Society of Robots - Robot Forum

Electronics => Electronics => Topic started by: visiontec on April 29, 2007, 05:19:52 PM

Title: thermistor calibraton... help!!
Post by: visiontec on April 29, 2007, 05:19:52 PM
i need to use a JT thermistor to measure the temperature.
i also need to calibrate it to show the temerature from 0C to 100C on an ammeter(0-100uA).

i have tried everything to make it work properly but all my efforts ended up in vain!

i have attached a diagram of my circuit. R5 is the thermistor.

ne help will be apriciated. Thanks!
Title: Re: thermistor calibraton... help!!
Post by: Kohanbash on April 29, 2007, 06:14:29 PM
What are you trying to do with it. that seems like a lot of circuitry for just a thermistor reading.
Title: Re: thermistor calibraton... help!!
Post by: ed1380 on April 29, 2007, 06:40:54 PM
measure at 0c and plot the point on a graph, same with 50 and 100. it might have a curve
Title: Re: thermistor calibraton... help!!
Post by: visiontec on April 30, 2007, 04:17:07 AM
i have measured the readings which are as below and yes it has a curve.
i am trying to measure the current from the thermistor to show it on an ammeter(0-100uA).


Temp(C)  Resistance(k-ohm)
0   27.7
10   18.07
20   12.11
30   8.301
40   5.811
50   4.147
60   3.011
70   2.224
80   1.668
90   1.267
Title: Re: thermistor calibraton... help!!
Post by: Admin on April 30, 2007, 05:05:41 PM
what is the expected current? it might be so low (or at least the difference so low) that your ammeter wont register it . . .

have you considered using a current sense IC? it will convert current directly to voltage for you.

and why measure current if you can measure voltage? by knowing the resistance and voltage, you can calculate for current . . .
Title: Re: thermistor calibraton... help!!
Post by: visiontec on April 30, 2007, 05:37:17 PM
the reason is that i have to design it that way!
this is an assignment n im running out of time.

im using opamps to amplify the current.

i need to measure the temp and show it on an ammeter(0-100uA)
i cant use any dedicated ICs for this assignment. just basic stuff can be used.

plz guys help me out!

Title: Re: thermistor calibraton... help!!
Post by: ed1380 on April 30, 2007, 05:54:34 PM
get a thermistor that is made to show 0-100c  with 0-100mA
Title: Re: thermistor calibraton... help!!
Post by: Admin on April 30, 2007, 06:00:30 PM
Quote
the reason is that i have to design it that way!
what are you allowed to change?

how sensitive is your ammeter? appears you need at least a 1000x amplification . . .
Title: Re: thermistor calibraton... help!!
Post by: visiontec on May 01, 2007, 04:44:32 PM
i cant get any stuff that will make my life easier!

im stuck wid the components im given:

im given a JT thermistor(10K 025 SHAPE (RC)) to measure the temperature

datasheet for the thermistor (103JT) :
http://www.rapidonline.com/netalogue/specs/61-0430e.pdf

im also given an ammeter (reading range:0-100uA) to measure the output current.

i have to linearize it and scale it so that a temperature of:

0C = 0uA, 25C = 25uA, 100C = 100uA

i can use any number of opamps(LM741CN) and all the resistors i want to scale it according to my requirements.

i made a couple of circuits for this but none of them give me the desired results.

plz help me im nearly out of time n my mind!!!
Title: Re: thermistor calibraton... help!!
Post by: Admin on May 01, 2007, 07:35:36 PM
Ok I suck at op-amps for this kind of thing, but I believe what you need is to apply the natural log to the output.

A logarithmic amplifer will linearize your output.
http://www.play-hookey.com/analog/logarithmic_amplifier.html

(http://www.play-hookey.com/analog/images/logarithmic_amp.gif)

And after it is linearized, you just do a basic non-inverting amplifier to bump it up to the 0->100 range.
(http://www.societyofrobots.com/images/schematics_voltampnoninvert.jpg)

hope that helped . . .
Title: Re: thermistor calibraton... help!!
Post by: ed1380 on May 01, 2007, 07:43:42 PM
without a programmed IC my noobiness can't come up with a way to do it, cause of the curve. if it was linear then it could be amplified and whatever as necesary
Title: Re: thermistor calibraton... help!!
Post by: visiontec on May 02, 2007, 05:45:03 PM
ive managed to get my circuit working with my orignal circuit.
the problem was that i was analysing it wrongly.

now i just have to make it on a pcb.

thanks everyone for helping.