Temperature Measurement System / Quiz 1

QUIZ:
The day began with a quiz. This is how my table attempted to answer it:

LAB:

The purpose of this experiment was to build a circuit that had an output voltage signal that provided a crude temperature measurement. This was accomplished using a NTC 10K @ 25 degrees celsius thermistor and a resistor. The goal was to make the output voltage vary by .5V when the ambient temperature varied from 25 to approximately 37 degrees celsius. Unfortunately, the resistance of the thermistor is NOT linear, so the value of it's resistance had to be observed from a chart, and then the necessary resistance of the circuits resistor had to be calculated. The following were our calculations to determine the necessary resistance: 

As the calculations above state, the necessary accompanying resistance value needed to be either 4.37K Ohms or 17.63K Ohms. These were calculated using values of resistance for the thermistor rated at 7K Ohms and 11K Ohms for 37 degrees Celsius and 25 degrees Celsius respectively.

With that completed, we began constructing the simple circuit, which had a 5V power supply connected to our thermistor, and then our fixed resistance resistor in series, with a voltmeter across the fixed resistance resistor to measure signal voltage. Before we constructed the circuit though, we measured the resistance of each element of the circuit, the results were as follows:

Resistance of resistor: 4.65K Ohms

Resistance of thermistor at ~25 Celsius: 11.40K Ohms

Resistance of thermistor at ~37 Celsius: 8.29K Ohms

*NOTE* It's important to note that the measured values of resistance for the thermistor varied from the expected quite a bit. The variation of resistance across the temperature range differed from the expected range of resistance by 22.3%. That was expected to be a huge variation in circuit performance. It should also be observed that the resistance of the fixed resistance resistor varied from the required resistance (due to limited selection of available resistors) by 6.4%. While that isn't a tremendous variation from the desired value, it too was expected to affect the circuits performance. 

The completed circuit looked as follows:

We then tested the circuit and got the following results:

Voltage at 25 degrees Celsius: 1.49V

Voltage at 37 degrees Celsius:1.89V

Delta Voltage: .40V

% Difference between desired delta Voltage and actual delta Voltage: 20%

As expected, the output voltage varied quite significantly from the expected value. Not surprisingly, it varied by 20%, which is consistent with the 22.3% variation of resistance range of the thermistor. Accepting that however, wasn't an option. We decided to try several other resistors in place of the 4.65K Ohm resistor, in hopes of achieving a full .5V variation in output voltage.

Results using a 3.9K Ohm rated resistor (measured at 3.81K Ohms)

Voltage at 25 degrees Celsius:1.26V

Voltage at 37 degrees Celsius:1.67V

Delta Voltage: .41V

% Difference between desired delta Voltage and actual delta Voltage: 18%

These results were our best results.

A photo of the voltage measurement from our most successful circuit.

Results using a 22K Ohm rated resistor (measured at 21.3K Ohms)

Voltage at 25 degrees Celsius:3.24V

Voltage at 37 degrees Celsius:3.47V

Delta Voltage: .23V

% Difference between desired delta Voltage and actual delta Voltage: 54%

These were our least impressive results.

Results using a 12K Ohm rated resistor (measured at 11.96K Ohms)

Voltage at 25 degrees Celsius:2.55V

Voltage at 37 degrees Celsius:2.80V

 Delta Voltage: .25V
 % Difference between desired delta Voltage and actual delta Voltage: 50%

These were also quite terrible. The following is a video of this circuit working (poorly).

It is agreed throughout our lab group that our circuit would have likely been successful if the thermistor had a more predictable (and consequently wide) resistance range. However, considering the limitations of the thermistor, the circuit itself appeared to perform just as expected, and varied from the desired values by the same percentage that the thermistor itself varied.

Temperature Measurement System Re-Design:

To create a circuit whose output voltage increases as temperature increases, and whose output sensitivity is at least 100mV per degree Celsius (across the same temperature range), we would use the exact same circuit, but we would up the voltage supply to 17V.