This chapter focuses on the receiver section in detail and talks about the various components used in the design of the receiver circuit. The receiver was used to detect the signals that were being sent by the transmitter section. The receiver front end consists generally of a photo-detector and a current to voltage converter. However the front end that was used for the experiment consisted of a collector lens, an optical filter, a photo-detector and an operational amplifier used in a trans-impedance amplifier setup. The rest of the receiver section was composed of a two of operational amplifiers setup that helped in voltage amplification and voltage level setting. The voltage setting section was used to set the output voltage to the proper TTL level output.
The selection of the photo-detector was very crucial to the experiment and was selected so that it would detect signals more efficiently with help from the optical filter. The received signals were then processed and conditioned before being fed to a laptop at the receiver end to receive information that was sent.
The selection the photo-detector, optical filter and operational amplifiers are discussed in this chapter.
5.2 Photo-detector
Photo-detectors are of many types. The different types include photo-resistors; photo-multipliers, photo-transistors etc. But they can be classified into two different categories; devices without internal gain and devices with internal gain. Devices such as p-n photodiode and p-i-n photodiode are those which do not have any internal gain. Whereas the Avalanche photodiode (APD) is a device which has an internal gain, this helps to amplify the detected signal. The experiment used a p-i-n photo-detector rather than an APD device. The different structures of the two devices, the working principles and basis of selection are explained further.
5.2.1 The p-i-n photodiode
The p-i-n photo diode works on one the principles of internal photo effect. As stated in B.A Saleh and Carl's book (1), this is the principle of photoconductivity. By this principle a photoconductor's electrical conductivity depends directly on the amount of light falling on the device. The structure of the p-i-n photodiode is shown in figs 5.1 and 5.2. As shown in the figure the device consists of a lightly doped n- epitaxial layer sandwiched between two heavily doped p and n regions. According to John Gowar(2), the preferred thickness of the depletion layer is in the order of tens of microns. This meant that the epitaxial region should be lightly doped, for the depletion region to 'reach through' to the heavily doped substrate at normal operating voltage (2).
Fig 5.1: Internal structure of the p-i-n diode
Source: "Optical Communication Systems", by John Gowar, Pg 446 (2)
Fig 5.2: Structure of the p-i-n photodiode
Source: "Fundamentals of Photonics", by Bahaa E.A.Saleh and Malvin Carl Teich, Pg 661 (1)
5.2.1.1 Working Principle
The working principle of the p-i-n diode can be explained with the help of the figure 5.3 shown below. When a photon is incident on the surface of an intrinsic semi-conductor, it gets absorbed by the semi-conductor; this results in the generation of a free electron. This electron has enough energy to be excited, and it moves from valance band to conduction band, and a hole is formed in the valance band (1). On application of an electric field the charge carriers start to move, which results in current being produced in the device. A p-i-n diode is generally connected in reverse bias in circuits. This increases the width of the depletion region that is formed in the epitaxial region.
Fig 5.3: Working of p-i-n photodiode
Source: "Fundamentals of Photonics", by Bahaa E.A.Saleh and Malvin Carl Teich, Pg 647 (1)
The p-i-n diode has a number of advantages over the p-n diode (1):
The increase in depletion region width increases the area which is available for capturing incident light.
The increase in width of the depletion region reduces the junction capacitance and so the RC time constant.
Reduction of the ratio between diffusion length and drift length of the device results in more parts of the generated current being carried by the faster drift process.
5.2.1.2 Equivalent circuit
The figure 5.4 shows the equivalent small signal model for a reverse biased photo detector. The photo detector basically acts as a current source while the shunt conductance represents the reverse bias slope characteristic (2). The series resistance represents that of the bulk semiconductor, the contacts and the leads and is in the range of 10Ω (2). The shunt capacitance, which is in parallel, represents the capacitance of the device. The device should be selected with this value of capacitance as low as possible.
Fig 5.4: Equivalent circuit of the p-i-n diode
Source: "Optical Communication Systems", by John Gowar, Pg 454 (2)
5.2.1.3 Properties of p-i-n photo detectors
Quantum Efficiency
According to B.A Saaleh and Malvin Carl (1), quantum efficiency of a photo detector is defined as the probability that a single photon incident on the photo detector generates an electron-hole pair that contributes to the overall current. It is denoted by η and can also be defined as the ratio of the number of photo carriers produced to the number of incident photons.
In most cases it is almost certain that all of the photons incident on the photo detector do not get completely converted into charge carriers. This is because some of the incident photons maybe reflected back and also because of electron-hole pair recombination taking place at the surface due to abundance of photons. (1). Quantum efficiency is expressed as:
……………………….. (5.1)
Where R is the responsivity of the device and can be expressed in terms of photocurrent generated to incident optical power as:
………………………… (5.2)
In equation 5.1, λ is the wavelength of the incident light beam. This equation shows the dependence of quantum efficiency on the wavelength of incident light. John Gowar (2), in his book, states that in order for a high value of quantum efficiency to be obtained three important steps should be carried out:
Minimizing the reflections taking place at the incident surface.
Maximizing the amount of absorption taking place within the depletion region.
Minimizing or completely avoiding the recombination of electron-hole pairs before they are collected.
Responsivity
Responsivity is defined as the ratio of the current flowing through the photo-diode to the incident optical power received by the photo-diode. The expression for responsivity was shown in equation 5.2. Responsivity can also be expressed in terms of quantum efficiency as:
……………….… (5.3)
The unit for responsivity is A/W. The responsivity of a device is susceptible to deterioration if exposed to very large amount of incident light or incident light having large optical power. This condition is called detector saturation (1).
Response Time
Response time of the photo-detector depended upon the transit time and the RC time-constant. The longer it takes for the charge carriers to travel from one end of the device to the external circuit; larger is the spreading with respect to time of these charges (1). This is known as transit-time spread. In the case of a p-i-n diode the carrier-transport response time is approximately equal to the recombination time Ï„. The response times for a p-i-n photodiode is just a few pico-seconds, corresponding to bandwidths of 50GHz could be achieved (1).
Spectral Response
The spectral response of a p-i-n photo-detector indicates the relationship between wavelength and responsivity of the detector. The spectral response of the detector depended upon the materials used in its fabrication. For example, devices which were made up of extrinsic semiconductors could be used to detect longer wavelengths of light, whereas those comprised up of intrinsic semiconductors were generally very good at detecting longer near infra-red wavelengths. The fig below shows the spectral response of silicon p-i-n photodiode (ideal) v/s the typical response of a silicon p-i-n photodiode.
Fig 5.5: Spectral Response
Source: "Fundamentals of Photonics", by Bahaa E.A.Saleh and Malvin Carl Teich, Pg 661 (1)
From the above fig it was observed that for a Si device, the response generally peaks at close to 800nm before decreasing rapidly in the infra-red region of the spectrum.
Noise
The sensitivity of a photo-diode was determined by the random voltage and current fluctuations that occur at its output, in the absence of optical power and in its presence (2). The random fluctuations of current about a mean value of photo-detector current were identified as noise. This is called shot noise. Shot noise is related to bandwidth (Δf) and dark current (Id) by the expression:
……………….. (5.4)
A small amount of current that flows through the device when no light is incident on it is called dark current. This was due to background radiation being picked up by the detector along with the junction saturation current (2).The dark current in most photo-detectors are in the range of a few nano amperes. The table below shows the different characteristics of a p-i-n diode including the dark current values:
Table 2: Characteristics of common p-i-n photodiodes
C:\Users\Mithun\Pictures\table 1.PNG
Source: "Fiber-optic communication systems", by Agrawal, G. P, Pg no 140 (4)
The signal to noise ratio (SNR) is defined as the ratio of the average signal power to the noise power. In case of a p-i-n photodiode is given by the following expression:
………. (5.5)
From the above equation the second term in the denominator indicates the thermal noise of the photo detector. It is mainly caused due to the movement of charge carriers, when a voltage is applied, on thermal excitation.
The three main figures of merit used to describe the performance of a photo-detector are:
Noise Equivalent Power(NEP)
Directivity (D)
Specific Directivity(D*)
5.2.2 The Avalanche Photodiode (APD)
Avalanche Photodiodes are devices which detect optical incident light, and due to a process known as avalanche multiplication increase the gain of the output of the device. As stated by John.M.Senior (3), APDs have a distinct advantage over photodiodes without internal gain for low light level detection. According to G. P. Agrawal (4), the responsivity of p-i-n photodiodes are limited and for practical cases detectors with large value of responsivity are preferred. APDs have much larger responsivity and provide an internal current gain in a method similar to the photo multiplier tubes (4).
5.2.2.1 Working Principle
The figure below shows the structure of an APD as well as its electric-field distribution. The principle of operation is similar to the p-i-n diode except for the inclusion of a current gain. This current gain is produced by an effect known as impact ionization. This additional current gain helped improve the SNR of the detector.
Fig 5.6: Structure of the Avalanche Photodiode
C:\Users\Mithun\Pictures\fig.PNG
Source: "Fiber-optic communication systems", by Agrawal, G. P, Pg no 143 (4)
As observed from the figure in the structure of an APD, an additional p region is present which provided the gain for the system. This layer is known as the multiplication layer (4).
In an APD an incident photon on the surface of the device was responsible for freeing up an electron which has, under certain conditions, enough kinetic energy to free another electron in the valance band. This meant that the primary electron was responsible for the creation of new electron-hole pairs which contribute to the overall current of the device. The process can also be explained with respect to the structure of the APD. The i-region was responsible for the collection of photons and creation of the primary electron-hole pairs, which move to the p region that was under the influence of a high electric field. This region was responsible for the generation of secondary electron-hole pairs. The table below shows the various properties of different types of APDs:
Table 3: Characteristics of common Avalanche photodiodes
C:\Users\Mithun\Pictures\table 2.PNG
Source: "Fiber-optic communication systems", by Agrawal, G. P, Pg no 145 (4)
The SNR of an APD could be, for the most part, described by equation (5.5) except for the numerator term which contains a squared amplification factor M. The SNR expression for an APD is shown below:
…… …..…. (5.6)
The optimum value of the amplification factor M to produce maximum SNR is given by:
………………………….…. (5.7)
The equation illustrates the dependence of the optimum value of M on factors such as the dark current and the responsivity of the device.
5.2.3 Selection of Photo-diode
For the experiment that was carried out one of the most essential requirements for the receiver front end design was the selection of the photo-detector. As explained previously the two main devices that were shortlisted for this purpose, amongst all the other photo-detectors, were the p-i-n diode and the avalanche photodiode. The table below makes a comparison between all the different types of photo-detectors that can be used for light detection.
Table 4: Comparison of different photo detectors
C:\Users\Mithun\Pictures\table blahhhh.PNG
Source: "Designing a wireless underwater communication system" (10)
The selection of the photodiode depended largely on the application. At a first glance the APD seemed to be a lot suited for selection because of the numerous advantages it had. For example APDs, as stated by Ta-Shing Chu.et.al (5), had a much greater sensitivity (generally 10-15 times greater) than silicon p-i-n diodes. Also p-i-n diodes required more complicated equalization circuits to improve the system performance. This additional circuitry i.e. use of two-stage automatic gain control, would add to the complication of receiver design (5). However, Hoa Le Minh.et.al (6) showed that pre or post equalization circuits could vastly improve the modulation bandwidth at the receiver section for p-i-n photodiodes; and thereby improved the overall system performance.
APDs however have a number of disadvantages that make its use in circuits, which do not operate at very high speeds, a little more problematic. APD fabrication is more complex and hence the cost of the device is more. The gains of APDs are not completely predictable due to its random nature and this contributes to additional noise in the receiver. APDs require very high reverse biasing voltage for its operation and while that might not be of much concern in indoor communications (5), it could prove problematic in outdoor situations. Also the gain of an APD depends on temperature variations and the receiver sections should therefore have necessary temperature compensation (2). APDs also add unnecessary noise to the signal as stated by J.R.Barry (7) . Taking into account these disadvantages of the APD, the p-i-n photodiode was preferred as the first choice as the receiver photo-detector over the APD.
The p-i-n photodiode had to be selected based on the type of transmitter LED employed i.e. based on the signal wavelength that it could detect. Fig 5.5 in section 5.2.1.3 represented the general spectral response of a p-i-n photodiode but, as mentioned in that section, the spectral response also depended upon the wavelength of the incident light.
The selection of a white LED as the optical transmitter was discussed in the earlier chapter. Since the colour white is composed of different wavelengths, the photodiode cannot be selected based on wavelength criteria unless it was tailored in some way to do so. For this purpose, and a few other advantages, the experiment used a blue filter before the receiver section. This helped by blocking off higher wavelengths and letting only a small range of wavelengths through. And so the receiver photodiode was selected on the basis of the wavelength detection criteria.
Another main criterion for the selection of the photo-detector was the area size. Larger areas of detection meant greater sensitivity, however it also meant larger values of diode capacitance. Larger values of capacitance restrict the bandwidth of the received incoming signal. So a trade-off between area size and capacitance had to be taken into consideration, during the selection of the p-i-n photo detector.
After taking into account all of the necessary factors, the p-i-n diode that was selected was a Centronic Series 5T blue sensitive photodiode. From the 5T range two photo-detectors were tested in the circuit. The first one, OSD15-5T had larger detection area, whereas the second one OSD5-5T had a smaller detection area. The 5T series, as mentioned in its datasheet (8), is well suited for low light level applications from 430-900nm. The spectral response of the 5T series diode is shown in the fig below:
Fig 5.7: Spectral Response of 5T series photodiodes
C:\Users\Mithun\Pictures\response.PNG
Source: http://www.farnell.com/datasheets/316996.pdf (8)
The above graph shows the spectral response of 5T series photodiodes in photovoltaic (0V) and photo conducting (12V) modes of operation.
5.3 Optical Filter
Filtering is used in a system in order to negate the effect of noise. The use of optical filters in the receiver design aided the system performance for two reasons, namely:
Improving Modulation Bandwidth
The white LED used as a transmitter had a very low modulation bandwidth of 2-3 MHz (6). An optical filter filters out the slower response of the yellowish phosphor, thereby improving the modulation bandwidth considerably. For a white LED, a blue filter was used to perform this task. With the help of a blue filter the modulation bandwidth could be improved to 10MHz (6).
Noise Reduction
The main reason for the use of the optical filter is to filter out the noise before the transmitted light reaches the photo-detector. In the lab experiment performed: atmospheric effects, which add substantial amount of noise to the signal, did not contribute a large extent to the noise figure. This was because of the small size of the communication link (1m) and due to the fact that the experiment was carried out indoors. The sources that contributed the most amount of noise in an indoor optical communication link, and hence affected the system performance, were that of ambient light (sunlight, fluorescent light and light from tungsten filament bulbs). Selection of the optical filter mainly depended on the criterion of noise reduction.
Optical filters that are designed could be of different types. O'Brien.et.al (9) makes use of a very thin (10nm thickness) interference filter that filters out the noise before the light waves are made to hit the optics. For the experiment a 2.5mm thick optical filter was used for filtering out the noise. The selection of the blue filter also depended on its filtering characteristics.
So for the experiment a Hoya blue B-390 band pass filter was selected to perform the task of filtering. This was selected based on its ability to pass wavelengths around the 300-500nm range and block all wavelengths beyond 500nm to the near infra-red range of 1200nm. As explained in the previous section, this would help the photo-detector which operated at a peak detection wavelength of 436nm. The figure below shows the filtering range of the B-390 filter.
Fig 5.8: Transmittance v/s wavelength graph of a B-390 filter
C:\Users\Mithun\Pictures\filter.png
Source: http://www.uqgoptics.com/pdf/Hoya%20B-390.pdf (14)
5.4 Current to Voltage converter
The photo-detector operates as a device which converts incident optical power to current in the external circuit. This generated current must be converted into voltage so that it represents the data voltage levels that were being transmitted. For this conversion it was an essential requirement for a current to voltage converter to be present in the system.
The simplest form of a current to voltage converter is a resistor. This acts as a passive current to voltage converter. Based on Ohm's law the resistor converts the current into a voltage level multiplied by its resistance value. However this setup could be used for only for testing the working of the photo-detector. In actual practical implementation resistors prove to be problematic when converting current to voltage. As Heather Brundage (10) explained in her thesis paper, that passive current to voltage converters assumes the load across the resistor to be of infinite resistance. But practically that is not the case and hence a small portion of current gets drawn by the load there by reducing the voltage conversion done by the resistor.
Fig 5.9: Passive current to voltage converter schematic
C:\Users\Mithun\Pictures\passive-current-to-voltage-converter.png
Source: http://www.circuitstoday.com/transimpedance-amplifier (11)
Also if a large value resistor is placed across the diode to improve the current to voltage conversion; has a disadvantage that, if the value is too large then the detected photocurrent flows back to the diode saturating it (10). Large resistances also cause response times of the circuit to slow down considerably; this is shown by the equation below:
……………………………. (5.8)
Where Ï„ is the time constant, R is the resistance value and is the parasitic capacitance of the diode.
As a result the experiment used an active current to voltage conversion setup which was performed by an operational amplifier acting as a Transimpedance amplifier (TIA). The figure below shows the Transimpedance amplifier setup:
Fig 5.10: Transimpedance amplifier setup using an operational amplifier
C:\Users\Mithun\Pictures\active-transimpedance-amplifier-using-opamp.png
Source: http://www.circuitstoday.com/transimpedance-amplifier (11)
As shown in the figure above, in an inverting TIA setup the non-inverting pin is grounded. When the anode of the reverse biased photodiode is connected to the inverting pin of the op-amp the parasitic capacitance of the photodiode is removed as it is connected to a virtual ground (10). The pull down load resistor was optional. The output of the TIA is given by:
………………………….. (5.9)
When the experiment was conducted a feedback capacitor was also included in the setup. This prevented noise from being amplified at the later voltage amplifier stage. The feedback capacitor reduced overshoot and helped to stabilize the circuit.
The selection of the feedback capacitor depended on the bandwidth that was available at the receiver and the feedback resistor values. The figure below shows the op-amp in connection with the photo detector and the required output.
Fig 5.11: THS4631 in Transimpedance Amplifier configuration
C:\Users\Mithun\Pictures\Capture3.PNG
5.5 Voltage Amplifier
Operational amplifiers can also serve as voltage amplifiers. They are used to amplify the voltage at its input pin by a gain G. The gain G depended upon the type of amplifier configuration employed. The two amplifier configurations are the inverting amplifier and non-inverting amplifier setup. The gains of both these systems are shown below:
(For inverting amplifiers) …………………….. (5.10)
(For non-inverting amplifiers) ……………….. (5.11)
In the experiment the inverting amplifier setup was used. This was because the output from the Transimpedance amplifier, as shown in figure 5.11, would be inverted. In order to amplify and get this signal back to its original format an inverting amplifier set up was used. This stage was useful as large gain amplification in the TIA section, by using large values for Rf, would cause the response of the system to slow down. Also the range of communication link could be increased further if the amplifier section was included in the receiver setup. The figure 5.12 shows the inverting amplifier setup using an LM7171 IC:
Fig 5.12: Inverting amplifier using LM7171
An additional feedback capacitor was added to negate the effect of ringing in the output waveforms. The selection of the feedback capacitor was based on the expression shown below:
…………………..….. (5.12)
5.6 Comparator
The third and final stage of the receiver section was comprised of a comparator section. This section consisted of another operational amplifier set up to operate as a comparator. This stage converts the output of the voltage amplifier into the required TTL logic level output. The voltage output from the inverting amplifier depended upon the supply provided to the amplifier. Since the experiment used a +5V and -5V supply for the amplifier stage, its output would be amplified to those voltage levels. The comparator performs the action of comparing the output of the voltage amplifier with a reference voltage; to get only the voltage levels required. The reference voltage that was supplied to the non-inverting pin of the comparator op-amp depended on how much the output signal from LM7171 IC, to the inverting pin of the comparator, was amplified.
The comparator works as such, when the voltage level on the inverting pin is more positive than the reference voltage on the non-inverting pin the output turns to logic "0". And if the voltage level on the inverting pin is less positive than the reference voltage on the non-inverting pin then the output turns to logic "1". The figure below shows a basic comparator configuration.
Fig 5.13: General Comparator configuration using an op-amp
C:\Users\Mithun\Pictures\comparator.jpg
Source: http://www.markallen.com/teaching/ucsd/147a/lectures/lecture5/5.php (12)
In the figure, the voltage being fed to the non-inverting pin is determined by the voltage divider circuit. So the input to the non-inverting pin is:
……………….. (5.13)
5.7 Receiver Circuit Design
The design of the receiver was not as simple as the transmitter section design that was discussed in the earlier chapter. This was due to the fact that the circuit had to be designed in such a way to minimize the use of high value capacitors, to avoid limiting the bandwidth of the received signal. The various sections of the receiver have already been discussed in this chapter. The block diagram of the receiver section shown in fig 5.14 illustrates the various signal conditioning blocks and the detector section that constitute the receiver.
Fig 5.14: Block Diagram of the Receiver Section
C:\Users\Mithun\Pictures\receiver.png
5.7.1 Selection of Components
Prior to the design of the circuit, the components that made up the various blocks in the receiver section as shown in figure 5.14 had to be selected. For each IC ordered to carry out a specific task, an alternate IC which could perform a similar function was also ordered. A comparison was made between the ICs and only then was the IC selected for the circuit design. To ensure minimum time delays while carrying out the experiment more than three numbers of each IC were ordered. This was done to avoid unnecessary delays in ordering equipment if any of the ICs were faulty or if they were damaged during the process of carrying out the experiment. The selection process of each IC is described below.
For the transimpedance amplifier section two TIA ICs were initially selected. One op-amp was part of the actual circuit design whereas the other one was an optional spare but was used for testing the circuit. Texas Instruments THS4631 and OPA380 were the ICs that were used. OPA380 was used for testing the photo-detector in photovoltaic mode of operation. It was selected due to the fact that it had a Transimpedance bandwidth greater than 1MHz and had a gain bandwidth product of about 90 MHz; and both these exceed the bandwidth of the information being received which is in the range of 1-2 MHz . The THS4631 which is a high voltage wideband FET input op-amp with a gain bandwidth product of 210 MHz was initially selected for the receiver design. THS4631 was connected to the photo-detector operating under reverse bias conditions. The THS4631 IC was used for testing the circuit. The final designed operated with the OPA380 IC.
For the purpose of voltage amplification, a very high speed voltage feedback amplifier was used. This was the LM7171 IC. This was preferred over the alternate LM318 IC as it had a propagation delay of just 6ns and because its response was much quicker (42ns) as compared to LM318 (200ns) (10). The LM7171 also had a gain bandwidth product was 200 MHz which was similar to the TIA used in the first stage.
The final comparator section consisted of a specialized comparator IC in favour of a regular operational amplifier. The IC used was the AD790. This was selected on the basis that it could operate on the same ± 5V rail that was used to supply the LM7171 IC. But most importantly it was selected because it had a delay of just 42ns.
5.7.2 Circuit Diagram and Theory
Fig 5.15: Receiver circuit
C:\Users\Mithun\Pictures\receiver circuit.png
The circuit diagram in figure 5.15 shows a rough design of the receiver section using the OPA380 operational amplifier. Firstly the current to voltage converter section was set up. This consisted of the OPA380 IC connected to the photo-detector in zero bias mode. The feedback capacitance value () and the feedback resistor value () was then calculated. Various resistor values were tested to ensure a high gain was produced without jeopardizing the bandwidth of the communication link. At first a 150kΩ resistor was selected which provided large gain but because of the high capacitance value of the photo-detector, the bandwidth of the output signal was severely limited and thus saturated the output signal.
The feedback resistor value finally selected was a 33kΩ resistor. This made sure that ample gain was provided to the output signal of the photo-detector. The feedback capacitor, which formed a part of the final circuit schematic, was calculated using transfer function analysis of the transimpedance amplifier as (13):
…………………………. (5.14)
Where is the photo-detector capacitance, is the unity gain bandwidth of the amplifier and is the feedback resistor value of 33kΩ. The photo-detector capacitance from the datasheet (8) is taken as 390 pf for 0V bias. The value of for the OPA380 is 90MHz. Equation 14 works under the assumption that >>. Substituting the above values in equation 14, we have
A standard value of 3 pf was selected for the feedback capacitor.
In the voltage amplifier section a gain value of 10 or greater was preferred. Based on equation (5.10) described earlier the resistor values were selected to provide a gain of 10. The resistor was arbitrarily selected as 1.5kΩ. This resistor value was selected taking into account the bandwidth limitations. So using equation 5.10 we have:
The value of A was selected to be 10.
So was selected as 15kΩ. The value of the feedback capacitance was determined taking into account the bandwidth for the communication link as shown in equation 5.12. The capacitance value selected in the voltage amplifier section was 10pf. Using equation 5.12 the bandwidth of the system was calculated as:
This value corresponded to the bandwidth of the transmitted signal which was 1 MHz. As a result the selection of the 10pf capacitor proved to be adequate for the bandwidth requirements of the communication link.
Finally the resistor value of 470Ω in the comparator section, between the latch (pin 5) and V-logic (pin 8) pins, was selected to limit the input currents during the power-up of the comparator. The circuit design which was modified and used for testing is shown below. This however, was not the finalized version of the circuit as certain design modifications were made during the testing process.
