Building a Spectrometer

1
Building a Spectrometer

• Robyn Badon
• Michael Barnes
• Terrell Black
• Ryan Bloodworth
• Robyn Mackey

2
Lecture Outline

• Introduction
• What is a Spectrometer?
• How it works/ Project relevance
• Spectrometer Design Overview
• LED Geometry
• Hardware
• Software
• Summary

3
Introduction

• What is a spectrometer?
• Optical instrument to automatically measure light
  intensity/ detect color changes
• What are we using it for?
• Urinalysis
• More accurate than human eye alone
• The urinalysis test strip is compared to
  predefined samples by color
• Computer can eliminate human error

4
Overview of Project Relevance

• Medical screening device with 5 modules
• Thermometer
• Blood Oxygen Saturation
• Blood Pressure
• ECG
• Urine Glucose ?SPECTROMETER
• For safety urine is sampled
• Reduce risk of disease transmission
• Much more sanitary
• 
  

5
How it works

• Collect sample using test strip
• Light reflects off the colored strip
• Light intensity is picked up by the photoresistor
• Different reading for each LED
• Based on concentration of glucose in urine sample

6
Light back to Physics...

• Human perception of color
• Composition of light an object reflects
• A quick example
• If an object appears red, the red light is
  reflected while all other colors are absorbed

7
Trivia

• If an object appears to be white, what is
  happening?

8
Light
Figure 3. Electromagnetic radiation. The visible
wavelength range extends from approximately
350-700nm. Image source http//www.andor.com/lib
rary/light/
9
Our Spectrometer

• Reflectance spectrometer
• Intensity measured by photoresistor
• Interpret results

10
Photoresistor

• A photoresistor or LDR is an electronic component
  whose resistance decreases with increasing
  incident light
• made of a high-resistance semiconductor
  intensity.
• If light falling on the object is high enough,
  electrons get energy to jump into the conduction
  band
• The resulting free electron and its hole conduct
  electricity, therefore lowering resistance

11
Application

• Glucose concentration
• Human analysis by color matching

12
Basic Reflection Spectrometer Design

• Three LEDS blue, green and red.
• LEDs are controlled by the PIC
• Each LED is turned on one at a time
• LED is reflected down on the photoresistor
  causing the resistance value to change
• The voltage value is read in by PIC
• Data sent back to PC and compared

13
Basic Reflection Spectrometer Design
Figure 4. Block diagram of mini reflectance
spectrometer
14
LED geometry

• LEDs are arranged at a 30 angle
• The light is directed in a narrow beam towards
  the sample
• Prevents light from entering photoresistor
  directly, which would obscure the true signal
• The entire circuit is put in an opaque box to
  preventing interfering light from entering

15
Spectrometer geometry
Variety of LEDs Boxed in green is most typically
used, estimated at 80 of world production Boxed
in red is the LED used for our spectrometer
16
(No Transcript)
17
(No Transcript)
18
Powering the LEDs

• Red and Green powered by PIC
• Resistors used as voltage dividers
• Blue LED needs more power

19
Blue LED

• Transistor used as switch to power blue LED

20
Spice Analysis of
Transistor and Diode Voltage
21
(No Transcript)
22
The detector

• NSL-6110 photoresistor
• Change is voltage is very small
• Must be amplified
• Run through 2 stages
• Voltage divider
• Non-inverting Amplifier

23
The detector
24
Controlling Software

• PIC code for reading reflectance
• GUI
• KNN Algorithms

25
Central Unit the 18F4520

• Input/Output port
• A/D conversion
• Adaptation for RS232 communication

26
PIC Code

• PIC must activate LEDs and read in Voltage from
  the photoresistor
• Set port D, pins 0-2 to output
• Loop program until RX interrupt
• Run functions based on char received

27
MeasureR()

• Turn on red LED, turn off green and blue
• Wait short time
• Read in voltage
• Convert analog to digital
• Divide by 256, multiply by 5
• Save value
• Turn off red LED

28
GUI Software

• Using Visual Basic
• Button and text box for each LED
• Button sends char to PIC
• PIC returns data which gets displayed in text box

29
Controlling Software
Figure 8. Example Visual Basic GUI
30
Color Matching Software

• Training phase
• Classification phase

31
Training Phase

• Create sample data by scanning known colors
• Urinalysis Strips with known concentrations

32
Classification Phase

• Compare current sample to test data by computing
  distance
• Create a list of K closest training values
• Take the average to find concentration

33
Summary

• Our spectrometer has been designed to read urine
  glucose strips
• However, it is versatile enough to be used in any
  application where color detection is needed
• It can be adapted to be used for anything from
  protein concentration to pregnancy tests

34
Questions?