Presentation is loading. Please wait.

Presentation is loading. Please wait.

Instructor: Dr Aleksandar Milenkovic Lecture Notes

Similar presentations


Presentation on theme: "Instructor: Dr Aleksandar Milenkovic Lecture Notes"— Presentation transcript:

1 Instructor: Dr Aleksandar Milenkovic Lecture Notes
CPE 323 Introduction to Embedded Computer Systems: The MSP430 System Architecture Instructor: Dr Aleksandar Milenkovic Lecture Notes Alex Milenkovich

2 MSP430: System Architecture
Outline MSP430: System Architecture System Resets, Interrupts, and Operating Modes Basic Clock Module Watchdog Timer CPE 323 Alex Milenkovich

3 MSP430: System Resets, Interrupts, and Operating Modes
Alex Milenkovich

4 Resets Reset – a sequence of operations that put device into a well-defined state (from which the user’s program may start) Performed when Power is first applied and Device detects serious fault in hardware or software from which the user’s program cannot be expected to recover MSP430 supports two types of reset (HW and SW controlled) Power-on Reset (POR) Power-up Clear (PUC) CPE 323

5 Power-on Reset (POR) Generated by the following severe conditions related to hardware Device is powered-up. POR us raised if the supply voltage drops to so low a value that the device may not work correctly. Include brownout detector. A low external signal on the #RST/NMI pin (if the pin is configured for the reset function rather than the nonmaskable interrupt). Active by default Some MSP430s have a more comprehensive supply voltage supervisor (SVS). It sets the SVSFG flag if the voltage falls below the programmed level and can optionally reset the device CPE 323

6 Power-Up Clear (PUC) It always follows the POR. It is generated when software appears to be out of control Watchdog time overflows in watchdog mode Write into the watchdog control register (WDTCTL) with incorrect password in the upper byte. Can be triggered even if the WDT is disabled or operates in the interval mode Correct password is 0x5A available as symbol WDTPW Write an incorrect password into the flash memory controller registers (FCTLn) Protects the stored program from a runaway software In newer devices, a PUC is triggered when we try to fetch an instruction from the range of addresses reserved for peripheral I/O or for unimplemented memory CPE 323

7 System Reset Power-up Clear Power-on Reset (POR) A POR signal
Powering up the device A low signal on the RST/NMI pin when configured in the reset mode An SVS low condition when PORON=1. Power-up Clear A POR signal Watchdog timer expiration when in watchdog mode only Watchdog timer security key violation A Flash memory security key violation CPE 323 Alex Milenkovich

8 Power-On Reset (POR) CPE 323 Alex Milenkovich

9 Brownout Reset CPE 323 Alex Milenkovich

10 Conditions after Reset
Initial conditions for all registers and peripherals after POR and PUC are specified in the family’s user guides; some common effects #RST/NMI pin is configured for reset Most I/O pins are configured as digital inputs For registers see the manual. Notation is as follows rw-0: means that a bit can be read and written and is initialized to 0 after a PUC rw-(0): means that a bit can be read and written and is initialized to 0 after a POR and retains its value after a PUC Status register is cleared (R2=0): active mode WDT starts in watchdog mode PC is loaded with the reset vector which CPE 323

11 Reset related flags How to identify a source of the reset when debugging IFG – Interrupt Flag Register (IFG1, IFG2) WDTIFG: shows that the WDT timed out or its security key is violated OFIFG: indicates an oscillator fault (causes a nonmaskable interrupt, not reset) RSTIFG: indicates a reset caused by a signal on the #RST/NMI pin PORIFG: is set on power-on reset NMIIFG: flags a non-maskable interrupt caused by a signal on #RST/NMI These bits are not cleared by a PUC, so they can be tested to identify the source of the PUC CPE 323

12 Software configuration
Your SW must configure the MSP430 Initialize the SP, typically to the top of RAM Configure the watchdog to the requirements of the application Setup the clock (clocks) Configure all ports (unused pins should never be left as floating inputs). Configure peripheral modules to the requirements of the application (e.g. TimerA, ADC12, ...) Finally enable interrupts if needed Note: Additionally, the watchdog timer, oscillator fault, and flash memory flags can be evaluated to determine the source of the reset CPE 323 Alex Milenkovich

13 Exceptions/Interrupts (CHAPTER 6 in Textbook)
Events caused by hardware or software that require an urgent response E.g., a packet of data has been received, a new sample from ADC is ready, etc. What do we need when an interrupt request is raised Processor stops its current task Stores enough information to be able to resume the task later (PC and SR) Executes an Interrupt Service Routine (ISR) or Interrupt Handler Like a subroutine to handle a specific request. It can be invoked at hardware at UNPREDICTABLE times Exception processing What are possible sources of interrupt requests (HW internal/external, SW)? What do we do in presence of multiple requests? How do we decide which request to accept (priorities, masking, selective masking)? Where do we find the starting address of an ISR (vector table) CPE 323

14 When Do We Use Interrupts
Crucial in embedded systems – tasks are often triggered by external events Handle urgent tasks that need to be executed at higher priority than the main code E.g., a received data packet should be read from a communication devices before it gets overwritten by another one Handle infrequent tasks E.g., reading slow input devices (key pressed). An alternative is I/O polling – CPU waits for an event – would be extremely inefficient Wake the CPU from sleep CPU is in a low-power mode to conserve energy Calls to an operating system (SW interrupt) CPE 323

15 Interrupt Service Routines
Look like a subroutine Unique characteristics: Interrupts arise at unpredictable times and require prompt response => ISR is carried out in such a way to allow the main code to resume without any error (like it has never occurred) Triggered by events in the CPU, peripherals, busses (hardware) or software (interrupt instruction) CPE 323

16 Tracking Interrupts and Masking
Interrupt Flags Each interrupt has a flag E.g., TAIFG – Timer A Interrupt Flag It is set when a the condition for interrupt occur Can be read from SW at any time (polling) Cleared when an interrupt is accepted by HW automatically in case of single-sourced interrupts by SW in the ISR in case of multiple-sourced interrupts Maskable Interrupts Each flag has a corresponding enable bit E.g, TAIE – Timer A Interrupt Enable Allows us to selectively enable/disable interrupts GIE – General Interrupt Enable in SR; when cleared all MASKABLE interrupt requests are ignored; non-maskable are not affected (though they may have their own mask bits) CPE 323

17 Interrupt Vector Table
Address of an ISR (vector) is stored in a table – interrupt vector table Starts at defined address in memory Vector is determined based on the source of the interrupt Single sourced interrupts: each vector is associated with a unique interrupt Multi-sourced interrupts: multiple sources may share a single vector Priority: each vector has a distinct priority Defines which vector is taken if more than one request exists CPE 323

18 MSP430 Interrupts 3 types System reset (Non)-maskable NMI Maskable Interrupt priorities are fixed and defined by the arrangement of modules CPE 323 Alex Milenkovich

19 (Non)-Maskable Interrupts (NMI)
Sources An edge on the RST/NMI pin when configured in NMI mode An oscillator fault occurs An access violation to the flash memory Are not masked by GIE (General Interrupt Enable), but are enabled by individual interrupt enable bits (NMIIE, OFIE, ACCVIE) CPE 323 Alex Milenkovich

20 NMI Interrupt Handler CPE 323 Alex Milenkovich

21 What Happens When an Interrupt Is Requested?
1) Any currently executing instruction is completed. 2) The PC, which points to the next instruction, is pushed onto the stack. 3) The SR is pushed onto the stack. 4) The interrupt with the highest priority is selected if multiple interrupts occurred during the last instruction and are pending for service. 5) The interrupt request flag resets automatically on single-source flags. Multiple source flags remain set for servicing by software. 6) The SR is cleared with the exception of SCG0, which is left unchanged. This terminates any low-power mode. Because the GIE bit is cleared, further interrupts are disabled. 7) The content of the interrupt vector is loaded into the PC: the program continues with the interrupt service routine at that address. Takes 6 cc to execute CPE 323 Alex Milenkovich

22 What Happens on Return from Interrupt?
RETI - Return from Interrupt Service Routine 1) The SR with all previous settings pops from the stack. All previous settings of GIE, CPUOFF, etc. are now in effect, regardless of the settings used during the interrupt service routine. 2) The PC pops from the stack and begins execution at the point where it was interrupted. Takes 5 cc to execute CPE 323 Alex Milenkovich

23 Interrupt Vectors CPE 323 Alex Milenkovich

24 Interrupt Vectors /************************************************************ * Interrupt Vectors (offset from 0xFFE0) ************************************************************/ #define PORT2_VECTOR * 2 /* 0xFFE2 Port 2 */ #define UART1TX_VECTOR * 2 /* 0xFFE4 UART 1 Transmit */ #define UART1RX_VECTOR * 2 /* 0xFFE6 UART 1 Receive */ #define PORT1_VECTOR * 2 /* 0xFFE8 Port 1 */ #define TIMERA1_VECTOR * 2 /* 0xFFEA Timer A CC1-2, TA */ #define TIMERA0_VECTOR * 2 /* 0xFFEC Timer A CC0 */ #define ADC_VECTOR * 2 /* 0xFFEE ADC */ #define UART0TX_VECTOR * 2 /* 0xFFF0 UART 0 Transmit */ #define UART0RX_VECTOR * 2 /* 0xFFF2 UART 0 Receive */ #define WDT_VECTOR * 2 /* 0xFFF4 Watchdog Timer */ #define COMPARATORA_VECTOR 11 * 2 /* 0xFFF6 Comparator A */ #define TIMERB1_VECTOR * 2 /* 0xFFF8 Timer B 1-7 */ #define TIMERB0_VECTOR * 2 /* 0xFFFA Timer B 0 */ #define NMI_VECTOR * 2 /* 0xFFFC Non-maskable */ #define RESET_VECTOR * 2 /* 0xFFFE Reset [Highest Pr.] */ CPE 323 Alex Milenkovich

25 Interrupt Service Routines
ASM EXAMPLE Reset: MOV.W #SFE(CSTACK),SP ; main program .... JMP $ ; loop forever TA0_ISR: ... ; ISR RETI ; return from interrupt ; COMMON INTVEC ; segment for vectors ORG TIMERA_VECTOR DW TA0_ISR ; ISR for TimerA ORG RESET_VECTOR ; address to start ex. DW Reset END C EXAMPLE #pragma vector = TIMERA0_VECTOR __interrupt void TA0_ISR (void){ // ISR code } CPE 323 Alex Milenkovich

26 Operating Modes (to be discussed later)
CPE 323 Alex Milenkovich

27 MSP430: Clock System Alex Milenkovich

28 Clock System: An Introduction
Clock – square wave whose edges trigger hardware Traditional clocks: a crystal with frequency of a few MHz is connected to two C pins; internally the clock may be divided by 2 or 4. Typical application cycle in embedded systems C stays in a low-power mode until An event wakes up C to handle it Often need multiple clocks (fast for CPU, slow for peripherals) Power consumption: P ~ CV2f CPE 323 Alex Milenkovich

29 Clock System: An Introduction
Crystal clocks Accurate (the frequency is typically within 11 part in 100,000), stable (do not change with time or temperature) High-frequency (a few MHz) or low-frequency (32,768 Hz) for a real-time clock Expensive, delicate, draw a relatively large current, require additional components (capacitors), take long time to start up and stabilize Resistor and capacitor (RC) clocks Cheap, quick to start Poor accuracy and stability Can be external or integrated into a chip CPE 323 Alex Milenkovich

30 MS430 Clock System Flexible to address conflicting demands for high-performance, low-power, and a precise frequency 3 internal clocks from 4 possible sources: MCLK, SMCLK, ACLK Master clock, MCLK: used by the CPU and a few peripherals (e.g., ADC12, DMA, ...) Subsystem master clock, SMCLK: distributed to peripherals Auxiliary clock, ACLK: distributed to peripherals Typical configuration: MCLK and SMCLK are in the megahertz range, ACLK is 32 KHz CPE 323

31 Clock System: MSP430 Digitally controlled Oscillator, DCO: available in all devices; highly-controllable oscillator Generated on-chip RC-type frequency controlled by SW + HW Low- or high-frequency crystal oscillator, LFXT1 LF: 32768Hz XT: 450kHz MHz High-frequency crystal oscillator, XT2 Internal very low-power, low-frequency oscillator, VLO: available in more recent MSP430F2xx devices; provides an alternative to LFXT1 when accuracy is not needed CPE 323 Alex Milenkovich

32 Basic Clock System: MSP430x1xx
DCOCLK Generated on-chip with 6s start-up 32KHz Watch Crystal - or - High Speed Crystal / Resonator to 8MHz (our system is 4MHz/8MHz high Speed Crystal) Flexible clock distribution tree for CPU and peripherals Programmable open-loop DCO Clock with internal and external current source LFXT1 oscillator 32kHz 8MHz XIN LFXT1CLK ACLK Auxiliary Clock to peripherals XOUT LFXT2CLK Clock Distribution MCLK Main System Clock to CPU 100kHz - 5MHZ Digital Controlled Oscillator DCO DCOCLK Rosc SMCLK Sub-System Clock to peripherals CPE 323 Alex Milenkovich

33 Basic Clock System – Block Diagram
DIVA 2 LFXTCLK ACLK /1, /2, /4, /8 OscOff XTS Auxiliary Clock ACLKGEN SELM DIVM CPUOff High frequency 2 2 XT oscillator, XTS=1 3 0,1 MCLK /1, /2, /4, /8, off Low power 2 MCLKGEN Main System Clock Vcc LF oscillator, XTS=0 Vcc Rsel SCG0 DCO MOD SELS DIVS SCG1 3 5 DCOCLK 2 DC- Digital Controlled Oscillator DCO SMCLK + /1, /2, /4, /8, off Generator 1 Modulator MOD P2.5 /Rosc 1 Sub-System Clock DCOR DCGEN DCOMOD SMCLKGEN The DCO-Generator is connected to pin P2.5/Rosc if DCOR control bit is set. The port pin P2.5/Rosc is selected if DCOR control bit is reset (initial state). CPE 323 Alex Milenkovich

34 Basic clock block diagram (MSP430x13x/14x/15x/16x)
CPE 323 Alex Milenkovich

35 Basic operation After POC (Power Up Clear) MCLK and SMCLK are sourced by DCOCLK (approx. 800KHz) and ACLK is sourced by LFXT1 in LF mode Status register control bits SCG0, SCG1, OSCOFF, and CPUOFF configure the MSP430 operating modes and enable or disable portions of the basic clock module SCG1 - when set, turns off the SMCLK SCG0 - when set, turns off the DCO dc generator (if DCOCLK is not used for MCLK or SMCLK) OSCOFF - when set, turns off the LFXT1 crystal oscillator (if LFXT1CLK is not use for MCLK or SMCLK) CPUOFF - when set, turns off the CPU DCOCTL, BCSCTL1, and BCSCTL2 registers configure the basic clock module The basic clock can be configured or reconfigured by software at any time during program execution CPE 323 Alex Milenkovich

36 Basic Clock Module - Control Registers
The Basic Clock Module is configured using control registers DCOCTL, BCSCTL1, and BCSCTL2, and four bits from the CPU status register: SCG1, SCG0, OscOff, and CPUOFF. User software can modify these control registers from their default condition at any time. The Basic Clock Module control registers are located in the byte-wide peripheral map and should be accessed with byte (.B) instructions. Register State Short Form Register Type Address Initial State DCO control register DCOCTL Read/write 056h 060h Basic clock system control BCSCTL Read/write 057h 084h system control BCSCTL Read/write 058h reset CPE 323 Alex Milenkovich

37 Basic Clock Module - Control Registers
rw-0 SELM.1 058h BCSCTL2 SELM.0 DIVM.1 DIVM.0 SELS DIVS.1 DIVS.0 DCOR Direct SW Control DCOCLK can be Set - Stabilized Stable DCOCLK over Temp/Vcc. rw-(1) rw-(0) rw-0 rw-1 057h BCSCTL1 XT2Off Rsel.0 Rsel.1 Rsel.2 XTS DIVA.1 DIVA.0 XT5V rw-0 rw-1 DCO.2 056h DCOCTL DCO.1 DCO.0 MOD.4 MOD.3 MOD.2 MOD.1 MOD.0 Selection of DCO nominal frequency Which of eight discrete DCO frequencies is selected Define how often frequency fDCO+1 within the period of 32 DCOCLK cycles is used. Remaining clock cycles (32-MOD) the frequency fDCO is mixed RSEL.x Select DCO nominal frequency DCO.x and MOD.x set exact DCOCLK … select other clock tree options CPE 323 Alex Milenkovich

38 DCOCTL Digitally-Controlled Oscillator (DCO) Clock-Frequency Control DCOCTL is loaded with a value of 060h with a valid PUC condition. DCOCTL DCO.2 DCO.1 DCO.0 MOD.4 MOD.3 MOD.2 MOD.1 MOD.0 056H MOD.0 .. MOD.4: The MOD constant defines how often the discrete frequency fDCO+1 is used within a period of 32 DCOCLK cycles. During the remaining clock cycles (32–MOD) the discrete frequency f DCO is used. When the DCO constant is set to seven, no modulation is possible since the highest feasible frequency has then been selected. DCO.0 .. DCO.2: The DCO constant defines which one of the eight discrete frequencies is selected. The frequency is defined by the current injected into the dc generator. CPE 323 Alex Milenkovich

39 BCSCTL1 Oscillator and Clock Control Register
BCSCTL1 is affected by a valid PUC or POR condition. BCSCTL1 XT2Off XTS DIVA.1 DIVA.0 XT5V Rsel.2 Rsel.1 Rsel.0 057h Bit0 to Bit2: The internal resistor is selected in eight different steps. Rsel.0 to Rsel.2 The value of the resistor defines the nominal frequency. The lowest nominal frequency is selected by setting Rsel=0. Bit3, XT5V: XT5V should always be reset. Bit4 to Bit5: The selected source for ACLK is divided by: DIVA = 0: 1 DIVA = 1: 2 DIVA = 2: 4 DIVA = 3: 8 CPE 323 Alex Milenkovich

40 BCSCTL1 Bit6, XTS: The LFXT1 oscillator operates with a low-frequency or with a high-frequency crystal: XTS = 0: The low-frequency oscillator is selected. XTS = 1: The high-frequency oscillator is selected. The oscillator selection must meet the external crystal’s operating condition. Bit7, XT2Off: The XT2 oscillator is switched on or off: XT2Off = 0: the oscillator is on XT2Off = 1: the oscillator is off if it is not used for MCLK or SMCLK. CPE 323 Alex Milenkovich

41 BCSCTL2 BCSCTL2 SELM.1 SELM.0 DIVM.1 DIVM.0 SELS DIVS.1 DIVS.0 DCOR
BCSCTL2 is affected by a valid PUC or POR condition. BCSCTL2 SELM.1 SELM.0 DIVM.1 DIVM.0 SELS DIVS.1 DIVS.0 DCOR 058h Bit0, DCOR: The DCOR bit selects the resistor for injecting current into the dc generator. Based on this current, the oscillator operates if activated. DCOR = 0: Internal resistor on, the oscillator can operate. The fail-safe mode is on. DCOR = 1: Internal resistor off, the current must be injected externally if the DCO output drives any clock using the DCOCLK. Bit1, Bit2: The selected source for SMCLK is divided by: DIVS.1 .. DIVS DIVS = 0:1 DIVS = 1: 2 DIVS = 2: 4 DIVS = 3: 8 CPE 323 Alex Milenkovich

42 BCSCTL2 or Bit3, SELS: Selects the source for generating SMCLK:
SELS = 0: Use the DCOCLK SELS = 1: Use the XT2CLK signal (in three-oscillator systems) or LFXT1CLK signal (in two-oscillator systems) Bit4, Bit5: The selected source for MCLK is divided by DIVM.0 .. DIVM.1 DIVM = 0: 1 DIVM = 1: 2 DIVM = 2: 4 DIVM = 3: 8 Bit6, Bit7: Selects the source for generating MCLK: SELM.0 .. SELM.1 SELM = 0: Use the DCOCLK SELM = 1: Use the DCOCLK SELM = 2: Use the XT2CLK (x13x and x14x devices) Use the LFXT1CLK (x11x(1) devices) SELM = 3: Use the LFXT1CLK CPE 323 Alex Milenkovich

43 Range (RSELx) and Steps (DCOx)
CPE 323 Alex Milenkovich

44 F149 default DCO clock setting
CPE 323 slas272c/page 46 Alex Milenkovich

45 External Resistor The DCO temperature coefficient can be reduced by using an external resistor ROSC to source the current for the DC generator. ROSC also allows the DCO to operate at higher frequencies. Internal resistor nominal value is approximately 200 kOhm => DCO to operate up to 5 MHz. External ROSC of approximately 100 kOhm => the DCO can operate up to approximately 10 MHz. CPE 323 Alex Milenkovich

46 Basic Clock Systems-DCO TAPS
DCOCLK frequency control nominal - injected current into DC generator 1) internal resistors Rsel2, Rsel1 and Rsel0 2) an external resistor at Rosc (P2.5/11x) Control bits DCO0 to DCO2 set fDCO tap Modulation bits MOD0 to MOD4 allow mixing of fDCO and fDCO+1 for precise frequency generation f 1 2 3 4 5 6 7 DCOCLK fDCO nominal nominal+1 Selected nominal-1 Example Selected: f3: f4: 1000kHz 943kHz 1042kHz Frequency Cycle time 1000 nsec 1060 nsec 960 nsec MOD=19 DCOCLK DCO +0 +1 Modulation Period To produce an intermediate effective frequency between fDCO and fDCO+1 Cycle_time = ((32-MOD)*tDCO+MOD*tDCO+1)/32 = ns, selected frequency  1 MHz. CPE 323 Alex Milenkovich

47 Software FLL Basic Clock DCO is an open loop - close with SW+HW
A reference frequency e.g. ACLK or 50/60Hz can be used to measure DCOCLK’s Initialization or Periodic software set and stabilizes DCOCLK over reference clock DCOCLK is programmable 100kHz - 5Mhz and stable over voltage and temperature Digital Controlled Oscillator DCO + DCO 3 DCOMOD MOD 5 Modulator MOD DCOCLK Vcc DCOR 1 P2.5 /Rosc DC- Generator SCG0 Rsel DCGEN reference clock e.g. ACLK or 50/60Hz SW+HW Controls the DCOCLK CPE 323 Alex Milenkovich

48 Software FLL Implementation
Example: Set DCOCLK= , ACLK= 32768 ACLK/4 captured on CCI2B, DCOCLK is clock source for Timer_A Comparator2 HW captures SMCLK ( Hz) in one ACLK/4 (8192Hz) period Target Delta = /8192= 150 CCI2BInt … ; Compute Delta cmp #150,Delta ; Delta= /8192 jlo IncDCO ; JMP to IncDCO DecDCO dec &DCOCTL ; Decrease DCOCLK reti IncDCO inc &DCOCTL ; Increase DCOCLK 15 Capture/Compare 1 2 3 Register CCR2 CCI2B Capture Mode Comparator 2 Target Hz DCOCLK source for timer Stable reference ACLK/4, 8192Hz source Delta CPE 323 Alex Milenkovich

49 Fail Safe Operation Basic module incorporates an oscillator-fault detection fail-safe feature. The oscillator fault detector is an analog circuit that monitors the LFXT1CLK (in HF mode) and the XT2CLK. An oscillator fault is detected when either clock signal is not present for approximately 50 us. When an oscillator fault is detected, and when MCLK is sourced from either LFXT1 in HF mode or XT2, MCLK is automatically switched to the DCO for its clock source. When OFIFG is set and OFIE is set, an NMI interrupt is requested. The NMI interrupt service routine can test the OFIFG flag to determine if an oscillator fault occurred. The OFIFG flag must be cleared by software. CPE 323 Alex Milenkovich

50 Synchronization of clock signals
When switching MCLK and SMCLK from one clock source to another => avoid race conditions The current clock cycle continues until the next rising edge The clock remains high until the next rising edge of the new clock The new clock source is selected and continues with a full high period CPE 323 Alex Milenkovich

51 Basic Clock Module - Examples
How to select the Crystal Clock BCSCTL1 |= XTS; // ACLK = LFXT1 = HF XTAL BCSCTL1 |= DIVA0; // ACLK = XT1 / 8 BCSCTL1 |= DIVA1; do { IFG1 &= ~OFIFG; // Clear OSCFault flag from SW for (i = 0xFF; i > 0; i--); // Time for flag to set by HW } while ((IFG1 & OFIFG)); // OSCFault flag still set? // clock is stable BCSCTL2 |= SELM_3; // MCLK = LFXT1 (safe) CPE 323 Alex Milenkovich

52 Basic Clock Systems-Examples
Adjusting the Basic Clock The control registers of the Basic Clock are under full software control. If clock requirements other than those of the default from PUC are necessary, the Basic Clock can be configured or reconfigured by software at any time during program execution. ACLKGEN from LFXT1 crystal, resonator, or external-clock source and divided by 1, 2, 4, or 8. If no LFXTCLK clock signal is needed in the application, the OscOff bit should be set in the status register. SCLKGEN from LFXTCLK, DCOCLK, or XT2CLK (x13x and x14x only) and divided by 1, 2, 4, or 8. The SCG1 bit in the status register enables or disables SMCLK. MCLKGEN from LFXTCLK, DCOCLK, or XT2CLK (x13x and x14x only) and divided by 1, 2, 4, or 8. When set, the CPUOff bit in the status register enables or disables MCLK. DCOCLK frequency is adjusted using the RSEL, DCO, and MOD bits. The DCOCLK clock source is stopped when not used, and the dc generator can be disabled by the SCG0 bit in the status register (when set). The XT2 oscillator sources XT2CLK (x13x and x14x only) by clearing the XT2Off bit. CPE 323 Alex Milenkovich

53 FLL+ Clock Module (MSP430x4xx)
FLL+ clock module: frequency-locked loop clock module Low system cost Ultra-low power consumption Can operate with no external components Supports one or two external crystals or resonators (LFXT1 and XT2) Internal digitally-controlled oscillator with stabilization to a multiple of the LFXT1 watch crystal frequency Full software control over 4 output clocks: ACLK, ACLK/n, MCLK, and SMCLK CPE 323

54 MSP430x43x, MSP430x44x and MSP430x461x Frequency-Locked Loop
CPE 323

55 FLL+ Clock Module LFXT1CLK: Low-frequency/high-frequency oscillator that can be used either with low-frequency Hz watch crystals, or standard crystals or resonators in the 450-kHz to 8-MHz range. XT2CLK: Optional high-frequency oscillator that can be used with standard crystals, resonators, or external clock sources in the 450-kHz to 8-MHz range. In MSP430F47x devices the upper limit is 16 MHz. DCOCLK: Internal digitally controlled oscillator (DCO) with RC-type characteristics, stabilized by the FLL. Four clock signals are available from the FLL+ module: ACLK: Auxiliary clock. The ACLK is the LFXT1CLK clock source. ACLK is software selectable for individual peripheral modules. ACLK/n: Buffered output of the ACLK. The ACLK/n is ACLK divided by 1,2,4 or 8 and only used externally. MCLK: Master clock. MCLK is software selectable as LFXT1CLK, XT2CLK (if available), or DCOCLK. MCLK can be divided by 1, 2, 4, or 8 within the FLL block. MCLK is used by the CPU and system. SMCLK: Sub-main clock. SMCLK is software selectable as XT2CLK (if available), or DCOCLK. SMCLK is software selectable for individual peripheral modules. CPE 323

56 FLL+ Clock Module Operation
After a PUC, MCLK and SMCLK are sourced from DCOCLK at 32 times the ACLK frequency. When a 32,768-Hz crystal is used for ACLK, MCLK and SMCLK will stabilize to MHz. Status register control bits SCG0, SCG1, OSCOFF, and CPUOFF configure the MSP430 operating modes and enable or disable components of the FLL+ clock module. The SCFQCTL, SCFI0, SCFI1, FLL_CTL0, and FLL_CTL1 registers configure the FLL+ clock module. The FLL+ can be configured or reconfigured by software at any time during program execution. Example, MCLK = 64 × ACLK = BIC #GIE,SR ; Disable interrupts MOV.B #(64−1),&SCFQTL ; MCLK = 64 * ACLK, DCOPLUS=0 MOV.B #FN_2,&SCFIO ; Select DCO range BIS #GIE,SR ; Enable interrupts CPE 323

57 LFXT1 Oscillator Low-frequency (LF) mode (XTS_FLL=0) with 32,768 Hz watch crystal connected to XIN and XOUT High-frequency (HF) mode (XTS_FLL=1) with high-frequency crystals or resonators connected to XIN and XOUT (~450 KHz to 8 MHz) XCPxPF bits configure the internally provided load capacitance for the LFXT1 crystal (1, 6, 8, or 10 pF) OSCOFF bit can be set to disable LFXT1 CPE 323

58 XT2 Oscillator XT2 sources XT2CLK and its characteristics are identical to LFXT1 in HF mode, except it does not have internal load capacitors (must be provided externally) XT2OFF bit disables the XT2 oscillator if XT2CLK is not used for MCLK and SMCLK CPE 323

59 DCO Integrated ring oscillator with RC-type characteristics
DCO frequency is stabilized by the FLL to a multiple of ACLK as defined by N (the lowest 7 bits of the SCFQCTL register) DCOPLUS bit sets the fDCOCLK to fDCO or fDCO/D (divider). The FLLDx bits define the divider D to 1, 2, 4 or 8. By default DCOPLUS=0 and D=2, providing fDCOCLK= fDCO/2 DCOPLUS = 0: fDCOCLK = (N + 1) x fACLK DCOPLUS = 1: fDCOCLK = D x (N + 1) x fACLK CPE 323

60 DCO Frequency Range CPE 323

61 Frequency Locked Loop FLL continuously counts up or down a 10-bit frequency integrator. The output of the frequency integrator that drives the DCO can be read in SCFI1 and SCFI0. The count is adjusted +1 or −1 with each ACLK crystal period. Five of the integrator bits, SCFI1 bits 7-3, set the DCO frequency tap. Twenty-nine taps are implemented for the DCO (28, 29, 30, and 31 are equivalent), and each is approximately 10% higher than the previous. The modulator mixes two adjacent DCO frequencies to produce fractional taps. SCFI1 bits 2-0 and SCFI0 bits 1-0 are used for the modulator. The DCO starts at the lowest tap after a PUC or when SCFI0 and SCFI1 are cleared. Time must be allowed for the DCO to settle on the proper tap for normal operation. 32 ACLK cycles are required between taps requiring a worst case of 28 x 32 ACLK cycles for the DCO to settle CPE 323

62 FLL+ Clock Module Registers
CPE 323


Download ppt "Instructor: Dr Aleksandar Milenkovic Lecture Notes"

Similar presentations


Ads by Google