Home

Improved Reference Infrared Spectrometer - Aalto

1. 27 0 8 0 6 0 4 Output efficiency n 0 2 0 0 0 01 0 1 1 10 100 Resistor ratio R R Figure 2 14 The relation between the output efficiency and the resistance ratio R Rp As seen in equations 2 28 the output sensitivity of the detector is also directly proportional to the bias voltage Therefore it is usually beneficial to use as high bias voltage as the detector can withstand However there are some exceptions lead selenide and lead antimony detectors have a specific threshold level for bias voltage after which the noise levels suddenly rise So the bias voltage should not be any higher than necessary In some cases the heating of detector element due higher bias voltage causes a problem as the D decreases even the detector itself can endure 44 In such cases the manufacturer usually recommends the optimum biasing voltage range With mercury cadmium telluride MCT detectors the bias current is more important factor than bias voltage or load resistance matching Load resistance is usually in the range of kilo ohms and the bias voltage is adjusted to achieve the wanted bias current Figure 2 15 shows typical signal and noise behavior of MCT detectors as a function of bias current 44 28 S N SIGNAL S ARB UNIT ES NOISE N ARB UNIT 0 5 10 15 20 25 BIAS CURRENT mA Figure 2 15 Typical signal and noise behavior of MCT detectors 44 3 Setup description and improvements
2. suffix v Table Al SI units in radiometry 43 65 Quantity Symbol Unit Definition The total energy emitted Radiant Q oken s transferred or received as energy i TB radiation in a defined period of time Radiant di a Power emitted transferred or a o W kg m s gt gt a a a flux received as radiation ectral CRE I Do W m kg m s Radiant power per wavelength 5 TA lo 23 Padian Le Y ni Radiant power per unit solid angle intensity ST Spectral Wan ae radiant Lo 23 3 31 Radiant intensity per wavelength intensity HD Radiance LE W m sr Radiant power per unit solid angle kg m fs sr per unit projected source area Special L Wem egr Radiance per wavelength radiance 2 kg m gt s sr P 8 i 4 Radiant power incident on Irradiance Ee W m 2 kg m s HR i surface ctral pa f P e de Ea Wm gt kg m s Irradiance per wavelength Radiant xitan 4 E Li ja Fai M W m kg m s Power emitted per unit source emittance 79 Appendix B Photoconductive detector preamplifier schematics To uC D8 To uC D9 WLLVSNO LNY W6 1006 06 gt 6 006 00L N N E x x gt a N E gt o x ox EN o _ m A Q N ll El tT KN K lt N O o O 1 lol 6 e lt y ao J SS 5 E N5 N Tess cae cc AW To X1 1 To X1 2 X11 2 INPUT To X14 1 To X14 2 Figure BI Schematic of the preamplifier circuit 80 NOE E
3. a Visible spectrum 400 500 600 700 Increasing Wavelength A in nm gt Figure 2 1 The Electromagnetic spectrum 9 There are countless standards and conventions based on context and applications for subdividing the vast infrared region For the purposes of this study a division commonly used in astronomy was used This division also presented in Figure 2 2 was selected mainly for practical reasons as it was also used in all the previous works 3 4 and 5 The infrared spectrum is split into three regions Near infrared region NIR includes wavelengths from the end of the visible spectrum around 760 nm up to 5 um Mid infrared region MIR covers the range between 5 um and 30 um Far infrared region extends all the way to the edge of microwave freguencies from 30 um to 1 mm These regions are also Known as short wave mid wave and long wave infrared regions respectively The region overlapping far infrared and microwave radiations is often referred as the terahertz waves The measurement setup improved as a part of this thesis was originally used mainly in the near infrared region 3 but was later extended to the mid infrared regime covering wavelengths up to 16 um 5 VISIBLE LIGHT INFRARED MICROWAVE Figure 2 2 One commonly used subdivision for infrared region also used in the scope of this work 2 1 2 Spectrometer Any device that measures any kind of spectrum is a spectrometer per se In an optical
4. 14 and 15 2 2 1 Ideal black body Electromagnetic radiation is emitted from all matter with a temperature above absolute zero A black body is a theoretical idealization of a physical body that absorbs all incident electromagnetic radiation Since there is no reflected radiation the black body radiates incandescently in a continuous spectrum that depends only on the body s temperature The spectral radiance of an ideal black body that is power emitted per unit surface area that falls within a given solid angle as a function of wavelength is defined by Planck s law 2hc La te A 2 3 25 Ga where Kp is the Boltzmann constant 1 380 6503 10 J K and T is the absolute temperature Spectral radiance of a black body as a function of wavelength in different temperatures is plotted in Figure 2 3 A black body is also a perfect Lambertian radiator meaning that its radiance is independent of view angle and therefore its radiant intensity J is determined by the Lambert s cosine law I I 0 cos L A cos 2 4 where is the angle between the viewpoint and the source surface normal L is the source radiance and A is the source area 2 via 200 K 9 10 300 K E 400K z 1 600 K 8 10 1000 K S 1500 K 3 10 2000 K lt 10 3000 K E o 4000 K 5000 K n 10 6000 K 100 1000 10000 100000 Wavelength nm Figure 2 3 Spectral radiance
5. Figure 6 3 Linearity of the preamplifier measured at 80 Hz Values are normalized with the value at 1V 100 mV 10 mV and 1 mV at gains 0 20 40 and 60 dB respectively 6 2 3 Frequency response Low frequency response in the range of 10 to 200 Hz is shown in figure 6 4 This is the most interesting range since it is important in optical measurements In this range the three smallest gains have an very flat response and even the highest gain is flat within 0 1 dB Also the high frequency response was measured up to 10 kHz and is shown in figure 6 5 Due to the constant feedback capacitance of the amplifier circuit and the changing feedback resistance the cutoff frequency lowers when gain is increased But even with the highest gain the 6 dB cutoff frequency is around 6 KHz 66 Normalized gain dB Frequency Hz Figure 6 4 Frequency response of the preamplifier in the range of 10 to 200 Hz Values are normalized with gain at 75 Hz frequency n g c A O 3 5 D N E A G 1 e G 10 4 G 100 6 v G 1000 100 1000 10000 Frequency Hz Figure 6 5 Frequency response of the preamplifier in the range of 100 Hz to 10 kHz Values are normalized with gain at 100 Hz frequency 67 6 3 Optical measurements For comparison optical measurements were done with Hamamatsu Photonics P2038 03 lead selenide detector with which the previou
6. 3 1 05 0 04 um 1 2 2 0 um Si LOT Oriel 50829 FK 4 1 65 0 07 um 2 0 2 7 um Ge LOT Oriel 50830 FK 5 2 40 0 09 um 2 7 4 0 um Ge LOT Oriel 50831 FK 6 3 60 0 14 um 4 0 6 0 um Ge LOT Oriel 50832 FK 7 5 60 0 10 um 6 0 10 5um Ge BARR A36 8 10 30 0 10 um 10 5 16 0 um Ge BARR X0041 Terahertz Technologies C 955 optical chopper is placed in front of the monochromator It should be as close to the entrance slit as possible since this minimizes the loss caused by the finite aperture of the chopper This also guarantees sharp rectangular shape of modulation even with low chopping frequencies The chopper has two apertures inner and outer with frequency ranges of 4 500 Hz and 40 5000 Hz respectively Both provide the 0 01 accuracy and 1 mHz resolution Normally only the inner aperture is used since it provides smaller phase Jitter shorter rise and fall time of chopped optical signal and larger beam diameter The most important part of wavelength selection is the Czerny Turner type monochromator M IP 23 It was originally built to cover working range of 200 2000 nm 47 but the range has been extended by purchasing new gratings with smaller groove frequency All available gratings and their properties are listed in the table 3 4 36 Table 3 4 Available gratings and their properties Original accessories of the monochromator are underlined 5 Grating Groo
7. 6 Preamplifier for photoconductive detectors A preamplifier for photoconductive detectors was designed and built to replace a broken preamplifier The old device was made according to the manufacturer s general recommendations for photodiode preamplifier The design was mainly used with Hamamatsu Photonics P2038 03 lead selenide detector albeit the gain bias voltage and load resistance were optimized for MCT detectors and does not suit well for the specific detector 51 Detailed schematics of the previous amplifier and along with explanatory text can be found in 3 The design of the new amplifier is introduced in the first subchapter and the second focuses on the characterization of the device Optical measurements done with the new amplifier are discussed in the third Specifications of the amplifier are shown in table 6 1 The bandwidth of the amplifier is discussed more thoroughly in section 6 3 2 Table 6 1 Specifications of the amplifier Gain O 20 40 or 60 dB Internal load resistance 1 kO 32 MO two values per decade Input resistance 100 MQ Bandwidth 10 Hz 200 Hz within 0 1 dB 1 Hz 1 5 kHz within 3 dB Internal biasing voltage 0 64 V Internal biasing current Max 120 mA External biasing voltage Max 400 V 6 1 Design The new amplifier was designed so that it could be used with a wide variety of detectors The old design served as a starting point for the new device and the
8. Input voltage vii Vout DDD o S N SG o S S S 3 N 3 S cho SD esos o S S Do Or Noise voltage Extraction voltage of a photoconductive detector Output voltage Voltage in Alternating Current Virtual Instrument Virtual Instrument Software Architecture Lock in amplifier X output Width of a monochromator s exit slit Distance from the optical axis Lock in amplifier Y output Input impedance Output impedance Ratio of load resistance and dark resistance Output voltage sensitivity Maximum of output voltage sensitivity Net radiant flux Emissivity Load resistance efficiency of photoconductive resistor Angle between viewpoint and surface normal Diffraction angle Angle of the incident light Wavelength Photon wavelength in vacuum Cut off wavelenght Wavelength for maximum spectral radiance contrast Wavelength corresponding to diffraction order m Wavelength for maximum spectral radiance Nominal attenuation Frequency Wavenumber Reciprocal linear dispersion Stefan Boltzmann constant 5 670 373 10 J s m K 1 Phase shift Radiant flux Spectral power Reference frequency viii 1 Introduction We are constantly surrounded by electromagnetic radiation Most of it goes unnoticed since only a small part of the radiation spectrum is visible to human eye The vast region of optical radiation that has longer wavelength than visible light yet shorter than microwaves is known
9. Multimeter Monochromator 924M JONI Figure 3 1 The layout of the measurement setup and an illustration of the electrical signal paths The spherical mirror is used to collect and focus the radiation from the infrared source to the monochromator through a filter wheel and an optical chopper Light travels through a one inch wide zinc selenide window from a box to another The optical chopper is placed in front of the entrance slit of the monochromator The desired wavelength is then focused to the exit slit of monochromator The beam travels through another optical window to the detection box The two off axis parabolic mirrors are used to collimate and focus the beam to the detector plane Detectors are mounted on a linear detector stage so that the reference detector and detector under measurement can be in turn placed and fine adjusted to the beam This substitution method is thoroughly analyzed in 3 As a part of the setup improvement all measurement and control devices were mounted to a standard 19 inch rack cabinet shown in figure 3 2 The tag numbers refer to devices listed in the table 3 2 All devices except the computer and its display are powered via rack power strip This allows the setup to turned on and off with a single flick of a switch The computer and its accessories are connected to a different phase in the power 31 grid so that the disturbances caused by the switched mode power supply could be avoided A
10. the setup was automated for most parts The original automation software was written in Java but also used routines written in C and C external C libraries with Java Native Interface JNI dynamic link libraries DLL Windows ActiveX components and various native support packages All in all the software was rather complicated platform dependent sometimes unreliable and laborious to modify and update When the development of the setup began it became clear that various parts of the original code had to be rewritten and also the old computer had to be replaced As a part of the setup modernization the automation programs were decided to be rewritten with visual programming language LabVIEW from National Instruments NI It is a widely used system design platform and development environment very appropriate for device control and automation It is also de facto standard in the laboratory measurements since it provides a large selection of ready made device drivers and data processing libraries 58 Unlike a set of compiled Java programs LabVIEW allows very adaptive programming so that built tools can be used with ease for any type of measurement LabVIEW functions are coded as separate virtual instrument files or VIs that act as building blocks for new programs A VI used inside another is commonly referred as a subVI All code was written with LabVIEW 9 0f3 being the latest available version of the software available at the tim
11. was optimized The most essential measurement and calibration programs were written In addition the versatile facility provides a user friendly selection of tools for a wide variety of other measurement procedures The linearity of phase sensitive detection devices is not well known The problem was studied in order to gain better understanding of nonlinearities in lock in amplifier 69 measurements As a result a fully electrical method for linearity measurements of lock in amplifiers was developed and tested This method improves the accuracy in lock in comparison methods and for example the nonlinearity component in the uncertainty of responsivity measurements is reduced to 0 2 The method can also be utilized for characterization of other amplifiers The amplification of a photoconductive detector s signal reguires various parameters of the amplifier to be matched to the detector in guestion Commonly a preamplifier is dedicated to some type of detectors As a part of the thesis a universal preamplifier for photoconductive detectors was designed built characterized and tested It allows a single device to be used with a large variety of detectors The preamplifier was also used in measurements to confirm the decrease in the response of lead selenide detectors at high power levels 70 References 1 2 3 4 5 6 7 8 Mohr P J Taylor B N Newell D B The 2010 CODATA Recommended Va
12. 57 10 J s or 4 136 667 516 10 eV s Therefore the spectrum of electromagnetic radiation can be denoted eguivalently in terms of energy freguency or wavelength 6 It should be noted that the speed of light and the wavelength of a photon are dependent of the propagation medium and in general wavelength refers to photon s energy using the speed of light in vacuum co 299 792 458 m s and the corresponding wavelength Jo Often used guantities are also the angular wavenumber K and the spectroscopic wavenumber 7 atk 2 2 2 1 1 Infrared spectrum The electromagnetic spectrum with commonly used division to categories is presented in figure 2 1 Wavelengths ranging from 5 nm to 1 mm are traditionally considered to be the region of optical radiometry Only a tiny fraction of the whole range wavelengths between 380 nm and 760 nm are visible light Slightly different values for both definitions can be found in literature In any case optical radiation with shorter wavelength or correspondingly higher energy and optical radiation with longer wavelength and lower energy are called ultraviolet UV and infrared IR radiation respectively 8 lt Increasing Frequency v 10 10 10 10 10 10 10 10 10 10 10 10 10 v Hz Y rays X rays UV IR Microwave FM AM Long radio waves Radio waves I I I Uni I l 106 jo 102 10 10 50 103 10 109 10 10 10 10 X m bc Increasing Wavelength A gt
13. a transfer standard for infrared responsivity measurements In the left side picture a hemispherical reflector is used for reflectance correction 38 9400 9200 9000 8800 8600 Responsivity V W 8400 8200 3 0 3 5 4 0 4 5 5 0 5 5 6 0 Wavelength um Figure 3 5 Spectral responsivity of the transfer standard between 2 8 and 6 um measured at 15 Hz chopping frequency The data points shown as red dots were obtained by comparing the responses of the transfer standard and the primary reference 5 The frequency response of the transfer standard on the other hand is known to be very poor 50 To confirm this behavior and to acquire more accurate correction coefficients the relative responsivity was measured at low chopping frequencies from 4 to 15 Hz shown in figure 3 6 Values are compared to the responsivity at 15 Hz chopping frequencies which is also used in the comparison measurements with the primary standard More detailed documentation about the transfer standard is available in 5 39 Relative responsivity Chopping freguency Hz Figure 3 6 Freguency response of the transfer standard at low freguencies Relative values are compared to the responsivity at 15 Hz chopping freguency Strong noise signal level at line freguency and its multiples has been reported previously to cause problems in measurements around same freguency range This was confirmed by measuring the
14. applications with emission ranging from visible light to 2 5 um are available 5 Gas discharge lamps as the name suggests send electric discharge through ionized noble gas or mixture of gases usually mixed with other materials like mercury Collisions between free electrons and atoms excite some of the electrons in atomic orbitals to a higher energy level which then emit light at a material specific emission 12 lines A special argon arc source has been designed for radiometric purposes It is superior to incandescent infrared sources in the 1 10 um wavelength region 29 Cryogenic application can extend wavelengths up to 20 um 30 2 3 Considerations on infrared optics The principles of optics are fundamentally the same in infrared region than in any other Still some special features have to be taken into account for instance in component and material selection The infrared absorption of normal atmosphere is one major concern This problem can be solved by using only suitable wavelengths or performing the measurements in vacuum or in a purging gas such as nitrogen The latter solution of purging is also used in the measurement setup related to this thesis and is examined thoroughly in 4 Absorption is also a problem in optical components Lens materials commonly used in the visible region are rather opague at longer wavelengths There are many alternative materials in the near and mid infrared regions many of which
15. four air fans and a variable transformer so that the rate of the airflow can be adjusted The signal generator 13 and the precision attenuator 12 are actually not part of the spectral responsivity measurement setup Originally they were used to characterize the electrical properties of the setup However they were left as a part of the setup as they provide a simple and fast auto calibration method for the phase sensitive measurement 33 3 2 Radiation sources and input optics The primary source of infrared radiation in the setup is a ceramic glower model Mini Igniter 301 manufactured by the Saint Gobain Ceramics Individual glowers have significant differences in the material properties at the nominal 12 V operating voltage the steady state current can vary from 1 to 2 4 A This in turn leads to the temperature range of 1150 1455 C More important factor is the power dissipation of the glower which can be adjusted by fine tuning the operation voltage The maximum rated temperature of the device is 1580 C 21 Assuming that the glower is roughly a gray body the Wien s displacement law in eguation 2 5 predicts the spectral radiance to peak around 2 um region when the glower is used in the nominal temperature range The previously used glower was damaged and had to be replaced with a new one The dimensions of the new version glower were unfortunately changed even though the model number is the same The actual glowing t
16. in decades to measure lock in amplifiers linearity Frequency response is measured by changing the frequency with the signal generator 50 within a given freguency range All values are then saved to a comma separated values CSV file the name and folder of which are either given in advance or prompted at the end of the measurement Lock in linearity VI measures the output of the lock in amplifier using manually selected parameters such as the time constant and settling time at the given signal generator voltages and freguencies The possible usage of attenuator is also specified By default the sensitivity range is optimized by the measurement VI but it can also be selected manually or by using the auto gain function of the lock in amplifier Freguency response VI can be used in a wide variety of measurements that reguire freguency sweeping A given number of points are measured within the user specified freguency range or a manual list of measurement freguencies can be given A linear or exponential sweeping can be selected the latter being often more rational when large frequency range is used Measurements are done by using either the lock in amplifier or the multimeter When high accuracy is reguired also the freguency response of the signal generator should be measured for comparison Most accurate AC measurements can be done with the multimeter using a so called synchronous sub sampled computed true RMS technigue It provid
17. merely using electrically heated resistance wire as an infrared source This of course has a very limited temperature range and therefore high radiances can only be attained briefly A ceramic element on the other hand can operate well above 1000 K temperatures and have acceptable emissivity of 0 8 20 Spectral radiance of a commercial ceramic element is presented in figure 2 5 Some ceramic sources used originally for ignition purposes reach temperatures around 1800 K 21 Nernst glower is a similar nowadays widely obsolete device composed of a mixture of certain oxides It does not conduct electricity at room temperature but reguires external start up heating One of its main replacements is a device alike but made out of silicon carbine and commonly known as a globar These devices make external heating unnecessary function at a high temperature and have an emissivity as good as 0 88 5 100000 10000 100 Spectral Radiance Wi m sr um Wavelength um Figure 2 5 Spectral radiance of a ceramic element in different temperatures 20 The peaks are noticeably narrower compared to a black body radiator 11 Higher emissivity can be achieved with cavity blackbody radiators They are based on a closed cavity with a small hole for output radiation This structure captures almost all incident radiation and when it is Kept at a constant temperature it almost resembles an ideal black body although for best performance the
18. of an ideal blackbody as a function of wavelength in different temperatures From figure 2 3 we can distinctly see that the wavelength at which the spectral radiance reaches its maximum decreases when temperature increases Wavelength for maximum spectral radiance at a given temperature can be found by setting the partial derivative of eguation 2 3 egual to zero a b cha 0 gt Amax a nE 2 5 This results is known as Wien s displacement law where b is called Wien s displacement constant 2 8977685 10 m K Using this law we can observe that black or grey bodies having a temperature around 3 3800 K will peak of emitted power in the infrared region Another result somewhat similar to Wien s displacement law is the law for maximum temperature sensitivity of spectral radiance also known as maximum contrast law By taking both partial derivatives and the setting the result to zero we can obtain De 8 2 6 0 agi 0 gt A asiain T al aa Lor where b is a maximum contrast constant 2 410 10 m K So for instance at a temperature of 300 K the spectral radiance has a maximum around 9 7 um but the spectral radiance is most sensitive to temperature changes around 8 0 um 2 The ratio of energy radiated by a physical body to energy radiated by an ideal black body at the same temperature is called emissivity e All real materials have emissivity less than one and it is among other things a function of wavele
19. of microporous titanosilicate materials Doctoral Thesis University of Delaware Department of Chemical Engineering Neward 2008 231 pp Malacara D Thompson B J Handbook of Optical Engineering New York Marcel Dekker Inc 2001 978 pp ISBN 978 0 8247 9960 1 Pedrotti F L Pedrotti L S Pedrotti L M Introduction to Optics 3 edition New Jersey Pearson Prentice Hall 2007 622 pp ISBN 978 0131499331 Kingston R H Optical Sources Detectors and Systems San Diego Academic Press 1995 198 pp ISBN 978 0124086555 Wolfe W L Zissis G J The Infrared Handbook 4 printing of revised edition Michigan Environmental Research Institute of Michigan 1993 1700 pp ISBN 0 9603590 1 X Shumaker D L Accetta J S The Infrared and Electro Optical Systems Handbook Bellingham SPIE Optical Engineering Press 1993 3524 pp ISBN 978 0819410726 Jones R W Infrared Technology Kirk Othmer Encyclopedia of Chemical Technology 2000 Referred 2 11 2011 Internet http onlinelibrary wiley com doi 10 1002 0471238961 0914061810151405 a01 full Ballico M A simple technigue for measuring the infrared emissivity of black body radiators Metrologia 2000 Vol 37 Pp 295 300 ISSN 1681 7575 Hecht E Optics 4 edition San Fransisco Addison Wesley 2002 698 pp ISBN 0 321 18878 0 Product datasheet ncandescent reflector infrared heat lamps industrial Philips 2002 72 19 20 21
20. signal multiplication itn dla cos Pm cos 2w t Pm Un sin w 2 21 a FAME Uin sin om sin 2tw t Pm Un cos w 2 22 2 24 The multiplied signals are then being filtered with low pass filter When the cut off freguency is much lower than the reference freguency ideally only DC signal is left The outputs X and Y can then be expressed as X COS m 2 23 Y En sin Pm 2 24 2 So called magnitude R which is directly proportional to the amplitude of the measured signal and phase of the measurement signal can now be calculated from the output values RWY 4Y frim 2 25 m tan Y X 2 26 Usually these values are computed in the lock in amplifier but if very rapid sampling is needed the calculation from X and Y values can be done later For most devices X Y and R are scaled so that they represent the RMS value of measured signal Usually one can also measure the signal level at harmonic multiples of the reference freguency 41 The phase sensitive detection can be done either analogically or digitally Modern digital lock in amplifiers outperform analog counterparts in virtually all respects The main problems in analog multiplication are related to insufficient harmonic rejection output offsets errors in the gain and rather limited dynamic reserve Instead of gain the term sensitivity is used to characterize the amplifications of a lock in amplifier It is simp
21. though are fragile hygroscopic or toxic Many metals on the other hand have well known and flat reflectance over wide spectral range Therefore reflective components are often a preferable option in the mid and far infrared regions The reflectance spectrum of various metal coated mirrors is shown in figures 2 6 and 2 7 13 Coated Mirrors 0 Angle 85 Reflectance 90 80 75 0 4 1 10 20 Wavelength um Figure 2 6 The reflectance spectra of aluminum red gold green and silver blue coated IR mirrors in 0 angle 31 Coated Mirrors 45 Angle Reflectance 9 1 10 20 Wavelength um Figure 2 7 The reflectance spectra of aluminum red gold green and silver blue coated IR mirrors in 45 angle 31 With black coating materials the problem is completely opposite These materials are used in components when reflection is unwanted such as baffles and cavities Materials that are very black in visible light usually are not in the infrared region For most purposes commonly used optical blackout sheets metal blacks carbon based paints and polymers are good enough For more demanding applications a surface treatment known as the super black can be used This technigue is based upon chemically etched nickel phosphorus alloy The reflectance is around one perce
22. via Signal Recovery 3830 multiplexer It has six floating channels that can be used as either inputs or outputs and are coupled to one of two common buses 54 The temperature and humidity of the detection box can be monitored during the measurement with a Vaisala HMI41 thermohygrometer The temperature is measured with 0 2 C accuracy and relative humidity with 2 accuracy 55 More important factor here is the stability of both values in comparison measurements Some detectors have an internal temperature control but in other cases the temperature of detector is controller with a Thorlabs TED200 temperature controller A step motor driven linear translator isel LES 5 is used in the detector plane It was previously driven with a self made controller based on the escap EDM 453 stepper motor drive circuit that allowed microstep operation up to 64 microsteps per full step resulting in 25600 microsteps per revolution 56 At this resolution the movement was understandably agonizingly slow and because the stepping mode could not be changed it was typically used without microstepping The former controller had broken down so a new controller was purchased This new isel IT 116 Flash controller provides the same maximum resolution of 25600 microsteps per turn 57 and the resolution can easily be changed 44 4 Setup automation The first version of the measurement setup was controlled manually but when nitrogen purging was introduced
23. with purely electrically generated signals in order to minimize all other nonlinearity sources from the measurement Unfortunately very low signal levels are reguired which makes this method somewhat complicated Compared to DC voltage very low and precise AC voltage levels are hard to produce Typical signal generators usually operate in volt and millivolt scale yet measurements well below microvolt are often reguired in spectroscopy In order to solve the problem a precision attenuator was built Its specifications and performance are explained in the first subchapter the actual measurement setup is introduced in the second and the results of the linearity measurements are shown in the third 53 5 1 Precision attenuator In order to produce very small AC signals a precision attenuator was designed and built This allows a signal generator to be used at constant signal level when the performance of the lock in amplifier is measured at different voltage levels and freguencies The attenuator was left as a part of the setup since it provides a handy tool for calibrations 5 1 1 Design Because the attenuator is only used with low frequency signals a straightforward method of attenuation based on resistance dividers was considered as the most suitable The schematic of the device is presented in figure 5 1 The resistor network construction guarantees that the minor contact resistance of the rotary switch has virtually no effect to the a
24. 200 V 400 V 100 V 100 V 100 V 100 V 25V 25V 25V 25V 50 V 50 V 100 V 63 V 100 V 63 V 50 V 100 V 63 V 100 V 100 V 25V 25V 25V 25V 25V 25V 1250 V 50 V 50 V 50 V 1000 V 1000 V 1000 V 1000 V 1000 V 1000 V 800 V C Photoconductive detector amplifier 10 10 10 20 20 10 10 10 10 20 20 20 20 10 10 20 10 10 20 20 20 20 10 10 10 10 10 10 5 10 10 20 20 20 20 20 20 20 20 10 10 10 5 list of 85 C 85 C 105 C 105 C 85 C D8 D9 DIS1 F1 FILI IC1 IC2 IC3 1C4 IC5 IC6 IC7 1C8 LED1 PTC1 POT1 POT2 Q1 02 R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 RN1 S1 Diode Zener Diode Zener LCD module Fuse Slow blow Filter Power line suppression Linear regulator Linear regulator Linear regulator Linear regulator Voltage reference Operational amplifier High voltage Operational amplifier Low noise Analog multiplexer Light emitting diode Resettable fuse Potentiometer Precision Potentiometer Transistor Bipolar NPN Transistor Bipolar PNP Resistor Metal film Resistor Metal film flameproof Resistor Metal film flameproof Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Carbo
25. 22 23 24 25 26 27 Product datasheet Tungsten ribbon filament strip lamp The Pyrometer Instrument Company Inc 2008 Referred 2 11 2011 Internet http www pyrometer com PDF_files Optical 20Strip 20Lamp 2001 2008 pdf Product datasheet IR 12 Series Miniature 8 to 11 Watt Infrared Emitter Boston Electronics Corporation 2004 Referred 2 11 2011 Internet http www boselec com documents CWSourcesWWW5 3 05 pdf Product datasheet Mini Igniter model 301 Saint Gobain Ceramics 2007 Sperfeld P Metzdorf J Yousef Galal S Stock K D M ller W Improvement and extension of the black body based spectral irradiance scale Metrologia 1998 Vol 35 Pp 267 271 ISSN 1681 7575 Yousef Galal S Sperfeld P Metzdorf J Measurement and calculation of the emissivity of a high temperature black body Metrologia 2000 Vol 37 Pp 365 368 ISSN 1681 7575 Yang R O Lin C H Murry S J Pei S S Jnterband cascade light emitting diodes in the 5 8 um spectrum region Applied Physics Letters 1997 Vol 70 Pp 2013 1015 ISSN 0003 6951 Zibik E A Ng W H Revin D G Wilson L R Cockburn J W Groom K M Hopkinson M Broadband 6 um lt A4 lt 8 um superluminescent quantum cascade light emitting diodes Applied Physics Letters 2006 Vol 88 121109 ISSN 0003 6951 Jayaraman V Strand T A Leonard D B Patent Multi wavelength light source for spectroscopy US Patent No 7880882
26. 26386 2 726386 2 726386 2 726386 2 726386 2 MCFE015 15 MCFE015 25 MSB03005 MSB02005 MSB02005 MSB03005 MSB02005 MSB03004 MSB03005 1 1337543 0 1 1337543 0 MSB02004 MSB03005 MSB02004 1 1337452 0 MSB02004 Arduino Uno 86 6A 250 MA 3A 3A 6 3 x 0 8 mm 6 3 x 0 8 mm 6 3 x 0 8 mm 6 3 x 0 8 mm 6 3 x 0 8 mm 2x15V 2x25V 13A 13A 13A 13A 13A 12A 13A 500 500 12A 13A 12A 500 12A 125 VAC 300 VAC 125 VAC 125 VAC 15 VA 15 VA 250 V 250 V 250 V 250 V 250 V 150 V 250 V 1500 V 4 GHz 1500 V 4 GHz 150 V 250 V 150 V 1500 V 4 GHz 150 V Appendix D Photoconductive detector amplifier program code include lt LiquidCrystal h gt Mapping pins define BIAS VOLTAGE AO define TOGGLE A4 define EXT BIAS A2 define INT BIAS 12 define POWER 11 define GAINO 8 define GAIN 9 LCD related pins define ENABLE 3 define RS 2 define DB4 6 define DB5 4 define DB6 7 define DB7 5 Voltage calculation coefficient define GAIN COEFF 159 84375 Object for LCD LiguidCrystal lcd RS ENABLE DB4 DB5 DB6 DB7 byte that carries gain information Gain bits are the signicant bits of the byte byte gain 0 Boolean value for the toggle buttons last value boolean last_toggle false void setup Setting inputs and outputs pinMode BIAS VOLTAGE INPUT pinMode TOGGLE INPUT pin ode
27. Aalto University School of Electrical Engineering Department of Signal Processing and Acoustics Timo D nsberg Improved Reference Infrared Spectrometer Thesis submitted for examination for the degree of Master of Science in Technology Espoo DT of J anuary 2012 Supervisor Prof Erkki Ikonen Instructor Dr Farshid Manoocheri 55 Aalto University School of Electrical Engineering AALTO UNIVERSITY ABSTRACT OF THE SCHOOL OF ELECTRICAL ENGINEERING MASTER S THESIS Aalto University School of Electrical Engineering Department of Signal Processing and Acoustics Author Timo D nsberg Title Improved Reference Infrared Spectrometer Date 27 1 2012 Number of pages 8 90 Supervisor Prof Erkki Ikonen Instructor Dr Farshid Manoocheri Language English Infrared radiation is electromagnetic radiation with a spectrum ranging from 750 nm to 1 mm in wavelength It is invisible to the human eye but has significance in optical spectroscopy An optical spectrometer is a device that measures the emission absorption or fluorescence spectrum of a material In this work a reference infrared spectrometer facility at the Metrology Research Institute was upgraded The spectrometer can be used to measure spectral responsivity of detectors spectral transmittance of optical materials and spectral power distribution measurements of light sources in the wavelength range of 750 nm to 16 um The me
28. B2 2011 Ng W Zibik E A Soulby M R Wilson L R Cockburn J W Broadband quantum cascade laser emitting from 7 7 to 8 4 um operating up to 340K Journal of Applied Physics 2007 Vol 101 046103 ISSN 1089 7550 13 28 29 30 31 32 33 34 35 36 Kim J Chen M Yang C Lee J Yin S S Ruffin P Edwards E Brantley C Luo C Broadband IR supercontinuum generation using single crystal sapphire fibers Optics Express 2008 Vol 16 Pp 4085 4093 ISSN 1094 4087 Bridges J M Migdall A L Characterization of argon arc source in the infrared Metrologia 1995 96 Vol 32 Pp 625 628 ISSN 0026 1394 Schrama C A Bloembergen P van der Ham E W M Monochromator based cryogenic radiometry between 1 um and 20 um Metrologia 2000 Vol 37 Pp 567 570 ISSN 0026 1394 Product datasheet IR Mirrors Thorlabs 2009 Referred 22 11 2011 Internet http www thorlabs de Navigation cfm Guide_ID 2143 Brown R J C Brewer P J Milton M J T The physical and chemical properties of electroless nickel phosphorus alloys and low reflectance nickel phosphorus black surfaces Journal of Materials Chemistry 2002 Vol 12 Pp 2749 2754 ISSN 1364 5501 Mizuno K Ishii J Kishida H Hayamizu Y Yasuda S Futaba D N Yumura M Hata K A black body absorber from vertically aligned single walled carbon nanotubes Proceedings of the National Academy of Sciences of the
29. EXT BIAS INPUT pinMode INT BIAS INPUT pinMode POWER OUTPUT pinMode GAINO OUTPUT pinMode GAIN1 OUTPUT SIS SS 2 W Lighting up the power Ll digitalWrite POWER HIGH Setting pull up resistors digitalWrite TOGGLE HIGH digitalWrite EXT BIAS HIGH digitalWrite INT BIAS HIGH 87 two least Initializing LCD lcd begin 16 2 Setting ADC reference to analogReference EXTERNAL Printing start message to LCD for int i 0 i lt 15 i lcd setCursor 0 0 lcd print Powering up lcd setCursor i l led print delay 300 lcd clear void loop O xternal Checking the bias switch position calculating bias voltage and updating LCD values If using external bias if digitalRead EXT BIAS lcd setCursor 0 0 lcd print Bias External LOW myg If no bias at all else if digitalRead INT BIAS lcd setCursor 0 0 lcd print Bias Off If using internal bias else Reading bias voltage value the noise long bias value 0 for int i 0 i lt 9 i bias value analogRead BIAS VOLTAG Averaging oat bias value GAIN COE by taking 10 samples to average out El N H H float print value fl Rounding o
30. I41 RS 232 Self made Linear translator isel IT116 Flash RS 232 Techno Linear Motion Systems Signal generator Agilent 33521A USB NI certified driver Multimeter Hewlett Packard 3458A GPIB NI certified driver 4 1 1 RS 232 interface drivers LabVIEW s built in communication tool called the Virtual Instrument Software Architecture VISA was used for serial connection This takes care of the port settings data flow and error handling Most of device functions only reguire sending a command and then parsing the replied information Therefore all drivers share similar connection initialization and closing VIs as well as message send and read VIs In addition to VIs listed in this chapter many utility VIs were written to perform simple tasks These are considered self evident and are mainly excluded from the thesis 46 The initialization VI is used to assign the serial connection related parameters These are the speed of the connection bits s buffer flow control and the number of data parity check and stop bits This is done using VISA Configure Serial Port VI Also the possible usage of termination character is specified Some devices reguire some start up data or a command to enter remote mode These are also sent by the initialization VI when necessary In order to ensure stable operation an appropriate delay is set before exiting the initialization VI The send and read VI sends a message to the device and waits for a p
31. Notes AboutLIAs pdf User Manual Model SR830 DSP Lock In Amplifier revision 6 2005 Stanford Research Systems Inc 2005 McNaught A D Wilkinson A UPAC Compendium of Chemical Terminology Second Edition Hoboken New Jersey Blackwell Wiley 1997 464 pp ISBN 978 0865426849 Hamamatsu Photonics K K Technical Information SD 12 Characteristics and use of infrared detectors 2011 Referred 18 10 2011 Internet http sales hamamatsu com assets applications SSD infrared_kird9001e04 pdf User manual DC Power Supply HY3000 HY5000 double series Precision Mastech Enterprises Co 2006 Referred 16 11 2011 Internet http hades mech northwestern edu images f f7 Mastech_power_supply_manual pdf User manual DC Power Supply AL 924A revision 12 09 elc 2009 75 47 48 49 50 51 52 53 54 55 56 57 User manual M P 23 Technical Description and Operating Instructions FO 34 14 515 TO 1988 Henricson J Monochromator controller Special assignment Metrology laboratory HUT 2003 22 pp Manoochehri F K rh P Palva L Toivanen P Haapalinna A Ikonen E Characterisation of optical detectors using high accuracy instruments Analytica Chimica Acta 1999 Vol 380 Pp 327 337 ISSN 0003 2670 Product datasheet SPH 40 Series Hybrid Pyroelectric Detector Spectrum Detector Inc 2010 Referred 17 11 2011 Internet http www crazyfingers com spe
32. RC filter The attenuated disturbance signal has a peak to peak voltage of about 60 mV Also has a high frequency noise component around 900 KHz was observed both with and without the supply voltage filtering The signal shown in figure 3 10 has a peak to peak voltage of around 3 mV Fortunately this high frequency causes negligible error in the low frequency lock in measurements Frequency Period Figure 3 10 900 kHz disturbance signal measured at the pyroelectric detector output 42 3 4 3 Other detectors Various other detectors have been used in the setup of which the most important ones are introduced in this section The previous reference detector used in the setup was a pyroelectric radiometer RK 5700 from Laser Precision with an internal chopper This device can still be used for instance in comparison measurements An indium arsenide photodiode model J12TE1 66D R02M manufactured by the Judson Technologies was previously mounted to the integrating sphere but is now used separately The device has an internal preamplifier and thermoelectric cooling element The colder the photodiode is kept the higher the responsivity but also the wavelength of the peak responsivity shortens Typically measurement temperature is around 20 C corresponding to 7 7 kQ resistance in the thermoelectric cooling element This results in a peak wavelength of 3 3 um At longer wavelength the responsivity rapidly decreases Two photoconduc
33. This chapter introduces the infrared spectrometer measurement setup The main focus is in the improvements done as part of this thesis For more profound description of the setup the reader is advised to look at the previous theses 3 4 and 5 The developed computer control of the devices and the automation of the measurement setup are described in the next chapter 3 1 Setup overview Figure 3 1 presents the overall layout of the measurement setup and an illustration of the electrical signal paths Device controllers and computer connections are omitted from the figure for clarity Table 3 1 explains the abbreviations used in the schematic diagram The whole setup is built inside three airtight boxes made of polypropylene this enables nitrogen purging when it is needed to avoid atmospheric absorptions Cable glands are avoided by using airtight feed through connectors BNC connectors are used for measurement signals DE9 connectors for device control and banana connectors for high current power supply Table 3 1 Abbreviations used in the layout diagram of the measurement setup in figure 3 1 Abbreviation Meaning CM Concave mirror G Diffraction grating IRS Infrared source M Flat mirror OPM Off axis parabolic mirror PS Power supply RD Reference detector SM Spherical mirror TD Test detector T H Thermohygrometer 30 Wavelength selection Temperature monitor Detection
34. United States of America 2009 Vol 106 6044 7 ISSN 0027 8424 Product datasheet Optical Cast Infrared IR Longpass Filters Edmund Optics Ltd 2011 Referred 7 11 2011 Internet http www edmundoptics com products displayproduct cfm productid 1918 Product datasheet IR Long Pass Filters Wavelength technology Ltd 2011 Referred 7 11 2011 Internet http www wavelength tech com IR Optics IRLongPassFilter jsp Donati S Photodetectors Devices Circuits and Applications New Jersey Prentice Hall 1999 432 pp ISBN 978 0130203373 74 37 38 39 40 41 42 43 44 45 46 Rogalski A Infrared detectors an overview Infrared Physics amp Technology 2002 Vol 43 Pp 187 210 ISSN 1350 4495 Rogalski A Infrared detectors status and trends Progress in Quantum Electronics 2003 Vol 27 Pp 59 210 ISSN 0079 6727 Gu riaux V de PIsle N B Berurier A Huet O Manissadjian A Facoetti H Marcadet X Carras M Trinit V Nedelcu A Quantum well infrared photodetectors present and future Optical Engineering 2011 Vol 50 061013 ISSN 0091 3286 Liu H C Quantum dot infrared photodetector Opto Electronics Review 2003 Vol 11 Pp 1 5 ISSN 1896 3757 Stanford Research Systems Inc Application Note 3 About Lock In Amplifiers Stanford Research Systems Inc 2004 Referred 11 11 2011 Internet http www thinksrs com downloads PDFs Application
35. a resistor network are used to select the load resistance There are two built in resistor values per decade ranging from 1 kO to 32 MO The last position of the rotary switch short circuits internal load resistance network enabling an arbitrary value external component to be used 6 1 3 Amplifier circuit A very simplified model of the amplifier circuit is shown in figure 6 2 Among other things the model omits high pass filtering overvoltage protection and gain selection implemented in the actual device 63 L i i Vs Ro Figure 6 2 Simplified model of the amplifier circuit A large input resistance Ry is used to keep the DC component of the amplifier circuit zero The low pass cut off frequency fe is determined by the R and the filter capacitor C 1 jo 2mRINC 6 1 The gain G is determined by the resistors R and Ry G 1 2 6 2 1 In the preamplifier the resistors R and R gt are replaced with a high precision integrated thin film network A multiplexer controlled by the microcontroller is then used to change the feedback circuitry when gain is changed 6 2 Characterization Agilent 33521A signal generator was used with the attenuator introduced in chapter 5 1 as a signal source in the characterization measurements Output voltages were measured with a Hewlett Packard 3458A multimeter using the synchronous sub sampled computed true RMS technique 64 6 2 1 Gain The gain of the amplifier was deter
36. and a semiconductor junction that creates Schottky barrier An avalanche photodiode APD is a photoemissive detector that provides high gain by generation of secondary charge carriers PEM detectors exploit the photoelectromagnetic effect where the photon absorbed by the semiconductor in a magnetic field generates electric current 38 2 5 3 Thermal detectors Thermal detectors operate on the principle that incoming radiation increases their temperature Temperature changes are then measured by a temperature dependent mechanism such as thermoelectric voltage pressure or resistance change or pyroelectric voltage Due to this energy conversion process thermal detectors are generally slower and require higher power levels than photon detectors In turn they can operate at room temperature are relatively low cost and usually rugged reliable and in principle do not have spectral dependence in responsivity 38 Most common type of thermal detectors is the pyroelectric Detector makes use of pyroelectric materials that become polarized when heated much the same as piezoelectric material becomes polarized under pressure This generates a tiny electric current that can be amplified and measured The polarization charges will eventually drain making the pyroelectric material neutralized Therefore constant or very slowly modulated radiation cannot be measured 22 Thermopile detectors exploit the thermoelectric phenomenon where two differe
37. as determined by a comparison between attenuated and unattenuated signals in order to cancel out the nonlinearity of signal source Frequencies up to 10 kHz were measured and the nonlinearities were less than 1 In the range of optical measurement frequencies the attenuator performed better the change in attenuation was less than 200 ppm up to 200 Hz with all attenuations 5 2 Measurement setup The measurement setup is shown in figure 5 4 The signal generator can produce signals in the RMS amplitude range of 1 mV to 1V The attenuator then allows the 57 measurements to be done in the range of 10 nV to 1 V A separate BNC cable is used to provide the digital freguency reference to the lock in amplifier For comparison signals in the range of 1 uV to 1 mV were measured with four different combinations of generator output and attenuation The differences in results with different combinations were negligible NOISE GENERATOR SIGNAL 1mV 1V ATTENUATOR 10 nv 1V LOCK IN adi coer CI FREQUENCY REFERENCE Figure 5 4 Measurement setup for electrical linearity and frequency response A noise generator was added to the measurement setup both the noise and the signal generator have an output impedance of 50 so they were merely connected in series The noise generator allows measurements below the single bit level of the lock in amplifiers analog to digital converter ADC The noise causes a bit to change in the device and
38. as infrared radiation A major part of molecular electronic transitions as well as vibrations and rotations in molecules occur in the energy range of an infrared photon Therefore infrared radiation offers an interesting research field in the radiometric sciences Furthermore common applications of infrared radiation such as thermal imaging medical treatment surveillance or meteorology affect our everyday lives This work continues the infrared radiometry project at the Metrology Research Institute In the first phase of the project a facility for spectral responsivity measurements of infrared detectors was established 3 The facility was then adapted for transmittance measurements and the nitrogen purging was also introduced 4 The latest major improvement was the extension of the spectral responsivity scale using a pyroelectric reference detector 5 The versatile measurement facility is currently capable of measuring spectral responsivity spectral transmittance and spectral power distribution in the wavelength range of 750 nm to 16 um The second chapter outlines the essential theory of infrared radiometry whereas the improved measurement setup is introduced in the third The main improvements are related to stable sourcing of infrared radiation and refining the guality of measured electric signal Also the low freguency response of the pyroelectric reference detector was determined In addition all measurement and control devices w
39. asurement setup was improved for full automation using LabVIEW and the previous calibration and automation procedures were refined Automated facility is versatile for different measurement setups and is easy to operate Phase sensitive detection is utilized in the measurement setup by using a lock in amplifier It enables the detection of very small signals in the presence of overwhelming noise Optical chopping is used to modulate the measurement signal at a known reference frequency Typically the linearity of the measurement system is determined optically Also in this work a fully electronic method for linearity measurements of lock in amplifiers was developed and tested This method improves the accuracy in lock in comparison measurements In addition a preamplifier for photoconductive detectors was constructed and characterized The amplifier was designed so that it can be used with wide variety of detectors Keywords infrared spectrometer automation linearity photoconductive detector AALTO YLIOPISTO KANDIDAATINTY N SAHKOTEKNIIKAN KORKEAKOULU TIIVISTELMA Aalto yliopisto S hk tekniikan korkeakoulu Signaalink sittelyn ja akustiikan laitos Tekij Timo D nsberg Ty n nimi Uudistettu referenssi infrapunaspektrometri P iv ys 27 1 2012 Sivum r 8 90 Ty n valvoja Prof Erkki Ikonen Ty n ohjaaja TkT Farshid Manoocheri Kieli Englanti Infrapunas teily on s
40. ayer ceramic C41 Capacitor Ceramic Disc D1 Diode Standard rectifier D2 Diode Standard rectifier D3 Diode Standard rectifier D4 Diode Standard rectifier DS Diode Standard rectifier D6 Diode Standard rectifier D7 Diode Fast recovery rectifier GBU4K MMKS 103K400J01L4 MMKS5 103K400J01L4 MMK5 103K400J01L4 ECOS1VP472BA ECOS1VP472BA MKS0C026800COOKSSD MKS0C026800COOKSSD MKS0C026800COOKSSD MKS0C026800COOKSSD RE3 25V471M RE3 25V471M EETED2D102CA EETED2D102CA MMK10 473K400A01L4 MMK15 105K100B04L4 MCGPR100V107M13X21 MKS2D031001AOOKSSD MKS2D031001AOOKSSD RE3 25V471M RE3 25V471M RE3 25V471M CB1E106M2CCB B37984M5105K K104K15X7RF5TH5V MMK15 105K100B04L4 MKS0C026800COOKSSD MMK15 105K100B04L4 MKS0C026800COOKSSD MCCHU7101J5 MMK15 105K100B04L4 MKS0C026800COOKSSD MCGPR100V226M8X11 MCGPR100V107M13X21 RE3 25V471M RE3 25V471M RE3 25V471M RE3 25V471M CB1E106M2CCB CB1E106M2CCB FKP1R021005DOOKSSD B37984M5105K B37984M5105K MCCHU7101J5 1N4007 1N4007 1N4007 1N4007 1N4007 1N4007 MR858 84 4A 10nF 10nF 10nF 4700 uF 4700 uF 680 nF 680 nF 680 nF 680 nF 470 uF 470 uF 1000 uF 1000 uF 47 nF 1 uF 100 uF 100 nF 100 nF 470 uF 470 uF 470 uF 10 uF 1 UF 100 nF 1 uF 680 nF 1 UF 680 nF 100 pF 1 UF 680 nF 22 uF 100 uF 470 uF 470 uF 470 uF 470 uF 10 uF 10 uF 10 nF 1 UF 1 uF 100 pF 1A 1A 1A 1A 1A 1A 3A 800 V 400 V 400 V 400 V 35V 35V 63 V 63 V 63 V 63 V 25V 25V 200 V
41. basic amplifier topology is similar to the one demonstrated in 44 The main feature of the new amplifier is the possibility to select gain biasing voltage and load resistance which 61 makes it a truly universal design Complete schematics of the device and the list of components are in the appendices B and C respectively The block diagram of the device is shown in figure 6 1 In addition to the actual preamplifier circuit a biasing circuit is used to provide very stable biasing voltage through appropriate load resistance Both the amplifier and the biasing circuit are controlled and monitored by the Arduino Uno microprocessor and front panel circuitry The microcontroller is also used to control the front panel LCD The complete program code along with comment text is listed in appendix D BIASING CIRCUIT BOWER W AMPLIFIER 7 DETECTOR SUPPLIES x MICROPROCESSOR Figure 6 1 Block diagram of the preamplifier Analog digital and power w o lt LL ui lt Lu lt a 2 o cc TE signals are shown green black and red respectively 6 1 1 Power supplies Common linear regulators are used to maintain stable operating voltages of 15 V Input voltages for the regulators are provided from a single 15 VAC secondary coil using two parallel half wave rectifiers in opposite polarity and reservoir capacitors Third linear regulator is used to decrease the 15 V operating voltage to 9 V for the microco
42. be photon detectors They all use electron excitation to detect incident radiation and have an absolute minimum energy level for detectable photon The detection of small energy photons also requires cooling in order to reduce thermal oscillations in detection material A simplified model of a general photon detector is presented in figure 2 11 Usually the optical area A and electrical area A are tried to be manufactured equal in size but when the use of concentrator is possible the ratio A Ae can be increased 20 Optical area Radiation Heterojunction contacts Electrical area N Concenirator EX Substrate Reflector Figure 2 11 Model of a photon detector 38 Intrinsic detectors utilize the generation of electron hole pairs across the semiconductor band gap in single semiconductor or a junction of semiconductors Extrinsic detectors used doped semiconductor materials creating impurity states in the band gap where charge carriers can transit This allows photons with smaller energy to excite an electron The disadvantage here is that thermal excitations are also easier and cooling to a low temperature is necessary In free carrier detector the photon energy is absorbed by a free carrier in either conduction or valence band Advanced manufacturing technigues enable complicated guantum structures to be grown on semiconductors These are based on bound states of an electron and hole known as excitons Confi
43. ble peaks Gas discharge lamps have narrow emission lines which depend on the gas or gas mixture in guestion whereas light emitting diodes typically emit monochromatic light A broadband spectrum is usually needed in monochromator based radiometry but in some cases also monochromatic or narrowband sources can be useful Stimulated emission on the other hand involves strong energy pumping that excites atoms or molecules in a lasing media The excited electron is then stimulated by photon causing it to drop to a lower energy level This creates in a new photon with the same energy phase polarization and direction as the stimulating photon When the excited state has more electrons than the lower energy state the rate of stimulated emission exceeds that of absorption resulting in optical amplification Lasers masers and optical amplifiers use stimulated emission The theory of an ideal black body radiation source is very useful in many infrared applications and it is described extensively in the first subchapter The second subchapter deals with incandescent light sources which are the most commonly used broadband sources in infrared spectroscopy Various other light sources are introduced in the third subchapter main focus being in those of broad spectrum More information on optical sources in general is widely available for example in 10 11 and 12 whereas detailed descriptions of infrared sources can be found for instance in 13
44. cd print Gain if digital digital 10 Read GAINO Write GAINO Read GAIN1 100 Read GAINO Write GAINO if digital digitalWrite GAIN1 break case 3 lcd print if digital digital if digital digitalWrite GAIN1 break default delay 50 Read GAIN1 Gain Read GAINO Write GAINI1 Read GAIN1 lcd print Undefined 1000 LOW HIGH LOW me LOW HIGH HIGH LOW my HIGH LOW LOW HIGH Ue LOW HIGH LOW HIGH gain 6 90
45. ctrum pdf datasheets SPH 40 pdf Product datasheet PbSe photoconductive detector P791 P2038 P2680 series P3207 05 Hamamatsu Photonics K K 2008 Referred 2 11 2011 Internet http sales hamamatsu com assets pdf parts_P p791 11_etc_kird1020e07 pdf Product datasheet Lead Sulfide Lead Seledine Detectors J13 J14 Series Judson Technologies LLC 2002 User manual Series PS300 High Voltage Power Supplies Stanford Research System Inc 2007 Referred 23 11 2011 Internet http www thinksrs com downloads PDFs Manuals PS300m pdf User manual Signal Recovery Model 3830 multiplexer Ametek Advanced Measurement Technology Inc 2004 User manual HMI41 Indicator and HPM41 45 46 Probes Vaisala 1998 Talvitie H Narrowing the spectral linewidth of a semiconductor laser Master s Thesis HUT Department of Electrical and Communications Engineering Espoo 1993 48 pp User manual 7716 flash IT116 mini isel 2009 76 58 59 60 61 62 63 64 65 User manual LabVIEW 2077 National Instruments 2011 Referred 16 1 2012 Internet http www ni com pdf manuals 371361h zip User manual HP 3458A Multimeter edition 3 Hewlett Packard 1994 Theocharous E Absolute linearity characterization of lock in amplifiers Applied Optics 2008 Vol 47 Pp 1090 1096 ISSN 0003 6935 Product datasheet Metal Film Thin Film Chip Resistors Type ERA 2A 3A 6A SA Panasonic 2011 Re
46. dark current p The output signal from a photoconductive detector is commonly extracted as voltage Vo by using a load resistor Rz and bias voltage Vg as shown in figure 2 13 44 Ro Ve Ri Vo Figure 2 13 Model for extracting the output signal of a photoconductive detector using load resistor R and bias voltage Vg 26 When the detector is not illuminated the output voltage Vo is simply s 2 27 RL Rp A change in output voltage 4Vo due changes in Rp when detector is exposed to light can be approximated with differentiation RLVB Rp RL AV So ARp AR 2 28 The output voltage sensitivity has a maximum value AV max when Rp and Rz are equal v AVo max AVolrp r zp R 2 29 In order to optimize the output signal level of the photoconductive detector the load resistance should be well matched By dividing equation 2 28 with equation 2 29 the efficiency of the detector due to resistor matching 7 can be calculated 4a iS 1ra 2 2 30 where a is the ratio R Rp The relation between the output efficiency and the ratio a is shown in figure 2 14 Because the dark resistance values of the detectors can vary enormously the amplifier has to either be dedicated to one type of detectors or the load resistance should be selectable Typical decade selection of resistors provides the minimum efficiency of about 73 90 while two resistors per decade increases the minimum efficiency to about 92
47. e 4 1 LabVIEW device drivers With the exception of preamplifiers two power supplies and the cooling unit all devices in the measurement setup are computer controllable These devices along with their connection interfaces and used LabVIEW drivers are presented in table 4 1 For four devices LabVIEW drivers were not available and they were self written The 45 multimeter Hewlett Packard 3458A is not part of the measurement setup per se but it was used in various thesis related measurements and still provides a good tool for calibration measurements All devices lacking a LabVIEW driver were coincidentally connectable via RS 232 serial port This allowed many of the same type VIs to be used in all the drivers with some modifications The section 4 1 1 concerns general aspects of RS 232 interfacing and the other sections focus on the actual device drivers that were written Table 4 1 Device connections and used LabVIEW drivers Drivers that were modified are underlined Device Model Interface LabVIEW driver Filter wheel Optec IFW RS 232 Self made Optical Chopper Terahertz Technologies C 995 RS 232 Self made Monochromator MAP 23 RS 232 Self made High voltage supply Stanford PS325 GPIB NI certified driver Multiplexer Signal Recovery 3830 USB Signal Recovery Lock in amplifier Stanford SR830 GPIB NI certified driver Multimeter Agilent 34410A GPIB NI certified driver Thermohygrometer Vaisala HM
48. ecision attenuator assembled The chassis ground marked GND and the signal ground are galvanically isolated 56 5 1 2 Performance The value R was measured using four point measurement method for each attenuation setting Keithley 263 calibrator was used as a current source and voltage was measured with Hewlett Packard 3458A multimeter The input resistance of the device was measured in the same way which actually corresponds to value R R gt It was measured to be 199995 8 1 1 O From these values also the nominal attenuations A and parallel resistances R were calculated The results are shown in table 5 2 Table 5 2 Measured resistances and calculated parameters of the attenuator The uncertainty of the nominal attenuation is 400 ppm for the 10 setting and 150 ppm for the rest Setting Resistance R O Nominal attenuation A parallel resistance Rojo O 101 20000 52 0 1 1 0000E 01 18000 38 0 2 10 2000 213 0 007 1 0001E 02 1980 208 0 016 10 200 0327 0 0007 1 0002E 03 199 8326 0 0016 10 20 01394 0 00007 1 0007E 04 20 01194 0 00016 10 2 000625 0 000023 1 0003E 05 2 000605 0 000040 Since AC signals are attenuated the frequency response of the attenuator is also crucial It was measured using Agilent 33521A as signal source Measurements were done using the synchronous sub sampled computed true RMS technique of the Hewlett Packard 3458A multimeter Linearity w
49. eeeeeeeeeees 36 3 4 Output optics and detection ee Y iia 37 A O A aa 37 3 42 Pyroelectric reference detector metidos ia ewydd di 38 34 3 Other detectors PORC i Cd Ae ili 43 3 0 DIE NS A A dd DE O R RE 44 1v 4 Setup AULOMALION Losail ie aaa 45 4 1 cEabYIBEW devies Qrvers secta lieti lione lied 45 4 1 1 RS 2321nterface drivers origano ani iii 46 412 Monochrematotes chele lait ici ici 47 SA a orso eros E aloe Gu SAC 48 ATA Other devices siii 50 4 2 Programs for electrical measurements naa aa en n n aaaaaaeeen 50 4 3 Programs for optical measurements s issalss mas ss Tn S NS a haas 51 5 eck amplitier lime arty lt tc er telepatia 53 SA Pr cision atten atotn os ia uae O SESE TEKSTIN s ate ee 54 Sl Designer ana a ia 54 5 12 Performance AA A lia 57 3 2 Meas rement Setup 57 5 3 Linearity measurements A ira 58 6 Preamplifier for photoconductive detectors i 61 ol bestseller iranica 61 Odi POWER A oa 62 612 A TO 63 6 1 3 Amplifier Omo sioni ia 63 6 2 CHaracierizationi ossalico alal 64 02 dla 65 622 Linearity ascella iaia illo iaia illa 65 0 2 3 FREQUENCY TES DON Sd y NW aa 66 6 3 Optical measurements A A O e 68 Tn CONGIUSIONS lieta 69 A E ls Tananan 71 Appendices ers estee PATATA ATTRAENTE RITRARRE aku IRINA FR ANF 78 Appendix A Radiometric Quantities in SI Units noun 79 Appendix B Photoconductive detector preamplifi
50. en s displacement law and Stefan Boltzmann law described in equations 2 5 and 2 7 respectively In order to spectral radiance peak at longer wavelengths the temperature should be lower This in turn strongly reduces radiant emittance Nevertheless at shorter wavelengths incandescent sources are irreplaceable Most incandescent light sources such as the common light bulb are based on a tungsten filament The lamps are protected from air by glass envelope that is either evacuated or filled with inert gas Lamps usually operate in the area of 1400 2400 K 11 but for special purposes much higher temperatures can be used with inevitable cost of shortened lifespan the absolute upper limit being the melting point of tungsten at 3695 K 13 Spectral power of a common commercial infrared lamp is illustrated in figure 2 4 Tk 2450 K mW per 10 nm per W 700 1000 1400 2000 3000 4000 Wavelength in nm Figure 2 4 Spectral power distribution of Philips IR R125 series infrared lamps at the nominal filament temperature of 2450 K 18 The problem with standard tungsten filament lamps is evaporation on the filament The higher the temperature is the faster the process The evaporated tungsten builds up to the inner wall of the bulb and affects the output of the lamp by absorbing some of the radiation The filament itself also thins causing the temperature locally to raise filament to evaporate even more and
51. er schematics 80 Appendix C Photoconductive detector amplifier list of components 84 Appendix D Photoconductive detector amplifier program code 87 Symbols and Abbreviations a gt gt gt 2 gt gt gt E p fe ECPR Active area of a detector Electrical area of a detector Amplitude of a measurand signal Optical area of a detector Amplitude of a reference signal Source area Alternating Current Analog to Digital Converter Avalanche Photodiode Wien s displacement constant 2 8977685 10 m K 1 Maximum contrast constant z2 410 10 m K 2 Optical band pass Noise eguivalent bandwidth of measurement system Bayonet Nut Coupling Speed of light Capacitance Speed of light in vacuum 299 792 458 m s 1 Comma Separated Values Grating constant Detectivity Specific detectivity Direct Current Dynamic Link Libraries Irradiance Spectral irradiance Band gap energy Energy of a photon Cut off frequency Electrically Calibrated Pyroelectric Radiometer EEPROM Electronically Erasable Programmable Read Only Memory FPA G h HUT Ia Le le le Tout IR INI kg Focal Plane Array Gain Planck constant 6 626 069 57 10 J s 1 Helsinki University of Technology Dark current Detector current Radiant intensity Spectral radiant intensity Output current Infrared Java Native Interface Angular wavenumbe
52. ere mounted to a rack cabinet and the setup was completely automated The setup automation and measurement procedures are treated in the fourth chapter Phase sensitive detection is utilized in the measurements by using a device called a lock in amplifier It enables the detection of very small signals in the presence of overwhelming noise Optical chopping is reguired to modulate the measured light at a known reference freguency However the accuracy and linearity of lock in amplifiers are poorly reported and test measurements commonly make use of complicated optical setups Therefore development and testing of a purely electrical testing method for the lock in amplifier was taken as an objective of this thesis The method and the measurements are introduced in the fifth chapter Photoconductive detectors are commonly used in infrared measurements Different types of detectors have very different properties and are generally used with a matched preamplifier As a part of the thesis reguirements for a more universal preamplifier were studied and such a device was then designed built and characterized The preamplifier and optical measurement for which it was used are introduced in the sixth chapter 2 Theory of infrared radiometry 2 1 Fundamentals The energy of a photon E depends only on its frequency v or inversely its wavelength 2 via the relation 2 1 where c is the speed of light and h is the Planck constant 6 626 069
53. eri ilmaisimien kanssa Avainsanat infrapuna spektrometri automaatio lineaarisuus valojohtava ilmaisin 11 Acknowledgements This work has been done at the Metrology Research Institute MRI of the Aalto University School of Electrical Engineering The research project was funded by the Centre for Metrology and Accreditation MIKES I would like to thank Professor Erkki Ikonen the head of the Metrology Research Institute for supervising this work and for giving me the opportunity and encouragement to work on many challenging and interesting projects I am most grateful for my instructor Dr Farshid Manoocheri for the support comments and advices that helped me to overcome all the obstacles on the way After countless constructive and instructive conversations I can only admire his wisdom to always ask the right guestions and moreover his patience to listen to my answers It is a pleasure and a privilege to work among the personnel of the Metrology Research Institute The both relaxed and productive working atmosphere is something truly unigue of which I want to thank all my co workers I am deeply thankful to my parents who have always given me love understanding support and the freedom to make my own decisions and to my sister who still Keeps guiding her little brother in life I also want to thank all my friends who have given me aid trust joy and a change to follow my many passions in life F
54. ermanium substrate but in the near infrared region also silicon can be used New type of plastic filters that have good thermal characteristics and tolerance for harsh chemical environment can be used close to visible light wavelengths 34 The performance of a typical germanium based infrared filter is presented in figure 2 9 100 TN EU UN UU EN UN UU UU UU EN LIN WG EN N UN UU YN UN UU ed T a ed a led O DR Ri eee Wavelength um 2 9 The performance of a typical infrared long wave pass filter on a germanium substrate 35 2 5 Detectors Optical detectors measure optical power Based on their operation principle they are generally divided into photomultiplier tubes photon detectors and thermal detectors The latter two are briefly discussed in the second and third subchapter respectively whereas photomultiplier tubes are beyond the scope of this work All devices mentioned above are available for wide a range on radiation but this thesis highlights the infrared region Some general theory and important detector parameters are introduced in the first subchapter More information on optical detectors at large can be found for instance in 10 12 and 36 while 13 37 and 38 focus on the infrared range detection 17 2 5 1 Theory and definitions A detector converts incident optical power Pinc to either output voltage Vou or output current Iou Only detectors with voltage output will be discussed
55. es excellent linearity and a freguency range of 1 Hz to MHz However the method reguired the input signal to be repetitive so measurements using changing freguency or noise are impossible At freguencies below 1 kHz the accuracy is typically around 0 01 percent The VI Multimeter sub sampled was written to simplify the usage of the method The VI takes a given number of measurements at a given precision target precision It should be noted that the measurements at precision are very time consuming More details about the measurement technigue can be found in 59 4 3 Programs for optical measurements Typically a measurement setup reguires rather customized measurements procedure Therefore measurement programs were written only for the two most common 51 measurements where either wavelength or chopping freguency is controlled More complicated measurements can easily be built with LabVIEW using the building blocks mentioned above Spectral measurement VI controls the monochromator and the filter wheel to take measurements with the lock in amplifier at different wavelengths The wavelength range and step increment are specified by user as well as all measurement related parameters By default a suitable filter is selected automatically but manual selection is also possible This is useful for example when testing or calibrating devices Freguency measurement VI is very similar to electrical measurement Freguency response Vl except i
56. eventually to break A halogen lamp also known as the guartz tungsten halogen lamp or the tungsten halogen lamp offers a solution to this problem It has a similar tungsten filament but it is contained within a mixture of inert gas and some halogen gas This halogen compound for example iodine I gt methyl iodide CH3I or hydrogen bromide HBr both combines with tungsten attached to the bulb and redeposit it back to the filament This regeneration is known as the halogen cycle Higher temperature causes the tungsten halogen pair to decompose more rapidly so the process automatically targets the thinnest points of the filament Because higher filament temperatures can be used halogen lamps are brighter and more efficient compared to normal tungsten lamps Halogen lamps also have to be smaller in size since the halogen cycle requires glass surface temperature to be at least 470 K usually around 670 1300 K The glass envelope in a normal or tungsten filament lamp absorbs significantly radiation that has wavelength above 2 5 um The glass acts as a radiator itself but naturally the surface temperature is far less than the filaments The total spectrum of the lamp is 10 combination of these two spectra The spectrum of a filament lamp can be improved with a special sapphire glass which allows good spectral range up to 3 um 19 Incandescent sources don t necessarily reguire any kind of enveloping glass The simplest solution is
57. ew device performed very well the RMS noise was less than 100 uV The old cabling from the power supply to the infrared source was replaced with a thicker 4 mm cable in order to decrease the voltage drop In addition to that separate voltage sensing cables were installed so that the actual operating voltage of the source can be monitored with a multimeter to compensate the voltage drop in the cables A concave spherical mirror made of aluminum and a protective layer of magnesium fluoride is used to collect the radiation from the glower The diameter of the mirror is 93 mm and the focal length 125 mm The distance from the source to mirror is 165 mm and it focuses the light to the entrance slit of the monochromator at a 455 mm distance This magnifies the source by a factor of about 2 8 making the image of the 1 3 mm wide tip of the ceramic glower barely fulfill the 3 6 mm wide entrance slit of monochromator ES 3 3 Wavelength selection and chopping An eight position IFW filter wheel from Optec Inc with a set of long wave pass filters is used in front of the monochromator This prevents the pass of unwanted diffraction orders through the monochromator Table 3 3 presents the current order of the filters and their specification Table 3 3 Filter wheel arrangement and working regions 5 No Cut on wavelength 5 Working range Substrate Part number 1 Open aperture 2 Closed aperture
58. ferred 25 11 2011 Internet http industrial panasonic com www data pdf AOA0000 AOA0000CE26 pdf Product datasheet Surface Mount Resistors RN73H KOA Speer Electronics 2011 Referred 25 11 2011 Internet http www koaspeer com products resistors surface mount resistors rn73h Product datasheet Standard Thick Film Chip Resistors D CRCW e3 Vishay 2011 Referred 25 11 2011 Internet http www vishay com docs 20035 dcrcwe3 pdf Product datasheet CNS 471 Decade Divider Thin Film Resistor Networks Vishay 2006 Referred 12 1 2012 Internet http www vishay com docs 60043 cns47 st pdf Taylor B N Thompson A NIST Special publication 330 2008 edition The International System of Units SI National Institute of Standards and Technology 2008 Referred 31 10 2011 Internet http physics nist gov Pubs SP330 sp330 pdf TT Appendices Appendix A Radiometric Ouantities in SI units Appendix B Photoconductive detector preamplifier schematics Appendix C Photoconductive detector preamplifier list of components Appendix D Photoconductive detector preamplifier program code 78 Appendix A Radiometric Guantities in SI units Table A1 presents radiometric guantities in SI units It should be noted that a wide range of different symbols are commonly used in the literature Spectral guantities can also be given per unit freguency in which case they are commonly denoted with
59. ff the value print value 10 0 print value 0 5 bias value int print value print value float bias value 10 0 88 Printing text lcd setCursor 0 0 lcd print Bias Volts Moving cursor to right position depending on the number of digits if bias value lt 100 lcd setCursor 7 0 lcd setCursor 6 0 Printing the calculated value lcd print print value Masking off an extra zero lcd setCursor 10 0 leds printer m Checking if changing the gain is needed and updating LCD values Ej If the toggle button is being pushed yet without toggle operation if digitalRead TOGGLE LOW amp amp last_toggle false Wait for a moment and check again to avoid double contacts delay 25 if digitalRead TOGGLE LOW last_toggle true gain 1 If not pressing change last toggle to false if digitalRead TOGGLE HIGH Wait for a moment and check again to double contacts delay 25 if digitalRead TOGGLE HIGH last_toggle false Changing gain if needed and updating the LCD lcd setCursor 0 1 switch gain 4 case 0 lcd print Gain 1 my if digitalRead GAINO HIGH 89 digitalWrite GAINO if digitalRead GAIN1 digitalWrite GAIN1 break case 1 lcd print Gain if digital digital if digital digitalWrite GAINI break case 2 l
60. heel control are listed in Table 4 3 The VIs used for wheel movement are quite self explanatory Select filter position drives the wheel to a given position using shortest rotating route Select home moves the wheel always clockwise and stops when it detects the first wheel position This may take up to 20 seconds and is 48 necessary only if the device is moved both manually and in the remote mode Other option is to use the Get filter position function to ensure current wheel position Table 4 3 Description of VIs related to controlling the filter wheel VI name Description Initialize Initialize connection Send and read Send commands and read replies manually Close Close connection Select home Driver filter wheel to home position Get filter position Read current position of the filter wheel Select filter position Driver filter wheel to a given position Identify wheel Identify current filter wheel Identify filter Identify a given filter Read wheel data Read the whole data of the current filter wheel Write wheel data Write the whole data of the current filter wheel Calibrate Calibrate filter wheel movement Many wheels can be used with one controller and therefore all wheels are given a letter identifier Current wheel can be read with Identify wheel VI which is useful if multiple wheels are being used Also each filter is given a name wh
61. here since all the equations apply when voltage terms are replaced by current terms The responsivity of the detector S also known as the photo sensitivity is defined Sec 2 14 Pinc If the irradiance E on the active area of the detector Ag is constant one can simply assume Pinc EcAa i 2 15 In general responsivity does not take the radiation spectrum into account Therefore it is defined in terms of suitability for application in guestion For instance the responsivities of infrared detectors are often measured with the spectrum of a 500 K black body Spectral responsivity R instead measures the wavelength dependence of detectors response and is defined in terms of spectral power The output of an ideal thermal detector is only dependent on the incident optical power and has therefore flat spectral response The output of a photon detector on the other hand is dependent on the amount of incoming photons and has linearly increasing spectral response up to the cut off wavelength 4 If the detector material has a band gap energy Ez according to eguation 2 1 we can determine the cut off wavelength gati 2 16 Noise equivalent power NEP is the quantity of incident power needed to equal the output and the intrinsic noise level of the detector or in other words it is the input power resulting to signal to noise ratio of one 18 Pinc _ Vn NEP ros 2 17 where V is the noise voltage Similar term noise equivalent irradia
62. hk magneettista s teily jonka aallonpituus on 750 nn 1 mm Se on t rke ty v line optisessa spektroskopiassa vaikka onkin ihmissilm lle n kym t nt Optisella spektrometrill tarkoitetaan laitetta joka mittaa materiaalin emissio absorptio tai fluoresenssispektri T ss ty ss uudistettiin Aalto yliopiston Mittaustekniikan ryhm n referenssi infrapunaspektrometri Laitetta k ytet n ilmaisimen vasteen aineen l p isyn ja l hteen tehojakauman spektrisiin mittauksiin aallonpituusalueella 750 nm 16 um Mittauslaitteisto automatisoitiin k ytt en LabVIEW ohjelmistoa Lis ksi kalibrointi ja mittausmenetelmi paranneltiin Automatisoitu mittauslaitteisto soveltuu monipuolisesti erilaisiin mittausj rjestelyihin ja on aiempaa helpompi k ytt Mittauksissa k ytet n vaihelukittua vahvistinta joka mahdollistaa hyvin heikkojen signaalien havaitsemisen suuren kohinasignaalin l sn olosta huolimatta Mitattavaa signaalia moduloidaan katkomalla valol hdett referenssitaajuudella Optisen menetelm n sijaan vaihelukitun vahvistimen lineaarisuusmittauksiin kehitettiin t ysin s hk inen menetelm joka parantaa ulkoista tarkkuutta vertailumittauksissa Lis ksi osana ty t suunniteltiin rakennettiin ja karakterisoitiin esivahvistin valojohtaville ilmaisimille Tavanomaisesta sovitetusta esivahvistimesta poiketen laitteen keskeiset parametrit ovat s dett vi joten sit voidaan k ytt lukuisien
63. ich can be read with Identify filter VI All filter wheel data can be read at once with Read wheel data VI The wheel data is stored in the controller in a non volatile EEPROM memory The data has to be stored one byte at a time and storing one character takes about 25 ms For this purpose a standalone Write wheel data VI was coded It modifies the given text data to a suitable format and takes care of the storing process For practical reasons the wavelength ranges of the filters were saved as their names The filter wheel has an offset correction for both counterclockwise and clockwise movement Basically this is a small angle correction to make sure that the filter always stays in the center of the device aperture This is done using the Calibrate VI To simplify the usage correction values to factory presets are given instead of absolute values The device used in the setup was verified working properly with correction value 2 clockwise and zero correction counterclockwise 49 4 1 4 Other devices The optical chopper is rather easy to control In an addition to common initialization manual command sending and closing VI typical user only needs two additional VIS The current status of the chopper is read with a VI Get status which returns the selected frequency range possible usage of external frequency source and current frequency The first two parameters are only selectable via device front panel but frequency can be set with
64. inally I wish to thank my dear girlfriend Julia for making my life the one I want to live Espoo 23 of January 2012 Timo D nsberg ill Table of Contents Acknowledgements vasaa ssa ydi FD Yd Dld FLYNN ROTAIA NATA riti iii Table O CONS A FOD A AR ENDS iV Symbols and Abbreviationis sirena vi 1 Tntroductiolteee ei eu dw Hu Gu niet 1 Zo Theory ofinfrared radio MEET y ue du GG gd ica rod 3 24 cPumdgmental8e iai i ci do ia do GL YF GN ila 3 2AM lt Anirared Spec rum a a 3 PA EPE A0 a EE i EEE SE A GO EE 5 2 2 Sources of infraredradiationi dalicnansa ila nale idad id 3 2 25 Ideal black body see a alal 6 2 22 Incandescentsourcesi cls aglio alito A IA aseena 9 2 2 3 Other SQUICES aiii did 12 2 3 Considerations on infrared optics e naan aa a n n n ana naa aa aeeeeen 13 2 4 WAVE length selection free 15 25 Dose GN DY aovd 17 2 38 Theory and ACTINTIONS Lire A NC nL 18 25 2 Photon detectors citrico siii pa 20 23 37 Thernialid lecio A a Pat ka Stasin Eua dakk av E E E auie 22 2 6 Signal amplification Lidia 23 2 6 1 Lock in amplihier visir ON Ydd osaa na en p YND FN Ea Ldi 23 2 6 2 Amplifier for photoconductive detectors ei 26 3 Setup description and improvementsS ii 30 dl SUPE lle ela ella 30 3 2 Radiation sources and input OpticS iii 34 3 3 Wavelength selection and Chopping cccccceeesssssneeeeeeeeeeeeeenennee
65. ing 83 Appendix components B1 Diode bridge C1 Capacitor Polyester film C2 Capacitor Polyester film C3 Capacitor Polyester film C4 Capacitor Aluminum electrolytic C5 Capacitor Aluminum electrolytic C6 Capacitor Polyester film C7 Capacitor Polyester film C8 Capacitor Polyester film C9 Capacitor Polyester film C10 Capacitor Aluminum electrolytic C11 Capacitor Aluminum electrolytic C12 Capacitor Aluminum electrolytic C13 Capacitor Aluminum electrolytic C14 Capacitor Polyester film C15 Capacitor Polyester film C16 Capacitor Aluminum electrolytic C17 Capacitor Polyester film C18 Capacitor Polyester film C19 Capacitor Aluminum electrolytic C20 Capacitor Aluminum electrolytic C21 Capacitor Aluminum electrolytic C22 Capacitor Tantalum electrolytic C23 Capacitor Multilayer ceramic C24 Capacitor Multilayer ceramic C25 Capacitor Polyester film C26 Capacitor Polyester film C25 Capacitor Polyester film C26 Capacitor Polyester film C27 Capacitor Ceramic Disc C28 Capacitor Polyester film C29 Capacitor Polyester film C30 Capacitor Aluminum electrolytic C31 Capacitor Aluminum electrolytic C32 Capacitor Aluminum electrolytic C33 Capacitor Aluminum electrolytic C34 Capacitor Aluminum electrolytic c35 Capacitor Aluminum electrolytic c36 Capacitor Tantalum electrolytic C37 Capacitor Tantalum electrolytic C38 Capacitor Polypropylene film C39 Capacitor Multilayer ceramic C40 Capacitor Multil
66. ion data for the grating in guestion which is read from user specified text file Each line of the file must contain data in the form lt expected wavelength gt _ lt measured wavelength gt where both values are in nanometers and comma is used as a decimal mark Table 4 2 Descriptions of VIs related to controlling the monochromator VI name Description Initialize Initialize connection Send and read Send commands and read replies manually Close Close connection Get position Read current position Move distance Move a given distance Move pulses Move the stepper motor a given number of individual pulses Move to position Move to a given position Stop Force all movement to stop Get gain Read the linear correction gain of the controller Get offset Read the linear correction offset of the controller Set gain Write the linear correction gain of the controller Set offset Write the linear correction offset of the controller Calibrate controller Read all linear correction data and if desired write new values Polynomial correction VI uses least sguares method to fit polynomial correction function of given degree to calibration data or manually given coefficients can be used with Manual correction Spline correction VI on the other hand creates a spline function from given calibration data and uses it to interpolate corrected wavelengths 4 1 3 Filter wheel VIs used for filter w
67. ip has the same dimensions but the holder part is significantly larger Due to this a new holder was machined on the basis of the old one Also a fitting piece was made that enables the smaller version of the glower to be used as well The new holder and the glower in operation are shown in figure 3 3 Figure 3 3 The new holder and the ceramic glower in operation 34 The ceramic glower can be used as an infrared radiation source up to 16 um For near infrared operation a halogen lamp can be used The model commonly used in the setup is a 50 W Osram Halostar that peaks its optical power around 1 2 um 3 It provides a good source of infrared radiation up to 2 5 um after which the glass envelope starts to absorb the radiation 4 Both radiation sources use the same power supply Mastech HY3005D regulated power supply was previously used in the setup However the device in guestion was measured to be surprisingly noisy having a root mean sguare ripple well above the nominal 500 uV 45 The test measurement was done driving 12 V into an 8 O load It was then replaced with an elc AL 924A power supply which is capable of providing voltage up to 30 V and current up to 10 A Higher current rating enables higher power light sources to be used in the future According to specifications the absolute maximum ripple of the device is 1 8 mV RMS and 5 mV peak to peak even at maximum current 46 A similar ripple measurement was conducted and the n
68. it can then be filtered as explained in section 2 6 1 5 3 Linearity measurements Linearity measurements were done within sensitivity ranges from 100 uV to 1 V The sensitivity at the full range input voltage of a particular sensitivity ranges was used as a comparison value for relative sensitivity For all ranges the nonlinearity was less than 1 when the input voltage was more than 1 of the full range voltage and less than 2 90 when the input voltage was more than 0 1 of the full range voltage Measuring signals below 1 of the signal level is highly discouraged for the sheer problem of limited bits in the output value As an example the relative sensitivity of the lock in 58 amplifier at 1 V range is shown in figure 5 5 For this range the measurements were also repeated by using the signal generator without the attenuator The results for both measurements converge 1 01 ai o o Relative sensitivity 1 0 99 a m Signal generator with attenuator m Signal generator 0 98 1E 4 1E 3 0 01 0 1 1 Input voltage V Figure 5 5 Relative sensitivity of the lock in amplifier at 1 V range Also a comparison measurement between sensitivity ranges from 100 nV to 1 V was done The sensitivity for every range was measured using 50 of the full range voltage as the input voltage These values were then scaled with the nominal sensitivity and compared to the 1 V sensitivity range The dete
69. lY 88 MZLOUN LO ld LOOPNI sa L007NI L007NI za ZOO NI La AGL ea ZOOVNL bavi Z XO Hoy uoo ti ANI 619 L ExX A 6 6082 vol ANI no NI ZOOVNL 9a H TPEO0 XI bd SNIVIA iring Schematic of the power supply and mains voltage w 81 Figure B2 ed On op sa on or LNH 9VIZVSNO 46 006 00L NG 1006 06 LX L ZLX Oz Or MOZZ 9TU MOzz LAdLNO 6X d2ZcVdO Ol raeoroa 891 i STU lt N3 N9 A l A A GED YED ECD ASPESNI 60 ASPESNI 8a U00L 8e9 ETU 8X LX OL b VLX OL LAdNI f lasing circul Figure B3 Schematic of the b 82 AOZZ 089 CWS 8IN9 Noz CWS peu sey DEU ey 8ey 0 2 9 ZX OL CA 89 AZZ 89 ZAZ AL CC O 1 2 oL ceu Ley 0 y BCM 8cMu i o iE L ZLX OL ZP LX Sz cT HAMOd 1 ZIX OL pix LI Nu0cc gb Jc ma an 00 XY SV ds 10 XL vv z mm ZC EV mm N m 0 ZV m 319901 mm VO LV mm SS sa ov I 9d p e gX OL po 70 UA mm L EX OL mm 80 ONO mm si a hg gen 08821 254 AQ mm 010 AS mm a N LLd ACE m mm CLO 13578 mm les koskei l n kl i q a oy a coooooomnonD lt 9X OL ay a earo RIS TZA b 9X OL oun ounpiy o PN AZO9LOW LSIO Figure B4 Schematic of the front panel circuitry and microcontroller wir
70. lso most data and power cables run behind the devices while signal cables are in front so as to minimize noise In order to simplify data wirings all device communications were implemented with a USB hub and various interface adapters that were placed behind the display These connections are further explained together with setup automation in chapter four Table 3 2 List of measurement and control devices mounted to the rack shown in figure 3 2 Tag Device Model 1 Cooling fan unit Self made 2 Linear translator controller Self made 3 Rack main power switch Adam Hall 87471 4 Monochromator controller MAP 23 5 Power supply elc AL924A 6 High voltage supply Stanford PS325 7 Multimeter Agilent 34410A 8 Lock in amplifier Stanford SR830 9 Filter wheel controller Optec IFW 10 Optical chopper controller Terahertz Technologies C 995 11 Power supply Instoma TL201 12 Precision attenuator Self made 13 Signal generator Agilent 33521A 14 Linear translator controller isel IT116 flash 15 Computerand accessories 32 Figure 3 2 Measurement and control devices mounted to a standard 19 inch rack cabinet The monochromator controller number 4 in figure 3 2 and the high current power supply 5 tend to heat a lot even when they are not active Therefore they were placed under a cooling fan unit 1 which circulates air through the whole rack cabinet The cooling unit consists of
71. lues of the Fundamental Physical Constants Web Version 6 0 National Institute of Standards and Technology 2011 Referred 19 11 2011 Internet http physics nist gov constants Dereniak E L Boreman G D Infrared Detectors and Systems New York Wiley 1996 592 pp ISBN 978 0471122098 Lano J Facility for Spectral Responsivity Measurements of Infrared Detectors Master s Thesis HUT Department of Electrical and Communications Engineering Espoo 2006 54 pp Valkonen M Development of Dry Nitrogen Purged Spectrometer for Infrared Region Master s Thesis HUT Department of Electrical and Communications Engineering Espoo 2007 64 pp Ahonen P Extension of Spectral Responsivity Scale for Infrared Detectors Master s Thesis HUT Department of Electrical and Communications Engineering Espoo 2009 92 pp Phillips A C Introduction to Quantum Mechanics Chichester John Wiley amp Sons Ltd 2003 282 pp ISBN 0 470 85323 9 Settle F A Handbook of Instrumental Techniques for Analytical Chemistry Englewood Cliffs New Jersey Prentice Hall 1997 995 pp ISBN 978 0131773387 Parr A C Datla R U Gardner J L Experimental Methods in the Physical Sciences Vol 41 Optical Radiometry Amsterdam Elsevier Academic Press 2005 565 pp ISBN 978 0124759886 71 9 10 11 12 13 14 15 16 17 18 Nash M An investigation into the photocatalytic properties
72. ly the full scale reading of the measured signal somewhat similar to a range in a multimeter When using digital PSDs averaging allows measurements of signal levels even below one bit resolution If N is the amount of measurement samples uncorrelated noise is reduced by a factor of 1 N Oversampling an AC signal is rather complicated but since the filtered output of a PSD can be assumed constant output 25 resolution simply increases by a factor of N In essence doubling the amount of measurement samples adds one bit to output resolution 42 Traditionally dynamic reserve is defined as the ratio of the largest tolerable noise to the full scale signal However in case of digital PSD this definition is not practical because as mentioned above there is no unambiguous limit to resolution Instead dynamic reserve refers to the distribution of gain between the analog preamplifier and digital multiplication For optimum performance dynamic reserve should be as small possible Just enough to avoid noise voltage to overload any part of the device 42 2 6 2 Amplifier for photoconductive detectors When an electric potential is applied across the absorbing region of a photoconductive detector and the energy of an incoming photon exceeds the energy gap between the valence and the conduction band a current g proportional to the irradiance flows through the detector 43 The detector also has a certain dark resistance Rp and corresponding
73. mW 5 x 20 mm 50 Hz 1A 1A 700 MA 1A 0 296 70 kHz 400 Blue 240V 10 turn 40 W 40 W 600 mW 500 mW 500 mW 600 mW 600 mW 600 mW 600 mW 1W 1W 1W 1W 1W 1W 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 500 mW 600 mW 500 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW 600 mW TO 220 TO 220 TO 220 TO 220 DIP8 DIP8 DIP8 DIP16 TO 220 TO 220 12 15 15 12 12 12 12 15 15 15 15 15 15 12 12 12 11 11 12 15 15 15 12 15 12 12 12 11 11 11 11 11 11 11 11 11 11 11 25ppm CSIL8 250 VAC S2 S3 S4 S5 TAB1 TAB2 TAB3 TAB4 TAB5 TR1 TR2 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 pc Switch Toggle ON OFF ON Switch Rotary 1 x 12 Switch Pushbutton SPST NO Switch Pushbutton SPST NO Tab Straight Tab Straight Tab Straight Tab Straight Tab Straight Transformer Toroidal Transformer Toroidal Connector Terminal block Connector Terminal block Connector Terminal block Connector Terminal block Connector Terminal block Connector Terminal block Connector Terminal block Connector BNC socket PCB Connector BNC socket PCB Connector Terminal block Connector Terminal block Connector Terminal block Connector BNC socket bulkhead Connector Terminal block Microcontroller MS 500H CK1059 R13 24AL 05 BB R13 24AL 05 BR 7
74. match the impedance of coaxial cables This causes a small voltage drop when the generator is loaded with the 200 kQ input impedance of the attenuator that has to be taken into account The first position in the attenuator s rotary switch only loads the signal source but does not attenuate the output so the actual voltage over the attenuator V can be measured and used as a reference value ATTENUATOR LOCK IN Vat Raro Figure 5 2 Model for attenuator connection The voltage over the attenuator V can be measured using the first position of the attenuator Also the input impedance of the measurement device connected to the attenuator causes an error to actual attenuated voltage Var According to Th venin s theorem a resistive attenuator is eguivalent to a single voltage source and a single series resistor Figure 5 2 illustrates the principle Parameter A is the nominal attenuation 55 A 5 1 Ra Rb and the resistance R is the parallel connection of R and R RaR Rallb Rasp 5 2 The actual attenuated voltage Van can now be determined basi 5 3 att in Zin Rajjb The assembled attenuator is presented in Figure 5 3 The attenuator s enclosure is galvanically isolated from the signal ground This allows separate grounding to be used for noise cancellation That is done most conveniently by using a floating signal generator with a separate guard wire for case grounding Figure 5 3 Pr
75. mined by comparing the input voltage from the signal generator and the output voltage from the preamplifier Table 6 2 shows the measurement results The difference between the nominal and measured gain is most likely due to the accuracy of the resistor network with is 0 1 90 for each separate resistor element However much more important factors are the high stability and low temperature coefficient of the divider network The ratio between the decade dividing resistor and thus the gain of the device changes less than 50 ppm per year and the difference in temperature coefficients between two resistors is less than 2 5 ppm C 64 Table 6 2 Measured gains of the preamplifier measured at 80 Hz Nominal gain Input Output Measured Relative expanded voltage voltage V gain uncertainty of measured my gain 96 K 2 0 dB 1001 471 1 000937 0 99947 0 015 20 dB 100 1426 0 998642 9 9722 0 03 40 dB 9 99342 0 99885 99 951 0 09 60 dB 0 99936 0 99754 998 17 0 1 6 2 2 Linearity Linearity of the amplifier was measured by keeping the amplitude of the signal generator constant and measuring the output voltage at different attenuations The results are shown in figure 6 3 At all nominal gains smaller input voltage resulted in smaller measured gain 65 1 0005 1 0000 0 9995 Relative gain o wo O O o 0 9985 0 9980 0 9975 1E 6 1E 5 1E 4 1E 3 0 01 0 1 1 Input voltage V
76. mplifier is a device that uses a phase sensitive detection method and enables the detection of very small signals in the presence of overwhelming noise The method requires the signal source to be modulated at a known constant reference frequency A simplified block diagram of a lock in amplifier is presented in figure 2 12 For clarity various filtering and amplifications stages found in real devices have been omitted in 23 this theoretical model The multiplier stage of the lock in amplifier is commonly referred as a phase sensitive detector PSD 41 Multiplier Low pass Input y filter signal Phase O shifter Output Reference X Output signal Multiplier Low pass filter Figure 2 12 Simplified block diagram of a lock in amplifier The reference signal u is a sine wave with an amplitude A and angular frequency o The phase shifter is used to generate another reference signal i with 90 or 7 2 radians phase difference N u A sin w t t A sin ot 2 19 The input signal uin is a sum of a possible very large noise component u and the desired measurand signal um that has an unknown amplitude Am and is being modulated with reference frequency The measurement setup may cause a delay between reference and measured signal which is seen as phase shift m Uin Um Un Am sin o t Pm Un 2 20 Using trigonometric identities one can obtain the result from input and reference
77. n composition Resistor Carbon composition Resistor Carbon composition Resistor Carbon composition Resistor Carbon composition Resistor Carbon composition Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Thick film Resistor Metal film Resistor Thick film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor Metal film Resistor network 6 decade divider Switch Toggle ON ON 1N5245B 1N5245B MC1602F SYR RIX 0342 H MC7815CT MC7915CT TL783CKCSE3 MC7815CT REFO2APG4 OPA445AP OPA27GP DG409DJZ LVRO12K 3590S 2 103L 3310C 1 203L TIP31C TIP31C SPRX1 2C SPRX1 2C HTS00635006JPB17 HTS00631006JPB17 CNS471A6 RT 556 85 15V 15V 16 x 2 CH 500 mA 3A 15V 15V 1 25 125 V 9V 5 V 45V 8 MHz 4x2 CH 5 mm 120 mA 10 K 20 K 3A 3A 1kO 330 MQ 330 MQ 100 kQ 100 kQ 330 kQ 820 30 kQ 30 kO 30 kQ 30 kO 30 kO 30 kO 220 kO 10 MQ 10 kQ 27 kQ 330 kQ 1000 220 MQ 220 MQ 500 MQ 1000 100 MO 220 kO 220 kO 2200 1kO 2 2 KQ 6 8 kQ 22 kQ 68 kO 220 kO 680 kO 22MO 6 8 MO 20 MO 22MO 0 196 2A 500 mW 500
78. nce NEI describes the performance of the whole system and is defined as the irradiance needed at the entrance aperture for the system output to have the signal to noise ratio of one The inverse of NEP is called detectivity D which is proportional to the square root of detector active area and noise equivalent bandwidth of the measurement system B Therefore for detector comparison normalized detectivity is more convenient The most used quantity is the specific detectivity D that is defined as detectivity normalized to an effective area of 1 cm and noise equivalent bandwidth of 1 Hz It is expressed as EA A VAaBn D DJAaBn ne 2 18 and is typically measured in the unit of cm sW 2 A comparison of the specific detectivity of various commercially available infrared detectors and theoretical limits is presented in figure 2 10 The chopping frequencies used in the figure are 10 Hz and 1 kHz for thermal and photon detectors respectively 19 citate SARO sai N ee perizie gt cc 1 15 2 3 4 5 6 78 910 15 20 30 40 Wavelength um Figure 2 10 Comparison of the specific detectivity of various commercially available infrared detectors solid lines and theoretical limits dashed lines 38 2 5 2 Photon detectors Intrinsic extrinsic and free carrier based semiconductor detectors together with quantum well and quantum dot structure based detectors are considered to
79. nement in one dimension is called a guantum well and confinement in all three dimensions is a guantum dot Especially well structures have very important applications for example in space technology and thermal imaging A guantum well infrared photodetector OWIP has been demonstrated in the range of 3 to 80 um A focal plane array FPA construction of OWIPs allows high performance imaging at long wavelengths 39 Ouantum dot infrared photodetectors ODIP have also been commercialized for infrared imaging devices and are feasible in high speed operations 40 Photon detectors can also be divided into groups by operational mode which are photovoltaic photoconductive photoemissive and photoelectromagnetic PEM Most 21 materials mentioned above can be used in different modes of operation Photovoltaic detectors have a semiconductor junction where a photon absorbed in the depletion region excites an electron from the valence band to the conduction band This creates a potential difference at the junction and current can flow in a closed circuit Photoconductive detectors on the other hand do not have a semiconductor junction and reguire external voltage across the detector The electron excitation by absorbed photon then changes the measured resistance of the detector Photoemissive detectors are based on the photoelectric effect where electrons are emitted from matter as a conseguence of photon absorption Typical construction is a metal
80. ngth angle of radiation temperature and surface roughness but in practice it can often be treated as a constant Especially when the emissivity is assumed to be independent of the wavelength the shape of the object s emission spectrum is eguivalent to the spectrum of a black body This is Known as the gray body assumption The emissivity of real black body cavities is typically greater than 0 99 16 Total radiant emittance of a black body in other words the total power radiated per unit surface area of a black body can be derived by integrating Planck law over all wavelengths and over the half sphere solid angle The result obtained is known as the Stefan Boltzmann law M 0T1 2 7 where o is the Stefan Boltzmann constant defined as 2r k 15c2h3 5 670 373 1078 J s lm K 2 8 Real objects however do not emit all the power and therefore the emissivity of the material has to be taken into account In the case of a grey body Stefan Boltzmann law takes the form M eoT 2 9 If the body is in an enclosure such as a cavity or a room having a temperature Ty it will both radiate and absorb energy The net radiant flux or the net radiant power of a source area of A becomes then 17 AO ec A T T amp J 2 10 2 2 2 Incandescent sources Incandescent light sources are most used in the near and mid infrared region At longer wavelengths a fundamental problem raises from the conflict between Wi
81. nt conductors generate a voltage proportional to a temperature gradient Many thermocouple junctions can be furthermore connected in series to multiply the output level The response time of a thermopile is typically around milliseconds but it can be improved by reducing the active element volume Semiconductor thermopiles have a higher sensitivity but the traditional metal connectors are found to be more suitable for photometry Bolometers use temperature dependent resistance for detection Semiconductors or biased active component are typically used to achieve high temperature coefficient of the resistance Also some kind of heat absorbing mechanism is needed The response time of a bolometer can be adjusted with the design of the heat sinking but it is around milliseconds Cryogenics cooling of the detector improves sensitivity and noise characteristics but is not mandatory 2 6 Signal amplification Signal amplification itself is a vast science so the subject has been narrowed down to two amplifiers most relevant to the thesis The basic theory of lock in amplifiers is discussed in section 2 6 1 and the various lock in related electrical measurements that were conducted are presented in chapter 4 Section 2 6 2 deals with the theory related to the amplification of photoconductive detector s signal Such a device was designed and built as part of this thesis and is closely examined in chapter 6 2 6 1 Lock in amplifier A lock in a
82. nt in the near infrared 14 region but rapidly increases thereafter 32 The best blackbody material so far in a wide spectral range is a recently discovered nanotube structure that has a remarkable reflectance of less than two percent in the 0 2 200 um wavelength range It is manufactured by growing a forest of vertically aligned single walled carbon nanotubes on a silicon wafer 33 A comprehensive overview of material selection regarding lenses mirrors and black coatings in the infrared region can be found in 5 2 4 Wavelength selection This chapter outlines the most important principles and features of diffraction grating monochromator and filter based wavelength selection More profound review on the matter can be found for instance in 10 or 11 A monochromator uses a diffraction grating or a prism to separate a monochromatic beam from a broadband radiation It acts as an optical band pass filter by focusing a small part of the spatially distributed light The diffraction grating has multiples slits or grooves in a periodic configuration that creates a diffraction pattern of the incident light This can be expressed mathematically with a so called grating eguation d sin 0 sin03 mAn MEZ 2 11 where d is grating constant or the distance between the grooves on the grating 6 is the angle of the incident light 0 is the diffraction angle m is the diffraction order and Am is the corresponding wavelength The
83. ntroller The microcontroller s own regulator is used to provide the 5 V logic voltages However a separate 5 V precision voltage reference is used with the microcontroller s ADC Two 25 VAC and one 15 VAC secondary coils are used in series to obtain a 65 VAC voltage This is then rectified to a DC voltage of approximately 90 V A high voltage regulator is then used to regulate a stable 75 V voltage for the biasing circuit A 62 polymeric positive temperature coefficient device PPTC is placed before the regulator to limit the maximum current of the high voltage supply to around 120 mA Toroidal transformers are used in the power supply to minimize electromagnetic interference while a suppression filter is used to prevent radio freguency interference from the power grid 6 1 2 Biasing circuit A multi turn potentiometer is used to adjust the biasing voltage A 0 to 5 V setpoint voltage from the potentiometer is amplified with a high voltage operational amplifier to correspond to 0 to 64 V in the biasing output NPN and PNP power transistors are used as a common collector amplifier in front of the operational amplifier to boost the output current The setpoint voltage is also measured with the microcontroller s ADC and the biasing voltage is displayed in the front panel LCD The front panel switch allows the user to select internal or external biasing voltage or to use the amplifier without biasing at all A rotary switch and
84. ossible reply The reading of the reply may also be ignored when it is unnecessary Before sending anything to the device the serial read buffer is emptied since it tends to fill from unread replies A specific timeout is also defined which cancels the execution of the VI in case the connection is lost The connection closing VI sets the device to initial state and sends a command to exit remote mode when needed Finally VISA Close subVI is used to terminate the connection and check whether any errors occurred 4 1 2 Monochromator Short descriptions of VIs related to controlling the monochromator are presented in table 4 2 Get position VI gives position of the monochromator and Move to position moves the monochromator to a given position Also a given distance or a given number of stepper motor pulses can be moved with VIs Move distance and Move pulses respectively The Stop VI forces the movement to stop and is the only command that can be given while other command is being executed The monochromator controller uses linear correction the parameters of which can be read with Get gain and Get offset and written with Set gain and Set offset Calibrate controller is a standalone VI to read and write both parameters simultaneously The VIs listed above only utilize the linear correction done by the monochromator controller Additional three VIs are available if more precise nonlinearity correction is 47 needed Both reguire calibrat
85. output signal of the detector with an oscilloscope when the input aperture was closed Figure 3 7 shows the measured disturbance signal having a peak to peak voltage of over 500 mV 40 RMS 186mY Peak Peak 581mY E Frequency 5 z Per iod 19 89ms Figure 3 7 Output signal of the pyroelectric detector with blocked input aperture The preamplifier of the detector was supplied by a typical bench top power supply which seemed to cause most of the noise signal Multiple power supplies were tested but the proximity of the devices to the detector seemed to cause similar effect in any case This is most likely due to the fact that the casing of the detector is connected straight to amplifier s ground potential The problem was eventually solved by placing the power supply two meters away from the detector and placing an external RC filter network to supply lines close to the detector The schematic of the filter is shown in figure 3 8 and the resulting output with the improved arrangement in figure 3 9 The level of the line freguency disturbance attenuated over 24 dB Filter box Pi og yen EEE 1 To detector lemo connector Power supply 22R 10R 12 V t 12 V To detector chassis GND I I I I U I I I I Figure 3 8 RC filter for the pyroelectric detector 41 RMS Peak Peak Frequency Period Figure 3 9 Output signal of the pyroelectric detector with blocked input aperture after installing the
86. r Boltzmann constant 1 3806503 10 J K 1 vi Le Lea LASER LCD LED MASER MCT NEB NEI NEP NI P inc PCB PEM PSD PPTC ODIP OWIP R R2 Raw R Rp Rin RMS SI SNR Text Um Uin Un Ur Ur UV Var Vg Radiance Spectral radiance Light Amplification by Stimulated Emission of Radiation Liquid Crystal Display Light Emitting Diode Diffraction order Radiant exitance or radiant emittance Microwave Amplification by Stimulated Emission of Radiation Mercury Cadmium Telluride Number of measurement samples Noise Equivalent Bandwidth Noise Equivalent Irradiance Noise Equivalent Power National Instruments Incident optical power Printed Circuit Board Photoelectromagnetic Phase Sensitive Detector Polymeric positive temperature coefficient Radiant energy Quantum Dot Infrared Photodetector Quantum Well Infrared Photodetector Magnitude Resistance in amplifier model Resistance in amplifier model Attenuator resistance to input Parallel connection of R and R Attenuator resistance to ground Dark resistance Input resistance Load resistance Root Mean Square Responsivity International System of Units French Syst me international d unit s Signal to Noise Ratio Time variable Absolute temperature Temperature of an external enclosure Measurand signal Input signal Noise component Reference signal Phase shifted reference signal Ultraviolet Attenuated voltage Bias voltage
87. reciprocal of the groove constant is known as the groove density Figure 2 8 illustrates how multiple wavelengths with different diffraction order satisfy the eguation 10 15 Incident light Grating 0 3 0 um 6 0 um 1 5 um 9 0 um e 1 0 um mei SN 3 0 um 4 5 um a 2 0 um Diffraction order ni Sy a 3 0 um m 3 Figure 2 8 Illustration of the multiple solutions to grating equations with different diffraction order 5 The monochromator has a narrow but finite band pass The spatial density of wavelengths in the diffracted distribution is called reciprocal linear dispersion py It is measured in the plane of the exit slit transverse to the optical axis and is defined as _ aa Py ay 2 12 where y is the distance from the optical axis The band pass B of the monochromator can now be approximated d B wp Wo 2 13 where w is the width of the exit slit Since the reciprocal linear dispersion is a function of diffraction angle the approximation works best when the exit slit width is small As a conclusion smaller exit slit leads to narrower band pass but of course the overall optical power decreases As a conseguence of multiple wavelengths passing the monochromator a set of filters is reguired to eliminate the unwanted wavelengths Usually this is done using long wave 16 pass filters but in some cases a band pass filter might be beneficial Infrared filters are typically made on a g
88. rmined values can be used as correction coefficients when different sensitivities are used in comparison measurements The results are shown in figure 5 6 An error of almost 15 was measured in the 100 nV sensitivity range This suspicious result was confirmed with two different signal generator signal levels using two different attenuations The possibility of noise causing the measurement error was canceled out by repeating the two measurements with very long time constant of 30 s Based on this result one can conclude that measurements using this sensitivity range 59 must be avoided Fortunately omitting one sensitivity range is not a problem because the measurements can be done with 1 uV sensitivity range as well 1 00 x x u i 1 008 4 gt 0 95 gt 1 006 4 E op 1 004 4 O 090 AU a 1 002 4 O o PA X 1 000 4 El 0 85 1V 100 mV 10 mV 1 mV 100 pV 1V 100mV 10mV 1mV 100pV 10pV 1MV 100nV Sensitivity Range V Figure 5 6 Comparison of sensitivity ranges The sensitivities are scaled with nominal sensitivity of the range and compared to the value of the 1 V sensitivity range The test setup used for electrical linearity measurements was left to the spectrometer measurement setup The linearity related uncertainty in comparison measurements is thus reduced from previous 0 35 to 0 2 as correction coefficients for signals of different magnitude can easily be determined 60
89. s preamplifier was also used With nominal biasing voltage of 50 V and with well matched load resistance the responsivity of the detector increased approximately by a factor of 5 as compared to the values measured with the old preamplifier The new amplifier also provides over 40 dB larger gain if needed which improves the measurement of small signals According to the manufacturer s specifications the responsivity of the P2038 series detector is linear up to an incident energy level of around 1 mW cm after which it starts to decrease approximately by a factor of five per decade 51 This behavior was confirmed when the responsivity of the detector was measured with laser at power levels of 10 uW to 1 mW Absolute values were obtained by comparison measurement with the pyroelectric reference detector The results are shown in figure 6 5 230 N N gt N o o Responsivity V W N 8 190 180 10 100 1000 Power level uW Figure 6 5 Responsivity of the PbSe photoconductive detector measured with 1 523 um laser at 15 Hz chopping freguency The detector was kept at a temperature just above 0 C and a bias voltage of 30 V and a load resistance of 320 k O were used 68 7 Conclusions The fundamentals of optical radiometry and the theory of infrared sourcing filtering and detection were considered The current state of infrared radiation sources and detectors was reviewed Furthermore two amplification methods pha
90. se sensitive detection and amplification of photoconductive detectors using load resistance for voltage extraction were studied A facility for infrared measurements was extensively improved The whole setup was refurbished and measurement and control devices were mounted to a rack to enhance and simplify the usability of the setup The stability of the infrared sourcing and the quality of the measured signal from the reference detector were greatly enhanced Also the previously uncharted low frequency response of the reference detector was determined In addition the improved facility enables a larger repertoire of infrared sources to be used The upgraded infrared spectrometer measurement setup is capable of measuring spectral responsivity of detectors spectral transmittance of optical materials and spectral power distribution of light sources Infrared detectors can be calibrated using the available solid state detectors as working standards in the wavelength range of 1 to 5 1 um and the pyroelectric reference detector in wavelength range of 750 nm to 16 um The expanded uncertainty of spectral responsivity calibrations is less than 3 7 The automation software of the spectrometer measurement setup was fully updated Virtually all measurement and control devices are now computer controlled Many aspects relating to computing and signal processing for example the compensation for the nonlinearity in the monochromator s wavelength scale
91. spectrometer however the emission absorption or fluorescence spectrum of a material is measured The wavelength selection can be accomplished for example by using optical filters prisms gratings Fourier transform techniques tunable lasers or combinations of the methods mentioned above 10 The measurement setup concerning this thesis uses optical filters and a monochromator with a single grating in a so called Czerny Turner mount both discussed later in chapters 2 4 and 3 3 2 2 Sources of infrared radiation In general any device transforming energy into the optical part of the electromagnetic spectrum is a light source but for all practical purposes a light source should provide enough constant radiant power in the wavelength region in hand The emission of an optical source can be either spontaneous or stimulated and both types of radiation sources are available in infrared region For the purposes of this work only electrically powered light sources were studied In spontaneous emission a transition from an excited energy state to a lower energy state occurs This can happen in an atom molecule or nanomaterial that is excited for instance by heat electric arc electromagnetic radiation or electron hole pair recombination Incandescent lamps are based on thermal excitation and have a broad output spectrum Fluorescent lamps usually use ultraviolet light for excitation of fluorescent material and produce wide spectrum with nota
92. t controls the optical chopper instead the signal generator Combined measurement VI combines these two measurements so that both the wavelength and the chopping freguency are varied 52 5 Lock in amplifier linearity There are not that many linearity measurements of a lock in amplifier reported Even though signals are processed digitally in modern lock in amplifiers they still reguire analog amplification filtering and signal generation These analog components are the main cause of nonlinearities in the amplifier For many devices the linearity factor can change more than 0 1 over a factor of two change in the output However the nonlinearity characteristics are found to be independent of the used reference freguency and time constants 60 Comparison measurements done between different amplifiers suggest that the linearity characteristics vary from one instrument to another 60 When signals of very different magnitude are measured the sensitivity of the lock in amplifier has to be changed between measurements This automatically changes the gain of the device s preamplifier During the measurements it was found that this has a similar effect to the linearity characteristics as for changing from lock in amplifier to another Typical linearity measurements involve an optical setup where comparison is done between different lock in amplifiers However in this thesis a different approach was developed The measurements were done
93. temperature and wavelength ranges might be rather limited Nevertheless black bodies with emissivity well above 0 999 have been reported 22 23 2 2 3 Other sources Light emitting diodes for near infrared region have been long available In fact the very first LEDs operated in the infrared region Due advances in guantum technology and manufacturing wavelengths up to mid infrared region are reachable So called guantum cascade technology enables broadband emission in the infrared region Originally these devices reguired low temperatures to operate but LEDs have been reported to have emission in the spectral range of 5 8 um at room temperature 24 and a flat band from 6 to 8 um at low temperatures 25 There is also a recent patent for multi wavelength light source using various semiconductor elements that is specially designed for spectroscopy in the range of 650 2500 nm 26 The very narrow spectrum of a common laser can be used for instance for calibration purposes in spectroscopy and there are devices available in the whole infrared region Also broadband lasers are available exploiting various guantum structures Devices operating in the room temperature have been reported having continuous spectrum from 7 7 to 8 4 um 27 Supercontinuum generation on the other hand offers broadband radiation in optical fibers Using this method a broadband infrared source with emission up to 3 2 um is demonstrated 28 while commercial
94. the VI Set frequency If the given frequency is not within the devices limit the highest or lowest possible frequency is set The remote interface of the thermohygrometer is somewhat complicated but using the device with LabVIEW was made really simple When the connection is initialized the device is automatically set to suitable measurement mode and kept there until connection is closed All data handling is done using the Get data VI The device sends all the data in one long string of which all the values are parsed Most used are temperature and relative humidity but also available are absolute humidity dew point temperature wet bulb temperature mixing ration and device battery voltage The high voltage supply uses NI certified drivers that are suitable for controlling all Stanford PS300 series voltage supplies All except one VI of the driver worked seamlessly the function for setting the voltage was evidently meant for different version of the device The problem was remedied by partially rewriting the VI So instead of using the original VI Set voltage one should use the modified version Sef voltage fix 4 2 Programs for electrical measurements Various programs were written for electrical testing methods The programs were also used to characterize the lock in amplifier and these measurements are further discussed in chapter 5 Linearity measurements use Agilent 33521A signal generator and an attenuator to sweep voltage levels
95. tive photodiodes model P2038 03 lead selenide detector by Hamamatsu Photonics and model J13TE1 3CN SOIM lead sulfide detector by Judson Technologies are also being used These devices are compared in table 3 5 Table 3 5 Comparison of the two photoconductive detectors used in the setup 51 52 Manufacturer Hamamatsu Judson Tech Model P2038 03 J13TE1 3CN SO1M Material lead selenide lead sulfide Temperature range C 30 to 50 30 to 50 Nomimal temperature C 10 10 Wavelenght range um 1 5 to 5 1 1 0 to 3 5 Peak wavelength um 4 1 2 5 Peak D cm s W 3 10 1 5 10 Typical D cm s w 3105 1 5 10 Active area mm 3x3 10x10 Dark resistance MO 1 7 7 0 0 5 8 0 Nominal bias voltage V 50 240 Maximum bias voltage V 100 500 43 3 5 Other eguipment Whenever possible built in preamplifiers of the photovoltaic detectors are used In other cases common current to voltage converter such as Vinculum SP042 or Stanford SR570 are used For photoconductive detectors on the other hand a versatile preamplifier was built that is discussed in chapter 6 This device has an internal biasing possibility of up to 64 volts For higher voltages Stanford Research PS325 supply is used as an external bias which can provide DC voltage up to 2 5 kV and current up to 10 mA 53 The output signals of various detectors and preamplifiers are routed to the lock in amplifier
96. ttenuation The circuit provides attenuations in decades from 1 to 10 while the input resistance of the device is deliberately constant 200 KQ at all attenuations OUTPUT S1 Lorlin CK1059 X1 X2 TE Connectivity 1 1337541 0 Figure 5 1 Schematic of the precision attenuator The right selection of resistors was the main issue in the design In order to minimize the stray capacitances and impedances 3 2 x 1 6 mm surface mount resistors were used For values between 18 O to 180 kO high performance metal film resistors were used but for the smallest 2 O resistance only thick film resistors were available The key parameters of these components are shown in table 5 1 54 Table 5 1 List of resistors used in the attenuator 61 62 63 No Tech Mfr Part number Value Toler Temp coeff R1 Thin film Panasonic ERASAEB184V 180kQ 0 1 25 ppm C R2 Thin film Panasonic ERASAEB183V 18kO 0 1 25 ppm C R3 Thin film Panasonic ERASAEB182V 1 8 kQ 0 1 25 ppm C R4 Thin film Panasonic ERA8AEB181V 1800 0 1 25 ppm C R5 Thin film Koa Speer RN73H2BTTD18ROB25 180 0 1 25 ppm C R6 Thickfilm Vishay CRCW12062ROOFNEA 20 1 100 ppm C Figure 5 2 illustrates the model for attenuator connection All signal sources have nonzero output impedance Zout In signal generators the practice is to use 50 Q or 75 Q resistances in the output to
97. ugh a zinc selenide window to the third polypropylene box The window has an antireflection coating which improves the transmittance to approximately 80 percent in the wavelength range of 1 to 15 um The 31 beam is diverging at this point but it is then collimated with a bare gold coated off axis parabolic mirror placed at its focal length 152 mm from the monochromator exit slit Another similar mirror is used to focus the beam to the detector plane A linear translator is used to change detectors in the detection plane 3 4 2 Pyroelectric reference detector A hybrid pyroelectric detector SPH 49 from Spectrum Detector shown in figure 3 4 is used as a transfer standard for infrared responsivity measurements The device consists of a large lithium tantalate LiTaO pyroelectric element and a 100 GO transimpedance amplifier The detector was previously characterized up to 13 um and was found to be very linear as expected Detector s spectral responsivity in the wavelength range of 2 8 to 6 um is shown in figure 3 5 In order to link the spectral responsivity of the pyroelectric detector the method of substitution was used to compare the pyroelectric detector with the primary reference at wavelengths of 3 and 5 um 5 The primary reference is an electrically calibrated pyroelectric radiometer ECPR that is traceable to a cryogenic radiometer via comparison with silicon trap detectors 49 Figure 3 4 Hybrid pyroelectric detector used as
98. ve freguency Working Type Scaling lines mm range um factor 612001 1200 0 2 0 5 Ruled 1 612001 1200 0 35 1 0 Ruled 1 G600 600 0 7 2 0 Ruled 2 G300 300 1 5 4 0 Holographic 4 G150 150 3 0 8 0 Ruled 8 G75 75 6 0 16 0 Ruled 16 The monochromator can be driven manually or with an external controller The external controlling is simply done by giving clock pulses to the driver of the stepper motor rotating the grating With the scaling factor of one the theoretical resolution is 3 pm per clock pulse For computer controlling via serial port a custom made controller is used 48 It was working inconsistently until a large filtering capacitor was added to the supply voltage taken from the monochromator The device uses a first degree polynomial function to compute the needed amount of pulses to drive the monochromator to a wanted wavelength although separate clock pulses can also be driven A more sophisticated nonlinearity correction of the monochromator is done in the computer program either by a higher degree polynomial function or spline function For instance a narrow bandwidth laser can be used as a calibration source Without filters all diffraction orders are seen in the output and the whole scale can be calibrated The nonlinearity correction functions are further discussed in section 4 1 2 3 4 Output optics and detection 3 4 1 Output optics The output beam of the monochromator travels thro

Improved Reference Infrared Spectrometer - Aalto

Contents

Download Pdf Manuals

Related Search

Related Contents