know where to start with any other one. From my knowledge base it seemed as good a place to start as any.

and why are you trying to measure voltage drop across R3? if you measure across NTC, then one of opamp inputs and the virtual ground connects to negative rail of the 5V (USB) supply. can't get any simpler...

Your right this is a much better idea.

all you need is amplifier that will produce reasonable output swing when NTC is subjected to various temperatures. T room temperature and 470k for R3, means that voltage across NTC would be about 0.1V. you need opamp with gain sufficient to bring this to say 3-5V (gain 30..50). as temperature increases, NTC resistance drops and so does the voltage across it. perhaps having R8 as trimmer wouldn't hurt.

Could you suggest an op-amp?

Thank you very much!