Personal Warning Systems of Forklifts in an Industrial Environment

Environment

Bachelor thesis in electronics

(CEL304)

Per Spaak

psk03003@student.mdh.se

March 31, 2011

Mälardalens högskola, IDT Examinor, MDH: Mikael Ekström Supervisor, MDH: Mikael Ekström

I would like to dedicate this thesis to my father, as proof of a promise I made a long time ago on a pile of lumber.

Abstract English

I have in this report studied the problem of losing awareness when using pro-tective earmuffs in noisy industrial environments. Losing awerness can lead to smal and major injuris, but also to fatalities.

In this essay I have presented a potential solution to this problem and also a prototype. The prototype consists of a transmitter part attached to the vehicle and then a receiver part that is attached to the persons protective earmuffs.

This report is a thesis work for 15 credits, Basic level 300, for a Bachelor degree in Mechatronics at Mälardalen University.

I found this project very educational and also very interesting. I can see this as a potential solution for a already existing problem on this market.

Abstract Swedish

Jag har i denna uppsats studerat problemet med att man förlorar mycket upp-märksamhet när man använder hörselskydd i bullriga industrimiljöer. Förlusten av uppärmsamheten kan leda till små och stora skador, men även i vissa fall dödsolockor.

I denna uppsats har jag presenterat en potentiell lösning till detta problem och även en prototyp av denna lösning. Prototypen består av en sändardel som fästes på fordonet och sedan en mottagardel som fästes på personens härselkå-por.

Denna uppsats är ett examensarbete på 15hp, Basic level 300, för examen som Högskoleingengör inom Mekatronik på Mälardalens Högskola.

Jag fann detta projekt mycket lärorikt och även intressant. Jag ser detta som en potentiell lösning på ett befintligt problem ute på marknaden.

Acknowledgements

I would like to take this opportunity to thank my supervisor Mikael Ekström and my good friend Robin Lilja for all the good input and advise.

I would also like to thank Robin Pettersson and Sahag Normanian for putting up with my antics in the office.

Last but not least I like to thank Ralf ”Kip” Strömberg and Stig-Åke Svens-son for the good atmosphere in our small and cramped lab.

Contents

3.1.1 Acceleration assessment stage . . . 4

3.1.2 Signal modulator stage . . . 5

3.1.3 Signal output stage . . . 5

3.2 Receiver . . . 6

3.2.1 IR-Receiver stage . . . 6

3.2.2 DC offset removal stage . . . 8

3.2.3 Summarizing and Amplification stage . . . 10

3.2.4 Speaker stage . . . 11 4 Results 13 4.1 Transmitter . . . 13 4.2 Receiver . . . 13 4.3 Prototype . . . 13 5 Discussion 15 5.1 The work process . . . 15

5.2 Issues and Problems . . . 16

6 Conclusions 17 6.1 Summary . . . 17

6.2 Recommendations and Future work . . . 17

6.2.1 Transmitter . . . 17

6.2.2 Reveicer . . . 17

References 19

A Abbreviations I

B Circuit Board Diagrams III

B.1 Transmitter . . . III B.2 Reciver . . . IV

C Simulated test data V C.1 Transmitter . . . V C.1.1 Function Test . . . V C.2 Receiver . . . VI C.2.1 Function Test . . . VI

D Raw test data IX

D.1 Transmitter . . . IX D.1.1 Number of LEDs . . . IX D.2 Receiver . . . X D.2.1 Voltage follower test . . . X D.2.2 Final Breadboard Test . . . XI

1

Introduction

1.1

Background

The background for this thesis came from me (the author) working for several years in different industrial environments, and where protective earmuffs have been ignored for the benefit of alert- and awareness. Often you do not use protective earmuffs because it limits your field of alertness to only what can see and not hear. The problem with this is all the vehicles that will approach you from the side or behind.

1.2

Problem formulation

”Construct a device that can detect a forklift that the user may not see.”

1.2.1 Analysis of problem

• The first problem is to protect the ears without losing the perception of the surrounding area.

• The second problem is to make an application so foolproof that it can not fail.

• The third problem is to make it annoying enough to attract the users attention, but not so much that he or she will no use it.

1.3

Objective

The objective of this thesis is divided into two parts:

• First and foremost, to find theories applicable to the problem. • Second, to construct a working prototype.

1.4

Constraints

Transmitter:

• Change sound frequency depending on the speed. • No physical alterations to the forklift.

• Sound amplitude rising with shorter distance. Receiver:

• Be able to detect a forklift at 20m. • Battery powered.

• No physical alterations to the protective earmuffs. • Completely analog circuitry.

Prototype:

• 1 Receiver (one side only) • 1 Transmitter with:

– 1 LED cluster module – 1 speed assessment module – 1 signal modifier module

2

Method

2.1

Theory

The working theory in this thesis is "if I pulse a directed IR-LED in the direction the forklift is traveling, then with a IR-PhotoTransistor I will be able to pick up that oscillation and then translate it into sound, thereby alerting the subject."

2.2

Process

First in the project comes a research period, to find out as much useable infor-mation as possible about IR-light, IR-light in applications and also if there are any similar solutions out in the market.

Second comes the development part of the project, were a process of ”Trail and Error” will be applied to find the best solution.

When this is done, a prototype circuit board will be manufactured and last the report will be produced.

3

Solution

This solution is divided into two parts, ”Transmitter” and ”Receiver”, each part will explain the inner workings of each individual module.

3.1

Transmitter

The Transmitter consists of three parts:

• Acceleration assessment stage • Signal modulator stage • Signal output stage

These three parts work together as the following flowchart will show:

Figure 1: Flowchart for the transmitter module

The Acceleration assessment stage module will send a approximation of the instantaneous acceleration to the Signal modulator stage. The Signal modulator stage will then take this input and integrate it and then change the output frequency accordingly and send that frequency to the Signal output stage. The Signal output stage then sends that signal through a cluster of IR-LED’s [23].

3.1.1 Acceleration assessment stage

This stage consists of a accelerometer that sends instantaneous acceleration to the ”Signal modulator stage”. Since this kind of part is quite expensive, it will not be used in the prototype. In the prototype a three point switch will be constructed to simulate the accelerometer. The three point switch will consist of three voltage levels 0.5V , 2.5V and 4.5V [10]. This to simulate either full deceleration, no acceleration or full acceleration. The supply voltage will be provided by the ”Signal modulator stage” circuit board.

3.1.2 Signal modulator stage

The signal modulating state consists of a ATTINY13A-PU [2](from here on re-ferred to as MPU), driven by VDD of +5V produced by the positive voltage

regulator L7805CV [20]. The MPU is programed to integrate the instantaneous acceleration from the accelerometer. After integrating the instantaneous accel-eration to assess the speed of the forklift, the MPU will then send out a frequency dependent PWM-signal, with a duty cycle of 50%, ranging from 500Hz up to 1000Hz depending on the speed, to the output stage.

3.1.3 Signal output stage

According to the TSAL-6100 [23] data-sheet, the IR-LED’s maximum rating for continuous forward current is 100mA, at this current (fig.4 page 3 in the data-sheet [23]) the forward voltage is 1.35V . The drain-sourse voltage loss of the BUZ11A MOSFET transistor [18](from here on referred to as BUZ11A) is approximately 0V [18]. With a VCC of 8.4V , R1 should be chosen to

approxi-mately 57Ω. UR1= VCC− 2 ∗ VT SAL6100 UR1= 8.4 − (2 ∗ 1.35) UR1= 5.7 R1 = UR1 IC R1 = 57Ω

The 100kΩ resistor between the gate and the source, is there to prevent the gate from fluctuating when the PWM-signal is turned low/off.

After simulations in Multisim1, It was found that the ”on”-current will be 91mA, and this is within the specified limitations.

Figure 2: Signal output stage circuit diagram

The Vccof 8.4V was chosen because of the 8.4V LI-PO battery [11] that will

drive this entire circuit. The battery was chosen because of its low price and small size.

3.2

Receiver

The Receiver is constructed into three main parts, which are:

• IR-Receiver (right, left and back) stage • DC offset removal stage

• Summarizer and Amplification stage • Speaker stage

These three subparts work together as the following flowchart will show:

Figure 3: Flowchart for the receiver module

The reasoning behind splitting the IR-Receivers up is to achieve a stereo-sound effect, for the user to get more information and be able to hear in ”3D” (both direction, speed and distance).

This hole circuit will also be driven by a VCC+of 5V produced by the positive

voltage regulator L7805CV [20] and a VCC− of −5V produced by the negative

voltage regulator L7905CV [19]. These two voltage regulators vill then be driven by one 8.4V LI-PO battery [11]. In the prototype the operational amplifiers will be supplied with a ±15V , due to the fact that the OPA277PA [3] do not have rail-to-rail capability, and can not operate at input higher than approxemetly 3.5V .

3.2.1 IR-Receiver stage

The IR-Receiver are three ”BPW-76B Silicon NPN Phototransistor” [22] (from here on referred to as IR-transitor).

Figure 4: Relative Radiant Sensitivity vs. Angular Displacement [22]

As seen in the figure above, at about 40◦ away from centerline the ”Relative Radiant Sensitivity” is only at 50%, this means that at a angle of 40◦ the IR-trasistor will only register a in-signal of approximately half of the full intensity. If then the IR-transistors is place next to each other at an angle of 80◦away from the first IR-transistor’s centerline, they will overlap at 40◦ and both of them will then register a signal of 50% and the total will be a full signal strength as provided in the figure below.

Figure 5: Overlapping radiant sensitivity

To produce a signal from the IR-transistor, a emitter follower circuit is cre-ated. After extensive testing it was found that a resistor, R1, of approximately

This leads to a maximum current of about 0.315mA through the IR-transistor. VCC+= (R1//R2) ∗ IC+ VDiode 5V =  ₁ 15kΩ+ 1 150kΩ −1 ∗ IC+ 0.7V 5V − 0.7V =  1 15kΩ+ 1 150kΩ −1 ∗ IC IC= 4.3V 1 15kΩ+ 1 150kΩ
−1 IC= 0.315333...mA

The signal for the next stage, DC offset removal stage, is taken at the point between the resistor and the IR-transitor.

Figure 6: IR-Receiver stage

3.2.2 DC offset removal stage

In this stage the circuit eliminates any offset voltage by integrating the produced signal and then subtracting it from the produced signal. Before the integrator, a voltage follower is connected so that the integrator do not effect the input signal. The integrator [6] will produce a DC voltage equal to the offset but with reversed polarity. Both the voltage follower and the integrator is constructed with OPA277PA [3] operational amplifier.

Figure 7: Integrator [6] and Summarizer [8]

With a RC time constant τ of 4.7ms [24, p. 320] the integrator has a rise time trof approximately 10ms [13, p. 314]. The approximate DC amplification

AVDC is -1 [6] and the approximate AC amplification AVAC is -0.043 [6].

τ = R1∗ C1 τ = 10kΩ ∗ 0.47µF τ = 4.7ms tr= τ ∗ ln(9) tr≈ τ ∗ 2.2 tr≈ 10.3ms AVDC = − R2 R1 AVDC = − 10kΩ 10kΩ AVDC = −1 AVAC = − R2 R1 ∗ 1 1 + (2πf CR2) AVAC = − 10kΩ 10kΩ∗ 1 1 + (2πf ∗ 0.47µF ∗ 10kΩ) f ≈ 750Hz AVAC = −1 ∗ 1 1 + (2π750Hz ∗ 0.47µF ∗ 10kΩ) AVAC = −1 ∗ 1 1 + (2πf ∗ 0.47µF ∗ 10kΩ) AVAC ≈ −1 ∗ 1 23.1 AVAC ≈ −0.043

The third resistor R3 is also a bias current resistor of 5kΩ. R3= R1//R2 1 R3 = 1 R1 + 1 R2 1 R3 = 1 10kΩ+ 1 10kΩ R3= 5kΩ

The summarizer then adds the integrated signal to the original signal and thereby eliminates any offset voltage, making the detected signal oscillate around 0v. At this stage in the process the amplification Av1 is made to be −6. The

summarizer is constructed with the operational amplifier OPA277PA [3].

Av1 = − R6 R5 Av1 = − 900kΩ 150kΩ Av1 = −6

The fourth resistor R7 is as always a bias current resistor, chosen to 70kΩ.

R7= R4//R5//R6 1 R7 = 1 R4 + 1 R5 + 1 R6 1 R3 = 1 150kΩ+ 1 150kΩ+ 1 900kΩ R3≈ 69.2kΩ

Since the sensor directed to the back is supposed to send input signals both to left and right channel, a voltage follower is added after the summarizer to protect the signal from interference between the left and right canal. The voltage follower will also be built by the operational amplifier OPA277PA [3].

3.2.3 Summarizing and Amplification stage

Next step in the process is to add the left- (or right-) and back signal to create a left (or right) channel. In this process both signals, left and back, or right and back, are added in the second summarizer through two 150kΩ resistors.

The maximum amplitude that the chosen speakers can handle is approxi-mately ±0.8V (the specifics and calculations will be explained in the next section called ”Speakers”) and after several tests (See Appendix ”D.2 Receiver”) it was found that the signal will go from a swing of 4mV (distance of 20m) up to about 45mV (distance of 5m).

Since the ”DC offset removal stage” had a amplification Av1 of −6, this

last stage needs a amplification Av2 of −6 to get a total amplification AvT ot of

approximately 36. The AvT ot is sett so that the the output produced will be

Figure 8: 2nd Summarizer [8]

To achieve a Av2 of −6 the feedback resistor R3 in the 2nd summarizer

should be 900kΩ. This summarizer is also constructed out of a OPA277PA [3].

AvT ot = 1600mV 45mV AvT ot = 35.555... AvT ot = Av1∗ Av2 AvT ot = −6 ∗ −6 AvT ot = 36 R3= AvT ot∗ R2 R3= 900kΩ

The bias current eliminating resistor ,R4, was chosen to 70kΩ.

R4= R1//R2//R3 1 R4 = 1 R1 + 1 R2 + 1 R3 1 R4 = 1 150kΩ + 1 150kΩ+ 1 900kΩ R4≈ 70kΩ 3.2.4 Speaker stage

The speakers used in the prototype are KDSG20008 [15] from Kingstate. The rated input effect is 0.08W, this effect over the impedance of 8Ω gives a

maxi-mum continues input voltage of 0.8V . P = U ∗ I 0.08W = U ∗ U 8Ω 0.08W =U 2 8Ω U2= 0.08W ∗ 8Ω U =√0.08W ∗ 8Ω U = ±0.8V

Since the input to the speaker varies around 0V , it follows that the input signal can have a variance of 1.6V . But this is only true if the first half of the pulse has a maximum amplitude of 0.8V above GND and the second half have a maximum amplitude of −0.8V below GND.

To make sure this will be the case, after the second summarizer there will be a protection stage with two diodes. The first diode is directed from GND to the signal, to ensure the voltage never goes below −0.7V . The second diode is then placed directed to the GND, having the same effect but at the voltage 0.7V .

4

Results

4.1

Transmitter

Because the accelerometer is so expensive, it was not possible to determine if it was a valid solution to the problem. It also was not possible to determine if a accelerometer would work satisfactory in the industrial environment. As earlier mentioned the MPU code was not finished due to time management issues, and because of this there where no reason to construct the ”three point switch”.

After extensive testing it is clear that the ”Signal output stage” prototype do work exactly as designed, and after testing it was found that an array of 5 entities, with 2 LEDs in series in every entity, were needed to produce big enough output to transmit a useable signal at a distance of 20m.

4.2

Receiver

The receiver prototype built in this project do also work, though with some improvements it will work a lot better.

• Be able to detect a forklift at 20m

– After extensive testing, results provided in Appendix D, it was found that it is possible to detect the transmitter unit at a distance of 20m.

• Battery powered.

– It where provided in the report that the receiver will be supplied by the 8.4V LI-PO battery [11].

• Stereo sound (direction estimating).

– By using three BPW-76B [22], a stereo sound effect is produced.

• No physical alterations of the protective earmuffs.

– This part of the project where not addressed in this report due to time management issues.

• Completely analog.

– By using only operational amplifiers the receiver was kept completely analog.

4.3

Prototype

• 1 Receiver (one side only)

– This prototype was invented, produced and tested satisfactory. The prototype consists of the back and left sensor and the following stages.

• 1 Transmitter with:

∗ By using 5 arrays with 2 LEDs in in every array, a working cluster was produced.

– 1 speed assessment module

∗ As mentioned, This stage was not produced due to the econom-ical factor.

– 1 signal modifier module

5

Discussion

5.1

The work process

The process used when going in to this project was to start off with an intense research period where I tried to find out as much useable information as possible about IR-light, IR-light in applications and also if there are other areas where IR-light is being used.

During this process I also tried to find out if there already exist a solution like this on the market, or any similar solution but used in other areas. But I could not find anything similar to my solution, neither on the problem formulated in this report or in other similar areas.

The second stage in the process where to try to design a initial solution proposal, so that I could order parts and components for the testing process. This part of the process became ultimately the largest part, this do to major delays in the shipping of the vital components.

A major part of this project has been spent in computer simulations and datasheets, were I could approach the problem from a ”Trail and Error” prin-cipal. To me, this approach means that I first get me some basic knowledge around the problem, then I test my way forward to a good result, and then try to confirm this results with equations and theory.

This way of working works really well together with a simulations program, because then you do not have to worry about maybe breaking and possibly destroying the prototype. There is also a time saving aspect to it, since it is relatively easy to make changes or even just ”cut n’ paste” some parts of the circuit to test it individually.

The programs I have utilized during this process of simulations and calcula-tions, and also during the process of making the drawings for the circuit boards is:

• Matlab2

• Multisim3

• eagleCAD4

When I finally found a solution that I were satisfied with, and the compo-nents had arrived. Then I started to construct the circuit in a so called ”bread-board5_{”, and perform live test such as distance test and functionally tests.}

Although this design was good, it was not my final design, since I changed the design in the very end of the project due to a unthought of complication. This later design is the one I will present in this report and it is also the design on which the tests, in the part appendix, were preformed if nothing else is stated. As I now were satisfied with the solution I had made, I started to draw the actual circuit boards. The circuit boards were design in the program EagleCAD 5.11.

2_{MathWorks: Matlab R2009b}

3_{National Instruments: Circuit Design Suite 11.0} 4_{CadSoft Computer: EAGLE 5.11}

I have, during the major part of this project, been working with this report, and have been changing it as the project progressed. To my help, I have been using this programs:

• TeXShop6

• TeXworks7

• Photoshop CS3 and CS58

5.2

Issues and Problems

Due to time management problems the final code for the MPU was not produced in time for this report, and due to economical issues the ”Acceleration assessment stage” was not produced nor tested.

Some other problems that occurred during the testing phase were that some of the operational amplifiers were broken and there fore made for some inter-esting and puzzling results.

During the breadboard phase there where noise on the signal, this is probably due to long wiring and bad grounding of the breadboard itself.

6_{TeXShop 2.37} 7_{TeXworks 0.2.3}

6

Conclusions

6.1

Summary

This report has addressed the potential solution to the problem of losing a lot of alertness when using protective earmuffs. It handles my thoughts and theories on the problem and also a prototype for the solution.

6.2

Recommendations and Future work

Since this is a report on a prototype, there are a lot of things that with more money and time can be improved, and now I will state some of this points.

6.2.1 Transmitter

• Changing to surface mounted components • Making some sort of battery warning

• Finding a good way of attaching the component to the vehicles • Finish the code for the MPU

6.2.2 Reveicer

• Changing to surface mounted components • Making some sort of battery warning

• Replace the operational amplifiers with some that have ”rail-to-rail” capa-bilities

• Replacing some of or all the operational amplifiers for some built in a ”Quad” capsule [1]

References

[1] Analog Devices. Dual/quad rail-to-rail operational amplifiers op295/op495. https://www1.elfa.se/data1/wwwroot/assets/datasheets/cpOP_ 495_Datasheet_EN.pdf, 2008. Rev. F

Visited 2011-03-20 13:47.

[2] ATMEL. Attiny13a. http://www.atmel.com/dyn/resources/prod_ documents/doc8126.pdf, Jul 2010. Rev. 8126E–AVR–07/10

Visited 2011-02-02 14:30.

[3] BURR-BROWN. High precision operational amplifiers. https://www1. elfa.se/data1/wwwroot/assets/datasheets/dyOPA277_e.pdf, March 1999. SBOS079.

Visited 2011-02-24 17:02.

[4] ECE Lab. The bipolar transistor as a switch. http://ecelab.com/ switch-bjt.htm.

Visited 2011-02-28 11:01.

[5] ECE Lab. Switching circuits using bipolar transistors. http://ecelab. com/switch-bjt2.htm.

Visited 2011-02-28 11:01.

[6] Electonics-Tutorials. The integrator amplifier. http://www. electronics-tutorials.ws/opamp/opamp_6.html.

Visited 2011-02-28 11:03.

[7] Electonics-Tutorials. Mosfet as a switch. http://www. electronics-tutorials.ws/transistor/tran_7.html.

Visited 2011-02-28 11:06.

[8] Electonics-Tutorials. The summing amplifier. http://www. electronics-tutorials.ws/opamp/opamp_4.html.

Visited 2011-03-16 17:32.

[9] Electonics-Tutorials. Transistor as a switch. http://www. electronics-tutorials.ws/transistor/tran_4.html.

Visited 2011-02-28 11:06.

[10] Elfa. Accelerationsgivare, kas901-04a. https://www.elfa.se/elfa3~se_ sv/elfa/init.do?item=73-218-00&toc=19312.

Visited 2011-03-20 14:39.

[11] Elfa. Battery pack 8.4v 600 mah, 9v li-po 600mah. https://www.elfa. se/elfa3~se_sv/elfa/init.do?item=69-204-66&toc=19088, xxxx. Visited 2010-12-21 13:14.

[12] Universitetslektor Mattias Hammar. Föreläsningsanteckningarelektronik, del 2( 2b1520), vt2006. http://www.imit.kth.se/courses/2B1520/ del2/Lecture%20notes/Lecture%209.pdf, 2006.

[13] Lennart Harnefors, Jonny Holmberg, and Joop Lundqvist. Signaler och system med tillämpningar. Liber AB, Stockholm, Sweden, first edition, 2004.

[14] KELAG. Single axis acceleration sensor. https://www1.elfa.se/ data1/wwwroot/assets/datasheets/KAS901-04_eng_TDS.pdf, January 2011. Revision January 2011 Visited 2011-03-20 14:33. [15] KINGSTATE. Kdsg20008. https://www1.elfa.se/data1/wwwroot/ assets/datasheets/xhKDSG20008_Data_E.pdf. Visited 2011-03-02 16:12.

[16] Philips Semiconductors. 2n3904 npn switching transistor. http:// ciberia.ya.com/ksys_net/2n3904.pdf, Apr 1999.

Visited 2010-12-16 16:22.

[17] Philips Semiconductors. Bat85 schottky barrier diode. https://www1. elfa.se/data1/wwwroot/assets/datasheets/tbBAT85_data_e.pdf, May 2000. Visited 2010-12-16 19:14. [18] ST. Buz11a. http://www.datasheetcatalog.org/datasheet/ stmicroelectronics/2948.pdf, July 1999. Visited 2011-02-15 10:39.

[19] ST. L7900 series negative voltage regulators. https://www1.elfa.se/ data1/wwwroot/assets/datasheets/tgL79xx_Data_E.pdf, August 2005. Rev. 10

Visited 2011-03-20 13:57.

[20] ST. L78xx l78xxc positive voltage regulators. https://www1.elfa. se/data1/wwwroot/assets/datasheets/teL78xx-L78xxC_e.pdf, March 2008. Rev. 19

Visited 2011-03-20 13:55.

[21] Vishay Semiconductors. Bzx55-series. https://www1.elfa.se/data1/ wwwroot/assets/datasheets/07005309.pdf, July 2007. Rev.1.5

Visited 2011-02-28 17:05.

[22] Vishay Semiconductors. Bpw76a, bpw76b. http://www.vishay.com/ docs/81526/81526.pdf, Sep 2008. Rev. 1.4

Visited 2010-12-13 22:99.

[23] Vishay Semiconductors. Tsal6100. http://www.vishay.com/docs/81009/ tsal6100.pdf, Jun 2009. Rev. 1.6

Visited 2010-12-13 22:00.

[24] Ron Walls and Wes Johnstone. Introduction to Circuit Analysis. West Publishing Company, Kellog Boulevard, St. Paul, 8th edition, 1999.

A

Abbreviations

MPU Micro Proccesing Unit IR-light Infra Red light LI-PO Lithium Polymer GND Ground

B

Circuit Board Diagrams

B.1

Transmitter

B.2

Reciver

C

Simulated test data

C.1

Transmitter

C.1.1 Function Test

Figure 12: Output signal and Control signal

C.2

Receiver

C.2.1 Function Test

In the simulations the BPW-76B were replaced with a 2N3904 bipolar transistor [16] to simulate the input signal. The 2N3904 bipolar transistor was driven by a PWM signal that in the figure is driven by the function generator. The input signal is a PWM signal with a amplitude of 20mA and a offset of 1.2V

Figure 14: Input signal

D

Raw test data

D.1

Transmitter

D.1.1 Number of LEDs

Distance 4 LEDs 8 LEDs 10 LEDs 5m 20mV 40mV 50mV 10m 7mV 15mV ˜17mV 15m <5mV9 _{7mV 10mV}

20m <2.5mV10 5mV >5mV

D.2

Receiver

D.2.1 Voltage follower test

This test was preformed to try to estimate the need for a voltage follower before the integrator (see 3.2.2 DC offset stage). This test were made without bias current compensation in the operational amplifiers.

Distance Without Range With Range 5m 20mV −5mV → +15mV 45mV −20mV → +25mV 10m 5mV +0mV → +5mV 15mV −5mV → +10mV 15m ˜4mV +2mV → +6mV 7mV −1mV → 6mV 20m ˜2.5mV 0mV → 2.5mV 5mV +7mV → +12mV

2.5mV → 5mV

At 20m the signal without voltage follower swung too much to make a definite reading so I took two readings at the ends of the swings.

D.2.2 Final Breadboard Test

This test was made with the final design except that there was a amplification of x1 (R = 150kΩ) and of x2 (R = 300kΩ) in the final summarizing stage, and that the back signal was connected to GND11_.

Distance x1 Range x2 Range 5m 45mV −22.5mV → +22.5mV 90mV −45mV → +45mV 10m 15mV −7mV → +8mV 28mV −14mV → +14mV 15m 6mV −3mV → +3mV ˜_{13mV −6mV → 7mV} 20m 4mV −2mV → +2mV 10mV −5mV → +5mV