Temperature Sensor Amplifier using opamp

Go To Last Post
16 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello guys,
i have a board with ADC input rated 0-10V.
The power supply of the system is 24V (big truck).

I would like to interface a temperature sensor on the board, i have choosed TMP36 with positive output for all temperature range, about 150mv-1500mv.

What i need is to amplify the sensor output about 5-6 times before i feed the signal to the ADC input.

I have tried a very common circuit, an opamp with single supply, the sensor signal connected to the positive input and two resistor with a ratio of x5.

It works for input signal > 500mv but less than that the output starts to go higher that getting lower till it gets to voltage rail.

I tried many opamps, cheap and expensive . JFET input and Rail-to-Rail with common results.

Do you have any idea how can i amplify the signal even at low ranges of 200-300mv or what am i doing wrong?

thank you for your time!
Kostas

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello,

Have you tried it without amplification? Why do you need to amplify the output? The ADC is not an AVR then?

John

Just some guy

Last Edited: Mon. May 24, 2010 - 02:58 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The OP states an ADC board with 0-10V input range, not an AVR ADC. The sensor outputs only 10mV/degC.

Do you have an exact schematic?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ok here is the schematic..

what do you think is the problem that cant amplify in near rail input signals? (200mv-500mv)

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Supply the opamp with dual supply,enabling the opamp to amplify signals near zero and avoid the 741 device.
OP07 is a perfect choice.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

i cant have dual supply, because it will work in a vehicle i have 0-24V only

also i tried several opamps, some with JFET input and some Rail-to-Rail opamps that said to work very close to rail voltage.

Any ideas?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What are the values of the feedback resistors? (Though I would expect this to have more effect at higher outputs than lower).

Is your power supply connection well bypassed? It might be oscillating. You might add a small capacitor (say 470pf to 1000pf) from op-amp output to negative input.

When it is in the mode where the output voltage is wrong, what are the various node voltages?

You need to pay attention to the common-mode input range as well as the output range. Not all rail-rail devices have a common-mode input range that does close to ground. You DEFINITELY need a device with rail-rail output!

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello Jim.You wrote about a capacitor between negative input and output.What exactly is the purpose of this capacitor?Speeds up the opamp?Eliminates any offset?
I put it in an opamp used as a comparator and the result was excellent.
PS.A negative voltage charge pump could solve the negative voltage supply problem.Even a MAX232.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Run the op-amp on 24V - you'll not have much of a problem hitting 10V. You'll most likely need to run a virtual gnd and subtract the offset on the output.

Putting a capcitor in the feedback loop of an op-amp creates a pole - net result is the op-amp response is slowed down.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

the values are typical.. on the opamp i have a 10Kohm potensiometer with 1Kohm safety resistor in series, and a 1Kohm resistor. so i can multiply between 1x-10x.

I think i should try the method with Virtual GND but i can find a good circuit somewhere.. i try to search deeper.

I am not sure how i will remove the offset on the output using this method.

if anyone knows of a simple example of VGND for opamp amplifier and can share it.. thanks..

regards

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The cap is to suppress oscillation in some op-amps.

A virtual ground should work but, IMHO, its over-kill. One voltage divider between +24V and ground, one unity gain non-inverting op-amp and you have a virtual ground at Vcc/2.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

unity gain non inverting opamp? hmm i am not sure if i understand what i should do..

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Google may help with the terminology!

Perhaps, more correctly, I should have written: "unity gain, non-inverting op-amp CIRCUIT" (that is, the op-amp, itself, is not unit gain). Also called "unity gain buffer". It is the simplest possible op-amp circuit.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello,
thanks for your answer!

i tried to create a virtual ground using two resistors.. also the unity gain needs just to connect the inverted input to the output and the feed the signal to the non-inv input...

how do i use the VGND? Also i have another issue.. the temperature sensor in powered using a LM08l05 that takes the +12V common opamp supply and regulates down to 5V.. so the output is relative to normal GND.. how can i feed the output of the sensor to the opamp relative to the VGND..

i am a bit confused!!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ok.. this is what i have done to the above circuit and it seems usuable now..

i put a zener diode on between the ground of the sensor and the real gnd of the circuit..

what i managed to do is that now the output of the sensor has an offset of what the zener voltage is.. my case 2.4V zener..

so now the output of the sensor is minimum 2V maximum 3+ volts.. and the opamp amplifies with a factor of 2.

Can you see any flaws in that solution?
I calibrate the sensor the first time using the trimmer so i get a x2 amplification.

The only thing i am worried is the stability of the zener on temperature changes.. i will check it next week in the frigde :-)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Again, I am talking out of school here, but would a small transformer work??

I might try wrapping some enameled .030 copper around a small diameter wooden dowel and then adding a top layer of .018 Cu and I get a ratio of 6.25 to 50 figuring 1.5 volt primary and 12 volt secondary. Then, I would take a look at the secondary output on my scope for some know values...if this worked you could consider a different core material...
OK, you can laugh now:0)

John

Just some guy