CHAPTER 1 INTRODUCTION TO CHARGE PUMP BASED PLL

1 CHAPTER 1 INTRODUCTION TO CHARGE PUMP BASED PLL 1.1 INTRODUCTION Phase Locked Loop (PLL) is a simple feedback system (Dan Wolaver, 1991) that compares the output phase with the input phase and produces the output frequency which is proportional to the input phase difference. It is widely used in wireless frequency synthesis, clock data recovery and clock generation. In all the PLL applications, it is required to generate low noise and low spur signals, while achieving fast settling time. In a PLL, the phase difference between the reference signal (often from a crystal oscillator) and the output signal is translated into two signals known as UP signal from pmos and DOWN (DN) signal from nmos often called as control signals. These two control signals are used to steer current into or out of a capacitor causing the voltage across the capacitor to increase or decrease. In each cycle, the time during which the switch is turned ON is proportional to the phase difference. Hence, the charge delivered is dependent on the phase difference also. The voltage on the capacitor is used to tune a Voltage Controlled Oscillator (VCO), generating the desired output signal frequency. The use of a Charge Pump (CP) naturally adds a pole at the origin in the loop transfer function of the PLL, since the CP current (I CP ) is driven into a capacitor to generate a voltage V, (V=I CP /(sc)). The additional pole at the origin is desirable, when considering the closed-loop transfer function of the PLL. This pole (S) at the origin integrates the error signal and causes the

2 system to track the input with one more order. The CP in a PLL design is constructed in an Integrated-Circuit (IC) technology, consisting of pull-up, pull-down transistors and on-chip capacitors. A resistor is also added to stabilize the closed-loop PLL. The detailed study of simple PLL architecture is discussed in section 1.12. 1.2 BASIC CHARGE PUMP PLL The Charge Pump PLL (CPPLL) is an extension of the basic PLL which requires the addition of a CP between the phase detector and loop-filter. The CP converts the voltage fluctuation in the Phase detector to corresponding current signal thereby reduces the static error. Figure 1.1 shows the usage of CP between Phase Detector and Loop Filter (LF). Figure 1.1 Charge Pump Based PLL The CP shown in Figure 1.2 consists of a set of current sources with magnitudes of I P1 and I P2 amps respectively. In most cases, the current sources are symmetrical. Thus I P1 = I P2 = I P.

3 Figure 1.2 Charge Pump In the above circuit, one source (I P ) is connected to the positive supply rail while the other (-I P ) is connected to the negative supply rail. The sources are separated by two switches S 1 and S 2. The output of the phase detector provides the gating signals U (UP) and D (DN) which turn on S 1 and S 2 respectively. The Phase Detector is designed such that, the switches are never ON simultaneously. When U is high and D is low then, S 1 is ON and S 2 is OFF. This causes current to flow out of the pump and into the loop-filter. When U is low and D is high then Q 1 is OFF and Q 2 is ON which causes current to flow out of loop-filter and into the pump. A representative Complementary Metal Oxide Semiconductor (CMOS) CP circuit is shown in Figure 1.3. The VP BIAS and VN BIAS voltages set the positive and negative CP currents respectively. An equal UP/DN current over entire control voltage range reduces the static phase error. Care should be taken that the minimal coupling to the control voltage during switching and leakage. It is very insensitive to power-supply noise and process variations and the loop stability.

4 Figure 1.3 Charge pump biasing 1.3 THEORY OF CHARGE PUMP PLL The theory of basic CPPLL is discussed here. Figure 1.4 shows the construction of CPPLL Figure 1.4 Simple CPPLL The reference input is applied to the one of the PFD and VCO output is given to another input. This implementation senses the transition at

5 the input and output detects phase or frequency difference and activates the CP accordingly. When loop is turned on, out may be far in, and the PFD and CP vary the control voltage such that out approaches in. When input and output frequencies are sufficiently close, the PFD operates as phase detector, performing phase lock. Now consider a case, that out in drops to zero. In this case PFD simply produce Q A = Q B = 0. The CP thus remains idle and CP sustains a constant control voltage. But this does not mean that PFD and CP are no longer needed. If V cont remains constant for a long time, the VCO frequency and phase begin to drift. In particular, the VCO create random variations in the oscillation frequency that can result in large accumulation of phase error. Then, the PFD detects the phase difference, produces corrective pulses on Q A or Q B that adjusts the VCO frequency through CP and filter. Also, as phase comparison is performed in every cycle, the VCO phase and frequency cannot drift substantially. Let s construct the mathematical model for simple CPPLL. Let the two different signals arriving at A and B have equal frequency but unequal phase. Let T ref is time period of reference input and t is the time difference between signal A and signal B. The phase difference (or phase error) between two input signals is given by: = t/t ref (1.1) The phase difference is zero when loop is locked. Hence, the output voltage of PFD is given by: V PFD =[(V DD -0) /4 ][ ] (1.2)

6 Hence, the gain of PFD is given by: K PFD = V DD /4 [volts/rad] (1.3) The output of the PFD is then given to the CP, then the characteristic of I P (CP current, Up or Down) is of Signum function is I p =I p = sgn( ) That is, I p is +I p if is positive and I p is -I p if this phase error is negative. Now in locked condition of PLL, the ON time of UP or DOWN switch is given by: t p = [ /2 f in ] s (1.4) cycle is given by: Then the current delivered to the filter Cp for the time t p on each I d =[I p -(I p )/ 4 ][ ] (1.5) K PFD = I P /2 (1.6) Thus the control voltage generated across the C P is given by: V c (s) =I d (s)/ Zc(s) =[I P /2 ] [Zc(s)] ]= K (1.7) PFD C P K PFD C P =I P /2 C p [Volts/rad] (1.8) in a locked condition, suddenly = o u(t) phase difference is introduced. Q A will produce the pulses which are t = T/2 sec, which leads to output to rise by (I P /C P ) (T/2 ) ) in every period. Approximating this to a ramp voltage we can write:

7 V c (t)=[i P /2 C p ][ t u(t)] (1.9) This leads to impulse response: Hence, the transfer function of PFD-CP-Filter combination is given by: h(t)= [I P /2 C p ] u(t)] (1.10) V cont / (S)=(I P /2 C p ) (1/S) (1.11) This output of PFD-CP-Filter combination is then given to the VCO with transfer function as (K VCO /S). Then the open loop transfer function of simple CPPLL as: out/ in) (S) open =(I P /2 C p ) (K VCO /S 2 ) (1.12) Since the open loop gain has two poles at origin, this topology is called as type II PLL. The closed loop transfer function is given by: H(S) s IPKVCO 2 CP I K 2 C 2 P VCO P (1.13) This result is alarming, because closed loop system contains two imaginary poles and therefore unstable. In order to stabilize the system, a zero is added in the loop gain by adding a resistor R p in series with the loop filter capacitor.

8 The PFD-CP-Filter now has the transfer function: Vcont IP 1 (s) R P 2 C s P (1.14) Thus, the closed loop transfer function of this system becomes: H(S) IPKVCO R pcp s 1 2 CP I I S K R s K 2 P P VCO p VCO 2 2 Cp (1.15) The closed loop system contains a zero at s z = -1/(R p C p ). The natural frequency and the damping ratio are given as: n IPK 2 C VCO p (1.16) R p IPCpKVCO 2 2 (1.17) As expected, if R p =0, then time constant is given by1/( n) = 4 /(R p I P K VCO ). =0. With complex poles, the decay As seen from the Equation (1.15), if the value of I P K VCO decreases, the gain crossover frequency decreases (or shifts toward the origin), degrading the phase margin. 1.4 NON IDEAL EFFECTS IN CHARGE PUMP 1. As shown in Figure 1.4, switches are constructed using PMOS and NMOS. The inherent mismatches between these two switches result in mismatch in charging and discharging

9 current in addition to timing mismatch. Though there is a variation in control voltage at the output and the W/L ratios are adjusted so as to have equal UP and DOWN currents. Even though about 0.6% of mismatching is observed between these currents in simulation, means that the two current sources are mismatched and the control voltage experiences the random changes in it. 2. There is also a Charge Sharing problem at the output node of CP (in fact between filter capacitor) and the parasitic capacitances between Drain and Source of switch transistors. This causes a sudden change in control voltage which may disturb the VCO. 3. Another effect found in CP is Clock Feed Through. The high frequency signal provided at the gate of switch transistor passes to the output node via gate to drain parasitic capacitor C gd. This also results in jumps in control voltage. Since the VCO sensitivity is high, even a small jump in control voltage results a large jump in output frequency. 4. One more effect is limited output voltage swing. If the higher output voltage is needed the current source value must be increased. This is not possible in every condition, because it increases power consumption. Apart from these, a reference spur in PLL is also one of the major problems which arise due to current mismatches in CP. By considering all the non ideal effects, many researchers are trying to remove it from CP by using various architectures which is discussed in Chapter 2 in detail.

10 1.5 LIMITATION AND SCOPE FOR IMPROVEMENT PLL plays an indispensable place in the electronics and wireless/ Radio Frequency (RF) based communication network design. There is always a huge demand for a PLL design with high lock range, low current mismatch in the CP, PLL with high stability and gain. The idea of error amplifier based CP design is proposed Xuan Xiangguang et al (2008), which claims that there is scope for improvement in the following aspects To increase the stability and reduce the jitter problem in the higher order frequency. To explore the methods for reducing the current mismatch in the CP due to transistor Channel Length Modulation (CLM) and the process variation. To derive a circuit which could provide high output voltage, high driving capability, high power efficiency, and/or small silicon area occupation. To find a method to arrest the process and temperature variation in the PLL. 1.6 STATE OF ART After making a thorough study of the literature mentioned in Chapter 2, it is found that, many researchers have been done for developing a CPPLL which could withstand the process and temperature compensation. Some researchers have concentrated on the VCO design for gain compensation and increasing the stability of the PLL. Few works have been reported for the development of gain boosting structure and current matching characteristics. The previous works have reported a mismatch reduction of

11 0.6%. Gobbi et al (2006) tried to develop a four phase Dickson CP where the output impedance is independent of output current. Even if the Charge Sharing problem could be reduced but still there is a big challenge for the mismatch reduction in the CP design. Some researchers have concentrated on the development of Low noise based VCO, to have higher PLL locking range in their design. 1.7 MOTIVATION BEHIND THIS RESEARCH PLL is the heart of the many modern electronics as well as communication system. Recently plenty of the researches have conducted on the design of PLL circuit and still research is going on this topic. Most of the researches have conducted to realize a higher lock range PLL with lesser lock time and have tolerable phase noise. The most versatile application of the PLL is for clock generation and clock recovery in microprocessor, networking, communication systems, and frequency synthesizers. Phase Locked-Loops (PLLs) are commonly used to generate well-timed on-chip clocks in high-performance digital systems. Modern wireless communication systems employ PLL mainly for synchronization, clock synthesis, skew and jitter reduction. It finds wide application in various advanced communication and instrumentation systems. Recent advances in integrated circuit design techniques have led to the development of high performance PLL which has become more economical and reliable. Now a whole PLL circuit can be integrated as a part of a larger circuit on a single chip. There are mainly five blocks in a PLL. These are Phase Frequency Detector (PFD), CP, Loop Filter (LF), Voltage Controlled Oscillator (VCO) and Frequency Divider (FD). Presently almost all communication and electronics devices operate at a higher frequency; a

12 faster locking PLL is needed. Hence there are lot of challenges in designing a CP with reduced mismatch in the CP current and thereby providing an appropriate control voltage for the VCO and thereby it tracks the incoming frequency. This work mainly concentrates on the development of low mismatch in the CP current and increases the lock range of the PLL. This research focuses on the design and development of CP based PLL with the objective of boosting the gain, reducing the current mismatch of the CP, increasing the output impedance, reducing the CLM effect and Charge Sharing problem. It also aimed for reduction of glitches at higher frequency. 1.8 OBJECTIVE OF THIS RESEARCH The main objectives of this research is To achieve less than 1% difference of the Up/Down current and to reduce the process variation like pressure, temperatures and operating voltage variations which could be achieved during manufacturing. To increase the output resistance without adding more cascode devices. To design a circuit to avoid the CLM effect and Charge sharing problem. To design a circuit in order to minimize the high speed glitches.

13 1.9 METHODOLOGY FOLLOWED The methodology followed for this research work is as follows: The complete design including the transmission gate and inclusion of charge removal have been made in the Cadence Design Environment and the simulation (both dc and ac) is done using Spectre and the model file used from the foundry TSMC with 180nm technology file. The locking range of the design is verified by setting up a hardware environment with PLL design network, Cathode Ray Oscilloscope (CRO) and Function Generator (FG). Since the Spectre supports the foundry file simulation, it is being used for all our design work. The complete design methodology and the techniques applied for our design is shown in the Figure 5.3. The design of CP based PLL with the objective of improving the output impedance and hence the gain, reducing the current mismatch of the CP, reducing the CLM effect and charge sharing problem. The technique is to be applied for the sake of reducing the current mismatch by the use of cascode and self biasing. The output impedance of the design is to be improved by the concept of cascade structure and the CLM can be reduced by the technique of increasing the length of the transistor. The technique of Charge Sharing is applied by using the charge removal transistor. Transmission gates are applied for the removal of glitches in this design. The detailed discussion of different techniques are used for the development of CPPLL is shown in Figure 1.5.

Figure 1.5 Approach Flow Diagram of the Proposed Research Work 14

15 1.10 TOOLS AND TECHNOLOGY FILE USED The architecture design is being done in the Transistor level and it involves the manual calculation and the Tool which supports Transistor level design. The Cadence Virtuoso 5.1 is used for the transistor level design and its simulation design is done using Spectre Tool. The working procedure with Cadence ADE has been attached in the Appendix 1 and the data sheet of TL082 is attached in the Appendix 2. the TL082 is a general purpose Junction Field Effect Transistor (JFET) operational amplifier (opamp), which is consider to be an error amplifier for this work in order to get good current matching characteristics. There are few tools which support the design simulation like HSpice, Spectre and PSpice. It is preferred to use Spectre for simulation for its accuracy and technology support. Also a foundry support is from Taiwan Semiconductor Manufacturing Corporation (TSMC) (i.e) technology model files 180nm from TSMC for simulation. Since this is foundry dependent, it is also preferred to use this for simulation and in case backend layout design is required we could take this work and the tool could be taken for finishing the design. 1.11 EXPECTED OUTCOMES OF THIS RESEARCH This research is expected to provide the following advantages over the existing PLL system design. The expected outcomes of this research are: To reduce the current matching characteristics in the CPPLL which is less than 0.6 % by using an opamp with reference current source. To obtain high output impedance of the CP with the help of cascode structure and hence high gain is to be achieved,

16 Low voltage cascode current mirror and charge removal transistors are being used to eliminate CLM effect and Charge Sharing problem respectively. The high speed glitches are to be eliminated with the help of matched transmission gates. 1.12 OVERVIEW OF PLL Since its invention in1932, the basic PLL has remained nearly the same but its implementation in different technologies and for different applications continues to challenge designers. This topic deals with basics of PLL. Figure 1.6 shows the block diagram of PLL. Figure 1.6 Basic Block Diagram of PLL A phase detector is a circuit whose average output voltage is proportional to the phase difference, between two inputs. In an ideal case, the relation between the average output voltage and the input phase difference being linear, it crosses the origin for =0 as shown in Figure 1.7.

17 Figure 1.7 Phase Detector Characteristics in V/rad. Called the gain of PD is the slope of line, K PD, which is expressed The output of PD is then passed through a low pass filter, so as to remove the high frequency content in PD output voltage. This is required because the control voltage of oscillator must remain quiet in steady state. Filter also provides a memory for the loop in case lock is momentarily lost due to large interference transient. This filtered control voltage is then applied to the input of Voltage Controlled Oscillator. Control voltage forces the VCO to change the frequency in the direction that reduces the difference between input frequency and output frequency. If two frequencies are sufficiently close, the PLL feedback mechanism forces the two PD input frequency frequencies to be equal and the VCO is locked with incoming frequency. This is called as locked state of PLL.

18 Figure 1.8 depicts the basic operation of PLL. Figure 1.8 Basic Operation of PLL Once the loop is in locked state, there will be small phase difference between the two PD input phase signals. This phase difference results in a dc voltage at the phase detector output which is required to shift the VCO from its free running frequency to input frequency and keeps the loop in locked state. 1.12.1 Dynamics of Simple PLL A linear model of PLL can be constructed mathematically by considering Figure 1.9 which shows the linear model of type I PLL. Low pass filter is assumed to be of first order for simplicity.

19 Figure 1.9 Linear Model of Type I PLL The PD output contains a dc component equal to K PD ( out - in) as well as high frequency components which are filtered by the LPF. PD is simply modeled as a subtractor whose output is amplified by K PD. The overall PLL model consists of the phase subtractor, the LPF transfer function 1/(1+ s/ LPF), where LPF is the 3 db bandwidth and the VCO transfer function K VCO /S. Here, in and out are the excess phases of input and output waveforms, respectively. The open loop transfer function is given by out H(s) open (S) open in K PD 1 s 1 LPF K s VCO (1.18) where, K PD is the gain of Phase detector, K VCO is the gain of VCO and LPF is the 3 db bandwidth of Low Pass Filter.

20 as: From Equation (1.18) closed loop transfer function can be obtained H(s) s closed 2 LPF K PD K VCO s K K PD VCO (1.19) Here H(s) closed is simply denoted by out/ in. Further, since the frequency and phase are related by a linear operator, the transfer function of Equation (1.19) can be expressed as: (S) K K out PD VCO 2 s in PD LPF s K K VCO (1.20) This is second order transfer function of type I PLL. Using the control theory approach the natural frequency and damping ratio are given by: n LPFK PDK VCO (1.21) 1 2 K K PD LPF VCO (1.22) The step response is given by: out 1 n (t) 1 e t sin 2 2 n 1 t u(t) 1 (1.23) where out denotes the change in output frequency and 1 2 sin 1. Thus, as per control theory approach, the step response will contain a

21 sinusoidal component with frequency constant ( n) -1. 2 n 1 that will decay with time Referring to above discussion it can be concluded that: 1. Settling speed of PLL is of great concern in most applications. Equation (1.23) thus, shows that the exponential decay determines how fast the output approaches its final value, provided that n is maximized. Equations (1.21) and (1.22), yield, n 1 2 LPF (1.24) This result shows the critical tradeoff between settling speed and ripple on the VCO control line. If the cutoff frequency of filter is reduce the, greater high frequency components are suppressed but at the same time pull in time increases. 2. In addition to the product of n and the value of is also important. If is less than typically 0.5, step response exhibits high amplitude oscillations before settling. Hence in order to avoid this ringing, the value of damping ratio is normally kept 0.707 or even greater than or equal to 1. 3. Equation (1.22) shows that both phase error and are inversely proportional to K PD and K VCO. Hence lowering the phase error makes the system less stable. Thus in summary the simple PLL (type I) has a drawback of trade off between the pull in time, the ripple on the control voltage, the phase error and the stability.

22 1.12.2 Types of PLL Several types of PLL (Floyd Gardner 1999) architectures are available in market. The architectures broadly range according to the application. These different architectures of PLL can be considered as different types of PLL. Following types of PLL are classified according to their application. 1. Programmable PLL: This type of PLL can be programmed for wide range of signals. 2. Single and multi-phase PLL: These can control a single or many phases. They are used in digital clock networks. 3. Digital Phase Locked Loop: They are used digital input signals for application like Manchester coding. 4. PLL with lock detector: It uses a lock on one of the pins and is used in frequency modulation. 5. PLL frequency synthesizer: These are used to synthesize the frequency of different range and band. 6. PLL FM/AM demodulator: The FM/AM radio frequencies are modulated and demodulated using this type of PLL. 7. Single RF/ Multi RF PLL: It is used for controlling single or multiple radio frequencies. 8. Super PLL: It is used for frequency synthesizing of radios, networks of GSM, cordless phones, etc. PLLs are also classified according to the type of loop filter used in architecture. The order of loop filter is the type of PLL. For example, if first

23 order loop filter is used, then it is called as type I PLL. If second order filter is used, it is called as type II PLL and so on. If PLL uses simple Phase detector in its architecture, it is called as simple PLL. But if PLL uses Phase Frequency Detector accompanied with CP, it is called as CPPLL. 1.13 NON IDEAL EFFECTS IN PLL So many imperfections always remain in practical PLL circuit. These lead to high ripple on the control voltage even when the loop is locked. These ripples modulate the VCO frequency, which results in non periodic waveform. This section considers these non ideal effects in PLL (Dan Wolaver 1991, Jakob Baker et al 2003, Behzad Razavi 2002). 1.13.1 Jitter in PLL A jitter is the short term-term variations of a signal with respect to its ideal position in time. This problem negatively impacts the data transmission quality. Deviation from the ideal position can occur on either leading edge or trailing edge of signal. Jitter may be induced and coupled onto a clock signal from several different sources and is not uniform over all frequencies. Excessive jitter can increase Bit Error Rate (BER) of communication signal. In digital system Jitter leads to violation in time margins, causing circuits to behave improperly. Common sources of jitter include: Internal circuitry of PLL Random Thermal noise from crystal Other resonation devices Random mechanical noise from crystal vibration

24 Signal transmitters Traces and cables Connectors Receivers The response of PLL to jitter is very important in most applications. Figure 1.10 explains the jitter in PLL. As shown in Figure 1.10, a strictly periodic waveform, x 1 (t), contains zero crossings that are evenly spaced in time. Now consider nearly periodic signal x 2 (t), whose period experiences a small changes, deviating the zero crossing from their ideal points. Hence we can say that x 2 (t) suffers from jitter. If the instantaneous frequency of signal varies slowly from one period to next period, then it is called as slow jitter, and if the variation is fast, it is called as fast jitter. Figure 1.10 Ideal and Jittery Waveforms

25 In PLL two types of phenomena are considered. a) The input exhibits jitter and b) The VCO produces jitter. In first case, the transfer function derived for type I and type II PLLs have a low-pass characteristics, indicating that if in(t) varies rapidly, then out(t) does not fully track the variations. That means, slow jitter at the input propagates to the output unattenuated but fast jitter does not. That is, PLL low pass filters in(t). If PLL is modelled for transfer function of out VCO for type II, the transfer function depicts the high pass characteristics. That is, slow jitter components generated by VCO are suppressed but fast jitter components are not. If VCO changes slowly, then the comparison with perfectly periodic input waveform generates slowly varying error that propagates through LPF and adjusts the VCO frequency, thereby counteracting the change in VCO. On other hand if VCO varies rapidly, then error produced by the phase detector is heavily attenuated by the poles in loop, failing to correct the change. 1.13.2 Phase Noise Phase noise is random variation of phase of the signal. It is the frequency domain representation of rapid, short term fluctuations in the phase of the wave, caused by time domain instabilities ( jitter ). Generally the phase noise and jitter are closely related. Or more specifically, radio engineer call it as phase noise, but digital system engineer call it as jitter of the clock. Phase noise is of very much concern in PLL, since it directly affects the entire performance of the system. Following are the common sources of phase noise in PLL.

26 i) Oscillator noise: There are two oscillators that contribute to the phase noise of the PLL. One is the reference oscillator and other is the VCO. Although both oscillators can be modelled similarly, their effects on the output noise are distinct just due to their position in the loop. If a noise less VCO is added with AWGN with DSPSD of No/2, then the output power spectrum is given by KV CO2 (No/2 2). Though it is very simplified equation, it clearly gives the idea of output noise of PLL in the presence of VCO noise. The reference oscillator is also assumed to have sufficient behavior with different constant of proportionality. ii) Frequency Divider noise: The excess noise of a digital divider can be modelled as additive noise source at its output. In a PLL, this noise directly appears at the input of phase detector and experiences the same transfer function as the noise on the input terminal. iii) Phase detector noise: Usually phase detectors are not major sources of noise in PLLs. As the work of PD is to detect the phase difference, any random variation in the phase of input signal makes the phase detector to produce wrong output, which is get transferred through filter and tunes the VCO wrongly. 1.13.3 Reference Spur Reference spurs are spurious emissions that occur from the carrier frequency at an offset equal to the channel spacing. These are usually caused by leakage and mismatch in CP of PLL. Though they occur outside the band

27 of interest, they can enter the mixers and be translated back onto band of interest. Reference spur mainly occurs in CPPLL. Though there is no phase difference between reference and feedback signal, in the locked state, the phase detector (or phase frequency detector) produces very narrow pulse width error voltage which drives the CP. Although these pulses have a very narrow width, the fact that they exist means that the dc voltage driving the VCO is modulated by a signal of frequency equal to input reference frequency. This produces reference spurs in the RF output occurring at offset frequencies that are integer multiples of input reference frequency. A spectrum analyzer can be used to detect reference spurs. Simply increasing the span to greater than twice the reference frequency reduces the spur. offset is given by: Let I cp is CP current, I leak is leakage current in CP then the phase I I leak 2 rad cp (1.25) Now if f REF is the input reference frequency, f BW is loop bandwidth, f pl is the frequency of pole in loop filter and N is the division value then the amount of reference spur in 3rd order PLL is given by: 1 f BW fref Pr 20log N 20log dbc (1.26) 2 f f REF PL If reference spur is not enough to meet the requirement, the loop bandwidth should be further narrowed or CP current should be increased. It is also helpful to reduce the division value to relax the CP design.

28 1.14 APPLICATIONS OF PLL Since its invention, PLL continues to find new applications in electronics, communication and instrumentation. Examples include memories, microprocessors, hard disk drive electronics, RF and wireless transceivers, clock recovery circuits on microcontroller boards and optical fibre receivers. Some of the applications are as follows (Dan Wolaver 2003). 1.14.1 Frequency Multiplication and Synthesis A PLL can be modified such that it multiplies its input frequency by factor of M. Figure 1.11 shows the basic frequency multiplication concept. Figure 1.11 Frequency Multiplication Just like a voltage divider is used in feedback in voltage amplifier, as shown in Figure 1.11, output frequency of PLL is divided by M and applied to the phase detector, we get, f out =M f in. Also, since f in and f D must be equal, PLL multiplies f in by M. Some systems require a periodic waveform whose frequency (a) must be very accurate and (b) can be varied in very fine stapes. Hence,to synthesize a required frequency, a channel control word (digital) is applied to

29 divider block in feedback that varies the value of M. Since f out = M f REF, the relative accuracy of f out is equal to that of f REF. It is also notable that f out varies in stapes equal to f REF if M changes by one each time. 1.14.2 Skew Reduction This is one of the very popular and earliest uses of PLL. Suppose synchronous pair of data and clock lines enter a large digital chip. Since clock typically drives a large number of transistors and logic interconnects, it is first applied to large buffer. Thus, the clock distributed on chip may suffer from substantial skew (delay due to buffer insertion) with respect to data. This is an undesirable effect which reduces the timing budget for on-chip operations. Now consider the circuit as shown in Figure 1.12. Here input clock CK in is applied to on chip PLL and buffer is placed inside the loop. Since PLL guarantees a nominally zero phase difference between CK in and CK B, the skew is eliminated. That is, the constant phase shift introduced by the buffer is divided by infinite loop gain of the feedback system. Alignment of V VCO with CK in is not important since V VCO is not used. Figure 1.12 Use of PLL to Eliminate Skew

30 1.15 ORGANISATION OF THE THESIS The thesis is organized as follows: Chapter 1 gave a brief introduction about the PLL and CP based PLL design and its application in the domain of wireless communication network and clock synthesis network. It also discusses the motivation behind this research, previous work on this topic and their result analysis, shortcoming in the previous work and the scope for improvement. The methodology followed for the research, Tools and model file usage from the foundry and the expected outcome of this research has been discussed. The conventional PLL design and its components have been discussed. Chapter 2 discusses the previous works done in the area of design a PLL with more stability, gain, low power design, increasing the lock range and its analysis. The interpretations made by different researchers have been analyzed and the way it helps our understanding of our research and improvement has been dealt in detail. Chapter 3 deals with the design of gain boosting CP, which could increase the output impedance with the help of cascode structure. A mathematical derivation which could justify the mismatch reduction and improving R out has been discussed in detail with appropriate illustration. Discussion is also made, to get high stability and its simulation using spectre tool has also been shown. The Locking of the reference signal with the incoming signal for the conventional and proposed design is shown in this chapter. Chapter 4 deals with the design of high performance CP which could reduce the current mismatch in the CP from 0.6% in existence to 0.08%.

31 It also deals with the appropriate selection of aspect ratio in the design of CP based PLL which could reduce the CLM and Charge Sharing problem. The recorded values of the mismatch between the UP and DOWN signal from the CP and their simulation results have been shown. It also shows the drawback of jitter existence in this design. Chapter 5 deals with technique proposed for the removal of jitter while operating with higher frequency range. A detailed discussion has been made for the modified CP design and its simulated output has also been shown. Chapter 6 discussed about the analysis, interpretation of the simulated results and the results obtained from the hardware setup. Also, the future scope for the improvement of this research work has been discussed in detail.