INTRODUCTION - Penn Engineering



Opto-electronics and the Measurement of Concentration

Group M2

Gavin Cheong

Jose Colon

Lucia Palant

Carrie Van Syckel

INTRODUCTION

The goal of this project is to build a colorimeter to determine concentration of hemoglobin in various samples. The colorimeter consists of a black box. Inside the box is a light bulb, a cherry-red filter, hemoglobin sample and photo-transistor circuit. The light bulb is powered by a 9V battery. The concept behind this device is that the light from the light bulb will pass through a filter which allows only cherry red light to pass through it. The hemoglobin at varying concentrations will absorb different amounts of light. The residual light that passes through the sample will be detected by the photoresistor. The transmitted light will be measured and the signal will then be amplified by a 741 op-amp circuit and read by the voltmeter. This will give the transmittance as a function of voltage. From this we can determine the absorbance and the concentration using the Beer Lambert law:

[pic]

where A is the absorbance, C is the concentration, ( is the molar absorption coefficient and b is the thickness or the length of light path in the medium. In order to get the transmittance as a function of voltage, a calibration curve will be made by measuring the voltage reading of known samples. Once these values are obtained, their accuracy can be checked against data obtained with the spectrophotometer.

BACKGROUND

A colorimeter was built in the ultimate attempt to measure the concentration of any given sample of hemoglobin. The goal of the project was extended to include the measurement of concentration for other solutions such as chlorophyll by the colorimeter. Colorimeters use filters as their wavelength selectors. Light is a form of energy called radiant energy. Radiant energy sends out waves of varying length. The range of these waves is called the Electromagnetic Energy Spectrum. Light waves are measured in wavelengths. The length of a light wave determines its color. Light waves can be reflected, they can be absorbed, or they can be transmitted. The visible spectrum can be broken into three predominant bands of light, namely red, green, and blue which are called the primary colors of light. Combining them in balanced amounts produces white light. (Miller, 300) The colorimeter could measure the concentrations of other solutions by simply changing the color filter inside the colorimeter with the appropriate one for the sample to be tested ( i.e. the one which transmits most at the wavelength for optimal absorption by the sample ).

Our colorimeter is an inexpensive and compact apparatus that can accurately measure the concentration of hemoglobin. It only requires two 9-Volt batteries in order to operate and these can easily be replaced by unscrewing the lid of the black box.. A switch turns the colorimeter on and off and a digital output displays the voltage which can then be related to a corresponding concentration of the hemoglobin. These corresponding values are found from the calibration curves that were developed by plotting the measured voltages against the corresponding known concentrations of the solution. This product can be marketed to teaching laboratories in either high schools or on the collegiate level to teach the basic concepts of finding the concentration of a solution. It is also convenient for laboratories which need a quick and inexpensive way to measure the concentration of hemoglobin. The color filter can very easily be changed to accommodate measuring the concentration of different solutions.

There are currently many different products on the market providing similar services. A computer-interfaced colorimeter monitors the absorbance of a certain colored solution with time using a colored light source. It is assumed that the absorbance is proportional to the concentration of the solution, as an application of Beer's Law. ()

Colortron from Light Source Computer Images Inc. is a $1,195 32-band spectrophotometer with 10-nanometer resolution. It measures light reflected from, or transmitted through, objects in such a way that it can describe the color of that object unambiguously. Colortron's software tools lets one translate measurements into a variety of color systems and allow the unit to function as a densitometer (reflective and transmissive) and a colorimeter. The main feature of this apparatus is that for its high accuracy it is not expensive in comparison to the instruments normally used in the industry that cost several tens of thousands of dollars. Even 10-nanometer spectrophotometers have traditionally carried five-figure price tags. Densitometers are the most common instrument in use. They simply measure the amount of light transmitted through or reflected from a specific object. They measure film density and dot percentages and are primarily used to calibrate imagesetters and to determine dot gain on press.

Spectrophotometers measure the spectral-power distribution of light. Colortron divides the visible spectrum, which lies between 400- and 720-nanometer wavelengths, into 32 10-nanometer bands and measures the light intensity of each band. The advantage of spectral data is that one can predict the color appearance of the target material under any light source. ()

The LaMotte Colorimeter is designed for simple and rapid measuring of a wide variety of water quality parameters. This colorimeter does not require the changing of filters or dialing wavelengths. This colorimeter automatically selects the proper wavelength.

Any of the 40 preprogrammed tests or user-entered tests (up to 25) can be easily accessed through alphabetical listing on the LCD display. Results are displayed in concentration (ppm), %T, or absorbance. The LaMotte Colorimeter comes with four sample tubes, AC adapter, rechargeable batteries, Instrument Manual including Test Procedures, a Quick Start Guide, and an RS232 port for computer interfacing. The price of this apparatus is $815.00 ()

The revolutionary aspect of our device is its ease of use, inexpensive price, and adaptability to accurately measure the concentrations of many different solutions.

The Beer-Lambert law is a fundamental equation when studying absorbance. I = I010-(cl is an expression of the Beer-Lambert law where ( is the molar absorption coefficient or the extinction coefficient. The concentration is mol/m3 is c. Io is the intensity of the incident beam and I is the intensity of the emergent beam after passing through the total cell length l. The Beer Lambert law is the basic equation for the various colorimetric and spectrophotometric methods of analysis. If the Beer-Lambert law holds then absorbance is given by A = -(cl.

Most of the oxygen in the blood is contained within the red blood cells, which is chemically bonded to hemoglobin. Each hemoglobin molecule consists of a protein globin part, composed of four polypeptide chains, and four nitrogen-containing, disc-shaped organic pigment molecules. One hemoglobin molecule can combine with four molecules of oxygen. Since there are about 280 million hemoglobin molecules per red blood cell, each red blood cell can carry over a billion molecules of oxygen.

The oxygen-carrying capacity of whole blood is determined by its concentration of normal hemoglobin. If hemoglobin concentration is below normal, anemia can occur which means that the oxygen concentration of the blood is reduced below normal. When the hemoglobin concentration is increased above the normal range then polycythemia occurs where the oxygen-carrying capacity of blood is increased accordingly. The production of hemoglobin and red blood cells in bone marrow is controlled by erthropoietin, produced by the kidneys. Its production is stimulated when the delivery of oxygen to the kidneys and other organs is lower than normal.

A normal hemoglobin concentration of 15 grams per 100 mL unloads about 4.5 mL per 100 mL. (Fox, 433)

The main components of the circuit are the 741 operational amplifier, photoresistor, color filters, and LED. The operational amplifier is a linear amplifier with an inverting and noninverting input and one output. The polarity of a signal applied to the inverting input is reversed at the output. A signal applied to the non-inverting input retains its polarity at the output. The gain is determined by a feedback resistor that feeds some of the amplified signal from the output to the inverting input. The smaller the resistor that is used, the smaller the gain. (Engineer's Mini-Notebook, Radio Shack)

The light-emitting diode is a semiconductor PN junction diode that emits visible light or near-infrared radiation when forward biased. Visible LED's emit relatively narrow bands of green, yellow, orange, or red light. LED's switch off and on rapidly, are very efficient, have a very long lifetime, and are easy to use. LED's are current dependent sources, and their light output is directly proportional to the forward current. A photoresistor is a very commonly used light sensor with a very high electrical resistance when dark and a very small resistance when illuminated. The most common semiconductor used to make photoresistors is cadmium sulfide. The following is the sensor spectral response graph. It explains the reason for switching from our original intention of using a phototransistor to a photoresistor. The photoresistor has very high accuracy for the range of wavelengths that we were trying to measure, in this case 500 - 700 nm.

[pic]

Roscolux color filters were used in this project. The spectral energy distribution curve of each Roscolux filter is located behind the translucent plastic color. The curve describes the wavelengths of color transmitted through the individual filter. The “Trans” percentage at the top of the curve sheet refers to overall light transmission that is allowed to pass through each individual filter.

In order to accurately measure the voltage of different solutions the cuvettes should be wiped clean and dry on the outside with a tissue, the cuvettes should be held by the top edge of the ribbed sides, and the solutions should be free of bubbles. The device should always be turned off before opening the lid to replace the batteries. The LED becomes very hot and can cause a burn. There is also the possibility of the cuvette spilling into the circuit since there is no isolated chamber holding it in place. The hole drilled in the box was made so that the cuvette exactly fit into it but it is possible that the sample could spill into the black box. In this case the colorimeter should be turned off, the lid removed, and the spill cleaned up as much as possible. The lid should be placed on the circuit again and the integrity of the circuit should be tested.

SYSTEM OVERVIEW

Various changes occurred throughout the different stages of product development. Originally it was planned to use a phototransistor circuit, light bulb and green filter. The first thing we opted to change was the light bulb. We realized that only a small amount of light was needed and that a small lamp would serve the purpose. We also decided against using the phototransistor because the photoresistor has a better response in the range of wavelengths we are working with (300nm to 700nm). The circuit below was the original circuit. Since we opted to use the photoresistor none of this circuit was used.

We then came up with a new circuit that utilized the photo-resistor. In our first tests, we placed a lens in front of the lamp. Its purpose was to columnate the light going into the sample.

[pic]

We decided that the lens was unnecessary and removed it from our final circuit. In the above circuit, the lamp is powered by a 9 volt battery. The circuit consisted of two major parts. The first was an inverting op amp. This was used to amplify the voltage change that was caused by the resistance change of the photo-resistor. The resistor that goes from pin 2 to pin 6 determines the gain of the op amp since the gain is equal to R2/R1.

[pic]

R1 = 1K

R2 = 100K

R3 = 1K

R3 is used to make sure that pin 3 is grounded. The input from the photo-resistor goes into pin 2 and the output from pin 6 is read by the multimeter. The voltage supply is positive in pin 7 (+9) and negative in pin 4 (-9). This voltage is created by the two nine volt batteries. The other major component is the voltage divider where R1 is the photoresistor. The electrical resistance of a dark photoresistor is ordinarily very high, up to 1,000,000 ohms or more. The resistance may fall to as little as a few hundred ohms when the photoresistor is illuminated. The most common semiconductor used to make photoresistors is cadmium sulfide. Photoresistors exhibit a “memory effect” in that they may require a second or more to return to their high-resistance state after a light source is removed. Through their slow response time, they are very sensitive and easy to use.

Photoresistors are photo-resistive detectors. They can be substituted for fixed or variable resistors to make a circuit sensitive to light. We used the voltage divider to change the variable resistance of the photoresistor into a variable voltage.

[pic]

Using this circuit the relationship between the amount of light going in and the output voltage will be linear.

Another change that was made from our proposed circuit was the filter we decided to use. Originally, we planned on using a green filter based on the assumption that hemoglobin will absorb green light. This assumption was incorrect since hemoglobin has a high percent transmittance between 500 and 700 nm. This was determined by bioengineering 210 students studying different blood concentrations and their percent transmittance.

Once the circuit was constructed and tested it was all placed inside a black box. The circuit board was mounted with Crazy Glue to two cuvettes so that it would be high enough to allow proper alignment of the cuvette, lamp and photoresistor. The cuvettes were also glued to the bottom of the box so that no shifting would occur. The batteries were taped inside the box so that they could be easily removed and replaced. The multimeter was the only component to remain outside the box. A double pull-single throw switch was added to the circuit so that the apparatus could be turned on and off to

extend battery life. The cherry red filter was placed in front of the lamp, and the photoresistor was placed facing the filter, so that the sample could be placed in between the filter and the photoresistor. The original design was again modified and black tubing was placed around the photoresistor so that it could only detect the light coming out of the sample. When the surrounding light was not eliminated there were insignificant changes in voltage with concentration. The final set-up of the inside of the box is shown below.

The dilutions of the chlorophyll were made but never tested. A pin from the 741 op amp in the circuit broke and therefore did not reach all the way into the breadboard. If this problem had been rectified then the circuit would have been tested by changing the cherry rose filter to a green filter. he following rule can be used to estimate the additional absorption due to chlorophyll:

Component Specifications

CDS Photocells:

120310 CDS002 Jameco 1-800-831-4242

Operating Temperature: -30 degrees celsius to 70 degrees celsius.

Cadmium sulfide

Soldering: 230 degrees celsius for 3 seconds

Generate variable signal based on detection of light levels.

Applications include: auto-focus lenses, exposure meters, contrast controls for TVs, dimmer or light switches, flame detectors, electronic toys, street lamp switches, opto couplers

125mW, 25K(light resistance,max light) 2M(dark resistance, min dark), .16 inches diameter, .1 inch lead spacing, 1,44 inch lead length 150V peak

Radio Shack Bi-Pin Lamp

cat no. 272-1154

12 volts 55mA

Replaces lamp used in 12V lighted switch

Pins also suited for IC-spacing (.1 inch) microfboard

The 741 Op-Amp

For device specifications see Radio Shack Catalog Number 62-5011. Engineer’s Mini-Notebook on Op Amps.

Roscolux

The spectral energy distribution curve of each Roscolux filter is located behind the translucent plastic color. The curve describes the wavelengths of color transmitted through the individual filter. The “Trans” percentage at the top of the curve sheet refers to overall light transmission that is allowed to pass through each individual filter.

#332 Cherry Rose

Trans = 38%

Roscolux #332 transmits approximately 60% of the violet and blue energy of the spectrum and 85 % of the orange and red energy. It blocks energy in the yellow and green range.

[pic]

Filter:

Rosco

26 Bush Ave. Port Chester, NY 10573 (914) 937-1300 (800) ROSCO NY

CONSTRUCTION PROCEDURE

CIRCUIT AND ITS COLLATERALS:

The first component placed on the circuit board was the 741 operational amplifier. It was positioned near a short edge of the bread board. This part was used to denote the bottom of the board. Having decided on an orientation for the circuit board, it was decided that the left-most red strip would be the ground for the circuit, the right-most red strip would be the positive voltage supply for the circuit, and the blue strips on either side of the board would be the negative voltage supply for the circuit. A wire was placed across the bottom of the circuit board in order to connect the two blue strips. The left side of the circuit board contains pins 1 through 4 of the op-amp and the right side contains pins 5 through 8 of the op-amp.

A wire was used to connect pin 4 of the op-amp to the left-side, negative voltage supply. Another wire was used to connect pin 7 of the op-amp to the positive voltage supply. A 1 kW resistor was connected to pin 2 of the op-amp and to an unoccupied row on the left side of the circuit board. At this end a 22 kW resistor was connected, with its other end attached to ground. One end of the photoresistor was placed at the connection between the 1 kW and 22 kW resistors. The other end was placed in an unoccupied row on the left side of the board. A wire was then used to connect this row to the positive voltage supply. The photoresistor was placed such that it was as far left on the circuit board as possible. A 100 kW resistor was then connected across pin 2 and pin 6 of the op-amp. A 1 kW resistor was connected from pin 3 of the op-amp to ground. On the right side of the circuit board, the lamp was placed as far right as possible and directly opposite the photoresistor. At one end of the lamp a wire was run to the positive voltage supply, while at the other end a wire was run to the right-side, negative voltage supply.

Battery caps were used to draw the current from the two 9 Volt batteries to the circuit board. A double-pole, single-throw switch was used to turn the device on and off. The negative leads from the battery caps were soldered to what was denoted as the two bottom leads of the switch. Two wires for the negative leads out of the switch were then soldered to the middle leads of the switch. The positive lead from one of the 9 Volt batteries was connected to the positive voltage supply strip on the circuit board. The negative lead from the switch corresponding to this battery was then connected to the ground strip on the circuit board. The positive lead of the second 9 Volt battery was connected to the ground strip of the circuit board. The negative lead from the switch corresponding to this battery was then connected to the right-side, negative voltage supply strip of the circuit board. The common wire from the multimeter was connected to the ground strip on the circuit board. The positive lead from the multimeter was connected to pin 6 of the op-amp.

The photoresistor leads were bent so that the face of the photoresistor pointed directly towards the lamp and in the direction perpendicular to the length of the leads. The lamp leads were also bent so that the top of the lamp pointed in the direction of the photoresistor and in the direction perpendicular to its own leads. It was then decided that the lamp and photoresistor should be placed higher. In order to accomplish this, rigid wire segments were soldered onto the end of the leads for both the photoresistor and the lamp. These leads were then bent in a double-L shape so as to have the tops of both components directly over the pin-holes in which they were inserted. This resulted in a wide enough gap between the photoresistor and the lamp for the cuvette to be placed in.

When taking readings, there was not enough disparity in voltage readings between samples of large different concentrations. It was determined that the reason for this was largely due to the light from the lamp being picked up on all sides by the photoresistor. In order to eliminate this problem, it was necessary to tunnel the light coming from the lamp and passing through the cuvette directly to the face of the photoresistor. To accomplish this, a short piece of rubber tubing was attached to the photoresistor using black, electrical tape. The tubing was cut to a length such that the open end was very close to the cuvette without actually touching it. A piece of rubber tubing was also attached to the lamp via black electrical tape in order to tunnel its light to the cuvette. However, it was found that the lamp became hot enough to cause the tape to melt. The tubing and tape were therefore removed.

The filter being used had to be made interchangeable and was to be placed in front of the lamp before the light hit the cuvette. The filters used were plastic, and therefore easily manipulable. Each of the filters used was cut into a square shape. Two rigid wire strips were then placed on either side of the lamp in adjacent pinholes. The filter was placed between the two wires on either side, holding it in place, perpendicular to the circuit board.

Initially it was thought that a lens might be needed to focus the light being emitted by the lamp before it hit the cuvette. For this purpose a plastic lens was to be used. Two rigid wire pieces were cut. In each piece a slit was made into which an edge of the lens was placed. Crazy glue was used to glue the lens to the wires. The lens was then placed between the lamp and the filter. However, it was found that there was no difference in readings taken with or without the lens. The lens was therefore removed from the device since it did not enhance its performance.

During testing it was discovered that the voltage readings being taken slowly shifted downward over time. It was found that this was due to the voltage drain on the battery. In order to remedy this problem, two voltage regulators were ordered; one having a rating of +8 Volts and the other a rating of -8 Volts. 8 Volt regulators were chosen since a readable voltage drop of at least 6 Volts was required. However, only the positive voltage regulator arrived on time. Therefore, the circuit was modified to contain only this voltage divider. The voltage divider was placed in the top right-hand corner of the board. It was oriented such that the input pin faced the bottom of the board and the output pin faced the top of the board.

Since the voltage divider required an input voltage of at least 14 Volts, a 9 Volt battery was placed in series with the battery whose positive lead went to the positive voltage supply strip on the circuit board. To do this the lead from the original battery was disconnected from the positive strip and placed in an unused row on the right side of the board. A battery cap was placed over the added battery. The negative lead of the additional battery was then connected to the positive lead of the original battery. The positive lead of the battery was connected to the input pin of the voltage regulator. A wire was used to connect the middle pin of the voltage regulator to ground. The output pin of the voltage regulator was then connected to the positive voltage supply strip of the circuit board.

This new set-up was tested and it was found that the same drift of voltage readings occurred over time. It was then determined that both the positive and the negative voltage supplies needed to be regulated in order to eliminate the drift in the readings. The circuit was therefore remodified to be as it was before the voltage regulator was inserted. Several tests resulted in readings which suggested a short in the circuit. The circuit was therefore taken apart and completely rebuilt as before.

Box:

A black, thick-walled, rectangular, plastic box was purchased form Radio Shack to house the components of the device. The orientation of the box was chosen such that the top of the box also contained the side walls and the bottom was just a flat piece. A small, circular hole was drilled into the top of the box in order to allow the wires form the multimeter to have access to the inside of the box. Next, another whole was drilled into the top of the box and then filed to the dimensions of the cuvette. To insure proper insertion of the cuvette through this hole, the hole was slightly enlarged by means of filing in order to offer a snug, but unrestrictng pathway for the cuvette to pass through. A third, circular hole was drilled into the top of the box for the switch. The top of the switch was inserted through this hole from the inside. The switch was then secured to the box by means of a washer and nut.

At this time the circuit had already been constructed. The position of placement of the cuvette over the circuit board had therefore already been determined by the placement of the lamp and the photoresistor. Measurements were made on both the box and the circuit board in order to determine where to mount the circuit board relative to the inside of the box in order to insure that insertion of the cuvette through the cuvette hole would place it in the correct position on the circuit board. For the cuvette to be able to rest on top of the circuit board, it was necessary to raise the height of the circuit board. In order to do this, two cuvettes were glued to the extremes of the bottom of the circuit board so that they were aligned with the short sides of the circuit board. The other sides of the cuvettes were then glued to the bottom of the box. The wires for the multimeter were then threaded through the whole made in the top of the box for this purpose. During testing, black electrical tape was used as a cover for the cuvette. The batteries were placed at the side of the circuit board and held in place with clear tape.

Device Malfunction:

In order to check the interchangeability of filters in the device, tests were to be made using chlorophyll. However, testing yielded readings which implied a short in the circuit. The different components of the circuit were then tested separately to try and isolate the problem. This did not solve the problem. The circuit was then rebuilt on a separate bread board and found to work correctly. Since it appeared that all of the wiring was correct, the set-up on the device circuit board was taken apart and rebuilt. Initially, the device did work as expected. However, after aligning the photoresistor and the lamp with each other, readings again indicated that there was a short somewhere in the circuit. Again the problem could not be isolated. Unfortunately, test time ran out at this point. The device was therefore dismantled and the bread board itself was checked for shorts. None were found. After further testing, it was finally ascertained that some of the pins of the op-amp did not reach far enough down into the bread board to make a strong connection. This resulted in the chip being unconnected in the circuit, although it was securely in place on top of the circuit board.

DILUTIONS :

We started with a stock solution and placed 45 mL into a container and centrifuged it for 30 minutes. This was done to completely separate the red blood cells from the plasma. We then poured off the plasma layer so that we were left with only red blood cells. Then, two sets of samples were made. For the first, 20 mL of the stock solution was added to 25 mL of water. This was then centrifuged to be sure that the solution was mixed well. Then dilutions were made using this new solution at 10 through 100 percent. Another solution was made by taking 20 mL of the solution just made and adding 30 mL of water to it. This was also centrifuged and dilutions were made at 10 through 100 percent. The reason we made two sets of samples was to compare the accuracy of our device for samples that were very dilute and very concentrated. Since we only knew the concentrations in terms of the percent blood in the mixture we devised a method of determining the concentration of the two samples. To do this we extracted 19 mL of sample 1 and 15 mL of sample two. These samples were then placed in weigh boats and put in the oven for two days at 70 degrees Celsius. The weigh boats were weighed before the blood was put. After the two day period the weigh boats were extracted from the oven and weighed again. then the weight of the boats was subtracted from the final weight to determine the amount of grams of each sample. Since the number of grams was known and the number of milli-liters was known we were able to determine the concentration in grams per milli-liter. For sample 1 the concentration was 0.091g/mL and for sample 2 it was 0.0275 g/mL. From the original concentration, all other concentrations were found by multiplying by the amount of blood in the sample and dividing by the total number of milli-liter which was always 4 mL.

The device was also tested using chlorophyll to determine if our design was applicable to solutions with absorbencies at different wavelengths. The dilutions of chlorophyll were made in a manner similar to the blood samples. They were tested using a pea green filter.

Preparation for dilutions from the stock solution

|Percent Dilution |Amount of |Amount of deionized water (ml) |

| |hemoglobin (ml) | |

| 0 % |0.0 |4.0 |

| 10 % |0.4 |3.6 |

| 20 % |0.8 |3.2 |

| 30 % |1.2 |2.8 |

| 40 % |1.6 |2.4 |

| 50 % |2.0 |2.0 |

| 60 % |2.4 |1.6 |

| 70 % |2.8 |1.2 |

| 80 % |3.2 |0.8 |

| 90 % |3.6 |0.4 |

|100 % |4.0 |0.0 |

Sample #1

Weight of weighing boat: .979 grams

Volume of sample: 19ml

weight of liquid sample: 20.2 grams

weight of dried sample: 1.721 grams

concentration of dried sample:

Sample #2

Weight of weighing boat: .988 grams

Volume of sample: 15ml

weight of liquid sample: 15.8 grams

weight of dried sample: 0.412 grams

concentration of sample #1 : 0.091 g/ml ( 0.6913%

concentration of sample #2 : 0.0275 g/ml ( 1.068%

TESTING PROCEDURE

First, the amplifier circuit was tested by using a function generator and an oscilloscope. The function generator was used to input a signal into the op-amp, and the output and original input signal were displayed on the oscilloscope to verify that the gain was the desired one.

The light bulb was tested by connecting it to the power supply and making sure that it turned on. The photoresistor was tested by connecting it to a multimeter and observing the resistance displayed. Since the photoresistor’s performance is dependent on the amount and type of light it receives, it was tested by covering it to see if there was any effect on the resistance.

After all of the components of the circuit were connected, it was tested with the function generator and oscilloscope. The function generator was used to input a signal into the circuit, and the oscilloscope was used to observe the output. This was done to ensure the gain was still the desired one, thereby ensuring there were no problems with any of the components in the circuit. It was also checked by measuring the output voltage of the op amp when the photoresistor was in the dark and completely exposed to light. We found that there was a significant voltage change. In the dark it was 1.9 volts, and in the light it was 7.3 volts.

The hemoglobin samples were placed in the colorimeter and the circuit was turned on. The voltage reading was taken from the voltmeter. This was done for every concentration of hemoglobin prepared and a plot of concentration vs. voltage was generated with the acquired data to create a calibration curve to obtain an empirical relationship between hemoglobin concentration in blood and output voltage. This relationship would later be used to determine the concentration of hemoglobin in a sample of blood with unknown concentration from the measured voltage using the colorimeter.

At first, a green filter was used in the circuit. Since the voltage readings did not vary significantly with varying concentrations of blood, it was determined that the green filter was not transmitting light at the wavelength for maximum absorption by hemoglobin. The cherry red filter was determined to be the most efficient choice, since it transmitted the most light at around 600 nm, which is in the range for the maximum absorption of light by hemoglobin. When the circuit was then tested with this filter, the voltage readings did vary significantly.

After obtaining readings from the different hemoglobin samples and creating the calibration curve, the circuit was going to be tested by placing a sample of a known concentration in the box. The concentration would be determined using the voltage reading and the calibration curve obtained before, and compared to the known value to determine how accurate the circuit was. This could not be done, since the circuit malfunctioned when this attempt was about to be made .

The problem which arose was that the readings being taken implied that there was a short in the circuit. Each of the components of the circuit was tested individually to try and isolate the problem, but the problem could not be found. The circuit was rebuilt on a separate bread board and the individual components were tested. These were found to work properly and the operation of the circuit as a whole was also tested. This also worked properly. The device circuit was rebuilt as an exact duplicate of the circuit on the separate bread board. The system at that point worked properly. After realigning the lamp with the photoresistor, readings which implied a short circuit in the system were again found. Every component in the system was then checked to try and isolate the problem. Again the source of the problem could not be found. The bread board itself was then checked for shorts. None were found. The 741 chip was then reinserted on the bread board and the connections between its pins and the bread board were checked. It was determined that pin 6, the output pin, did not reach far enough down into the bread board to make a connection.

OPERATIONAL SPECIFICATIONS

|Percent Dilution |Sample 1 Concentration g/mL |Sample 2 Concentration g/mL |

|10 |0.0091 ( 0.6931 % |0.0028 ( 1.069% |

|20 |0.0018 ( 0.6931 % |0.0055 ( 1.069% |

|30 |0.027 ( 0.6931 % |0.0083 ( 1.069% |

|40 |0.0364 ( 0.6931 % |0.011 ( 1.069% |

|50 |0.0455 ( 0.6931 % |0.0138 ( 1.069% |

|60 |0.0546 ( 0.6931 % |0.0165 ( 1.069% |

|70 |0.0637 ( 0.6931 % |0.0193 ( 1.069% |

|80 |0.0728 ( 0.6931 % |0.022 ( 1.069% |

|90 |0.0819 ( 0.6931 % |0.0248 ( 1.069% |

|100 |0.091 ( 0.6931 % |0.0275 ( 1.069% |

The above chart lists the concentrations of the solutions used in creating the calibration curves and their associated errors.

Sample #1

|Percent |Measurement #1 |Measurement #2 |Measurement #3 |Mean & Standard Deviation |

|dilution |(Volts) |(Volts) |(Volts) | |

|blank |7.05 |7.05 |7.05 |7.05(0 |

|10% |6.95 |6.94 |6.92 |6.94(.02 |

|20% |5.55 |5.53 |5.53 |5.54(.01 |

|30% |4.54 |4.53 |4.52 |4.53(.01 |

|40% |3.95 |3.94 |3.93 |3.94(.01 |

|50% |3.74 |3.74 |3.73 |3.74(.01 |

|60% |3.00 |2.99 |2.98 |2.99(.01 |

|70% |3.17 |3.17 |3.15 |3.16(.01 |

|80% |2.47 |2.47 |2.46 |2.47(.01 |

|90% |2.51 |2.50 |2.49 |2.50(.01 |

|100% |1.91 |1.90 |1.89 |1.9(.01 |

The preceding chart lists the three voltage readings taken for each solution from sample one used to make the first calibration curve. The means and standard deviations of the data are also listed.

Sample #2

|Percent |Measurement #1 |Measurement #2 |Measurement #3 |Mean & Standard Deviation |

|Dilution |(Volts) |(Volts) |(Volts) | |

|10% |7.0 |6.99 |6.98 |6.99(.01 |

|20% |6.97 |6.96 |6.95 |6.96(.01 |

|30% |6.94 |6.94 |6.93 |6.94(.01 |

|40% |6.9 |6.88 |6.87 |6.88(.02 |

|50% |6.84 |6.82 |6.80 |6.82(.02 |

|60% |6.63 |6.63 |6.60 |6.62(.02 |

|70% |5.87 |5.86 |5.84 |5.86(.02 |

|80% |5.50 |5.50 |5.47 |5.49(.02 |

|90% |5.88 |5.88 |5.86 |5.87(.01 |

|100% |4.05 |4.03 |4.02 |4.03(.02 |

The preceding chart lists the three voltage readings taken for each solution from sample two used to make the second calibration curve. The means and standard deviations of the data are also listed.

[pic]

[pic]

Two different calibration curves were made. The data used for the curves came from two different sets of readings, one for 'high concentration' dilutions and the other for 'low concentration' dilutions. It was expected that both sets of data would yield roughly the same calibration curves if the device was working properly. To create the calibration curves, three readings were taken for each solution in each set. The negative of the reading was taken since an inverting op-amp was used to amplify the signal. The means of these data points were then calculated and plotted in each set. The curves were graphed so as to create a relationship for concentration as a function of voltage. Furthermore, it was deduced from the Beer-Lambert Law that the relationship should yield a negative log curve. From the graphs, a relationship for concentration as a function of voltage was to be made.

The two calibration curves were not smooth negative log curves. It was found that the voltage readings tended to drift downwards with use of the device. This was determined by taking readings of the deionized zero concentration water at different points in time. This explained the downward shifts observed in the calibration curves. Because of these constant shifts, it was not feasible to derive a relationship for concentration as a function of voltage. To remedy this problem, voltage regulators need to be used in order to keep the voltage supply constant. This should result in a smooth negative log curve from which the desired relationship between concentration and voltage can be calculated.

ERROR ANALYSIS

In order to find the errors, the procedure delineated in the BE310 Lab Manual was used. When addition or subtraction was used in a calculation, the absolute errors of the two measurements were added together. When multiplication or division was used in a calculation, the individual absolute errors were converted to percent errors. The resultant error was then found by taking the square root of the sum of the squares of the two percent errors.

Stock I Stock II

20mL blood ( 0.1 20mL Stock I ( 0.1

25mL H2O ( 0.1 30mL H2O ( 0.1

Stock I Error = 0.44% Stock II Error = 0.798%

19mL Stock I ( 0.1 15mL Stock II ( 0.1

Error = 0.5263% Error = 0.6666%

Weight = 1.721 ( 0.001 Weight = 0.412 ( 0.001

Error = 0.0581% Error = 0.2427%

Total Error of Weighed Samples

Error = 0.5295% Error = 0.7095%

Total Error = 0.6913% Total Error = 1.068%

For each concentration made from stock I the error will be 0.6931% and for each concentration made from stock II the error will be 1.069%. This is found by combining the error of the volume measurement which is 0.050% and the error of each stock solution.

DEVICE SPECIFICATIONS

Power supply: two 9 Volt batteries

Sample concentration range (Hemoglobin): 0 - 0.100 g/mL

Output voltage range: 0 - 8 Volts

Precision: ( 0.02 Volts

FUTURE IMPROVEMENTS

In the future to improve our apparatus a calibration button would be included. A cuvette of deionized water could always be set to the same value for voltage in order to insure reproducibility of results. In our project a piece of black tape was used to cover the top of the cuvette to keep the light out. In the future a small black lid would be used.

There would also be an isolated chamber for the cuvette in the black box. There would be a small hole constructed in the chamber to allow the light source to shine through the sample. It was also seen that because of the draining of the batteries the base line of the voltage rapidly decreased affecting the accuracy of the calibration curve. Our future improvement would add two three terminal positive fixed voltage regulators to the circuit to correct this problem.

This family of fixed voltage regulators are monolithic integrated circuits capable of driving loads in excess of 1.0 A. These three-terminal regulators employ internal current limiting, thermal shutdown, and safe-area compensation. Devices are available with improved specifications, including a 2% output voltage tolerance, on A-suffix 5.0, 12 and 15 V device types. Although designed primarily as a fixed voltage regulator, these devices can be used with external components to obtain adjustable voltages and currents. This series of devices can be used with a series-pass transistor to boost current capability at the nominal output voltage. (Jameco catalog)

The following are the device specifications of the voltage regulators that would be used:

Motorola

Three Terminal Positive Fixed Voltage Regulators

This family of fixed voltage regulators are monolithic integrated circuits capable of driving loads in excess of 1.0 A. These three-terminal regulators employ internal current limiting, thermal shutdown, and safe-area compensation. Devices are available with improved specifications, including a 2% output voltage tolerance, on A-suffix 5.0, 12 and 15 V device types.

Although designed primarily as a fixed voltage regulator, these devices can be used with external components to obtain adjustable voltages and currents. This series of devices can be used with a series-pass transistor to boost current capability at the nominal output voltage.

LM340,A Series

Three-Terminal Positive Fixed Voltage Regulators

T Suffix

Plastic Package

Case 221A

Pin 1. Input

2. Ground

3. Output

Heatsink surface is connected to Pin 2

[pic]

Maximum Ratings:

Input Voltage (5.0 V to 18 V) Vin = 35Vdc

(24V) Vin = 40Vdc

Electrical Characteristics:

Output Voltage Min Vo = 7.7 Vdc

Typ Vo = 8.0 Vdc

Max Vo = 8.3 Vdc

Output current = 5.0mA to 1.0A

The following is a circuit diagram of how the circuit would be re-wired for the two additional voltage regulators.

[pic]

It was also desired to test different concentrations of chlorophyll to test the adaptability of the colorimeter for different solutions. The solution was made by grinding spinach and mixing it with water and then centrifuging and filtering out the spinach leaves until a solution was made. The dilutions of the chlorophyll were made but never tested. A pin from the 741 op amp in the circuit broke and therefore did not reach all the way into the breadboard. If this problem had been rectified then the circuit would have been tested by changing the cherry rose filter to a green filter. The following rule can be used to estimate the additional absorption due to chlorophyll:

Absorption due to chlorophyll=.0667(C(g/l).758m-1 where C(g/l) = chlorophyll concentration in micrograms per liter. The wavelength to which the equation applies is 670nm. Yentsch and Phinney have described the method for measuring the concentration of plankton absorption by using a filter to collect the plankton samples. (C.S. Yentsch and D.A. Phinney 1988, Relationship between cross-sectional absorption and chlorophyll content in natural populations of marine phytoplankton. Ocean Optics IX, SPIE, vol. 925. Bellingham, WA:SPIE,109).

-----------------------

Lens

C

Photo-transistor

741

3

2

6

Multimeter

R1

R2

+9V

Light In

Vout

LED

Sample

Cherry Rose

Filter

Photo-Resistor

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download