Mentzikof

MsITT THESIS TITLE:
Optical tunable filter control circuit design for applications in DWDM optical add/drop nodes.
by
George Mentzikof
Supervisor: Dr. Dimitrios Klonidis
AIT, December 2006
AIT / MsiTT student (2004): George Mentzikof AIT/ MsITT 2006 THESIS TITLE: Optical tunable filter control circuit design for applications in DWDM optical add/drop nodes.
TableofContents
INTRODUCTION ............................................................................................................... 3 CHAPTER1.TUNABLEFILTERAPPLICATIONS................................................................ 4
1.1 Wavelength-division multiplexing.............................................................................................. 4 1.2 ITU GRID SPECIFICATION..................................................................................................... 6 1.3 FILTER APPLICATIONS.......................................................................................................... 7 1.4 OADM (Optical Add-Drop Multiplexers) .................................................................................. 8 1.5 Fixed vs Reconfigurable OADM ................................................................................................ 9 1.6 Optical Layer Cross-Connects (OXCs)..................................................................................... 10 1.7 Band filtering ............................................................................................................................ 11 1.8 Why Reconfigurability Is Important [Error! Bookmark not defined. ....................................... 12
CHAPTER2.TYPESOFFILTERTECHNOLOGIES ............................................................. 15
2.1 Optical Filters in General.......................................................................................................... 15 2.2 Key Characteristics of Optical Filters. ...................................................................................... 17 2.3 OPTICAL FILTERS TECHNOLOGIES - TYPES .................................................................. 19 2.3.1 Gratings ................................................................................................................................. 19 2.3.2 Fiber Gratings ....................................................................................................................... 24 2.3.3 Fabry-Perot Filters ................................................................................................................ 26 2.3.4 Multilayer Dielectric Thin-Film Filters ................................................................................. 29 2.3.5 Mach-Zehnder Interferometers () ................................................................................. 30 2.3.6 Ferroelectric Liquid Crystal F-P Filters ................................................................................. 32 2.3.7 Acousto-Optic Tunable Filter ............................................................................................... 33 2.3.8 Electro-Optic Tunable Filters ............................................................................................... 35 2.3.9 Tunable Thermo-Optic Filter for WDM Applications........................................................... 35 2.3.10 Tunable Filters Based on Semiconductor Laser Structures.................................................. 36 2.4 Comments on the Various Filter Types Characteristics............................................................ 37 Table 2.1 Filter technologies characteristics overview-comparison .............................................. 37
CHAPTER3.THESANTECOTF655 ............................................................................... 39
3.1 Module Description .................................................................................................................. 39 3.1.1 Tuning Technology................................................................................................................ 40 3.1.2 Manufacturer Optical Specifications ..................................................................................... 41 3.1.3 Operational Parameters.......................................................................................................... 42 3.2 The OTF-655 Stepper Motor .................................................................................................. 45 3.2.1 Stepper Motor Theory ........................................................................................................... 45 3.2.2 OTF-655 STEPPER MOTOR................................................................................................ 49
CHAPTER4.OTF655DRIVINGCIRCUITDESIGN .......................................................... 50
4.1 The Driving Circuit Goals ........................................................................................................ 50 4.2 Circuit Analysis ........................................................................................................................ 51 4.2.1 ATMEL AVR ATtiny2313 ............................................................................................... 51 4.2.2 Stepper Motor Driver (L293D).......................................................................................... 52 4.2.3 The Voltage Comparator (LM339N) . ............................................................................... 55 4.2.4 The Power Supply Units.................................................................................................... 58 4.2.5 SANTEC OTF-655............................................................................................................ 58 4.2.6 The Circuit Electronic Schematic Diagram ........................................................................... 59 4.3 MICROCONTROLLER PROGRAM DESCRIPTION............................................................ 60 4.4 The PCB.................................................................................................................................... 62
CHAPTER5.FILTERMEASUREMENTS ........................................................................... 64
5.1 INPUT-OUTPUT FULL OPERATION RANGE CHARACTERISTICS COMPARISON .... 64 5.2 FILTER SHAPE OVER OPERATION RANGE ..................................................................... 65 5.2.1 Filter Shape at 1540nm .......................................................................................................... 65 5.2.2 Filter Shape at 1550nm .......................................................................................................... 67 5.2.3 Filter Shape at 1560nm .......................................................................................................... 69 5.2.4 COMPARISON ..................................................................................................................... 71 5.3 WDM and DWDM filter Stepping............................................................................................ 72 5.3.1 WDM (0.8nm) STEPPING................................................................................................... 72 5.3.2 DWDM (2x0.2nm) STEPPING ............................................................................................. 74 5.4 POLARIZATION LOSS MEASUREMENT. .......................................................................... 76 5.5 Channel Separation at 100GHz Spacing (WDM) ..................................................................... 77 5.6 Channel Separation at 50GHz Spacing (DWDM) .................................................................... 78 5.7 Measurements Conclusions ...................................................................................................... 79
CONCLUSION................................................................................................................. 81
APPENDIX A. MICROCONTROLER ASSEMBLY CODE. ...................................................... 82 References: ..................................................................................................................................... 86
INTRODUCTION
Optical filters are important components in optical communication systems, primarily for the rejection of the accumulated amplification noise in optical transmission systems. However, over the last years the development of tunable optical filters introduced another significant application in optical networking and more specifically in WDM switching nodes. Here, tunable devices are required in order to select a specific wavelength from a group of transmitted (WDM) wavelengths. The control mechanism to perform this operation is an important parameter and is the main objective of this research topic. The presented research activity focuses on the design and development of a control board able to operate a tunable optical filter module (Santec OTF655). The OTF-655 is a compact tunable filter device that incorporates a stepper motor and utilizes independent linear sliding technology to achieve accurate wavelength tuning while maintaining constant optical properties over the entire wavelength range. A series of measurements is performed in order to evaluate the operation of the driving circuit and the filters performance characteristics. As an addition to the above an overview of various optical filter technologies is done as well as a short presentation of their applications in optical communications.
CHAPTER 1. TUNABLE FILTER APPLICATIONS

1.1 Wavelength-division multiplexing In fiber-optic communications, wavelength-division multiplexing (WDM) is a technology which multiplexes multiple optical carrier signals on a single optical fiber by using different wavelengths (colors) of laser light to carry different signals. This allows for a multiplication in capacity, in addition to making it possible to perform bidirectional communications over one strand of fiber.
Fig.1.1 A simplified WDM system.
WDM systems A WDM system (fig.1.1) uses a multiplexer at the transmitter to join the signals together, and a demultiplexer at the receiver to split them apart. With the right type of fiber it is possible to have a device that does both simultaneously, and can function as an optical add-drop multiplexer. Various types of optical filtering devices are used to separate the various light frequencies like etalons, stable solid-state single-frequency FabryPerot interferometers in the form of thin-film-coated optical glass and others.
WDM systems are divided into two market segments, dense and coarse WDM. Systems with more than 8 active wavelengths per fiber are generally considered Dense WDM (DWDM) systems, while those with fewer than eight active wavelengths are classed as Coarse WDM (CWDM). CWDM and DWDM technology are based on the same concept of using multiple wavelengths of light on a single fiber, but the two technologies differ in the spacing of the wavelengths, number of channels, and the ability to amplify signals in the optical space. Channel: A single signal channel consists of a frequency band that has a finite pass bandwidth and is centered at a given frequency such as one specified by the ITU Grid.
1.2 ITU GRID SPECIFICATION Whether a WDM or DWDM system is used the channels used are not chosen randomly by they are predefined in a very precise way by the ITU grid specifications. An example can be seen at table 1.1.
Table 1.1
An ITU Grid Table specifies the following: - Channel Center Wavelength (nm): The wavelength at which a particular signal channel is centered. - Channel Spacing (GHz) The frequency difference between two neighboring channel center frequencies - Center Wavelength Offset (pm) A relative drift of the actual central wavelength of a particular channel with respect to the standard ITU Grid. - Channel Pass Bandwidth (nm) A maximum wavelength (or frequency) range around the corresponding center wavelength (or frequency) at a given power level. Now, the industry well accepts the definition at 0.5 dB down power level.
6
1.3 FILTER APPLICATIONS Optical Filters play an important role in Laser Application. They are as infrastructure components for various applications that have to do with their physical properties and as active parts of an Optical telecommunication system. Example for applications for infrastructure are: At laser sources to improve the accuracy and stability. For Gain equalization of fiber amplifiers. For Suppression of lasing at unwanted wavelengths. As Intermediate filters to improve the shape of a spreaded channel, correcting the channel spacing and BER value. For Optical Source Spectral Analysis (with a tunable filter and a broadband photodetector). For Balancing a photo detector response or the non-uniform spectrum of a light source. In transmission systems optical filter are necessary parts of some of the most important components needed to achieve optical telecommunications.
Fig.1.2 .A typical point to point optical network
They are inner parts of: - Multiplexers and Demultiplexers (fig.1.2), - Optical Add-Drop Multiplexers where one might choose a static or a dynamic selection of wavelengths to be add or dropped (fig1.2), - OXCs (Optical Crossconnects) , where wavelength selection is necessary to route information. In addition, they are used as safety switches on a SONET Ring network, quickly changing from one destroyed or lost wavelength-channel to another, thus restoring the communication.
1.4 OADM (Optical Add-Drop Multiplexers) The OADM node is the basic building block of an optical ring [1]. A generic overview of an OADM node is given in Fig. 1.3. An OADM consists of three stages, namely, optical demultiplexer, optical switch or add/drop stage, and optical multiplexer. Optical demultiplexer functions to separate wavelengths in an inlet fiber onto individual fibers. Then these fibers are dropped, connected to 2x2 optical switches, or directly connected to optical multiplexer. The last stage is optical multiplexer which functions to multiplex all those wavelengths either added, directly from optical demultiplexer, or 2x2 optical switches, into an outlet fiber. All the lightpaths that directly pass an OADM are termed cut-through lightpaths, while those lightpaths added/dropped at the OADM node termed added/dropped lightpaths.
Fig. 1.3 Reconfigurable OADM node
OADM nodes use passive optical filters to implement their fixed or variable wavelength relations. Here, various passive filtering technologies can be used, such as thin-film filters, fiber Bragg gratings, Mach-Zehnder filters, AWG devices, and diffraction gratings. OADM can be further classified into reconfigurable OADM if optical switches are deployed, or fixed OADM if not everything is reconfigurable. Physically, there are several ways to realize an OADM. One can use the traditional demultiplexer and multiplexer architecture to realize an OADM, or use a combination of fiber grating and optical circulators to realize an OADM.
Figure 1.4 displays the numerous architectural choices OADM provides.
Fig 1.4. OADM architecture choices.
An OADM can be seen as a special case of an OXC, i.e., a degree-2 OXC,, which is generally used for the construction of a ring-based optical transport network.
1.5 Fixed vs Reconfigurable OADM Fixed OADMs: add/drop and express (through) channels are predetermined and cannot be rearranged after installation the add/drop percentage is fixed reconfigurations after installation are performed manually. Fully reconfigurable OADMs: select add/drop s & offer connectivity between add/drop and through ports more complicated structures but more flexible offer flexible wavelength assignment with tuneable transmitters & receivers & remote path provisioning (provisioning on demand). more functional as they minimize the requirements in wavelengths and transponders potentially support protection accommodate any traffic churn & network expansion
1.6 Optical Layer Cross-Connects (OXCs) An optical cross-connect (OXC) is device used by telecommunications carriers to switch high-speed optical signals in a fiber optic network. There are several ways to realize an OXC. Opaque OXC: One can implement an OXC in the electronic domain; that is, all the inlet optical signals are converted into electronic format after they are demultiplexed by demultiplexers; then electronic signals are switches by a core large-scaled electronic switch module; finally the switched electronic signals are converted back into optical signals by using them to modulate optical lasers and then these optical signals are multiplexed by optical multiplexers onto outlet fibers. Transparent OXC: Second Approach is by switching optical signals in alloptical domain. Specifically, optical signals are demultiplexed by demultiplexers; then each demultiplexed wavelengths are switched by alloptical switch modules in the optical domain. After switching, those optical signals are multiplexed onto outlet fibers by optical multiplexers. Such a switch architecture keeps the features of speed and protocol transparency.
Fig 1.5. Basic OXC architecture.
As a compromise of opaque OXC and transparent OXC, there is a type of OXC called translucent OXC. In such a switch architecture, there is a switch stage which consists of an all-optical switch module, and an allelectronic switch module. Optical signals passing through the switch stage can be switched either by an optical switch module or an electronic switch module. Translucent OXC nodes provide a compromise of full optical signal transparency and comprehensive optical signal monitoring. It also provides signal regeneration possibility on each node. Thus, such node architecture seems to be the most practical one to deploy the nextgeneration optical transport networks. Optical filters play an important role in OXC wherever a wavelength selection must take place or at laser regenerators for narrow bandwidth.
10
1.7 Band filtering When distances of an Optical Ring topology are getting bigger then the necessity of Band Filtering comes up. Band filtering lowers insertion losses considerably (about 1-2 dB ), and thereby significantly improves link budgets. Moreover, this technique also prevents channel passband-narrowing effects associated with full-spectrum filters, i.e., since multihop filter concatenations (spectral function multiplication) induce time-domain signal distortion (e.g., 25% reduction in 3-dB bandwidth for four filters ). Note also that modular interleaving techniques can also be used to selectively increase channel densities (e.g., Fig. 1.6).
Fig 1.6. Interleaving/banding techniques for finer filter spacing.
11
1.8 Why Reconfigurability Is Important [3Error! Bookmark not defined.]
Fig 1.7a Simple point to point optical network.
Fig 1.7b Complex Ring topology optical network.
At figure 1.7a displays a simple point to point optical network between two Main sites. On the link two Data and storage servers are connected with the help of OADM. In such a network its relatively easy to have predefined wavelengths for each communication path. Problems start when a more real type of optical network appears like the one a figure 1.7b. At this figure many sites are connected in a ring topology and at least two rings are interconnected together. In that case a predefined wavelength-link relationship would be inefficient. There dynamic wavelength changing components are needed. In order to make this more clear see the example at figure 1.8
Fig 1.8 Larger domain of transparency (OAs not shown).
12
In the configuration, in figure. 1.8, there is an all-optical path, A-Y-X-U-Z, connecting A to Z. Transponders to optically isolate the domain would need to be present on the boundary of the domain and would serve to define the boundary of the domain. If a U-Y link were also added, the domain would turn from a topological "tree" into a more general "mesh." The alternate paths that result might be useful for restoration, but they also might complicate considerably the management of impairments. Returning to the example given in figure. 1.7a, for fixed wavelengths and a fixed configuration of the PADMs, each input port is in effect hard-wired to some distant output port. The connectivity between ports is fixed. In fact, the set of transmitters and receivers that are tuned to a specific frequency can be thought of as a plane that cannot be interconnected within the domain to planes defined by other frequencies. This can be a serious problem in some situations. In addition, the lack of reconfigurability can make it difficult to effectively use the capacity of a complex domain of transparency such as that in Fig. 1.8. In this figure, for example, if a specific frequency is in use between D and E, then this frequency cannot be used for the path A-Y-X-U-Z connecting A to Z it is blocked on the X-Y link. If it were possible to reconfigure this connectivity, the OTrS would in effect be turned into a distributed switch or cross-connect, which could be used for software controlled provisioning or for restoration after some types of failures. These types of functional capabilities are at the heart of the business rationale for deploying the optical network.
13
There are a number of ways in which reconfigurability may be achieved: Laser/receiver tunability. The lasers producing the LR wavelengths may have a fixed frequency, may be tunable over a limited range, or be tunable over the entire range of wavelengths supported by the DWDM. Tunability speeds may also vary. Tunability may give additional connectivity options, and allow ports that could not otherwise be connected to do so. Wavelength conversion. Internal to a domain of transparency, it might be possible to change the frequency of a connection, thereby in effect interconnecting the planes described earlier. This could be done by converting to the electrical domain and then modulating a laser (fixed frequency or tunable) with the signal. However, this is apt to be quite expensive and may degrade the signal. Conversion in the optical domain has been the subject of considerable research but products with functionality, reliability, performance, and cost adequate to make them attractive are not yet available. Switch fabrics. A switching fabric could be placed in the PADM or a crossconnect could be inside a domain of transparency or be placed between two domains. Adaptation grouping adjustments. If the boundaries between groupings are dynamically adjustable, bandwidth could be moved between groupings.
14
CHAPTER 2. TYPES OF FILTER TECHNOLOGIES

2.1 Optical Filters in General An optical filter is usually meant to be a component with a wavelengthdependent transmission or reflectivity, although there are also filters where the dependence is on polarization or spatial distribution. There are many different types of optical filters, based on different physical principles. Some examples for optical filters are: absorbing glass filters, dye filters, color filters: based on wavelength-dependent absorption in some material such as a glass dopant, dye, pigment or semiconductor Fabry-Perot interferometers, etalons, dielectric (dichroic) mirrors, fiber Bragg gratings, and arrayed waveguide gratings and photonic band gap devices: based on interference effects and wavelength-dependent phase shifts during propagation Lyot filters: also based on interferometric effects, but involving wavelength-dependent rotation of the polarization; used e.g. as birefringent tuners filters based on wavelength-dependent refraction in prisms (or prism pairs) or on wavelength-dependent diffraction at gratings, combined with an aperture Depending on the shape of the transmission curve, one distinguishes filters of the following types: bandpass filters, transmitting only a certain wavelength range notch filters, eliminating light of a certain wavelength range edge filters, transmitting only wavelengths above or below a certain value ( high pass and low pass filters) Next-generation high-speed communications systems will need to support the aggregate bandwidth and low latency requirements of current and future applications such as Internet access, high-quality videoconferencing, and multimedia traffic. Static dense wavelength division multiplexing (WDM) is currently becoming the leading technology in point-to-point transmission links. However in the next stage, dynamic WDM technology is expected to take over, and support full functionality on the system and network levels. Circuit switching moderate-speed functionalities such as optical filtering, optical add-drop multiplexing, wavelength conversion, and optical crossconnects seem to be the main goal in the near term. In the long term, ultrahigh-speed manipulations are expected to be needed for the packet and cell
15
switching required in IP and asynchronous transfer mode (ATM) applications. Here, we focus on one type of essential high-performance cost-effective devices, needed to enable widescale implementation of dynamic WDM systems and networks: tunable optical filters. Since in WDM systems each channel is related to a different wavelength, channel manipulations and particularly channel selection require optical wavelength selection (i.e., optical filtering). The various filter technologies are being designed in a way to best fit the above parameters in WDM applications. For example, a WDM International Telecommunication Union (ITU) standard grid for channel spacing of 0.8 nm (100 GHz) has already been defined, and in some systems 0.4 nm (50 GHz) of channel spacing is being implemented.
16
2.2 Key Characteristics of Optical Filters. [2] A variety of optical filtering technologies are available. Their desirable key characteristics for use in systems are the following: 1. Low insertion loss. The insertion loss is the input-to-output loss of the filter. 2. The loss should be independent of the state of polarization of the input signals (PDL value as close to 0db as possible). Since the state of polarization varies randomly with time in most systems, if the filter has a polarization-dependent loss, the output power will vary with time as wellan undesirable feature. 3. The passband of a filter should be insensitive to variations in ambient temperature. The temperature coefficient is measured by the amount of wavelength shift per unit degree change in temperature. The system requirement is that over the entire operating temperature range (about 100C typically), the wavelength shift should be much less than the wavelength spacing between adjacent channels in a WDM system. 4. As more and more filters are cascaded in a WDM system and DWDM, the passband becomes progressively narrower. To ensure reasonably broad passband at the end of the cascade, the individual filters should have very flat passband, so as to accommodate small changes in operating wavelengths of the lasers over time. This is measured by the 1 dB bandwidth, as shown in Figure 2.1. 5. Passband skirts should be sharp to reduce the amount of energy passed through from adjacent channels. This energy is seen as crosstalk and degrades the system performance. The lower the crosstalk point between two adjacent channels the lower the BER. - The crosstalk suppression, or isolation of the filter, which is defined as the relative power passed through from the adjacent channels, is an important parameter as well. A requirement for -20 or -30 dB channel isolation should be met by each filter technology. 6. Mass-production possibility and low price. Depending on manufacturing technology.
17
Additionally for tunable filters we can add the following characteristics: 7. Tuning speed (microseconds, preferably nanoseconds). Fast tuning is particularly important. For example packet-switched WDM networks require microsecond to nanosecond tuning time whereas circuit switched WDM networks can operate with millisecond switching. 8. Tuning Range. Wide tuning range is required to cover the entire 40 nm Erbium doped fiber amplifier (EDFA) spectral range. 9. 10. Power Consumption. Low power consumption is desirable. Control mechanism. Speed , accuracy and durability.
Generally, tunable optical filters can be divided application-wise into two main categories: Slow-speed tunable, with tuning times up to a few milliseconds, relevant for circuit-switched networks. High-speed tunable, in the microsecond and nanosecond range, relevant for packet- and cell-switching networks The first category is mostly based on temperature or mechanical effects, and is in the advanced development stage. At figure 2.1 below we can see the characteristic shape graph of an optical filter together with the characterization of some important spectral-shape parameters of optical filters. 0 is the center wavelength of the filter, and denotes the wavelength of the light signal.
Fig. 2.1 Shape Characteristic of a typical Bandpass filter.
18
2.3 OPTICAL FILTERS TECHNOLOGIES - TYPES In this part are going to be demonstrated various types of technologies to achieve filtering in optical spectrum as well as some of the ways approaches to make a tunable filter. 2.3.1 Gratings [3] Grating is almost any device whose operation involves interference among multiple optical signals originating from the same source but with different relative phase shifts. An electromagnetic wave (light) of angular frequency propagating, say, in the z direction has a dependence on z and t of the form cos(t-z). Here, is the propagation constant and depends on the medium. The phase of the wave is t-z thus a relative phase shift between two waves from the same source can be achieved if they traverse two paths of different lengths. Gratings have been widely used for centuries in optics to separate light into its constituent wavelengths. In WDM communication systems, gratings are used as demultiplexers to separate the individual wavelengths or as multiplexers to combine them. Two examples of gratings are shown in Figure 2.2 below.
Fig. 2.2 (a) A transmission grating and (b) a reflection grating. i is the angle of incidence of the light signal. The angle at which the signal is diffracted depends on the wavelength.
Based on figure 2.2 above here is a brief explanation on how it works. Multiple narrow slits are spaced equally apart on a plane, called the grating plane. The spacing between two adjacent slits is called the pitch of the grating.
19
Light incident from a source on one side of the grating is transmitted through these slits. Since each slit is narrow, by the phenomenon known as diffraction, the light transmitted through each slit spreads out in all directions. Thus each slit acts as a secondary source of light. Consider some other plane parallel to the grating plane at which the transmitted light from all the slits interferes. We will call this plane the imaging plane. Consider any point on this imaging plane. For wavelengths for which the individual interfering waves at this point are in phase, we have constructive interference and an enhancement of the light intensity at these wavelengths. For a large number of slits, which is the case usually encountered in practice, the interference is not constructive at other wavelengths, and there is little light intensity at this point from these wavelengths. Since different wavelengths interfere constructively at different points on the imaging plane, the grating effectively separates a WDM signal spatially into its constituent wavelengths. In a fiber optic system, optical fibers could be placed at different imaging points to collect light at the different wavelengths. Since multiple transmissions occur in the grating of figure 2.2(a), this grating is called a transmission grating. If the transmission slits are replaced by narrow reflecting surfaces, with the rest of the grating surface being nonreflecting, we get the reflection grating of figure 2.2(b). The principle of operation of this device is exactly analogous to that of the transmission grating. Majority of the gratings used in practice are reflection gratings since they are somewhat easier to fabricate. There are three types of gratings: classical (Stimax), arrayed waveguide, Bragg gratings and Fiber Bragg Gratings. [4]
20
In the classical grating, we use gratings in the Stimax configuration (Figure 2.3). The dispersive element was a grating embedded in a monoblock of silica, and the optical fibers were directly fixed to the block. The number of grooves on the grating (several tens to several thousands per millimeter) are obtained by using a diamond tool or by holographic photoetching.
Fig. 2.3 Stimax configuration. Grating muIti/demultiplexers consist of three main parts: entrance and exit elements (fiber array or transmission line fiber, emitters or receivers), focusing optics, and dispersive grating. The concave mirror transforms the diverging beam coming from any fiber into a parallel beam. The beam then hits the grating and is angularly dispersed back to the concave mirror where it is imaged, depending on its wavelength, onto the output fiber array.
The grating has the property of diffracting light in a direction related to its wavelength. Hence an incident beam with several wavelengths is angularly separated in different directions. Conversely, several wavelengths 1, 2 ... n coming from different directions can be combined in the same direction. The diffraction angle depends on the groove spacing and on the incidence angle. The arrayed waveguide grating (figure 2.4) was designed to increase the resolving power, i.e., the fine splitting of the wavelengths. It has increased optical path difference between the diffracting elements by using a waveguide structure equivalent to the well-known Michelson echelon gratings in classical optics. The advantage is a smaller channel spacing. The disadvantages are a much smaller free spectral range that will limit the total number of channels and near-end crosstalk that affects bidirectionality.
Fig 2.4 Arrayed waveguide grating
21
A Bragg grating is a transparent device with a periodic perturbation of the refractive index, so that a large reflectivity may be reached at a wavelength which fulfills a Bragg condition: the wavenumber of the grating matches the difference of the wavenumbers of the incident and reflected waves. Other wavelengths are only weakly affected by the Bragg grating, except for some side lobes in the reflection spectrum. Around the Bragg wavelength, even a weak index modulation can be sufficient to achieve nearly total reflection, if the grating is sufficiently long. Bragg gratings can be made in a bulk piece (e.g. of some glass or polymer) ( volume Bragg gratings) or in optical fibers ( fiber Bragg gratings), usually by irradiation with coherent ultraviolet light which is spatially modulated using an interference pattern. Volume Bragg gratings can be used e.g. as output couplers for laser diodes; the small reflection bandwidth (e.g. below 1 nm) can then lead to a narrow emission bandwidth and a low temperature dependence of the emission wavelength, which can facilitate the pumping of solid-state lasers. There are also laser diodes with built-in Bragg gratings for narrowing and stabilization of the emission wavelength. The reflection spectrum shown in Figure 2.5(a) is for a grating with a uniform refractive index pattern change across its length. In order to eliminate the undesirable side lobes, it is possible to obtain an apodized grating, where the refractive index change is made smaller toward the edges of the grating. (The term apodized means "to cut off the feet."). The reflection spectrum of an apodized grating is shown in Figure 2.5(b). Note that, for the apodized grating, the side lobes have been drastically reduced but at the expense of increasing the main lobe width.
Fig. 2.5 Reflection spectra of Bragg gratings with (a) uniform index profile and (b) apodized index profile. A is a measure of the bandwidth of the grating and is the wavelength separation between the peak wavelength and the first reflection minimum, in the uniform index profile case. A is inversely proportional to the length of the grating. A A. is the detuning from the phase-matching wavelength.
In figure 2.5(c) the Principle of operation of a Bragg grating Apodization is explained briefly.
22
Fig. 2.5c Principle of operation of a Bragg grating. Apodization can be achieved by gradually starting and ending the grating. This technique is similar to pulse shaping used in digital communication systems to reduce the side lobes in the transmitted spectrum of the signal. The bandwidth of the grating, which can be measured, for example, by the width of the main lobe, is inversely proportional to the length of the grating. Typically, the grating is a few millimeters long in order to achieve a bandwidth of 1 nm.
23
2.3.2 Fiber Gratings [5] Definition: a periodic or aperiodic perturbation of the effective refractive index in the core of an optical fiber, or a fiber device containing such a perturbation. A fiber grating (figure 2.6) is made by recording a Bragg grating in the core of single-mode fiber made photosensitive by doping with, for example, germanium. This grating can be used as an narrowband filter. It is necessary to use one grating per wavelength. So there is some limitation to the number of channels that can be obtained with these devices. A fiber Bragg grating is a periodic or aperiodic perturbation of the effective refractive index in the core of an optical fiber. Typically, the perturbation is approximately periodic over a certain length of e.g. a few millimeters or centimeters, and the period is of the order of hundreds of nanometers. This leads to the reflection of light (propagating along the fiber) in a narrow range of wavelengths, for which a Bragg condition is satisfied. Other wavelengths are nearly not affected by the Bragg grating, except for some side lobes which frequently occur in the reflection spectrum (but can be suppressed by apodization).
Fig. 2.6 : Schematic structure of a fiber Bragg grating (FBG).
The reflection bandwidth of a fiber grating, which is typically well below 1 nm, depends on both the length and the strength of the refractive index modulation. The narrowest bandwidth values, as are desirable e.g. for the construction of single-frequency fiber lasers or for certain optical filters, are obtained for long gratings with weak index modulation. As the wavelength of maximum reflectivity depends not only on the Bragg grating period but also on temperature and mechanical strain, Bragg gratings can be used in temperature and strain sensors. Transverse stress, as generated e.g. by squeezing a fiber grating between two flat plates, induces birefringence and thus polarization-dependent Bragg wavelengths. Fiber Bragg gratings can be fabricated with extremely low loss (0.1 dB), high wavelength accuracy ( 0.05 nm is easily achieved), high adjacent channel crosstalk suppression (40 dB), as well as flat tops. These properties of fiber Bragg gratings make them very useful devices for system applications. Fiber Bragg gratings are finding a variety of uses in
24
WDM systems, ranging from filters and optical add/drop elements to dispersion compensators. A simple optical drop element based on fiber Bragg gratings is shown in figure 2.7. It consists of a three-port circulator with a fiber Bragg grating. The circulator transmits light coming in on port 1 out on port 2 and transmits light coming in on port 2 out on port 3. In this case, the grating reflects the desired wavelength A2, which is then dropped at port 3. The remaining three wavelengths are passed through. Adding can be achieved by introducing a coupler to add the same wavelength that was dropped.
Fig. 2.7 Optical add/drop elements based on fiber Bragg gratings.
Many variations of this simple add/drop element can be realized by using gratings in combination with couplers and circulators. A major concern in these designs is that the reflection of these gratings is not perfect, and as a result, some power at the selected wavelength leaks through the grating. This can cause undesirable crosstalk. Fiber Bragg gratings can also be used to compensate for dispersion accumulated along the link.
25
2.3.3 Fabry-Perot Filters [6] A Fabry-Perot filter consists of the cavity formed by two highly reflective mirrors placed parallel to each other (figure 2.8). This filter is also called a Fabry-Perot interferometer or etalon. The input light beam to the filter enters the first mirror at right angles to its surface. The output of the filter is the light beam leaving the second mirror.
Fig. 2.8 Principle of operation of a Fabry-Perot filter.
This is a classical device that has been used widely in interferometric applications. Fabry-Perot filters have been used for WDM applications in several optical network testbeds. There are better filters today, such as the thin-film resonant multicavity filter. These latter filters can be viewed as Fabry-Perot filters with wavelength-dependent mirror reflectivities. Thus the fundamental principle of operation of these filters is the same as that of the Fabry-Perot filter. Compact Fabry-Perot filters are commercially available components. Their main advantage over some of the other devices is that they can be tuned to select different channels in a WDM system. Principle of Operation The principle of operation of the device is illustrated in Figure 2.8. The input signal is incident on the left surface of the cavity. After one pass through the cavity, as a part of the light leaves the cavity through the right facet and a part is reflected. A part of the reflected wave is again reflected by the left facet to the right facet. For those wavelengths for which the cavity length is an integral multiple of half the wavelength in the cavityso that a round trip through the cavity is an integral multiple of the wavelengthall the light waves transmitted through the right facet add in phase. Such wavelengths are called the resonant wavelengths of the cavity.
26
The power transfer function of a filter is the fraction of input light power that is transmitted by the filter as a function of optical frequency f, or wavelength. For the Fabry-Perot filter, this function is given by
This can also be expressed in terms of the optical free-space wavelength . as
The power transfer function of the Fabry-Perot filter is plotted in figure 2.9. for A = 0 and R = 0.75, 0.9, and 0.99. Note that very high mirror reflectivities are required to obtain good isolation of adjacent channels.
Figure 2.9. The transfer function of a Fabry-Perot filter. FSR denotes the free spectral range, / the frequency, and R the reflectivity.
The power transfer function Tfp (f) is periodic in f, and the peaks, or passbands, of the transfer function occur at frequencies f that satisfy f = k/2 for some positive integer k. Thus in a WDM system, even if the wavelengths are spaced efficiently far apart compared to the width of each passband of the filter transfer function, several frequencies (or wavelengths) may be transmitted by the filter if they coincide with different passbands. The spectral range between two successive passbands of the filter is called the free spectral range (FSR). A measure of the width of each passband is its full width at the point where the transfer function is half of its maximum (FWHM). In WDM systems, the separation between two adjacent wavelengths must be at least a FWHM in order to minimize crosstalk. (More precisely, as the transfer function is
27
periodic, adjacent wavelengths must be separated by a FWHM plus an integral multiple of the FSR.) Thus the ratio FSR/FWHM is an approximate (order-of-magnitude) measure of the number of wavelengths that can be accommodated by the system. This ratio is called the finesse (F), of the filter and is given by R
F= 1 R
But if the mirrors are highly reflective, won't virtually all the input light get reflected? Also, how does light get out of the cavity if the mirrors are highly reflective? To resolve this paradox, one must look at the light energy over all the frequencies. You will see that only a small fraction of the input light is transmitted through the cavity because of the high reflectivities of the input and output facets, but at the right frequency, all the power is transmitted. Perot Tunability A Fabry-Perot filter can be tuned to select different wavelengths in one of several ways. The simplest approach is to change the cavity length. The same effect can be achieved by varying the refractive index within the cavity. Mechanical tuning of the filter can be effected by moving one of the mirrors so that the cavity length changes. This permits tunability only in times of the order of a few milliseconds. For a mechanically tuned Fabry-Perot filter, a precise mechanism is needed in order to keep the mirrors parallel to each other in spite of their relative movement. The reliability of mechanical tuning mechanisms is also relatively poor. Another approach to tuning is to use a piezoelectric material within the cavity. A piezoelectric filter undergoes compression on the application of a voltage. Thus the length of the cavity filled with such a material can be changed by the application of a voltage, thereby effecting a change in the resonant frequency of the cavity. The piezo material, however, introduces undesirable effects such as thermal instability and hysteresis, making such a filter difficult to use in practical systems.
28
2.3.4 Multilayer Dielectric Thin-Film Filters A thin-film resonant cavity filter (TFF) is a Fabry-Perot interferometer, or etalon, where the mirrors surrounding the cavity are realized by using multiple reflective dielectric thin-film layers. This device acts as a bandpass filter, passing through a particular wavelength and reflecting all the other wavelengths. The wavelength that is passed through is determined by the cavity length. A thin-film resonant multicavity filter (TFMF) consists of two or more cavities separated by reflective dielectric thin-film layers, as shown in figure 2.10.
Fig. 2.10. A three-cavity thin-film resonant dielectric thin-film filter.
The effect of having multiple cavities on the response of the filter is illustrated in figure 2.11. As more cavities are added, the top of the passband becomes flatter and the skirts become sharper, both very desirable filter features.
Fig. 2.11. Transfer functions of single-cavity, two-cavity, and three-cavity dielectric thin-film filters. Note how the use of multiple cavities leads to a flatter passband and a sharper transition from the passband to the stop band.
29
2.3.5 Mach-Zehnder Interferometers () [7,8,9] A Mach-Zehnder interferometer (MZI) is an interferometric device that makes use of two interfering paths of different lengths to resolve different wavelengths. An incident beam of light is split in two ways and recombined after a short distance. The combining fields will interfere according to the phase difference between the two fields. If the two arms of the interferometer are of the same length, there will be no phase difference between the two fields, and they will interfere constructively. However, if a time delay is introduced in one arm, say by using a phase modulator, a periodic frequency-dependent phase difference is formed. The fields are combined according to the phase difference. Devices constructed on this principle have been around for some decades. Today, Mach-Zehnder interferometers are typically constructed in integrated optics and consist of two 3 dB directional couplers interconnected through two paths of differing lengths, as shown in figure 2.13(a). The substrate is usually silicon, and the waveguide and cladding regions are silica (Si2).
Figure 2.13. (a) An MZI constructed by interconnecting two 3 dB directional couplers. (b) A block diagram representation of the MZI in (a). AL denotes the path difference between the two arms, (c) A block diagram of a four-stage Mach-Zehnder interferometer, which uses different path length differences in each stage.
Mach-Zehnder interferometers are useful as both filters and (de)multiplexers. Even though there are better technologies for making narrow band filters, for example, dielectric multicavity thin-film filters, MZIs are still useful in realizing wide band filters. Narrow band MZI filters are fabricated by cascading a number of stages (figure 2.13c). In principle, very good crosstalk performance can be
30
achieved using MZIs if the wavelengths are spaced such that the undesired wavelengths occur at, or close to, the nulls of the power transfer function. However, in practice, the wavelengths cannot be fixed precisely (for example, the wavelengths drift because of temperature variations or age). MZIs are useful as two-input, two-output multiplexers and demultiplexers (figure 2.13b). The transfer function (figure 2.14) of the MZ filter is usually not narrow enough for WDM applications. Consequently, cascaded MZ configurations are used to produce narrower filters.(Figure 2.13c.)
Fig. 2.14. Transfer function of a multistage Mach-Zehnder interferometer.
However, since the MZ transfer function suffers from low finesse due to its inherent cosine shape, more attention is focused on combining additional technology within MZ configurations. Examples include MZ-based switches with thermo-electric or thermo-optic couplers, as well as Bragg grating filters designed in various MZ configurations. Tuning of such a filter is made through the electro-optic effect by applying a voltage to one arm of the interferometer, making these devices relatively fast , or by applying heat to one arm taking advantage of the heat sensitivity of such a device but it is rather slow and unstable process.
31
2.3.6 Ferroelectric Liquid Crystal F-P Filters[10] This type of filter is an Fabry Perrot interferometer with a ferroelectric liquid crystal material inserted in the resonant cavity (figure 2.15). Electrodes create an electric field over the cavity. This field causes the liquid crystal molecules to change their spatial orientation, which causes a change of the refractive index in the cavity. The peaks of transmission in an FP filter occur when the round-trip phase of the light in the resonator equals multiples of 2. The change of the refractive index of the liquid crystal causes a shift of the filter peak transmittance wavelength. The tuning time of the liquid crystal is several s. Insertion loss is small in this device (1-5 dB). he liquid crystal filter is essentially polarization-dependent, and different approaches to minimize this dependence (to about 0.3 dB) exist.
Fig. 2.15. Tuning a liquid crystal Fabry Perot filter.
32
2.3.7 Acousto-Optic Tunable Filter [11,12] The acousto-optic tunable filter is a versatile device. It is probably the only known tunable filter that is capable of selecting several wavelengths simultaneously. This capability can be used to construct a wavelength crossconnect (OXC). Operation The theory is based on utilizing diffraction of light by the refractive index gratings generated by the acoustic waves. The acoustic source which is usually an interdigital transducer is electronically controlled leading to an electronically tunable filter. An oscillating electric signal drives the transducer to vibrate, which creates sound waves in the glass. These can be thought of as moving periodic planes of expansion and compression that change the index of refraction. Incoming light scatters off the resulting periodic index modulation and interference occurs similar to in Bragg diffraction. Changing the driving acoustic frequency changes the band of optical frequencies that the filter passes (selects). Changing the power of the acoustic wave changes the length required for complete diffraction of the input wave. Figure 2.16, 2.17 present the operation basics of an AOTF The properties of the light exiting the AOTF can be controlled in five ways: 1. Deflection ,2. Intensity ,3. Frequency , 4. Phase , 5. Polarization The fact that the gratings are not permanent and acoustically generated. offers this filter its extended tuning range. Even from early reported filters, acousto-optic filters have shown an unmatched tuning range.
Figure 2.16. An acousto-optic modulator consists of a piezoelectric transducer which creates sound waves in a material like glass or quartz. A light beam is diffracted into several orders. By vibrating the material with a pure sinusoid and tilting the AOM so the light is reflected from the flat sound waves into the first diffraction order, up to 90% deflection efficiency can be achieved.
33
Fig. 2.17. A simple AOTF. An acoustic wave introduces a grating whose pitch depends on the frequency of the acoustic wave. The grating couples energy from one polarization mode to another at a wavelength that satisfies the Bragg condition.
Advantages: Acousto-Optic Filters have plenty of advantages. These are: 1) They have narrow-band filter characteristics with a full width at half maximum (FWHM) bandwidth of less than one nn (nanometer). 2) They have narrow channel spacing, a characteristic of importance for dense packing. The channel spacing of ATOFs can be in the order of 1 nm. 3) They offer an all-electronic broad tunability in the range of interest for guided wave optics ( 1.3 - 1.6m). This unmatched tuning range reaches hundreds of nanometers since it is only limited by the acoustic transducer's fractional bandwidth. ATOFs tuning range is about 20 times that of electro-optic filters. 4) The integrated version of the acousto-optic tunable filter requires low electrical drive power. 5) ATOFs are capable of simultaneous filtering of several wavelengths. Hence, they can perform parallel channel processing. The low radio frequency (RF) power requirement allows multiple wavelength selection by driving the filter with multiple RF frequencies simultaneously. 6) They are capable of more complex integration. For example, they can be configured as a four-port switch. 7) They have a moderately fast switching speed of few secs which is limited by the speed of the acoustic wave. 8) Also, ATOFs are important for use with narrow linewidth extended cavity semiconductor lasers. Disadvantages: The main disadvantage in acousto-optic filters is the secondary maxima (side lobes) in its optical transmission spectrum. This results in interchannel crosstalk in dense WDM (FDM) systems which may lead to a bit error rate (BER) penalty.
34
2.3.8 Electro-Optic Tunable Filters [2] Like their relatives the AOTFs, the filtering principle of electro-optical tunable filters (EOTFs) is TE (electric) TM (magnetic) mode coupling of a narrow and of incident light. The mode coupling occurs through interaction with a periodic perturbation of the refractive index. This perturbation is induced via the electro-optic effect using electrodes positioned in equal spaces along the waveguide. Only a narrow portion of the bandwidth will undergo a polarization change. The complete structure of an EOTF is similar to that of an AOTF. Tuning is achieved by a uniform electric field that tunes the average optical refraction index. The use of the electro-optic effect allows very small tuning times on the order of nanoseconds, still wide bandwidth, and strong sidelobes are a problem as in the AOTF.
2.3.9 Tunable Thermo-Optic Filter for WDM Applications [13,14] There is a class of Fabry-Perot based micromachined tunable filters which varies the resonator-length through physically adjusting the distance between the mirrors by electrostatic or thermal actuation. These elements are challenging to fabricate and fragile during operation.
Figure 2.18. Schematic setup of a Fabry-Perot filter with a solidstate cavity and dielectric mirrors.
The optical thickness of the resonator is modulated by thermal means. Variations in temperature give rise not only to changes in physical thickness but also alter the refractive index. This is known as the thermo-optic effect, which in this case dominates the filter behavior. Polycrystalline silicon is usually chosen as the resonators material, because of its high thermo-optic coefficient.
35
2.3.10 Tunable Filters Based on Semiconductor Laser Structures

[15]
Semiconductor laser structures inherently provide spectral filtering which is responsible for the longitudinal mode selectivity in these structures. Conventional multilongitudinal-mode lasers provide that selection through their FP cavity structure, whereas many single-longitudinal-mode lasers have built-in periodic structures (e.g., gratings in DFB and DBR lasers) to provide for the selection of a single frequency. Therefore, using these devices as optical amplifiers (i.e., biased below the lasing threshold) will result in devices that can selectively amplify signals only within a narrow spectral range, thus offering the required filtering. The resonant frequency can be varied by either current injection or temperature variation; however, temperature tuning is a slow mechanism.
36
2.4 Comments on the Various Filter Types Characteristics A summary of the main characteristics of the tunable filters discussed is given at table 2.1:
Filter Category Tuning Range (nm) Tuning Time Tuning Mechanism Comments Fiber Implementation Available, slow tuning, tuning range is typically less than EDFA. Can be used as a router as well. Suffer from wide passbands and high sidelobes. Thermo-Optic Tuning, Mostly as fixed in market Low power consumption Resolvable Channels. In research Temperature or mechanical stretching tuning. Mostly as fixed in market Insertion Loss can be an issue Small Number of Channels. Still suffer from a number of problems. Latest Technology
Fabry-Perrot
500
1-10ms
Mechanical /Others
Acousto Optic
250
10s
Acousto Optic
AWG Tunable Filter Liquid Crystal F-P Electro - Optic
40 30 16
10ms 0.5-10s 1-10ns
Crystal Orientation Electro-Otic
Fiber Bragg grating Mach-Zehnder Interferometers Tunable Filters based on semiconductor or laser structures Micromachined Fabry Perrot
10
1-10ms
Temperature, stretching Thermo electric, Thermo Optic -
50ns
0.1-1ns
<60
100s
Micromachined
Table 2.1 Filter technologies characteristics overview-comparison [16]
Tunable filters can be used in circuit- or packet switched networks; tuning time is therefore of the essence. An argument similar to that in the beginning of this chapter concerning which characteristics an optical filter is desirable to have derives. Fiber FP tunable filters are commercially available; however, given the slow tuning, they are most suitable for low-speed networks. Moreover, their tuning range is typically less or close to that of the erbium doped fiber amplifier (EDFA) gain range (3040 nm), and therefore they limit the number of available WDM channels. Acousto-optic tunable filters are also commercially available; however, they suffer from wide passbands and high sidelobes.
37
AWG filters and FBGs are both also commercially available, but mostly as fixed rather than tunable elements. Electro-optic tunable filters are still mostly in the research stage; they have fast tuning times, but also suffer from high sidelobes. Tunable filters based on semiconductor laser structures still suffer from a number of problems, including mode hopping, which have prevented them from being commercially available. Therefore, some more work needs to be done on tunable filters in order to achieve wide tuning ranges and fast switching. Packet-switching applications such as IP over WDM will typically call for switching speeds on the order of tens of nanoseconds. Coupling this with a wide tuning range (at least the EDFA 3040 nm gain band) then defines the requirements set for devices needed in the near future. A table such as table 2.1 is usable for a network designer because it briefly presents the main characteristics of each type. Tuning speed and range , but also the problems that come from physical properties of material and tuning mechanism of each filter are very critical for a proper choice.
38
CHAPTER 3. THE SANTEC OTF-655

3.1 Module Description The OTF-655 is a compact motorized tunable filter module designed for applications such as channel selection in DWDM systems, ASE noise reduction, and wavelength monitoring. The device incorporates a stepper motor and utilizes independent linear sliding technology to achieve accurate wavelength tuning while maintaining constant optical properties over the entire wavelength range. The OTF-655 offers a 35-nm tuning range to cover either the C- or L-band and is available with a wide range of filters. The device is hermetically sealed with a height of 7.1 mm-suitable for PCB mounting and system integration (figure 3.1).
Fig.3.1 OTF-655 mechanical drawing
39
3.1.1 Tuning Technology The OTF-655 utilizes the Fabry-Perrot Filter technology. Two competing technologies exist in motorized F-P tunable filters which make use of the varying cavity length technique. The angle-tuned filters achieve modifying the length of cavity by changing the angle of incidence of light at a specially designed filter element. These interference OTFs are stable, easy to mass-produce but have a serious PDL problem as well as variable characteristics over the tuning range. Figure 3.2 presents an angle tuned filter.
Fig. 3.2 Conceptual design of an angle tuned tunable filter. With increased angle of incidence, the filter transmits light of decreased wavelength.
Although there are solutions to the PDL problem like separating the S and P polarizations at the input and recombining them at the output. This type of filter is not the best. The OTF-655which is the filter AIT utilizes in its Optical Lab , uses a unique and patented technology of Santec. The linear sliding technique (figure 3.3 ).
Fig. 3.3 The linear sliding technology.
This technique according to Santec Engineers N. Mekada, A. AL-Hamdan, T. H. Chong and D. G. Daut [17] has many advantages such as polarization independence (PDL 0.1dB), low insertion loss, narrow 3-dB bandwidth ( as narrow as 0.3 nm), fine tuning resolution ( 0.1 nm), environmentally stable, suitable for mass production, low cost and offer nearly constant characteristics over their tuning range. In addition to all the advantages mentioned it allows cascading identical filter elements, the result will be narrower bandwidth characteristics.
40
The specially designed interference filter (multilayer dielectric-coated) is fabricated using electron beam assisted deposition of silicon oxide and titanium oxide with a quarter wavelength thickness. It is tuned by linear sliding. The sliding mechanism is performed using a stepper motor. The filter changes thickness in a linear pattern analogous to the length of it.
Fig. 3.4 OTF-655 simplified internal design block diagram.
As it can be seen at figure 3.4 the wavelength filter is directly attached to a step motor. This motor is responsible for sliding the linear filter and thus modifying its center (bandpass) frequency. A special electronic circuit with advanced complexity is required for driving the step motor. 3.1.2 Manufacturer Optical Specifications
Fig. 3.5 Optical Characteristics of OTF-655 from Santec
At figure 3.5, optical characteristic values are presented as measured by the manufacturer. Most important figures, which will be validated by the
41
measurements taking place on this thesis, are noted with an arrow next to them. So according to Santec the module can be manufactured to operate at C or L band (C in this case) with a maximum tuning range of 35nm. The filter bandwidth at -3dB(0.4nm max) and -20dB(1.2nm max) is such that allows DWDM channel separation. The insertion Loss is kept low (-6dB max) excluding the connectors. The maximum tuning resolution has a very low value (0.035nm max) and is going to be verified. The PDL has a maximum value of 0.2dB over the full tuning range of the filter showing that the filter is insensitive to polarization changes. Other characteristics have to do with how fast the filter can be tuned, its environment temperature insensitivity and what is the maximum light power it can handled. For the latter we rely on the manufacturer specifications. The others will be validated in testing process of the filter.
3.1.3 Operational Parameters The module needs very little electrical power to operate (less than 2.5W) which is mainly consumed only during the tuning process by the stepper motor. During the idle mode the power consumed is practically insignificant (less than 30mW). The control voltage is the voltage required for the motor movement, is also very low (3.6Volts max) (figure 3.6).
Fig. 3.6 Electrical-Operating specifications of the filter.
In order to drive the OTF-655 a pin connector coming out of the device allows the communication with the driving circuit. It also allow the mounting on a PCB(Printed Circuit Board). The pin description of this connector can be seen at figure 3.7 below.
42
Fig. 3.7 Pin Connector Description
More precisely : Pin 1-4: are the step motor drivers. Power supply and drive logic is given through this pins. Figure 3.8 displays the operation specifications of the step motor specifically. What is most important to keep out of this figure is the low voltage and low current the motor needs to operate and more precisely their maximum values. Most microelectronics stepper motors operate with 5V and can withstand high currents. For such king of motor there is a wide range of IC drivers. This specific motor presents a difficulty because its specification are met only in robotics applications. So a special IC had to be found in order to operate it properly.
Fig. 3.8 Stepper Motor Specifications
The stepper motor must be driven by a sequence of binary words in order to move in very precise and predetermined steps. This sequence given by Santec is depicted at the following table:
43
Pin 5-7: A potentiometer is mounted inside the case of OTF-655 directly attached to the sliding mechanism of the filter. Its purpose is to provide electrical indication of filters position and to monitor the operation of the filter so as the latter not to get out of the operation limits and be destroyed. The potentiometer is supplied with Vs(=+5) Volts and the operation range must be kept within the limits of 25% of Vs to 75% of Vs. Figure 3.9 below displays its operation specifications:
Fig. 3.9 Potentiometer operational specifications
Pin 8: Grounds the case of the OTF. Pin 9-11: These pins provide an optional monitoring of the operation temperature of the filter for calibration and special experiment purposes. Pin 12. Just a positioning pin.
44
3.2 The OTF-655 Stepper Motor Here some theory about stepper motors is presented and more information about the one the module has is given next. 3.2.1 Stepper Motor Theory [18,19] A stepper motor is a unique type of DC motor that rotates in fixed steps of a certain number of degrees. Step size can range from 0.9 to 90. Figure 3.10 illustrates a basic stepper motor, which consists of a rotor and stator. In this case, the rotor is a permanent magnet, and the stator is made up of electromagnets (field poles). The rotor will move (or step) to align itself with an energized field magnet. If the field magnets are energized one after the other around the circle, the rotor can be made to move in a complete circle.
Fig. 3.10 A PM 90 stepper motor.
Stepper motors are particularly useful in control applications because the controller can know the exact position of the motor shaft without the need of position sensors. This is done by simply counting the number of steps taken from a known reference position. Step size is determined by the number of rotor and stator poles, and there is no cumulative error (the angle error does not increase, regardless of the number of steps taken). In fact, most stepper motor systems operate open-loop that is, the controller sends the motor a determined number of step commands and assumes the motor goes to the right place. A common example is the positioning of the read/write head in a floppy disk drive. Steppers have inherently low velocity and therefore are frequently used without gear reductions. A typical unit driven at 500 pulses/second rotates at only 150 rpm. Stepper motors can easily be controlled to turn at 1 rpm or less with complete accuracy.
45
Excitation Modes for Permanent Magnet Stepper Motors

Bi-polar Stepper Motors
Stepper motors come with a variety of winding and rotor combinations. Those in common use can be divided into general classes, Unipolar driven motors and Bipolar driven motors. In the past Unipolar motors were common and preferred for their simple drive configurations. However, with the advent of cost effective integrated drivers, bipolar motors are now more common. These bipolar motors typically produce a higher torque in a given form factor.
Two-Phase (Bipolar) Stepper Motors
The two-phase (bipolar) stepper motor has only two circuits but actually consists of four field poles. Figure 3.11(a) shows the motor symbol, and Figure 3.11(b) shows how it is wired internally. In Figure 3.11(b), circuit AB consists of two opposing poles such that when voltage is applied (+A B), the top pole will present a north end to the rotor and the bottom pole will present a south end. The rotor would tend to align itself vertically (position 1) with its south pole up (because, of course, opposite magnetic poles attract).
Fig. 3.11 A two-phase (bipolar) stepper motor.
The simplest way to step this motor is to alternately energize either AB or CD in such a way as to pull the rotor from pole to pole. The term bipolar applies to this motor because the current is sometimes reversed. The voltage sequence needed to rotate the motor one full turn is shown below. Reading from top to bottom gives the sequence for turning CCW, reading from bottom up gives the CW sequence: Circuit Position A+ B1 C+ D2 AB+ 3 CD+ 4
46
Another way to operate the two-phase stepper is to energize both circuits at the same time. In this mode, the rotor is attracted to two adjacent poles and assumes a position in between. Figure 3.12(a) shows the four possible rotor positions. The excitation sequence for stepping in this dual mode is as follows:
Fig. 3.12 .Additional operating modes for stepper motors.
Having two circuits on at the same time produces considerably more torque than the single-excitation mode; however, more current is consumed, and the controller is more complex. Still, because it produces the greatest power-to-weight ratio, this bipolar, dual excitation mode is quite common. Both methods produce four-step drives, that is, four steps per cycle. By alternating the single- and dual-excitation modes, the motor can be directed to take half-steps, as shown in Figure 3.12(b). Positions 1, 2, 3, and 4 are from the single-excitation mode, and positions 1, 2, 3, and 4 are from the dual-excitation mode. When driven this way, the motor takes eight steps per revolution and is called an eight-step drive.
47
Four-Phase (Unipolar) Stepper Motors The four-phase (unipolar) is the most common type of stepper motor (Figure 3.13). The term four-phase is used because the motor has four field coils that can be energized independently, and the term unipolar is applied because the current always travels in the same direction through the coils.
Fig.3.13 .A four-phase (Unipolar) stepper motor.
The simplest way to operate the four-phase stepper motor is to energize one phase at a time in sequence (known as wave drive). To rotate CW, the following excitation sequence is used (for Figure 3.13a): A B A C D C E F E G H G Compared with the two-phase bipolar motor, the four-phase motor has the advantage of simplicity. The control circuit of the four-phase motor simply switches the poles on and off in sequence; it does not have to reverse the polarity of the field coils.(However, the two-phase motor produces more torque because it is pushing and pulling at the same time.) The torque of a four-phase stepper motor can be increased if two adjacent coils are energized at the same time, causing the rotor to align itself between the field poles [similar to that shown in Figure 3.12(a)]. Although twice the input energy is required, the motor torque is increased by about 40%, and the response rate is increased. By alternating single- and dualexcitation modes, the motor steps in half-steps, as shown in Figure 3.12(b). else by connecting common to ground
48
3.2.2 OTF-655 STEPPER MOTOR The stepper motor of OTF-655 is an Bipolar one. Its excitation sequence as given by the manufacturer is shown at table 3.1 below: OTF-655 motor Excitation Sequence Pin. No Pin Name 1 2 3 4 1 2 3 4 A(-) B(+) B(-) A(+) + + + + + + + +
Table 3.1 OTF-655 stepper motor excitation sequence
The +,- symbols in the table represent the polarity needed at each step for each specific coil. From that sequence is clear that two circuits (coil) at a time are energized, directly positioning the rotor in between of that coils. This stepping provides the accuracy provided by the manufacturer (Santec). The wiring diagram is as the one provided in figure 3.11
49
CHAPTER 4. OTF-655 DRIVING CIRCUIT DESIGN

4.1 The Driving Circuit Goals Before starting creating a driving circuit for such a component one must clearly set its operational goals. The following operational goals were set for the circuit designer. OPERATION Goals 1. Tuning ability of the filter with its maximum resolution. 2. The operator must be able to perform fast manual tuning throughout its operational range. 3. The circuit must be able to perform automatic tuning steps of 0.2nm or 0.8nm allowing WDM or DWDM channel slection. 4. OTFs safety feature is required to be utilized in order to protect the device from damage. From the operational demands the following list of circuit design quests and answer derives: DESIGN Goals - The circuit should be low cost and easy to assemble. A: This means that high integration was desirable. - The logic should be reprogrammable in order to support feature additions, logic corrections , various experiments on the filter operation parameters and future development. A: This is clearly one the case where a programmable microcontroller is used. - The motor requires low current and low voltage driving that is bellow the usual low power stepper motor specifications. A: A special IC H-bridge motor driver that finds use mainly in robotics applications had to be used. - Monitoring of potentiometer wiper should take place during operation for constant protection of filter from going beyond its operation limits. A: An IC Quad voltage comparator responsible for all circuit safety features.
50
4.2 Circuit Analysis
Fig. 4.1 The Driving Circuit Block Diagram. It consists of five main parts.
At figure 4.1 the block Diagram of the circuit is presented. It consist of 5 main elements which will be analyzed next: 4.2.1 ATMEL AVR ATtiny2313 [20] The heart of the circuit receiving all input signal and driving the filters stepper motor is the AVR ATtiny2313. This tiny 20pin ICs is a powerful microcontroller. It has two 8bit programmable ports (B and D). Port B is used as the MCs output. The four lower bit are used for driving the Filters stepper motor. The upper four are used for driving a 4 LED display which is there as an indicating the various operation modes. Port D is the input of the MC. The 5 operator buttons occupy five inputs of that port and are constantly monitored (see circuit schematic at figure 4.9). As it is clear from the goals set above the need of a microcontroller was a necessity, as with it the one can integrate in a single chip both the driving logic of a stepper motor as well as the operational interface. Moreover since it is a programmable device any change or adjustment can be made during the development phase by only changing the program inside avoiding the otherwise need for redesigning the circuit if district components were used. The microcontroller used is one of ATMELs 8bit RISC family. More specifically the model ATtiny2313 is used. ATtiny2313 OVERVIEW The ATtiny2313 is a low-power CMOS 8-bit microcontroller based on the AVR enhanced RISC architecture. By executing
51
powerful instructions in a single clock cycle, the ATtiny2313 achieves throughputs approaching 1 MIPS per MHz allowing the system designer to optimize power consumption versus processing speed. The AVR core combines a rich instruction set with 32 general purpose working registers. All the 32 registers are directly connected to the Arithmetic Logic Unit (ALU), allowing two independent registers to be accessed in one single instruction executed in one clock cycle. The resulting architecture is more code efficient while achieving throughputs up to ten times faster than conventional CISC microcontrollers. The ATtiny2313 provides the following features: 2K bytes of InSystem Programmable Flash, 128 bytes EEPROM, 128 bytes SRAM, 18 general purpose I/O lines, 32 general purpose working registers, a single-wire Interface for On-chip Debugging, two flexible Timer/Counters with compare modes, internal and external interrupts, a serial programmable USART, Universal Serial Interface with Start Condition Detector, a programmable Watchdog Timer with internal Oscillator, and three software selectable power saving modes. 4.2.2 Stepper Motor Driver (L293D)[21] Next is the L293D IC which is an H-Bridge Driver. [22] An H-Bridge Driver is an electronic circuit which enables DC electric motors to be run forwards or backwards or more specifically change the direction of current on a load such the coil of a motor. These circuits are often used in robotics. H-bridges are available as integrated circuits, or can be built from separate components. The term "H-bridge" is derived from the typical graphical representation of such a circuit. An H-bridge is built with four switches (solid-state or mechanical). When the switches S1 and S4 (see figure 4.2) are closed (and S2 and S3 are open) a positive voltage will be applied across the motor. By opening S1 and S4 switches and closing S2 and S3 switches, this voltage is reversed, allowing reverse operation of the motor.
Fig.4.2a H-Bridge Driver Topology
52
Fig.4.2b H-Bridge Driver. (Operation Basis)
Using the nomenclature above, the switches S1 and S2 should never be closed at the same time, as this would cause a short circuit on the input voltage source. The same applies to the switches S3 and S4. This condition is known as shoot-through. In Figure 4.2b the proper H-Brige operation is shown.
Fig.4.3 H-Bridge Driver Relay Circuit Operation Example
Semiconductor H-Bridges We can better control our motor by using transistors or Field Effect Transistors (FETs)*.These solid state circuits provide power and ground connections to the motor, as a relay circuit would (figure 4.3). The high side drivers need to be current "sources" which is what PNP transistors and P-channel FETs are good at. The low side drivers need to be current "sinks" which is what NPN transistors and N-channel FETs are good at (see figure 4.4).
Fig. 4.4 Semiconductor H-Brighe Operation

*
The field-effect transistor (FET) is a transistor that relies on an electric field to control the shape and hence the conductivity of a 'channel' in a semiconductor material.
53
Mosfets are much more efficient, they can provide much more current and not get as hot. They usually have the flyback diodes built in so you don't need special protection. This helps guard against flyback voltage frying your MCU. The L293D[23] is a quadruple high-current half-H drivers. It is designed to provide bidirectional drive currents of up to 600-mA at voltages from 4.5 V to 36 V. The device is designed to drive inductive loads such as relays, solenoids, dc and bipolar stepping motors, as well as other high-current/high-voltage loads in positivesupply applications. It has two supplies. One for the logic part of it and another for providing the needed voltage and current for driving the motor. All inputs are TTL compatible. Each output is a complete totempole drive circuit, with a Darlington transistor sink and a pseudoDarlington** source. At figure 4.5 the application info on how a Bi-polar motor is operated with L293D is displayed. Drivers are enabled in pairs, with drivers 1 and 2 enabled by 1,2EN and drivers 3 and 4 enabled by 3,4EN. When an enable input is high, the associated drivers are enabled, and their outputs are active and in phase with their inputs. When the enable input is low, those drivers are disabled, and their outputs are off and in the high-impedance state. With the proper data inputs, each pair of drivers forms a full-H (or bridge) reversible drive suitable for solenoid or motor applications.
The metal-oxide-semiconductor field-effect transistor (MOSFET, MOS-FET, or MOS FET), is by far the most common field-effect transistor in both digital and analog circuits. Special diodes placed to protect sensitive components like ICs from reverse currents cause by coils usually. The transistor-transistor-logic (TTL) family was developed in the use of transistor switches for logical operations and defines the binary values as 0 V to 0.8 V = logic 0 and 2 V to 5 V = logic 1. ** Darlington transistor is a semiconductor device which combines two bipolar transistors in tandem (often called a "Darlington pair") in a single device so that the current amplified by the first is amplified further by the second transistor.
54
Fig. 4.5 Two-Phase Motor Driver (L293D) Application Info [23]
4.2.3 The Voltage Comparator (LM339N) [24,25]. Comparator Operation in General Voltage comparator is an OP-AMP circuit which provides voltage or no voltage at its output by comparing the voltages at its inverting and non-inverting inputs. The following drawing (figure 4.6) shows the two simplest configurations for voltage comparators. The diagrams below the circuits give the output results in a graphical form. For these circuits the REFERENCE voltage is fixed at one-half of the supply voltage while the INPUT voltage is variable from zero to the supply voltage. In theory the REFERENCE and INPUT voltages can be anywhere between zero and the supply voltage but there are practical limitations on the actual range depending on the particular device used.
An operational amplifier, usually referred to as an 'op-amp' for brevity, is a DC-coupled highgain electronic voltage amplifier with differential inputs and, usually, a single output.
55
Fig.4.6 Basic Operation of Voltage Comparators.
The LM339N Operation This ICs is there for the safety of the Filters operation. It does that by continuously monitoring the output voltage of the potentiometer wiper which is directly connected to the filters main element. The voltage from the wiper has some high and low limits (25% to 75% of its supply voltage) directly connected with the high and low limits of the movement of the linear filter element. Exceeding these limits can possibly cause destruction of the OTF as the glass element will crash on the module internals. Whenever these limits are hit the LM339D is responsible for enabling an optical indication of which limit is hit and sending a reset signal to the microcontroller immediately stopping its operation and thus the stepper motor movement. Any operation at that point is prohibited. The alarm remains until filter is back to normal operation with the help of a bypass button.
56
LM339N Device Description. The device consists of four independent precision voltage comparators. It is designed specifically to operate from a single power supply and over a wide range of voltages.
Fig. 4.7 LM339N IC pinout.
Two of the four comparators of the IC are used in a circuit fashion as that of figure 4.8 below , where the two operators combine two of their inputs to be used as the common input of the to be compared signal. The other inputs are connected to the Reference voltage sources having the proper operation limits value. The transistor 2N2222 is used in saturation mode (ON/OFF switch) to protect the IC from the Load(Lamp) current. In the OTFs circuit the supply voltage is 6Volts.
Fig. 4.8 Limit Comparator Operation of LM339N.
The other two comparators of the IC are used to drive the limit indication Leds of the circuit .
57
4.2.4 The Power Supply Units. The whole circuit is supplied with a voltage of 9 to 12V. However all the components need supplies of 6Volts and 5Volts. Generally as it obvious from the circuit schematic diagram (figure 4.9) there are 4 supply units ,which belong in the of LM78xx26 IC voltage regulators family, and each one is responsible a part of the circuit. IC Voltage regulators are components which take as an input a DC voltage which has many variations at its amplitude, giving out a completely stabilized output voltage and proper current for our application. The four ICs (figure 4.9) are: - IC7 is an 6V low power(0.2A) voltage regulator which is responsible for the supply of the comparator as well as for the voltage limits references of the circuit in order for the comparator to compare the potentiometer output voltage. Also limit indicator leds are supplied from here. - IC2 is a 5V high power(1.5A) voltage regulator which is responsible for the supply of the MC and the logic circuit part of L293D. - IC5 is a 5V low power voltage regulator which is responsible for powering the potentiometer of the OTF-655 with a constant stabilized voltage so that the operation limits can always be monitored accurately. Also another reason is the possible need of measuring constantly the output voltage of potentiometer wiper with a voltmeter and assigning voltage values to corresponding bandpass frequencies. - IC6 is a 5V high power voltage regulator which is responsible for the supply of the L293D power Mosfets part and as a secondary job the supply of the indicator Leds. 4.2.5 SANTEC OTF-655. The last main component is the OTF-655 Filter which has already been analyzed in Chapter 3.
58
4.2.6 The Circuit Electronic Schematic Diagram
Fig. 4.9 Electronic Schematic Diagram. The main Circuit parts are highlighted. Notice the simplicity of the circuit. The use of smart ICs limits the complexity and the number of components needed which in other case would be at least triple.
4.3 MICROCONTROLLER PROGRAM DESCRIPTION
Fig. 4.10 Microcontroller Software Block Diagram
In figure 4.10 the code structure of the firmware loaded in microcontrollers memory is presented.
60
To start with the stepping sequence table of the microcontroller is stored in the EEPROM of the microcontroller. The software reads from there and makes the proper output to the motor. The stepping table is read top to bottom or reverse depending the desired rotation direction of the motor. The stepping speed is determined by a software clock routine which is a divider of the hardware clock of the MC(set at 8MHz). By altering a variable in this the software can determine how fast the motor will move. BLOCK DIAGRAM WALKTHROUGH When Circuit is powered on fistly the program loads all need initial values and parameters and then it falls in waiting state for some event to occur. As already told the MC waits instructions from the 5 ports-buttons available to the operator. The five operator inputs are: 1. UP 2. DOWN 3. SINGLE STEP (maximum resolution mode) 4. MULTI STEP 1 (0.2nm automatic filter variation mode) 5. MULTI STEP 2 (0.8nm automatic filter variation mode). When circuit initializes the default mode is that of the maximum resolution. The operator can make use of UP and DOWN button to command the software to make the filter tune at a higher or lower frequency. Single quick press of one of these two buttons will perform the command in the mode selected by mode select inputs. If Single Step Mode is selected and operator keeps UP or DOWN buttons pressed then stepping speed variable gets a very small value allowing the filter to move to a higher or lower frequency very fast until the button is released. If one of the Multistep modes is then only single press is allowed to UP and DOWN inputs. Each press then commands the automatic change of filter frequency in the preset values of 0.2 or 0,8nm. After performing each tuning process the software returns to waiting state for the next operation command. Filter - Software calibration. In order for each filter to tune accurately each circuit filter pair must go under the process of calibration. That means that it must be found how much each motor step changes the center frequency of the filter. This is critical for accurate tuning of filter since automatic feature of 0.2nm and 0.8nm directly depend on how many motor steps must be done in order to achieve these values. Even in single step mode operator must know what is the variation of center frequency for each button press. The modulation of center frequency for each motor step varies for each module , so each
61
complete circuit this process must take place. The process is displayed in Chapter 5 where filter measurements and circuit operation is validated. Security Features The security feature is a hardware part of the circuit. A special part of the circuit involving the Voltage Comparator IC is responsible for setting the reset port of the Microcontroller if the operation limits of the filter are reached. The Reset remains set until the limit warning is reset and prohibits any operation. The operator is responsible for pressing bypass limit control button at board setting the filter in a normal position and resetting the security feature. The complete code in assembly language can be seen commented in Appendix A. 4.4 The PCB At figure 4.11 you can see the final product completely assembled and operational.
Fig. 4.11 The OTF-655 Driving Circuit Board. Additionally to the main parts of the circuit described before the board features the following: 1. OTF-655 is attached on board making the PCB a single module by itself. 2. It has Power ON/OFF button, Limit Control Bypass Button, and motor Shut Down Button.
62
3. Wire connections for DC power protected from reverse and high voltage and for Voltometer connection for monitoring the Filter Potentiometer Wiper voltage value. 4. Several free pad connections around the microcontroller for attaching signal wires for future additions. 5. A 4 Led indicator providing indication of the Circuit operation modes and filter alteration frequency direction. 6. 3 more Leds for High or Low limit indicator and normal operation of the safety circuit. 7. Pad connector for future utilization of Temperature sensor existing in the OTF-655 module.
63
CHAPTER 5. FILTER MEASUREMENTS

In order to validate the circuit operation and filter performance a series of measurements took place after the creation of the final product. 5.1 INPUT-OUTPUT FULL OPERATION RANGE CHARACTERISTICS COMPARISON The first measurement that took place is the one that gives us the response graph of the filter compared to a broadband input signal throughout its operational range. For that reason the white noise source was chosen which is provided by an ETFA amplifier. An ETFA amplifier can amplify signals with an almost linear response at a range of 40nm which is enough to cover the almost 35nm of beneficial range of OTF-655. The response graph is displayed on figure 5.1. With blue color is the ETFA white noise response (filter input) and with red the filter output responce.
Fig. 5.1 OTF-655 Operation Range Response. Input - Output characteristics comparison.
The filter presents a quite linear behaviour presenting a response which follows exactly the input signal. This is obvious also because the loss (yellow colour in graph) throughout its operational range has a value of 4.5dBm.
64
5.2 FILTER SHAPE OVER OPERATION RANGE The next set of measurments was to prove that the OTF retains a constant behaviour over its operation range for that reason Filter shape characteristic graphs were taken at 1540nm , 1550nm and 1560nm. Then a comparison of the three graphs takes place. The source used is again an ETFA white noise output. 5.2.1 Filter Shape at 1540nm The shape of the filter at 1540nm center frequency is displayed at figure 5.2 and the attenuation over the bandwidth round the center frequency graph is displayed at figure 5.3.
Fig. 5.2 Filter shape graph at center frequency of 1540nm
Fig. 5.3 Filter attenuation compared to bandwidth around center frequency of 1540nm
65
The values of filter bandwidth measured at the important attenuation points of -3,-6,-9,-20dBm compared to the peak power value which is at the center frequency of filter are displayed at table 5.1 BW(nm) 0,872 0,512 0,416 0,28 Attenuation to peak(dBm) -20 -9 -6 -3
Table 5.1 Bandwidth values at 1540nm at characteristic attenuation points.
As it can be seen the filter presents fast falling side lobes and the bandwidth of filter at -20dBm is quite small (less than 0.9nm) making possible the use of filter for DWDM channel separation (see par. 5.6).
66
5.2.2 Filter Shape at 1550nm The shape of the filter at 1550nm center frequency is displayed at figure 5.4 and the attenuation over the bandwidth round the center frequency graph is displayed at figure 5.5.
67
It is obvious from graphs(figure 5.4, 5.5) and values measured (table 5.2) the filter presents the same shape behavior as in 1540nm being still capable for WDM and DWDM separation.
68
5.2.3 Filter Shape at 1560nm The shape of the filter at 1560nm center frequency is displayed at figure 5.6 and the attenuation over the bandwidth round the center frequency graph is displayed at figure 5.7.
69
It is obvious from graphs (figure 5.6, 5.7) and values measured (table 5.3) the filter presents the same shape behavior as in 1540nm being still capable for WDM and DWDM separation.
70
5.2.4 COMPARISON As a final step to this measurement all three attenuation graphs are compared (overlaid) in order to see if they fall on each other (figure 5.8). That way it is proved that the filter presents a constant shape behavior both at the middle and the edges of its operational range.
Fig. 5.8 Comparison of filter attenuation to bandwidth graphs around center frequency of 1540,1550 and 1560nm.
The measurements for the three measured points is displayed at table 5.4 and as it obvious the values are very close to its points. Attenuation to peak(dBm) -20 -9 -6 -3 0 BW(1540nm) 0,872 0,512 0,416 0,28 0 BW(1550nm) 0,888 0,528 0,424 0,296 0 BW(1560nm) 0,888 0,528 0,424 0,296 0
Table 5.4 Bandwidth values comparison of 1540, 1550 and 1560nm at characteristic attenuation points.
So at -3dBm the filter has a bandwidth of 0.3nm. At -6dBm has a BW value of 0.42nm, which is increased only to 0.5nm at -9dB and 0.88nm at -20dBm. From these values, it is clear that the filter shape has fast falling side lobes, which is good for prevention of crosstalk phenomenon between channels transmitted.
71
5.3 WDM and DWDM filter Stepping The purpose of this measurement is to demonstrate the ability of the filter - driving circuit combination to perform automatic and accurate WDM and DWDM channel selection, which means that it can perform accurate 0.8nm and 0.4nm jumps between channel. This step is also important for the calibration of the circuit because of that measurement it possible to calibrate the software parameter that has to do with the relation of step motor movement and filter center frequency alteration. That is how much the center frequency of each filter module is altered for each stepper motor step. This varies slightly for each OTF-655 module because of mechanical and physical characteristics of filter material and sliding mechanism. Since the stepper motor does not have cumulative error at its rotation it allows the calculation of the maximum resolution value and the proper setup of each circuit, by finding how many steps the motor must do to perform a jump to the next channel.
5.3.1 WDM (0.8nm) STEPPING In figure 5.9 a single 0.8nm filter jump is presented. To achieve that stepper motor must perform 57steps. Dividing the 0.8nm spectral width to 57 steps, it comes out that the maximum filter resolution this specific OTF655 module can achieve is 0.014nm. This value with small variations was apparent at other OTF modules also tested.
Fig. 5.9 Circuit Performing a 0.8nm automatic tuning (Agilent Analyzer Screen Capture)
72
In order to validate the calibrated now circuit a series of 0.8nm jumps were performed using the corresponding auto mode. In order to improve the credibility of this experiment the initial center frequency was set to be with a value set by ITU-grid specification about WDM channels. The table on the right of figure 5.10 has for each color the ITU grid value in nanometers. On the graph itself on top of most of shape graphs representing the channel selected each time, there is the measured value.
Fig. 5.10 Circuit Performing multiple WDM automatic tuning steps
The following table (table 5.6) presents comparisons between ITU Grid Channels and the corresponding Measured values achieved: ITU Grid WDM channel Measured value frequency 1551.72 1551.71 1552.52 1552.52 1554.13 15554.12 1554.94 1554.9 1556.52 1556.52 1557.36 1557.3
Table 5.6 ITU Grid Measured value comparison
As it is obvious the circuit with proper calibration for the filter follows the ITU Grid specifications. Measured center frequency value at each case is very close to the ITU Grid for the specific channel.
73
5.3.2 DWDM (2x0.2nm) STEPPING In figure 5.11 a double 0.2nm filter jump is presented. To achieve a 0.2nm jump stepper motor must perform 14steps. Dividing the 0.2nm spectral width to 14 steps, it comes out that the maximum filter resolution this specific OTF-655 module can achieve is 0.014nm, same as the one calculated at previous experiment. This is an indication of accurate and repeatable the tuning performance of the OTF-655 is.
Fig. 5.11 Circuit Performing a 2x0.2nm automatic tuning (Agilent Analyzer Screen Capture)
In order to validate the calibrated now circuit a series of 0.4nm (2x0.2nm) jumps were performed using the corresponding auto mode (figure 5.12). Again the filter perform continuously accurate 0.4nm jumps following the ITU specs for DWDM applications. The measured values are displayed on top of each filter shape graph.
74
Fig. 5.12 OTF-655 0.4(2x0.2)nm MULTI-STEP FILTER OPERATION DWDM CHANNELS AUTOMATIC TUNING
75
5.4 POLARIZATION LOSS MEASUREMENT. The purpose of this measurement is to validate the low PDL value the manufacture gives for linear sliding technology. In order to do this experiment we used a white noise source and a polarizer (figure 5.13) between the source and the input of the filter. The filter was tuned at various frequencies within its linear operational range.
Fig. 5.13 A 45 TFP Plate Polarizer example
At each wavelength the polarizer was tuned manually in order to change the polarization of input signal and the peak value of output of the filter was constantly monitored at the Agilent Analyzer. The output amplitude variation during the change of polarization was so small that was not possible for the analyzer to have a clear measurement. Since to do that measurement the lowest scale of instrument was used (0.1db/div), it is certain that PDL of OTF-655 of AIT Lab is less than 0.1dB. That is a fact for the entire tuning range of filter.
76
5.5 Channel Separation at 100GHz Spacing (WDM) The purpose of this measurement is to see how well is the channel separation achieved by the OTF-655 when neighboring channels are multiplexed in its input and only one of them is desired at its output. So for this experiment 5 neighboring WDM (100GHz spacing) channel sources are multiplexed and inserter at filters input. The middle channel is at 1552.52nm. All five sources are equalized to have the same peak power level. The input channels are depicted with red color on figure 5.14.
Fig. 5.14 100Ghz spacing channel separation by OTF-655.
The filter band pass frequency is set at 1552.52nm. The filter output is depicted with the blue color while the shape of the filter is with green color (figure 5.14). The difference of immediate neighboring channels to the center channel at the measured at the filter output is almost -30dBm. Two channels away from center frequency the output power is insignificant. Since the neighboring channels center to center amplitude difference at filter output is greater than the theoretical crosstalk limit of -20dBm the OTF-655 guaranties the WDM channel separation.
77
5.6 Channel Separation at 50GHz Spacing (DWDM) This measurement is the continuation of the one at paragraph 5.5. This time five neighboring DWDM (50GHz spacing) channel sources are multiplexed and inserter at filters input. The middle channel is at 1552.495nm. All five sources are equalized to have the same peak power level. The input channels are depicted with red color on figure 5.15.
Fig. 5.15 50Ghz spacing channel separation by OTF-655.
The filter band pass frequency is set at 1552.495nm. The filter output is depicted with the blue color while the shape of the filter is with green color (figure 5.15). The difference of power between the peak amplitude of immediate neighboring channels to the center channel at the measured at the filter output is almost -18dBm. Two channels away from center frequency the output power is insignificant (more than -30dBm). Since the neighboring channels center-to-center amplitude difference at filter output is quite close to the theoretical crosstalk limit of -20dBm, the OTF-655 guaranties the DWDM channel separation under laboratory environment.
78
5.7 Measurements Conclusions The filter specifications as those have derived from the measurements that took place up to now are concentrated in table 5.7 Parameter Wavelength Tuning Range Filter Bandwidth Filter Bandwidth Filter Bandwidth Filter Bandwidth Tuning Resolution PDL Units nm nm nm nm nm nm dB Value up to 35 0,888 OR LESS 0,528 0,424 0,296 0.014 <0.1 -4.5dB OR LESS Notes C Band Full Width@-20dB (Good for DWDM Applications) Full Width@-9 dB Full Width@-6 dB Full Width@-3 dB Exceeding Manufacturers Expectations. Filter Calibration value. Practically not measurable With Connectors! Almost stable throughout the tuning range.
Insertion Loss
dB
Table 5.7 OTF-655 Characteristic values derived from Laboratory measurements.
The specific family of OTF-655 filters that were used for these measurements are capable of covering the whole C-band having a tuning range up to 35nm. However, the filter has a completely linear behavior within a spectral range of 30nm at center of the tuning range. Exceeding this 30nm range a small loss occurs up to 35nm while much higher loss values are observed outside of the operational range. The filter presents fast falling sidelobes giving a very narrow filter shape characteristic. Passband skirts should be sharp to reduce the amount of energy passed through from adjacent channels. This energy is seen as crosstalk and degrades the system performance. The lower the crosstalk point between two adjacent channels the lower the BER. The 3db bandwidth is very narrow having a width of less than 0.3nm, which is increased to only 0.4nm at -6dB and -0.5nm at -9dB. The most important -20dB bandwidth has a very satisfying value of 0.88nm. All these bandwidths within the limits given by the manufacturer (maximum: 0.4nm at -3dB and 1.2nm at -20dB ) guarantying the ability of
79
the filter to select through a wavelength signal through a multiplexed signal of WDM or DWDM channels. The tuning resolution value is only 0.014nm revealing a well made tuning mechanism and filter elements construction and surely exceeds even Santecs expectations (max value: 0.035nm). Moreover the driving circuit filter combination presented an accurate tuning mechanism, with no cumulative errors making it easy to use the automatic tuning modes of the circuit. According to the manufacturer allegations was the PDL loss should be insignificant (max 0.2dB). This was completely certified by the PDL measurement done here at this thesis, where the value was less than 0.1dB (practically not measurable). Finally the insertion loss of the filter never exceeded the value of 4.5dB including the losses because of connectors (<0.5dB). The manufacturer had specified a value of 6dB excluding connectors. This time the filter not only is within the limits but performs almost ideally for its kind of technology. Generally the filter exceeds the manufacturers expectations in every characteristic. Also the filter can perform very accurate and repeatable wavelength selection. The bandwidth of filter at -3,-6,-9 and -20dB reveals the ability of the filter to easily discriminate neighboring ITU Grid channels with 100GHz or 50GHz spacing.
80
CONCLUSION
Optical filters are important components in optical communication systems, primarily for the rejection of the accumulated amplification noise in optical transmission systems. However, over the last years the development of tunable optical filters introduced another significant application in optical networking and specifically in WDM switching nodes. The control mechanism to perform this operation is an important parameter and is the main objective of this research topic. The initial purpose of this thesis was the design and development of a control board able to operate a tunable optical filter module (Santec OTF655). The OTF-655 is a compact tunable filter device that incorporates a stepper motor and utilizes independent linear sliding technology that claims to have very low PDL losses, to achieve accurate wavelength tuning while maintaining constant optical properties over the entire wavelength range. The circuit design presented various difficulties that mostly had to do with the type of stepper motor used in the module (Bipolar) as well as its special and rare characteristics (low voltage, low current motor). The heart of the circuit is an ATMEL 8bit microcontroller that performs a series of application that combine operation logic and stepper motor driving in one IC instead of two that normally are used for this kind of applications. The circuit provides operational features like tuning with maximum filter resolution, automatic filter tuning at 0.4nm and 0.8nm, device protection, (using the potentiometer provided by the OTF-655 module) and finally allows future expandability, since its a fully reprogrammable device and has free connection ports in design. The series of measurements for the evaluation of the systems operational characteristics (i.e. the driving circuit and the filters performance characteristics) were adequately performed. The filter matched and in many cases exceeded the manufacturers specifications. Accurate and repeatable filter tuning was achieved through the control circuit. The filters narrow bandwidth characteristics with the fast falling side lobes provides assurance that the designed system can easily discriminate neighboring ITU Grid channels with 100GHz or 50GHz spacing.
81
APPENDIX A. MICROCONTROLER ASSEMBLY CODE.

;*************************************************************************** ;* FILTERDRIVER.ASM: FULL STEP MOTOR DRIVER. ;* PROGRAMMER: GEORGE MENTZIKOF (gmen@ait.edu.gr) ;* FOR Santec OTF-655 S/N: 02070002 . ;* ;* Version :4.0 ;* Target MCU :1200,2313 or later ;* 31-10-2006. ;* 1) Corrected step routines. Now back forth back flawlesly! ;* with use of direction flag plus control of special cases! ;* ;* 2) Decrease delay if key is kept pressed down and return to initial ;* delay if released. ;* 3) Now Performs 1 step (0.014nm) ;* or 14 steps (0.196nm step) per button press. ;* 4) Perform 57 steps (0.798nm step)button ;* Changed button for 1,14,57 steps to input PD4 to PD5 ;* and PD6 in order to leave interupt inputs free ;* for future use! ;* ;* 5) External Comparator Triggers Reset ;***************************************************************************
.include "tn2313def.inc"
rjmp
RESET
;reset handle
;* ;* ;* ;* ;* ;*
longDelay Do a delay based on: 2 + temp * 256 * 3 * 256 + temp * 256 * 3 + temp * 3 + 4 = temp * 197379 + 6 clock cycles. At a clock frequency of 4MHz, this becomes about temp * 49.3ms
;* Register variables
.def .def .def .def .def .def .def .def .def .def
T1 T2 temp bitCount safecount direction speed speedflag stepsize temp2
= = = = =
r1 r2 r19 r20 r21
= r30 = r29 = r28 = r27 = r26
;* Register variables
;* Defines
82
;* Code software clock-delay routine
longDelay: clr clr delay_1: brne dec brne dec brne ldi out sbis ser sbis ser ret
T1 T2 dec T2 delay_1 T1 delay_1 speed delay_1 temp, 0xFF PORTB, temp PIND,0x00 speedflag PIND,0x01 speedflag
;T1 used as delay 2nd count ;T2 used as delay 3d count
; ; ; ; ; ; ;
;delay speed variable delay master count load temp with all 11 (off output) load temp to PortB (temporary off for testing) if at end of delay button 0 is still pressed then set the speedflag. if at end of delay button 1 is still pressed then set the speedflag
;* Read data from EEPROM routine walk: out sbi in out rcall ret
EEAR, bitCount EECR, 0 temp, EEDR PORTB, temp longDelay
;bitCount sets eeprom address ;set eeprom read strobe ;get data from eeprom ;display the data for a sec
;* Main program ;* Code RESET: ldi temp,RAMEND out SPL,temp ser temp out DDRB, temp ldi bitCount, 3 ldi safecount, 10 ldi direction ,0 clr speedflag ldi stepsize ,1 main: cpi speedflag, 0xFF breq fastmove ldi speed,15 mainnormal: sbis PIND,0x00 rcall fwd sbis PIND,0x01 rcall bkwd sbis PIND,0x04 rjmp Gap0 sbis PIND,0x05 rjmp Gap10 sbis PIND,0x06 rjmp Gap20 rjmp main fastmove: clr speedflag sbis PIND,0x00 ldi speed, 1 sbis PIND,0x01 ldi speed, 1 rjmp mainnormal
; Set stackptr to ram end
;all outputs ;initial eeprom position ;10 step for safety (not used) ;direction flag (0=forth, 1=back) ;clear speed flag ;set step size to 1
;Check if speedflag is set ; if yes go to fastmove tag ; else set walk speed to 15! ; MAINNORMAL POLLING OF BUTTONS ; But0 calls fwd routine ; But1 calls bkwd routine ; But4 calls Gap0 routine ; But5 calls Gap10 routine ; But6 calls Gap20 routine
; ; ; ; ;
Clear speedflaf Check if key still pressed then set walk speed to 3 (low delay) Check if key still pressed then set walk speed to 3 (low delay)
83
AIT / MsiTT student (2004): George Mentzikof AIT/ MsITT 2006 THESIS TITLE: Optical tunable filter control circuit design for applications in DWDM optical add/drop nodes. ; SINGLE GAP STEP ROUTINE Gap0: ldi stepsize, 1 ; set stepsize ldi speed, 20 ; call delay to avoid duplicate key press rcall longDelay ldi temp,0b11010000 ; display setting out PORTB,temp ; to port B rjmp main ; go back to main ; MULTIPLE STEP ROUTINES Gap10: ldi stepsize, 0x0E ; ldi speed, 20 ; rcall longDelay ldi temp,0b10010000 ; out PORTB,temp ; rjmp main ; Gap20: ldi stepsize, 0x39 ldi speed, 20 rcall longDelay ldi temp,0b00010000 out PORTB,temp rjmp main
set stepsize (14steps = 0.2nm) call delay to avoid duplicate key press display setting to port B go back to main
; set stepsize (57step = 0.8nm) ; call delay to avoid duplicate key press ; display setting ; to port B ; go back to main
; FWD ROUTINE (calls forth as many times as stepsize register) fwd: cpi stepsize, 1 breq forth mov temp2, stepsize ; copy stepsize to temp ldi speed, 10 ; set delay speed for single step rcall longDelay ; call delay fwd_loop: ldi speed, 3 ; set delay speed for multiple steps rcall forth ; call forth setted steps times dec temp2 ; temp -1 cpi temp2,0 ; if temp !=0 brne fwd_loop ; then do one more step. ldi speed, 10 ; restore delay speed ret
;STEP FORTH ROUTINE forth: cpi direction, 1 breq fwdcorrect rcall walk dec bitCount cpi bitCount, -1 breq clearfwd ret clearfwd: ldi bitCount, 3 ret fwdcorrect: ldi direction, 0 ;special cases for correction cpi bitCount, 0 breq fwdcorrect_sc1 cpi bitCount, 1 breq fwdcorrect_sc2 ; normal correction dec bitCount dec bitCount rjmp forth fwdcorrect_sc1: ldi bitCount, 2 rjmp forth fwdcorrect_sc2: ldi bitCount, 3 rjmp forth
; check if previous step was opposite direction ; if yes go to correction routine
; if bitCount = 0 -> ; walk backward
; reset counter ; Correct bitCount value when changing direction
; ; ; ;
special solving special solving
case 1 routine case 2 routine
84
AIT / MsiTT student (2004): George Mentzikof AIT/ MsITT 2006 THESIS TITLE: Optical tunable filter control circuit design for applications in DWDM optical add/drop nodes. ; BKWD ROUTINE (calls bkwd: cpi stepsize, 1 breq back mov temp2, stepsize ldi speed, 10 rcall longDelay bkwd_loop: ldi speed, 3 rcall back dec temp2 cpi temp2,0 brne bkwd_loop ldi speed, 10 ret forth as many times as stepsize register)
; copy stepsize to temp ; set delay speed for single step ; call delay ; ; ; ; ; ; set delay speed for multiple steps call forth setted steps times temp -1 if temp !=0 then do one more step. restore delay speed
;STEP BACK ROUTINE back: cpi direction, 0 ; check if previous step was opposite direction breq bkwdcorrect ; if yes go to correction routine rcall walk inc bitCount cpi bitCount, 4 ; if bitCount = 0 -> breq clearbck ; walk backward ret clearbck: ldi bitCount, 0 ;reset counter ret bkwdcorrect: ; Correct bitCount value when changing direction ldi direction, 1 ;special cases for correction cpi bitCount, 2 ; special case 1 breq bkwdcorrect_sc1 ; solving routine cpi bitCount, 3 ; special case 2 breq bkwdcorrect_sc2 ; solving routine ; normal correction inc bitCount inc bitCount rjmp back bkwdcorrect_sc1: ldi bitCount, 0 rjmp back bkwdcorrect_sc2: ldi bitCount, 1 rjmp back ; * SAFETY ROUTINE (future feature not used)* safe: rcall back dec safecount cpi safecount,-1 brne safe ldi safecount, 10 rjmp main
;* 4 steps stored in EEPROM ;* Data (steps must stored backwards. eg 3-2-1-0) .ESEG ; (bit7-4step no) table: .db 0b01110110 ;data .db 0b10110101 ;data .db 0b11011001 ;data .db 0b11101010 ;data
(bit3to0 B+B-A+A-) 0 1 2 3 (step (step (step (step 4) 3) 2) 1)
85
References:
1
ITU-T G.709 Network Node Interface For The Optical Transport Network,initial version ,Feb.2001 Dan Sadot and Efraim Boimovich , Ben Gurion University Tunable Optical Filters for Dense WDM Networks IEEE Communications Magazine(December 1998). p.50
Rajiv Ramaswami, Kumar N. Sivarajan Optical Networks: A Practical Perspective 2nd Edition , p.118.
Web site: OER 166 - Wavelength division multiplexing http://www.spie.org/web/oer/october/oct97/multiplex.html

5
Web site: Fiber Bragg gratings http://www.rp-photonics.com/fiber_bragg_gratings.html
Rajiv Ramaswami, Kumar N. Sivarajan Optical Networks: A Practical Perspective 2nd Edition , p.130 . 7 Dan Sadot and Efraim Boimovich , Ben Gurion University Tunable Optical Filters for Dense WDM Networks IEEE Communications Magazine(December 1998). p.51 L. Wooten et al., "Rapidly tunable narrowband wavelength filter using LiNb03 unbalanced Mach-Zehnder interferometers," J. Lightwave Tech.,vol. 14, no. 11, Nov. 1996, pp. 2530-36.
9 8
B. H. Verbeek et al., "Integrated four channel Mach-Zehnder multi/demultiplexer fabricated with phosphorous doped Si02 waveguides on Si," 1. Lightwave Tech., vol. 6, June 1988, pp. 101 1-15.
10
Jaafar M.H. Elmirghani, University of Wales Swansea Hussein T. Mouftah, Queens University Technologies and Architectures for Scalable Dynamic Dense WDM Networks, IEEE Communications Magazine February 2000, p.61 Jaafar M.H. Elmirghani, University of Wales Swansea Hussein T. Mouftah, Queens University Technologies and Architectures for Scalable Dynamic Dense WDM Networks, IEEE Communications Magazine February 2000, p.58
11
12
Rajiv Ramaswami, Kumar N. Sivarajan Optical Networks: A Practical Perspective 2nd Edition , p.144 M. Aziz, et al., A New and Simple Concept of Tunable Two-Chip Microcavities for Filter Applications in WDM Systems, IEEE Photonics Technology Letters, 12( 1 l), pp. 1522-1524, (2000). A. DehC, et al., III-V Compound semiconductor micromachined actuators for long resonator tunable Fabry-Perot detectors, Sensors and Actuators A, 68,pp.365-371(1998)
13
14
15
Jaafar M.H. Elmirghani, University of Wales Swansea Hussein T. Mouftah, Queens University Technologies and Architectures for Scalable Dynamic Dense WDM Networks, IEEE Communications Magazine February 2000, p.61 Jaafar M.H. Elmirghani, University of Wales Swansea Hussein T. Mouftah, Queens University Technologies and Architectures for Scalable Dynamic Dense WDM Networks, IEEE Communications Magazine February 2000, p.60-61 N. Mekada, A. AL-Hamdan, Member, IEEE, T. H. Chong, and D. G. Daut Polarization Independent, Linear-Tuned Interference Filter with Constant Transmission Characteristics Over
16
17
86
15301570-nm Tuning Range , IEEE PHOTONICS TECHNOLOGY LETTERS, VOL. 9, NO. 6, JUNE 1997, 10411135/97$10.00 1997 IEEE.
18
Delmar, Modern Control Technology--Components & Systems (2nd Ed.), CHAPTER 8 Stepper Motors. H. SAX ,APPLICATION NOTE Stepper Motor Driving , SGS Thomson Electronics
19 20
Web Link: ATMEL Products . AT tiny 2313. (http://www.atmel.com/dyn/products/product_card.asp?part_id=3229) Web Link: L293 Datasheet (http://www.st.com/stonline/products/literature/ds/1330/l293d.pdf)
21
Web Site:Written by Jim Brown , Brief H-Bridge Theory of Operation, (http://www.dprg.org/tutorials/1998-04a/ ) , April 1998
23
22
Texas Instruments , Datasheet: L293, L293D Quadraple H-Bridge Driver , revised November 2004. Web site: Voltage Comparators, ( http://home.cogeco.ca/~rpaisley4/Comparators.html)
24 25
ST Electronics , Datasheet: LM139,A - LM239,A - LM339,A LOW POWER QUAD VOLTAGE COMPARATORS , March 2003 Web Link: 78xx Voltage Regulators Datasheet (http://www.st.com/stonline/books/pdf/docs/2143.pdf)
26
87

Mentzikof

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mentzikof

Uploaded by

Copyright:

Available Formats

MsITT THESIS TITLE:

Supervisor: Dr. Dimitrios Klonidis

AIT, December 2006

CHAPTER 1. TUNABLE FILTER APPLICATIONS

Fig.1.1 A simplified WDM system.

Fig.1.2 .A typical point to point optical network

Fig. 1.3 Reconfigurable OADM node

Figure 1.4 displays the numerous architectural choices OADM provides.

Fig 1.4. OADM architecture choices.

Fig 1.5. Basic OXC architecture.

Fig 1.6. Interleaving/banding techniques for finer filter spacing.

1.8 Why Reconfigurability Is Important [3Error! Bookmark not defined.]

Fig 1.7a Simple point to point optical network.

Fig 1.7b Complex Ring topology optical network.

Fig 1.8 Larger domain of transparency (OAs not shown).

CHAPTER 2. TYPES OF FILTER TECHNOLOGIES

Fig. 2.1 Shape Characteristic of a typical Bandpass filter.

Fig 2.4 Arrayed waveguide grating

Fig. 2.6 : Schematic structure of a fiber Bragg grating (FBG).

Fig. 2.7 Optical add/drop elements based on fiber Bragg gratings.

Fig. 2.8 Principle of operation of a Fabry-Perot filter.

This can also be expressed in terms of the optical free-space wavelength . as

Fig. 2.10. A three-cavity thin-film resonant dielectric thin-film filter.

Fig. 2.14. Transfer function of a multistage Mach-Zehnder interferometer.

Fig. 2.15. Tuning a liquid crystal Fabry Perot filter.

2.3.10 Tunable Filters Based on Semiconductor Laser Structures

AWG Tunable Filter Liquid Crystal F-P Electro - Optic

10ms 0.5-10s 1-10ns

Crystal Orientation Electro-Otic

Temperature, stretching Thermo electric, Thermo Optic -

Table 2.1 Filter technologies characteristics overview-comparison [16]

CHAPTER 3. THE SANTEC OTF-655

Fig.3.1 OTF-655 mechanical drawing

Fig. 3.3 The linear sliding technology.

Fig. 3.4 OTF-655 simplified internal design block diagram.

Fig. 3.5 Optical Characteristics of OTF-655 from Santec

Fig. 3.6 Electrical-Operating specifications of the filter.

Fig. 3.7 Pin Connector Description

Fig. 3.8 Stepper Motor Specifications

Fig. 3.9 Potentiometer operational specifications

Fig. 3.10 A PM 90 stepper motor.

Excitation Modes for Permanent Magnet Stepper Motors

Fig. 3.11 A two-phase (bipolar) stepper motor.

Fig. 3.12 .Additional operating modes for stepper motors.

Fig.3.13 .A four-phase (Unipolar) stepper motor.

Table 3.1 OTF-655 stepper motor excitation sequence

CHAPTER 4. OTF-655 DRIVING CIRCUIT DESIGN

4.2 Circuit Analysis

Fig.4.2a H-Bridge Driver Topology

Fig.4.2b H-Bridge Driver. (Operation Basis)

Fig.4.3 H-Bridge Driver Relay Circuit Operation Example

Fig. 4.4 Semiconductor H-Brighe Operation

Fig. 4.5 Two-Phase Motor Driver (L293D) Application Info [23]

Fig.4.6 Basic Operation of Voltage Comparators.

Fig. 4.7 LM339N IC pinout.

Fig. 4.8 Limit Comparator Operation of LM339N.

4.2.6 The Circuit Electronic Schematic Diagram

4.3 MICROCONTROLLER PROGRAM DESCRIPTION

Fig. 4.10 Microcontroller Software Block Diagram

CHAPTER 5. FILTER MEASUREMENTS

Fig. 5.2 Filter shape graph at center frequency of 1540nm

Table 5.1 Bandwidth values at 1540nm at characteristic attenuation points.

Fig. 5.4 Filter shape graph at center frequency of 1550nm

Table 5.2 Bandwidth values at 1550nm at characteristic attenuation points.