SP8852E 2·7GHz Parallel Load Professional Synthesiser Preliminary Information FEATURES ■ 2·7 GHz Operating Frequency ■ Single 5V Supply ■ Low Power Consumption <1·3W ■ High Comparison Frequency : 20MHz ■ High Gain Phase Detector : 1mA/rad ■ Zero ‘Dead Band’ Phase Detector ■ Wide Range of RF and Reference Division Ratios ■ Programming by Dual Word Data Transfer ABSOLUTE MAXIMUM RATINGS Supply voltage Operating temperature Storage temperature Prescaler and reference input voltage Data inputs Junction temperature 1 44 B4 B3 B2 B1 B0 0V (PRESCALER) RF INPUT RF INPUT VCC (PRESCALER) VEE LOCK DETECT SP8852E C-LOCK DETECT RSET CHARGE PUMP OUTPUT CHARGE PUMP REF NC NC FPD* FREF* VCC REF OSC CAPACITOR REF IN/CRYSTAL The SP8852E is one of a family of parallel load synthesisers containing all the elements apart from the loop amplifier to fabricate a PLL synthesis loop. Other parts in the series are the SP8854E which has hard wired reference counter programming and requires only a single 16-bit programming word, and the SP8855E which is fully programmable using hard wired links or switches. The SP8852E is programmed using a 16-bit parallel data bus. Data can be stored in one of two internal buffers, selected by a single address bit on the input interface. In order to fully program the device, two 16-bit words are required, one to select the RF division ratio (A and M counters) and phase detector gain, and one to set the 10-bit reference divider count, phase detector state and sense. Once the reference divide ratio has been set, frequency changes can be made by a single 16-bit data load entry to the RF divider chain. DS4237 - 2.0 June 1998 B5 B6 B7 B8 B9 B10 B11 B12 B13 B14 B15 Supersedes January 1996 version, DS4237 - 1.2 ORDERING INFORMATION SP8852E KG HCAR Non-standard temperature range, 255°C to 1100°C, standard product screening SP8852E IG HCAR Industrial temperature range, 240°C to 185°C, standard product screening HC44 *FPD and FREF outputs are reversed by the phase detector sense bit in the F1/F2 programming word, bit 12. The above diagram is correct when bit 12 is high. Fig. 1 Pin connections - top view THERMAL DATA 20·3V to 16V 255°C to1100°C 265°C to 1150°C 2·5Vp-p VCC 10·3V VEE 20·3V 1175°C STROBE ADDRESS NC NC NC NC NC NC NC NC NC uJC = 5°C/W uJA = 53°C/W ESD PROTECTION 1000V, human body model SP8852E VCC PRESCALER RF INPUT RF INPUT MODULUS CONTROL 15 13 3-BIT A COUNTER 48/9 14 FPD 11-BIT M COUNTER 14 0V PRESCALER B0 STROBE ADDRESS B0 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 B13 B14 B15 B2 B3 B13 B14 RF BUFFER 38 20 11 21 10 17 9 PHASE DETECTOR 8 7 6 18 25 LOAD 5 4 19 24 CHARGE PUMP OUTPUT CHARGE PUMP REFERENCE LOCK DETECT OUTPUT RSET C-LOCK DETECT FPD* FREF* REFERENCE BUFFER INPUT INTERFACE B0 B9 B10 B12 3 2 10-BIT REFERENCE DIVIDER 1 FREF 44 43 42 *FREF and FPD outputs are reversed by the phase detector sense bit, bit 12 in the programming word. The pin allocations shown are correct when bit 12 is high. 41 40 28 REFERENCE CRYSTAL 27 REFERENCE CAPACITOR Fig. 2 Block diagram 2 B15 LOAD 39 26 16 VCC VEE SP8852E Description Pin 1-11, 40-44 These are the inputs to the 16-bit data bus. When pin 38 is high the data goes to the buffers for the A counter, M counter and phase detector gain. When pin 38 is low the data goes to the buffers for the reference counter and the phase detector state (see Table 4). Open circuit = 1 (high) on these pins. Data is transparent from pins to the selected buffers when pin 39 (STROBE) is high and frozen in buffers when pin 39 is low. 13 (RF INPUT) Balanced inputs to the RF preamplifier. For single-ended operation the signal is AC-coupled into pin 13 with pin 14 AC-decoupled to ground (or vice-versa). Pins 13 and 14 are internally DC biased. 14 (RF INPUT) 17 (LOCK DETECT INPUT) A current sink into this pin is enabled when the lock detect circuit indicates lock. Used to give an external indication of phase lock. 18 (C-LOCK DETECT) A capacitor connected to this point determines the lock detect integrator time constant and can be used to vary the sensitivity of the phase lock indicator. 19 (RSET) An external resistor from pin 19 to VCC sets the charge pump output current. 20 (CHARGE PUMP OUTPUT) The phase detector output is a single ended charge pump sourcing or sinking current to the inverting input of an external loop filter. The direction is controlled by bit 12 of the reference word. For bit 12 = 1 and FPD or RF phase leads Ref phase pin 20 will sink current (see Table 3). 21 (CHARGE PUMP REF) Connected to the non-inverting input of the loop filter to set the optimum DC bias. 22 Not Connected. 23 Not connected. 24 FPD if pin 23 is high FREF if pin 23 is low RF divider output pulses. FPD = RF input frequency/(M.N1A). Pulse width = 8 RF input cycles (1 cycle of the divide by 8 prescaler output). 25 FPD if pin 23 is low FREF if pin 23 is high Reference divider output pulses. FREF = reference input frequency/R. Pulse width = high period of Ref input. 27 (Ref. oscillator capacitor) Leave open circuit if an external reference is used. See Fig. 5 for typical connection for use as an onboard crystal oscillator. 28 (REF IN/XTAL) This pin is the input buffer amplifier for an external reference signal. This amplifier provides the active element if an onboard crystal oscillator is used. 29-37 Not connected. 38 (ADDRESS) Controls which buffer the data on the input bus goes to. Pin 38 high sends data to the RF divider group of functions. Pin 38 low sends data to the Ref divider group of functions (see Fig. 6). Open circuit = high. 39 (STROBE) When pin 39 is high the A, M, and R counters are held in the reset state and the charge pump output is disabled. The data on the input bus is loaded into the buffers selected by the ADDRESS input state (pin 38) when pin 39 goes low. When pin 39 is low the data is fixed in the buffers, the buffers are loaded into the counter and control register, all the counters are active, and the charge pump is enabled. Open circuit = high. Table 1 Pin descriptions 3 SP8852E ELECTRICAL CHARACTERISTICS The Electrical Characteristics are guaranteed over the following range of operating conditions unless otherwise stated TAMB = 2 55°C to 1100°C (KG parts), 2 40°C to 185°C (IG parts); VCC = 4·75V to 5·25V Value Characteristic Pin Min. Max. 180 240 mA dBm Supply current 18, 26 RF input sensitivity 13,14 25 17 13,14, 24 56 16383 Reference division ratio 28, 25 1 1023 Comparison frequency 28, 24, 25 RF division ratio Reference input frequency 28 10 Reference input voltage 28 0 Units Typ. 16 50 MHz 100 MHz 110 dBm Conditions 100MHz to 2·7GHz. See note 3. Ref division ratio >2. See note 1 FREF/FPD output voltage high 24, 25 20·8 V WRT VCC, 2·2kΩ to 0V FREF/FPD output voltage low 24, 25 21·4 V WRT VCC, 2·2kΩ to 0V 17 300 500 mV IOUT = 3mA 61·4 61·5 61·7 mA VPIN20 = VPIN21, IPIN19 = 1·6mA, 62·0 62·3 62·5 mA 63·4 63·8 64·1 mA 65·4 66·1 66·5 mA LOCK DETECT output voltage CHARGE PUMP current 19, 20, 21 multiplication factor = 1 VPIN20 = VPIN21, IPIN19 = 1·6mA, multiplication factor = 1·5 VPIN20 = VPIN21, IPIN19 = 1·6mA, multiplication factor = 2·5 VPIN20 = VPIN21, IPIN19 = 1·6mA, multiplication factor = 4·0 Input bus logic level high 1-11, 38-44 Input bus logic level low 1-11, 38-44 Input bus current source 1-11, 38-44 2200 Input bus current sink 1-11, 38-44 Up/down current matching CHARGE PUMP REFERENCE voltage V 3·5 1 V µA VIN = 0V 10 µA VIN = VCC 20 65 % VPIN20 = VPIN21, IPIN19 = 1·6mA 21 VCC20·5 V IPIN19 = 1·6mA, current multiplication factor = 1·0 V VCC21·6 IPIN19 = 1·6mA, current multiplication factor = 4·0 2 mA RSET current 19 RSET voltage 19 1·6 V IPIN19 = 1·6mA C-LOCK DETECT current 18 110 µA VPIN18 = 4·7V 0·5 Note 2 STROBE pulse width 50 ns Note 3 Data setup time 100 ns Note 3 NOTES 1. Lower frequencies may be used provided that slew rates are maintained. 2. Pin 19 current3multiplication factor must be less than 5mA if charge pump accuracy is to be maintained. 3. Guranteed but not tested. 4 SP8852E 120 RF INPUT TO PIN 13 (dBm) TYPICAL OVERLOAD 110 17 GUARANTEED OPERATING WINDOW 0 25 210 220 TYPICAL SENSITIVITY 230 100MHz 1GHz 2GHz 2·7GHz 10GHz FREQUENCY Fig. 3 Input sensitivity j1 j 0.5 j2 ZO = 50Ω j 0.2 j5 0.2 0 0.5 1 2 5 50MHz 1·1GHz 2·5GHz 2j 5 2j 0.2 2j 2 2j 0.5 2j 1 Fig. 4 RF input impedance 5 SP8852E ADDRESS STROBE CONTROL MICRO 15V 1k 7 6 5 4 3 1n 1 44 43 42 41 40 39 8 38 9 37 10 36 11 35 12 15V 2 34 SP8852E 13 33 14 32 15 31 16 30 17 29 18 19 20 21 22 23 24 25 26 27 28 1n SP8852E 1n 1n 100p REF IN 27 28 2·2k FREF LOOP FILTER * * 33p 100p * 1µ 130V 10MHZ FPD 15V 10n − Application using crystal reference VCO OP27 ETC + DEPEND * VALUES ON APPLICATION Fig. 5 Typical application diagram DESCRIPTION Prescaler and AM counter The programmable divider chain is of A and M counter construction and therefore contains a dual modulus front end prescaler, an A counter which controls the dual modulus ratio and an M counter which performs the bulk multi-modulus division. A programmable divider of this construction has a division ratio of MN1A and a minimum integer steppable division ratio of N(N21), where N is the prescaler value. Data Entry and Storage Data is loaded from the 16-bit bus into one of the internal buffers by applying a positive pulse to the STROBE input. The input bus can be driven from TTL or CMOS logic levels. When 6 STROBE is low, the inputs are isolated and the data can be changed without affecting the programmed state. The data is loaded into the RF buffer when the address input is high and into the reference buffer when low. When the STROBE input is taken high, the A and M and reference counters are reset and the input data is applied to the internal storage register. When STROBE is again taken low, the data on the input bus is stored in the selected register and the counters released. The STROBE input is level triggered so that if the data is changed whilst the input is high, the final value before STROBE goes low will be stored. In order to prevent disturbances on the VCO control voltage when frequency changes are made, the STROBE input disables SP8852E the charge pump outputs when high. During this period the VCO control voltage will be maintained by the loop filter components around the loop amplifier, but due to the combined effects of the amplifier input current and charge pump leakage a gradual change will occur. In order to reduce the change, the duration of the strobe pulse should be minimised. Selection of a loop amplifier with low input current will reduce the VCO voltage droop during the strobe pulse and result in minimum reference sidebands from the synthesiser. Output for RF phase lag Sense bit (bit 12) Pin 20 1 Current source 0 Current sink The SP8852E has a digital phase/frequency comparator driving a charge pump with programmable current output. The charge pump current level at the minimum gain setting is approximately equal to the current fed into the RSET input, pin 19, and can be increased by programming the bus according to Table 2 by up to 4 times. Table 3 The FPD and FREF signals to the phase detector are available on pins 24 and 25 and may be used to monitor the frequency input to the phase detector or used in conjunction with an external phase detector. These outputs may be programmed by bits 10 and 11 of word 0 according to Table 4. State 3, where the outputs are disabled by the lock detect circuit, is useful where the user wishes to use an external phase detector. The internal phase/frequency detector may be used to pull the loop into lock and an automatic switch-over to the external phase detector made. When the FPD and FREF outputs are to be used at high frequencies, an external pull down resistor of minimum value 330Ω may be connected to ground to reduce the fall time of the output pulse. Bit 15 Bit 14 Bit 11 Bit 10 Reference Input The reference source can be either driven from an external sine or square wave source of up to 100MHz or a crystal can be connected as shown in Fig. 5. Phase Comparator and Charge Pump Current multiplication factor Phase detector state 0 0 1·0 0 0 Enabled, FPD and FREF off 0 1 1·5 0 1 Enabled, FPD and FREF on 1 0 2·5 1 0 Disabled by lock detect, FPD and FREF on 1 1 4·0 1 1 Disabled, FPD and FREF on Table 2 Pin 19 current = Table 4 The charge pump connections to the loop amplifier consist of the charge pump output and the charge pump reference. The matching of the charge pump up and down currents will only be maintained if the charge pump output is held at a voltage equal to the charge pump reference using an operational amplifier to produce a virtual earth condition at pin 20. The lock detect circuit can drive an LED to give visual indication of phase lock or provide an indication to the control system if a pullup resistor is used in place of the LED. A small capacitor connected form the C-LOCK DETECTOR pin to ground may be used to delay lock detect indication and remove glitches produced by momentary phase coincidence during lock up. VCC21·6V RSET Phase detector gain = IPIN19 (mA)3multiplication factor mA/rad 2p To allow for control direction changes introduced by the design of the PLL, bit 12 on the input bus address 0 can be programmed to reverse the sense of the phase detector by transposing the FPD and FREF connections. In order that any external phase detector will also be reversed by this programming bit, the FPD and FREF outputs are also interchanged by bit 12 as shown in Table 3. PIN 40 BIT 15 PIN 11 BIT 0 ADDRESS NOT USED PHASE DETECTOR SENSE CONTROL (SEE TABLE 3) 28 27 26 25 24 23 22 21 20 29 0 PHASE DETECTOR STATE CONTROL (SEE TABLE 4) 10-BIT REFERENCE COUNTER Fig. 6a Reference word bit allocation PIN 40 BIT 15 PIN 11 BIT 0 213 212 211 210 29 27 26 25 24 23 22 21 20 ADDRESS 28 1 PHASE DETECTOR GAIN CONTROL (SEE TABLE 2) M COUNTER 3-BIT A COUNTER Fig. 6b RF division ratio bit allocation Fig. 6 Programming data format 7 SP8852E VCC VCC 40k 40k 5k 4k 5k 325 325 INPUT 500 500 13 RF INPUT 14 RF INPUT 50µA 3mA 3k 0V 0V Fig. 7b RF inputs Fig. 7a 16-bit input bus, strobe and address VCC C-LOCK DETECT (HIGH WHEN LOCKED) 2·5k 18 2·5k VCC 3k 3k LOCK DETECT OUTPUT (LOW WHEN LOCKED) VREF 4·7V 20µA 3k 17 3k 400µA 100µA 100 100 1k 1k 0V 0V Fig. 7d Lock detect output Fig. 7c Lock detect decouple VCC RSET 450 19 450 VCC f UP CHARGE PUMP CURRENT SOURCES 83 f UP VCC 83 20 OUTPUT 21 REFERENCE f DN f DN 130 0V 2mA 0V Fig. 7e RSET pin Fig. 7f Charge pump circuit Fig 7 Interface circuit diagrams 8 SP8852E VCC VCC 296 296 40k 3k 3k 40k 296 28 24, 25 FPD, FREF OUTPUTS CRYSTAL CAPACITOR 27 60k 60k 50µA 50µA 3·3mA 0V 0V Fig. 7g FPD and FREF outputs 100µA 100µA 100µA Fig. 7h Reference oscillator Fig. 7 Interface circuit diagrams (continued) APPLICATIONS RF Layout Lock Detect Circuit The SP8852E can operate with input frequencies up to 2·7GHz but to obtain optimum performance, good RF layout practices should be used. A suitable layout technique is to use double sided printed circuit board with through plated holes. Wherever possible the top surface on which the SP8852E is mounted should be left as a continuous sheet of copper to form a low impedance ground plane. The ground pins 12 and 16 should be connected directly to the ground plane. Pins such as VCC and the unused RF input should be decoupled with chip capacitors mounted as close to the device pin as possible, with a direct connection to the ground plane; suitable values are 10nF for the power supplies and <1nF for the RF input pin (a lower value should be used sufficient to give good decoupling at the RF frequency of operation). A larger decoupling capacitor mounted as close as possible to pin 26 should be used to prevent modulation of VCC by the charge pump pulses. The RSET resistor should also be mounted close to the RSET pin to prevent noise pickup. The capacitor connected from the charge pump output should be a chip component with short connections to the SP8852E. All signals such as the programming inputs, RF IN, REFERENCE IN and the connections to the op-amp are best taken through the pc board adjacent to the SP8852D with through plated holes allowing connections to remote points without fragmenting the ground plane. The input pins are designed to be compatible with TTL or CMOS logic with a switching threshold set at about 2·4V by three forward biased base-emitter diodes. The inputs will be taken high by an internal pull up resistor if left open circuit but for best noise immunity it is better to connect unused inputs directly to VCC or ground. The lock detect circuit uses the up and down correction pulses from the phase detector to determine whether the loop is in or out of lock. When the loop is locked, both up and down pulses are very narrow compared to the reference frequency, but the pulse width in the out of lock condition continuously varies, depending on the phase difference between the outputs of the reference and RF counters. The logical AND of the up and down pulses is used to switch a 20mA current sink to pin 18 and a 50kΩ resistor provides a load to VCC. The circuit is shown in Fig. 7c. When lock is established, the narrow pulses from the phase detector ensure that the current source is off for the majority of the time and so pin 18 will be pulled high by the 50kΩ resistor. A voltage comparator with a switching threshold at about 4·7V monitors the voltage at pin 18 and switches pin 17 low when pin 18 is more positive than the 4·7V threshold. When the loop is unlocked, the frequency difference at the counter outputs will produce a cyclic change in pulse width from the phase detector outputs with a frequency equal to the difference at the reference and RF counter outputs. A small capacitor connected to pin 18 prevents the indication of false phase lock conditions at pin 17 for momentary phase coincidence. Because of the variable width pulse nature of the signal at pin 18 the calculation of a suitable capacitor value is complex, but if an indication with a delay amounting to several times the expected lock up time is acceptable, the delay will be approximately equal to the time constant of the capacitor on pin 18 and the internal 50kΩ resistor. If a faster indication is required, comparable with the loop lock up time, the capacitor will need to be 2 to 3 times smaller than the time constant calculation suggests. The time to respond to an out of lock condition is 2 to 3 times less than that required to indicate lock. RF Inputs Charge Pump Circuit The prescaler has a differential input amplifier to improve input sensitivity. Generally the input drive will be single ended and the RF signal should be AC coupled to either of the inputs using a chip capacitor.The remaining input should be decoupled to ground, again using a chip capacitor. The inputs can be driven differentially but the input circuit should not provide a DC path between inputs or to ground. The charge pump circuit converts the variable width up and down pulses from the phase detector into adjustable current pulses which can be directly connected to the loop amplifier. The magnitude of the current and therefore the phase detector gain can be modified when new frequency data is entered to compensate for change in the VCO gain characteristic over Programming Bus 9 SP8852E its frequency band. The charge pump pulse current is determined by the current fed into pin 19 and is approximately equal to pin 19 current when the programmed multiplication ratio is 1. The circuit diagram Fig. 7e shows the internal components on pin 19 which mirror the input current into the charge pump. The voltage at pin 19 will be approximately 1·6V above ground due to two VBE drops in the current mirror. This voltage will exhibit a negative temperature coefficient, causing the charge pump current to change with chip temperature by up to 10% over the full military temperature range if the current programming resistor is connected to VCC as shown in the application diagram, Fig. 5. In critical applications where this change in charge pump current would be too large the resistor to pin 19 could be increased in value and connected to a higher supply to reduce the effect of VBE variation on the current level. A suitable resistor connected to a 30V supply would reduce the variation in pin 19 current due to temperature to less than 1·5%. Alternatively a stable current source could be used to set pin 19 current. The charge pump output on pin 20 will only produce symmetrical up and down currents if the voltage is equal to that on the voltage reference pin 21. In order to ensure that this voltage relationship is maintained, an operational amplifier must be used as shown in the typical application Fig. 5. Using this configuration pin 20 voltage will be forced to be equal to that on pin 21 since the operational amplifier differential input voltage will be no more than a few millivolts (the input offset voltage of the amplifier). When the synthesiser is first switched on or when a frequency outside the VCO range is programmed, the amplifier output will limit, allowing pin 20 voltage to differ from that on pin 21. As soon as an achievable frequency value is programmed and the amplifier output starts to slew the correct voltage relationship between pin 20 and 21 will be restored. Because of the importance of voltage equality between the charge pump reference and output pins, a resistor should never be connected in series with the operational amplifier inverting input and pin 20, as is the case with a phase detector giving voltage outputs. Any current drawn from the charge pump reference pin should be limited to the few microamps input current of a typical operational amplifier. A resistor between the charge pump reference and the noninverting input could be added to provide isolation but the value should not be so high that more than a few millivolts drop are produced by the amplifier input current. When selecting a suitable amplifier for the loop filter, a number of parameters are important; input offset voltage in most designs is only a few millivolts and an offset of 5mV will produce a mismatch in the up and down currents of about 4% with the charge pump multiplication factor set at 1. The mismatch in up and down currents caused by input offset voltage will be reduced in proportion to the charge pump multiplication factor in use. If the linearity of the phase detector about the normal phase locked operating point is critical, the input offset voltage of most amplifiers can be adjusted to near zero by means of a potentiometer. The charge pump reference voltage on pin 21 is about 1·3V below the positive supply and will change with temperature and with the programmed charge pump multiplication factor. In many cases it is convenient to operate the amplifier with the negative power supply pin connected to 0V as this removes the need for an additional power supply. The amplifier selected must have a common mode range to within 3·4V (minimum charge pump reference voltage) of the negative supply pin to operate correctly without a negative supply. Most popular amplifiers can be operated from a 30V positive supply to give a wide VCO voltage drive range and have adequate common 10 mode range to operate with inputs at 13·4V with respect to the negative supply. Input bias and offset current levels to most operational amplifiers are unlikely to be high enough to significantly affect the accuracy of the charge pump circuit currents but the bias current can be important in reducing reference side bands and local oscillator drift during frequency changes. When the loop is locked, the charge pump produces only very narrow pulses of sufficient width to make up for any charge lost from the loop filter components during the reference cycle. The charge lost will be due to leakage from the charge pump output pin and to the amplifier input bias current, the latter usually being more significant. The result of the lost charge is a sawtooth ripple on the VCO control line which frequency modulates the phase locked oscillator at the reference frequency and its harmonics. A similar effect will occur whenever the strobe input is taken high during a programming sequence. In this case the charge pump is disabled when the strobe input is high and any leakage current will cause the oscillator to drift off frequency. To reduce this effect, the duration of the strobe pulse should be minimised. FPD and FREF Outputs These outputs provide access to the outputs from the RF and reference dividers and are provided for monitoring purposes during product development or test, and for connection of an external phase detector if required. The output circuit is of ECL type, the circuit diagram being shown in Fig. 7g. The outputs can be enabled or disabled under software control by the address 0 control word but are best left in the disabled state when not required as the fast edge speeds on the output can increase the level of reference sidebands on the synthesised oscillator. The emitter follower outputs have no internal pulldown resistor to save current and if the outputs are required an external pulldown resistor should be fitted. The value should be kept as high as possible to reduce supply current, about 2·2kΩ being suitable for monitoring with a high impedance oscilloscope probe or for driving an AC-coupled 50Ω load. A minimum value for the pulldown resistor is 330Ω. When the FPD and FREF outputs are disabled the output level will be at the logic low level of about 3·5V so that the additional supply current due to the load resistors will be present even when the outputs are disabled. Reference Input The reference input circuit functions as an input amplifier or crystal oscillator. When an external reference signal is used this is simply AC-coupled to pin 28, the base of the input emitter follower. When a low phase noise synthesiser is required the reference signal is critical since any noise present here will be multiplied by the loop. To obtain the lowest possible phase noise from the SP8852E it is best to use the highest possible reference input frequency and to divide this down internally to obtain the required frequency at the phase detector. The amplitude of the reference input is also important, and a level close to the maximum will give the lowest noise. When the use of a low reference input frequency say 4 to 10MHz is essential some advantage may be gained by using a limiting amplifier such as a CMOS gate to square up the reference input. In cases where a suitable reference signal is not available, it may be more convenient to use the input buffer as a crystal oscillator in this case the emitter follower input transistor is connected as a Colpitts oscillator with the crystal connected from the base to ground and with the feedback necessary for oscillation provided by a capacitor tap at the emitter. The arrangement is shown inset in Fig. 5. SP8852E From equation (3): C2 1 2tan 45°1 cos 45° t3 = 100kHz32p C1 R2 FROM CHARGE PUMP FROM CHARGE PUMP REFERENCE − + From equation (2): Generally, the third order filter configuration shown in Fig. 8 gives better results than the more commonly used second order because the reference sidebands are reduced. Three equations are required to determine values for the three constants, where t1 = C1R1 t2 = R2 (C11C2) t3 = C2R2 KfK0 11vn2 t22 vn2N 11vn2 t32 1 2 1 ∴t2 = 3·84431026 Using these values in equation (1): 13102332p310MHz/V 1 2tan F0 1 cos F0 vn …(1) = Since the phase detector used is linear over a range of 2p radians, the phase detector gain is given by: mA/radian These values can now be substituted in equation (1) to obtain a value for C1 and in equations (2) and (3) to determine values for C2 and R2. Example Calculate values for a loop with the following parameters: 1000MHz 10MHz 1000MHz/100MHz = 100 2p310MHz/V 45° 6·3mA The phase detector gain factor Kf = 6·3/2p = 1mA/radian 1 2 3[A] 11vn2 t22 11vn2 t32 11(100kHz32p)23(3·84431026)2 11(100kHz32p)23(65931029)2 1 2 t1 = 62832 212 6·833 1·1714 39·48310 …(3) where Kf is the phase detector gain factor in mA/radian K0 is the VCO gain factor in radians/seconds/V N is the division ratio from VCO to reference frequency vn is the natural loop frequency F0 is the phase margin, normally set to 45° Phase comparator current setting Kf = 2p 1003(100kHz32p)2 where A = …(2) vn2t32 Frequency to be synthesised Reference frequency Division ratio K0 VCO gain factor F0 phase margin Phase comparator current (100kHz32p)2365931029 t1 = The equations are: t3 = 1 t2 = Loop Filter Design t2 = 628319 ∴t3 = 65931029 Fig. 8 Third order loop filter circuit diagram t1 = 0·4142 = TO VCO = 1·593102932·415 ∴t1 = 3·8431029 Now, t1 = C1 ∴ C1 = 3·84nF t2 = R2 (C11C2) t2 = C2R2 Substituting for C2: t2 = R2 C11t3 or, R2= R2 t22t3 C1 3·84431026265931029 0·015331026 ∴R2 = 829·4Ω = t3 = C2R2 = t3 R2 = 65931029 829·4 ∴C2 = 0·794nF 11 http://www.mitelsemi.com World Headquarters - Canada Tel: +1 (613) 592 2122 Fax: +1 (613) 592 6909 North America Tel: +1 (770) 486 0194 Fax: +1 (770) 631 8213 Asia/Pacific Tel: +65 333 6193 Fax: +65 333 6192 Europe, Middle East, and Africa (EMEA) Tel: +44 (0) 1793 518528 Fax: +44 (0) 1793 518581 Information relating to products and services furnished herein by Mitel Corporation or its subsidiaries (collectively “Mitel”) is believed to be reliable. However, Mitel assumes no liability for errors that may appear in this publication, or for liability otherwise arising from the application or use of any such information, product or service or for any infringement of patents or other intellectual property rights owned by third parties which may result from such application or use. Neither the supply of such information or purchase of product or service conveys any license, either express or implied, under patents or other intellectual property rights owned by Mitel or licensed from third parties by Mitel, whatsoever. Purchasers of products are also hereby notified that the use of product in certain ways or in combination with Mitel, or non-Mitel furnished goods or services may infringe patents or other intellectual property rights owned by Mitel. This publication is issued to provide information only and (unless agreed by Mitel in writing) may not be used, applied or reproduced for any purpose nor form part of any order or contract nor to be regarded as a representation relating to the products or services concerned. The products, their specifications, services and other information appearing in this publication are subject to change by Mitel without notice. No warranty or guarantee express or implied is made regarding the capability, performance or suitability of any product or service. Information concerning possible methods of use is provided as a guide only and does not constitute any guarantee that such methods of use will be satisfactory in a specific piece of equipment. It is the user’s responsibility to fully determine the performance and suitability of any equipment using such information and to ensure that any publication or data used is up to date and has not been superseded. Manufacturing does not necessarily include testing of all functions or parameters. These products are not suitable for use in any medical products whose failure to perform may result in significant injury or death to the user. All products and materials are sold and services provided subject to Mitel’s conditions of sale which are available on request. M Mitel (design) and ST-BUS are registered trademarks of MITEL Corporation Mitel Semiconductor is an ISO 9001 Registered Company Copyright 1999 MITEL Corporation All Rights Reserved Printed in CANADA TECHNICAL DOCUMENTATION - NOT FOR RESALE