SCANPSC100FMW [TI]

SPECIALTY MICROPROCESSOR CIRCUIT, PDSO28, SOIC-28;
SCANPSC100FMW
型号: SCANPSC100FMW
厂家: TEXAS INSTRUMENTS    TEXAS INSTRUMENTS
描述:

SPECIALTY MICROPROCESSOR CIRCUIT, PDSO28, SOIC-28

光电二极管 外围集成电路
文件: 总29页 (文件大小:416K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
September 1998  
SCANPSC100F  
Embedded Boundary Scan Controller  
(IEEE 1149.1 Support)  
General Description  
Features  
n Compatible with IEEE Std. 1149.1 (JTAG) Test Access  
Port and Boundary Scan Architecture  
The SCANPSC100F is designed to interface a generic paral-  
lel processor bus to a serial scan test bus. It is useful in im-  
proving scan throughput when applying serial vectors to sys-  
tem test circuitry and reduces the software overhead that is  
associated with applying serial patterns with a parallel pro-  
cessor. The ’PSC100F operates by serializing data from the  
parallel bus for shifting through the chain of 1149.1 compliant  
components (i.e., scan chain). Scan data returning from the  
scan chain is placed on the parallel port to be read by the  
host processor. Up to two scan chains can be directly con-  
trolled with the ’PSC100F via two independent TMS pins.  
Scan control is supplied with user specific patterns which  
makes the ’PSC100F protocol-independent. Overflow and  
underflow conditions are prevented by stopping the test  
clock. A 32-bit counter is used to program the number of TCK  
cycles required to complete a scan operation within the  
boundary scan chain or to complete a ’PSC100F Built-In Self  
Test (BIST) operation. SCANPSC100F device drivers and  
1149.1 embedded test application code are available with  
National’s SCANEase software tools.  
n Supported by National’s SCAN Ease (Embedded  
Application Software Enabler) Software  
n Uses generic, asynchronous processor interface;  
compatible with a wide range of processors and PCLK  
frequencies  
n Directly supports up to two 1149.1 scan chains  
n 16-bit Serial Signature Compaction (SSC) at the Test  
Data In (TDI) port  
n Automatically produces pseudo-random patterns at the  
Test Data Out (TDO) port  
n Fabricated on FACT 1.5 µm CMOS process  
n Supports 1149.1 test clock (TCK) frequencies up to  
25 MHz  
n TTL-compatible inputs; full-swing CMOS outputs with  
24 mA source/sink capability  
n Standard Microcircuit Drawing (SMD) 5962-9475001  
Connection Diagrams  
28-Pin DIP and Flatpak  
Pin Assignment for LCC  
DS100325-18  
DS100325-1  
FACT is a trademark of Fairchild Semiconductor Corporation.  
TRI-STATE® is a registered trademark of National Semiconductor Corporation.  
© 1999 National Semiconductor Corporation  
DS100325  
www.national.com  
provide user specific control data. Test Data In (TDI) re-  
ceives serial data from the scan chain. A local control block  
is associated with each Shifter/Buffer to provide shift and  
load control as well as providing full or empty status. The SSI  
also provides Test Clock (TCK) Control. TCK is stopped and  
started depending on the status of the Shifter/Buffers or the  
32-bit Counter. By stopping and starting TCK, scan opera-  
tions will proceed only when the enabled Shifter/Buffers are  
ready to send and/or receive serial data.  
Chip Architecture  
The ’PSC100 is designed to act together with a parallel bus  
host as a serial test bus master. Parallel data is written by the  
host to the ’PSC100, which serializes the data for application  
to a serial test bus. Serial data returning from the target scan  
chain(s) is placed on the processor port for parallel reads.  
Several features are included in the ’PSC100 which make  
scan test communication more convenient and efficient.  
Figure 1 shows the major functional blocks of the ’PSC100  
design. The Parallel Processor Interface (PPI) is an asyn-  
chronous, 8-bit parallel interface which is used by the host  
processor to write and read data. The PPI generates the  
necessary internal data, address, and control signals to  
complete internal write and read operations.  
The 32-bit Counter (CNT32) is a count-down binary counter  
included to assist in controlling the SSI. The initial state of  
CNT32 is loaded from the parallel port with four consecutive  
writes to its address. When enabled, CNT32 is used to pro-  
gram the number of TCKs applied by the SSI to the bound-  
ary scan chain(s). The value of CNT32 can also be used to  
generate interrupts (i.e., when CNT32 reaches terminal  
count) and to trigger ’PSC100 features, such as, Auto TMS  
High (discussed later within this datasheet).  
The Serial Scan Interface (SSI) consists of  
a bank of  
double-buffered parallel/serial shift registers (i.e., a 2 x 8 bit  
FIFO), or Shifter/Buffers. The double buffering improves effi-  
ciency by allowing parallel writes or reads to/from one of the  
two 8-bit FIFOs within the shifter/buffer while the other FIFO  
is shifting data to/from the scan chain. Three Shifter/Buffers  
are provided for outgoing serial data and one for incoming  
serial data. Test Data Out (TDO) is for scanning out test data  
while the two Test Mode Select signals (TMS0/1) are used to  
The Mode and Status Registers are used to control and ob-  
serve the operation of the SSI and CNT32. Each of the  
Shifter/Buffers and CNT32 have an associated mode bit  
which enables it for participation in on-going operations. Sta-  
tus bits can be used for polling operations.  
DS100325-2  
FIGURE 1. ’PSC100 Block Diagram  
Pin Descriptions  
Pin  
Description  
Name  
RST (Input)  
The Reset pin is an asynchronous input that, when low, initializes the ’PSC100. Mode bits, Shifter/Buffer  
and CNT32 control logic, TCK Control, and the PPI are all initialized to defined states. RST has hysteresis  
for improved noise immunity.  
SCK (Input)  
OE (Input)  
The System Clock drives all internal timing. The test clock, TCK, is a gated and buffered version of SCK.  
SCK has hysteresis for improved immunity.  
Output Enable TRI-STATEs all SSI outputs when high. A 20 kpull-up resistor is connected to  
automatically TRI-STATE® these outputs when this signal is floating.  
www.national.com  
2
Pin Descriptions (Continued)  
Pin  
Description  
Name  
CE (Input)  
R/W (Input)  
STB (Input)  
Chip Enable, when low, enables the PPI for byte transfers. D(7:0) and RDY are TRI-STATEd if CE is high.  
CE has hysteresis for improved noise immunity.  
Read/Write defines a PPI cycle — Read when high, Write when low. R/ W has hysteresis for improved  
noise immunity.  
Strobe is used for timing all PPI byte transfers. D(7:0) are TRI-STATEd when STB is high. All other PPI  
inputs must meet specified setup and hold times with respect to this signal. STB has hysteresis for  
improved noise immunity.  
A(2:0)  
The Address pins are used to select the register to be written to or read from.  
(Input)  
D(7:0) (I/O)  
INT  
Bidirectional pins used to transfer parallel data to and from the ’PSC100.  
Interrupt is used to trigger a host interrupt for any of the defined interrupt events. INT is active high.  
(Output)  
RDY  
Ready is used to synchronize asynchronous byte transfers between the host and the ’PSC100. When low,  
RDY signals that the addressed register is ready to be accessed RDY is enabled when CE is low.  
(TRI-STATE  
Output)  
TDO  
Test Data Out is the serial scan output from the ’PSC100. TDO is enabled when OE is low.  
(TRI-STATE  
Output)  
TMS(1:0)  
(TRI-STATE  
Output)  
The Test Mode Select pins are serial outputs used to supply control logic to the UUT. TMS(1:0) are  
enabled when OE is low.  
TCK  
The Test Clock output is a buffered version of SCK for distribution in the UUT. TCK Control logic starts  
and stops TCK to prevent overflow and underflow conditions. TCK is enabled when OE is low.  
(TRI-STATE  
Output)  
TDI (Input)  
Test Data In is the serial scan input to the ’PSC100. A 20 kpull-up resistor is connected to force TDI to  
a logic 1 when the TDO line from the UUT is floating.  
FRZ (Input)  
The Freeze pin is used to asynchronously generate a user-specific pulse on TCK. If the FRZ Enable Mode  
bit is set, TCK will be forced high if FRZ goes high. FRZ has hysteresis for improved noise immunity.  
Mode and Status Registers  
MODE REGISTER 0 (MODE0)  
Bit 7  
TDO  
Bit 6  
TDI  
Bit 5  
Bit 4  
TMS0  
Enable  
Bit 3  
TMS1  
Enable  
Bit 2  
Bit 1  
Auto TMS High  
Enable  
Bit 0  
Loop-  
CNT32  
Enable  
Enable  
Enable  
Reserved  
Around  
Enable  
This register is purely a mode register. All bits are writeable  
and readable. The value 00100000 is placed in this register  
upon RST low or a synchronous reset operation.  
Bit 4: This bit enables the TMS0 shifter/buffer for shift op-  
erations. If this bit is set, the TMS0 shifter/buffer will  
cause TCK to stop if it is empty.  
Bit 7: This bit enables the TDO shifter/buffer for shift op-  
erations. If this bit is set, the TDO shifter/buffer will  
cause TCK to stop if it is empty.  
Bit 3: This bit enables the TMS1 shifter/buffer for shift op-  
erations. If this bit is set, the TMS1 shifter/buffer will  
cause TCK to stop if it is empty.  
Bit 6: This bit enables the TDI shifter/buffer for shift op-  
erations. If this bit is set, the TDI shifter/buffer will  
cause TCK to stop if it is full.  
Bit 2: This bit is reserved and should remain as a logic 0  
during all ’PSC100 operations.  
Bit 1: If this bit is set, TMS will be forced high when the  
32-bit counter is at state (00000001)h.  
Bit 5: This bit enables the 32-bit counter. If this bit is set,  
the counter will cause TCK to stop if if has not been  
loaded or if it has reached terminal count.  
Bit 0: This bit causes TDI to be connected directly back  
through TDO for Loop-Around operations.  
3
www.national.com  
Mode and Status Registers (Continued)  
MODE REGISTER 1 (MODE1)  
Bit 7  
TDO  
Bit 6  
TDI  
Bit 5  
CNT32  
Interrupt  
Enable  
Bit 4  
PRPG  
Enable  
Bit 3  
SSC  
Bit 2  
Freeze  
Pin  
Bit 1  
Test  
Bit 0  
Test  
Interrupt  
Enable  
Interrrupt  
Enable  
Enable  
Loop-  
Back  
Loop-  
Back  
Enable  
This register is purely a mode register. All bits are writeable  
and readable. The value 00000000 is placed in this register  
upon RST low or a synchronous reset operation.  
is reconfigured as a 16-Bit Serial Signature  
Compactor. If set, and MODE0 Bit 6 is set,  
the TDI shifter/buffer will cause TCK to stop  
until a seed value has been written to the  
two TDI registers.  
Bit 7:  
Bit 6:  
Bit 5:  
Bit 4:  
If this bit is set and the TDO shifter/buffer is  
not full (i.e., one or both 8-bit TDO FIFOs  
are empty), the INT pin will go high.  
Bit 2:  
If this bit is set, a high value on FRZ will  
force TCK high (see TCK Control Section).  
If this bit is set and the TDI shifter/buffer is  
not empty (i.e., one or both 8-bit TDI FIFOs  
are full), the INT pin will go high.  
Bits 1 and 0: These bits are used to control  
Test Loop-Back operations according to the  
following table.  
If this bit is set, and the 32-bit counter is not  
loaded or has reached terminal count, the  
INT pin will go high.  
MODE1  
MODE1  
Function  
Bit 1  
Bit 0  
This bit signifies that the TD0 shifter/buffer  
is reconfigured as a 32-Bit Pseudo Random  
Pattern Generator. If set, and MODE0 Bit 7  
is set, the TDO shifter/buffer will stop TCK  
until a seed value has been written to all  
four of the 8-bit LFSR segments.  
0
0
1
1
0
1
0
1
Normal Operation  
Loop-Back TDO to TDI  
Loop-Back TMS0 to TDI  
Loop Back TMS1 to TDI  
Bit 3:  
This bit signifies that the TD1 shifter/buffer  
MODE REGISTER 2 (MODE2)  
Write:  
Bit 7  
Not  
Bit 6  
Not  
Bit 5  
Not  
Bit 4  
Not  
Bit 3  
Bit 2  
Update  
Status  
Bit 1  
Bit 0  
Continuous  
Update  
Single  
Step  
Used  
Used  
Used  
Used  
Reset  
CNT32  
Read:  
Bit 7  
TDO  
Bit 6  
TDI  
Bit 5  
Bit 4  
Bit 3  
TMS1  
Status  
Bit 2  
Continuous  
Update  
Bit 1  
Bit 0  
Single  
Step  
CNT32  
Status  
TMS0  
Status  
Status  
Status  
Reset  
CNT32  
This register contains both mode and status bits. Bits 4–7  
are status bits only. Bit 3 is a status bit during read opera-  
tions and a mode bit during write operations. Bits 0–2 are  
mode bits only. Upon RST low, or a synchronous reset, the  
value placed in MODE2 is 10111000 (Read mode). Latches  
used to update status bits 3–7 retain their last state upon  
RST and are in an “unknown” state after power-up. To initial-  
ize the latches to a known state, they need to be updated us-  
ing the Update Status bit (bit 2) or continuous update bit (bit  
3).  
Bit 5: Set high if the 32-bit counter has not been loaded,  
or has reached terminal count.  
Bit 4: Set high if the TMS0 shifter/buffer is not full, i.e.,  
one or both 8-bit TMS0 FIFOs are ready to be writ-  
ten to.  
Bit 3 (Read Cycle):  
Set high if the TMS1 shifter/buffer is not full, i.e.,  
one or both 8-bit TMS1 FIFOs are ready to be writ-  
ten to.  
Bit 3 (Write Cycle):  
Bit 7: Set high if the TDO shifter/buffer is not full, i.e., one  
or both 8-bit TDO FIFOs are ready to be written to.  
If set, will cause all status bits to be continuously  
updated.  
Bit 6: Set high if the TDI shifter/buffer is not empty, i.e.,  
one or both 8-bit TDI FIFOs are ready to be read  
from.  
Bit 2 (Read Cycle):  
Shows the state of the Continuous Update bit dur-  
ing read operations (Bit 3 during writes).  
www.national.com  
4
MODE1(4:3)  
MODE2(0)  
Mode and Status Registers (Continued)  
Bit 2 (Write Cycle):  
If set, will cause a pulse to be issued internally that  
will update all status bits. This bit will be reset upon  
completion of the pulse. The state of this bit is not  
readable. It is reset upon RST low.  
Parallel Processor Interface (PPI)  
ADDRESS ASSIGNMENT  
The following table defines which register is selected for ac-  
cess with the address lines, A(2:0).  
Bit 1: If set, will cause a synchronous reset of all func-  
tions except the parallel interface. The value of this  
bit will return to zero when the reset operation is  
complete.  
A2  
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
A1  
0
0
0
0
1
1
1
1
0
0
0
0
1
1
1
1
A0  
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
R/W  
0
Function  
TDO Shifter/Buffer  
Counter Register 1  
TDI Shifter/Buffer  
TDI Shifter/Buffer  
TMS0 Shifter/Buffer  
Counter Register 2  
TMS1 Shifter/Buffer  
Counter Register 3  
32-Bit Counter  
Counter Register 0  
MODE0  
Bit 0: If set, will cause the 32-bit counter to count for one  
SCK cycle (no TCK cycle will be generated). The  
value of this bit will return to zero when the single  
step operation is complete.  
1
0
1
0
PROGRAMMING RESTRICTIONS  
1
Because certain mode bits enable shift operations for certain  
functions, these mode bits should not be changed when shift  
operations are in progress. The alignment of all registers  
during shift operations is controlled by a 3-bit counter in the  
TCK control block. Enabling or disabling a function in the  
middle of a shift operation may disrupt the logic necessary to  
keep all shifter/buffers byte-aligned.  
0
1
0
1
0
1
MODE0  
For example, if the TDO shifter/buffer (already loaded) is en-  
abled while the 3-bit counter value is 3, the shifter/buffer will  
only shift out only five bits of the first byte loaded.  
0
MODE1  
1
MODE1  
The following bits should not be changed when shift opera-  
tions are in progress, i.e., when TCK is enabled (see section  
on TCK Control).  
0
MODE2  
1
MODE2  
MODE0(7:3)  
5
www.national.com  
Parallel Processor Interface (PPI) (Continued)  
TIMING WAVEFORMS  
DS100325-3  
FIGURE 2. Write Cycle  
DS100325-4  
FIGURE 3. Read Cycle  
Note 1: Valid data is provided on the RDY line a t  
pd1  
after R/W is asserted low or a t after valid data is decoded on A2:0. The RDY line will remain high until the  
pd2  
addressed register is ready to participate in the write operation. This condition only applies when writing to a shifter/buffer and is eliminated (i.e., RDY will go low  
immediately once valid) when using shifter/buffer status polling (discussed later in this datasheet).  
Note 2: Valid data will not appear on D7:0 (and RDY will remain high) until the addressed register is ready to participate in the read operation. When the addressed  
register becomes ready (i.e., a byte is available to be read), valid data will be placed on the D7:0 bus and the RDY pin will go low allowing the bus cycle to continue.  
This read cycle delay only applies when reading the TDI shifter/buffer and is eliminated when using shifter/buffer status poling.  
www.national.com  
6
Parallel Processor Interface (PPI) (Continued)  
TIMING WAVEFORMS (Continued)  
DS100325-5  
FIGURE 4. Consecutive Read/Writes (best case timing)  
DS100325-6  
FIGURE 5. Consecutive Read/Writes (worst case timing)  
Note 3: Figures 4, 5: Figure 4 shows the best case bus cycle timing for SCK and STB during consecutive read or write cycles. The rising edge of strobe occurs a  
setup time, t or before the falling edge of SCK. This allows the cycle to be completed within 1.5 clock SCK clock cycles. Figure 5 shows the worst case bus cycle  
s4  
timing for SCK and STB during consecutive read or write cycles. The rising edge of strobe does not meet the t requirement between STB and SCK. Therefore,  
s4  
the propagation of the internal PSC100 control and reset signals is delayed until the next falling edge of SCK. The bus cycle is then completed 1.5 SCK cycles later  
creating a total bus cycle time of 2.5 SCK cycles. If worst case timing is considered for bus cycle timing, t  
is not a mandatory timing specification.  
s4  
7
www.national.com  
Parallel Processor Interface (PPI) (Continued)  
TIMING WAVEFORMS (Continued)  
DS100325-20  
FIGURE 6. Read/Write or Write/Read (best case timing)  
DS100325-7  
FIGURE 7. Read/Write or Write/Read (worst case timing)  
Note 4: Figures 6, 7: This diagram shows the timing for a read followed by a write (or write followed by a read). Separate Read and Write data/address latches and  
control logic allow consecutive read/write or write/read operations to be overlapped (i.e., do not need to wait 2 or 3 SCK cycles between bus cycles). For the best  
case timing scenario (Figure 6: rising edge of STB to falling edge of SCK greater than t ), a new bus cycle can be performed each SCK cycle. For the worst timing  
s4  
scenario (Figure 7: rising edge of STB to falling edge of SCK is less than t ), a one SCK cycle delay must be included after each back to back read/write or write/  
s4  
read sequence.  
Note 5: Figures 4, 5, 6, 7 assume that the PSC100 register participating in the bus cycle is ready to accept/provide data. For bus cycles involving a PSC100 shifter/  
buffer(s), the ready status of a shifter/buffer can be checked using the status bits in Mode Register 2 prior to the start of the bus cycle. Polling is required when the  
RDY pin is not used to provide a processor “handshake”.  
READ AND WRITE CYCLES  
dress lines for a valid PSC100F register address. Once a  
valid address has been decoded and if the addressed  
PSC100F register is ready to be read (see Table 2), valid  
data is placed on the Data lines (D7:0) a propagation delay  
later and the ready line is asserted low. If the addressed reg-  
ister is not ready (e.g., the TDI shifter/buffer is empty), the  
ready line will remain high and hold the bus cycle until the  
register contains valid data. RDY will then go low allowing  
the read cycle to continue. With the high to low edge on RDY  
line, the processor can successfully read the valid data.  
However, the bus cycle is not completed within the PSC100F  
until the rising edge on STB which resets the PSC100F read  
logic (required prior to the start of the next read cycle).  
A Write cycle (see Figure 2) is initiated by asserting CE and  
R/W low followed by a low on STB a set time later. CE and  
STB are gated within the PSC100F and may be asserted  
concurrently (i.e., zero setup and hold time). The address is  
then asserted on A2:0 to indicate which internal address  
within the PSC100F will be written to by the processor. An  
address decoder within the PSC100F monitors the address  
lines for a valid PSC100F register address. Once a valid ad-  
dress has been decoded, the RDY line becomes active (a  
propagation delay time later). The active RDY line will go low  
immediately if the addressed register is ready to accept data.  
If the addressed register is not ready, the RDY pin will re-  
main high preventing the processor from completing the bus  
cycle. Once the register is ready to receive date (see Table  
2), the RDY pin will go low and processor can resume the  
write cycle. The processor then forces a high on STB (a wait  
time after RDY goes low) which latches the address (A2:0)  
and data (D7:0) completing the bus cycle. The RDY line is  
forced high a propagation delay later.  
Important note concerning the use of RDY : The RDY sig-  
nal provides a useful “handshake” between the PSC100F  
and the processor. However, care must be taken when using  
the PSC100F RDY signal to prevent a large (or indefinite)  
number of processor generated wait states. For example, if  
the TDO shifter/buffer is not enabled for shift operations and  
the processor writes to the TDO shifter/buffer address 3  
times, the two registers which make up the TDO shifter/  
buffer will accept the first two bytes of data, but since the  
data is not shifting out, the 3rd byte will be held off by the  
RDY signal indefinitely. An equally severe problem could re-  
sult with a finite number of wait states if the application uses  
A Read cycle (see Figure 3) is initiated by asserting CE low  
and R/W high followed by a low on STB a set time later. CE  
and STB are gated within the PSC100F and may be as-  
serted concurrently (i.e., zero setup and hold time). The ad-  
dress bits (A2:0) are then asserted to indicate which internal  
address within the PSC100F will be read by the processor.  
An address decoder within the PSC100F monitors the ad-  
www.national.com  
8
Consecutive Reads and Writes: Separate control logic and  
data/address latches are used for a read and write operation  
within the PSC100F. This allows a write to occur after a read  
(or conversely, a read to occur after a write) prior to the 1.5/  
2.5 SCK clock cycle requirements described above. The tim-  
ing for a read (or write) followed by a write (or read) is shown  
in Figure 5 and Figure 6.  
Parallel Processor Interface (PPI)  
(Continued)  
dynamic memories. Holding the local bus with the PSC100F  
RDY line long enough to violate a DRAM refresh time will re-  
sult in lost data within the dynamic memory.  
Writing and Reading without the use of RDY : With use of  
worst case PSC100F timing, Write and Read cycles can be  
successfully completed without the use of the RDY signal. All  
read and write cycles will complete within 2.5 SCK cycles  
(worst case). Therefore, by assuring at least 2.5 cycles occur  
after the rising edge of STB, bus cycles can be completed  
without using the RDY “handshake”. The critical timing rela-  
tionship within the PSC100F for write and read operation is  
between the rising edge of STB and the falling edge of SCK.  
The rising edge of strobe latches the address/data and also  
generates the internal signals required to complete read/  
write within the PSC100F (including a signal with resets the  
read/write logic and releases the RDY line). The propagation  
of these internal signals is initiated on the first falling edge of  
SCK after the STB pin is asserted high. If the rising edge on  
STB occurs an internal setup time (ts4) or greater before the  
falling edge of SCK, the bus cycle can be completed within  
1.5 SCK cycles (see Figure 4). However, if the internal setup  
time is not met, the propagation of internal control/reset sig-  
nals is delayed until the next falling edge of SCK (1 SCK  
cycle later) which effectively completes the read/write opera-  
tion and reset the logic for the next bus cycle within 2.5  
cycles (see Figure 5). Synchronizing the rising edge of STB  
with the falling edge of SCK to assure that ts4 is met provides  
the maximum performance for a read/write operation. How-  
ever, the asynchronous interface can be used effectively with  
software delays, hardware delays or programmed wait  
states (to assure 2.5 SCK cycles are completed) to avoid the  
need for synchronization.  
SYNCHRONIZATION  
Writes and reads can be synchronized by using any of three  
methods: polling, interrupts, or wait state generation:  
Status bits may be polled to see if a register is ready to be  
written to or read from. To stabilize the status bits for read  
operations, the Update Status bit must be set in MODE2  
to latch the status.  
Note: The status bits only provide the state of the shifter/buffers and do not  
indicate that an internal write or read is complete. Therefore, for appli-  
cations not using the RDY signal to monitor the internal write/read sta-  
tus, timing must be controlled to assure that at least 2.5 SCK cycles  
are completed between consecutive read or consecutive write cycles.  
Any of three different events can be used to generate in-  
terrupts by forcing the INT pin high, see Table 1.  
The RDY pin can be used to hold off the host until the ad-  
dressed register is ready to be accessed. As described  
above, this pin can also be used to hold off additional  
reads/writes until the synchronizer has recovered from  
=
the previous read/write. RDY  
0 signifies that the  
’PSC100F is ready to complete the current PPI cycle.  
The logic that determines the state of RDY is summa-  
rized in Table 2.  
Reading from CNT32 can be synchronized for testing by us-  
ing the Single Step Counter mode bit.  
TABLE 1. Interrupt Logic  
=
=
=
MODE1(7) 1 and TDO  
MODE1(6) 1 and TDI  
MODE1(5) 1 and  
Shifter/Buffer Not Full  
Shifter/Buffer Not Empty  
CNT32 Not Loaded, or at  
INT  
Terminal Count  
1
X
X
0
X
1
X
0
X
X
1
1
1
1
0
0
Note 6: Interrupts are generated using the INT pin. Three events trigger INT high. Each event has its own mode bit associated with it for masking or enabling these  
interrupts.  
9
www.national.com  
Parallel Processor Interface (PPI) (Continued)  
TABLE 2. Ready State Logic  
Write  
Synchronizer  
Busy  
TDO  
TMS0  
Shifter/Buffer  
Full and  
TMS1  
Shifter/Buffer  
Full and  
Read  
Synchronizer  
Busy  
TDI  
Shifter/Buffer  
Full and  
Shifter/Buffer  
Empty and  
R/W  
RDY  
=
=
=
=
A(2:0) 1  
A(2:0)  
0
A(2:0)  
2
A(2:0)  
3
0
0
0
0
0
1
1
1
1
X
X
X
0
X
1
X
X
1
X
X
X
1
X
X
X
X
X
1
X
X
X
X
X
X
1
1
1
1
1
0
1
1
0
X
X
0
X
0
0
X
X
X
X
X
X
X
X
X
X
X
X
X
0
0
DS100325-12  
FIGURE 8. 1149.1 (JTAG) TAP Controller State Diagram  
Serial Scan Interface (SSI)  
TCK CONTROL  
to change only when SCK is low; therefore, TCK always  
stops low. TCK does not toggle (remains low) under the fol-  
lowing conditions:  
TCK CONTROL is the central control block that enables or  
disables shift operations and provides byte alignment for the  
shifter/buffers. The state of all shifter/buffers and the 32-bit  
counter (CNT32) is evaluated here and TCK is stopped and  
started. A clock enable circuit allows the “TCK enable” signal  
TDO Shifter/Buffer is enabled and empty.  
TDO Shifter/Buffer is enabled in PRPG mode and is not  
fully loaded.  
TDI Shifter/Buffer is enabled and full.  
www.national.com  
10  
asynchronously creating the TCK pulse via an external  
PSC100 pin. When the Freeze Pin Enable bit (bit 2 in Mode  
Register 1) is set, a logic high on the PSC100 FRZ input pin  
will cause TCK to go high. Once the transition is complete,  
the Freeze Mode can be removed (i.e. Freeze Pin Enable bit  
returned to logic 0 or Freeze pin forced low) and the sampled  
data can be shifted out/evaluated using the “standard”  
PSC100 protocol. Figure 9 illustrates the logic implementa-  
tion of the Freeze feature. It should be noted that Freeze  
mode is simply gated with the TCK output and does not dis-  
able shift operations within the shifter/buffers or disable  
CNT32. Therefore, no shifting or TCK counting using CNT32  
should be performed when Freeze mode is enabled.  
Serial Scan Interface (SSI) (Continued)  
TDI Shifter/Buffer is enabled in SSC mode but is not fully  
loaded with an initial value.  
TMS0 Shifter/Buffer is enabled and empty.  
TMS1 Shifter/Buffer is enabled and empty.  
CNT32 is enabled but not loaded.  
CNT32 is enabled and has reached terminal count.  
Also included within the TCK control block in CNT3, a 3-bit  
count up counter. CNT3 is included to maintain byte align-  
ment within the shifter/buffers by providing a signal to toggle  
between the two 8-bit FIFOs which comprise the shifter/  
buffer. The toggling operation occurs, in an enabled shifter/  
buffer, each time CNT3 counts 8 TCK cycles or when CNT32  
reaches terminal count. The CNT3 is reset to 0 when CNT32  
reaches terminal count or after a PSC100 reset condition.  
The “standard” mode of TCK control uses CNT32 in conjunc-  
tion with the status registers to start and stop TCK. For this  
mode, CNT32 is enabled and loaded with the number of  
TCK cycles required to shift the desired bits to/from the scan  
chain. The shifter/buffer(s) participating in the shift operation  
is enabled and provides the necessary full/empty status to  
stop TCK for processor writes/reads. This mode of TCK con-  
trol provides a systematic protocol for managing PSC100  
operations (specifically, handling partial bytes). Another op-  
tion for TCK control relies solely on the status of the shifter/  
buffers (i.e., CNT32 is disabled) to start and stop TCK. This  
option eliminates the time required to load CNT32, but  
makes management of partial bytes (see shifter/buffer de-  
scription section) more cumbersome.  
FREEZE MODE. This mode is included in the TCK control  
block to support the 1149.1 SAMPLE operation. The intent of  
the SAMPLE instruction is to allow device input and output  
levels to be observed during normal system operation. Data  
is latched (or “sampled”) into the boundary scan registers  
when the TAP controller (see Figure 8 on previous page)  
transitions from the Capture-DR state to the Shift-DR state (if  
SAMPLE/ PRELOAD is the active instruction). Synchroniz-  
ing this “transition” (rising edge of TCK with TMS at logic low)  
with a known system state is imperative to an accurate pass/  
fail assessment. The Freeze Mode provides a means of  
DS100325-9  
FIGURE 9. TCK Logic  
DS100325-10  
FIGURE 10. TMS Shifter/Buffer Block Diagram  
TMS(1:0) SHIFTER/BUFFERS  
of shifting is least significant bit first. TMS(1:0) are forced  
high upon RST low. TMS(1:0) are TRI-STATEd when OE is  
high.  
The TMS Shifter/Buffer block diagram is shown in Figure 10.  
These two blocks take parallel data and serialize it for shift  
operations through the serial port pins TMS0 and TMS1.  
Write operations are completed if the shifter/buffer is not full  
(independent of whether shifter/buffer is enabled or dis-  
abled). Otherwise they are ignored. Shifting occurs when the  
following conditions are all true:  
Double-buffering is achieved by configuring the shifter/buffer  
as a 2 x 8 FIFO. Write and shift operations are controlled by  
a local state machine that accepts stimulus from the PPI,  
Mode Registers, CNT32 and TCK Control section. The TMS  
outputs always change on the falling edge of SCK. The order  
TMS is enabled with its respective mode bit.  
TMS shifter/buffer is not empty.  
11  
www.national.com  
with two additional 8-bit parallel-to-serial shift registers. Write  
and shift operations are controlled by a local state machine  
that accepts stimulus from the PPI, Mode Registers, CNT32,  
and the TCK Control section. The TDO output always  
changes on the falling edge of SCK. The order of shifting is  
least significant bit first. TDO is forced high upon RST low.  
TDO is TRI-STATEd when OE is high.  
Serial Scan Interface (SSI) (Continued)  
TCK is enabled according to the logic in TCK Control.  
When shift operations are not enabled, the TMS output re-  
tains its last state. During long shift sequences, the TMS  
shifter/buffer can be disabled and held static so that shift op-  
erations are concentrated only on TDI and TDO. The TMS  
output also retains its last state when Test Loop-Back opera-  
tions are in progress.  
Write operations are completed if the shifter/buffer is not full  
(independent of whether shifter/buffer is enabled or dis-  
abled). Otherwise they are ignored.  
Local select circuitry is used to toggle back and forth be-  
tween the two registers of the “FIFO” when shifting. At any  
given time, one register is selected for shift operations. The  
other holds its previous state or can accept new parallel  
data. Shift register selection changes due to the following  
two events:  
Shifting occurs when the following conditions are all true:  
TDO is enabled with its respective mode bit.  
TDO shifter/buffer is not empty.  
TCK is enabled according to the logic in TCK Control.  
When shift operations are not enabled, the TDO output re-  
tains its last state. The TDO output also retains its last state  
when Test Loop-Back operations are in progress.  
CNT3 in TCK Control signals that 8 bits have been  
shifted. This event is used for basic toggling between  
each of the two shift registers.  
Local select circuitry is used to toggle back and forth be-  
tween the two registers of the “FIFO” when shifting. At any  
given time, one register is selected for shift operations. The  
other holds its previous state or can accept new parallel  
data. Shift register selection changes due to the following  
two events:  
CNT32 enabled and at terminal count. This event is used  
to account for scan lengths which are not multiples of  
eight. When shift register selection changes due to this  
signal, any data remaining in the shift register is unused.  
AUTO TMS HIGH MODE . This feature is included in the  
TMS shifter/buffer block to improve the efficiency of the  
PSC100 in supporting shift operations within the 1149.1 de-  
vices connected to the SSI. Shifting data and instructions  
into 1149.1 compliant devices requires that their TAP control-  
lers be sequenced to the Shift-DR or Shift-IR states (see Fig-  
ure 8). Once in this state, shifting occurs by holding TMS low  
and clocking TCK. The last bit is shifted when the TAP con-  
troller transitions to the EXIT1 state. This transition requires  
a logic 1 on TMS. The Auto TMS High feature, enabled by  
setting bit 1 of Mode Register 0, automatically creates a logic  
CNT3 in TCK Control signals that 8 bits have been  
shifted. This event is used for basic toggling between  
each of the two shift registers.  
CNT32 enabled and at terminal count. This event is used  
to account for scan lengths which are not multiples of  
eight. When shift register selection changes due to this  
signal, any data remaining in the shift register is unused.  
PRPG MODE . By setting MODE1(4), the TDO Shifter/Buffer  
is reconfigured as a 32-bit PRPG (Pseudo Random Pattern  
Generator) using the primitive polynomial:  
=
1 on the TMS lines of the PSC100 when CNT32 1. Conse-  
quently, the last bit is shifted out without having to load spe-  
cific TMS data into the shifter/buffer.  
32  
+ X22 + X2 + X + 1  
=
F(X)  
X
The PSC100 was developed to support both 1149.1 and  
non-1149.1 serial test methodologies. Since 1149.1 compli-  
ant devices include boundary scan registers on control pins  
(i.e. OE), which must remain fixed during boundary scan in-  
terconnect testing, generating pseudo-random patterns with  
PRPG mode provides limited usefulness for boundary scan  
test operations. PRPG mode may provide usefulness in  
other serial test or non-test related implementations which  
do not require fixed bits in the serial chain.  
Note: Auto TMS High mode creates a logic 1 on both TMS lines (i.e., TMS0  
and TMS1). Therefore, when using the Auto TMS High feature, all  
1149.1 devices connected to the TMS line not participating in the cur-  
rent JTAG test operations should be placed in the Test-Logic-Reset  
TAP controller state to prevent inadvertent TAP controller transitions.  
TDO SHIFTER/BUFFER  
The TDO Shifter/Buffer block diagram is shown in Figure 11.  
This block takes parallel data and serializes it for shift opera-  
tions through the serial port pin TDO. During normal shift  
modes, double-buffering is achieved by configuring the  
shifter/buffer as a 2 x 8 FIFO. This block can also be config-  
ured as a 32-bit Pseudo Random Pattern Generator (PRPG)  
Figure 12 shows a block diagram of the Linear Feedback  
Shift Register hookup.  
www.national.com  
12  
Serial Scan Interface (SSI) (Continued)  
DS100325-11  
FIGURE 11. TDO Shifter/Buffer Block Diagram Register Hookup  
DS100325-14  
FIGURE 12. TDO PRPG Block Diagram  
The PRPG is loaded by four PPI writes to the TDO address.  
When the PRPG enable bit is set, a pulse is issued internally  
that initializes the local parallel load logic such that the  
PRPG is loaded sequentially, least significant byte first, most  
significant byte last. When in PRPG mode, writes can be  
completed at any time; however, shift operations will be dis-  
abled until the PRPG is fully loaded.  
ture can be used for read-only scan operations where data is  
shifted into TDI while returning the scan chain to its previous  
state when shifting is completed. It can also be used to by-  
pass PSC100 devices connected within a boundary scan  
chain (i.e., a PSC100 located within a chain, but not provid-  
ing the JTAG TAP data). Loop around has limited usefulness  
in most boundary scan applications since, typically, data in  
the scan chain is shifted out and evaluated as new data is  
shifted into the chain for the next test.  
LOOP AROUND MODE . This mode, enabled by setting bit  
0 in Mode Register 0, will cause data appearing at the TDI in-  
put to be placed directly back on the TDO output. This fea-  
13  
www.national.com  
TDI is enabled with its respective mode bit.  
TDI shifter/buffer is not full.  
Serial Scan Interface (SSI) (Continued)  
WRITING A PARTIAL BYTE TO THE TMS0, TMS1 OR  
TDO SHIFTER/BUFFER. Since the TMS0, TMS1 and TDO  
shifter/buffers shift out least significant bit first, the valid  
(meaningful) bits within a partial byte (i.e., byte containing  
TCK is enabled according to the logic in TCK Control.  
Local select circuitry is used to toggle back and forth be-  
tween the two registers of the “FIFO” when shifting. At any  
given time, one register is selected for shift operations. The  
other holds its previous state or can accept new serial data.  
Shift register selection changes due to the following two  
events:  
<
8 valid bits to be shifted to the scan chain) must be stored  
and written into the shifter/buffer as the least significant bits.  
This will assure that the desired bits will be accurately shifted  
to the boundary scan chain. For example, moving the TAP  
controllers within the boundary scan chain connected to  
TMS0 from the Pause-DR state to the Run-Test/Idle state re-  
quires a 3-bit (110) sequence on TMS0. To provide correct  
3-bit sequence on TMS0, the partial byte would be written to  
the TMS0 shifter/buffer as:  
CNT3 in TCK Control signals that 8 bits have been  
shifted in. This event is used for basic toggling between  
each of the two shift registers.  
CNT32 enabled and at terminal count. This event is used  
to account for scan lengths which are not multiples of  
eight. When shift register selection changes due to this  
MSB  
LSB  
<
signal, a partial byte (i.e., byte with eight valid data bits  
x
x
x
x
x
0
1
1
TMS0  
shifted from the scan chain) will exist in the correspond-  
ing shift register. The embedded test software functions  
written to support the evaluation of data read from the  
TDI shifter/buffer must consider bit placement when  
reading and evaluating a partial byte.  
A subsequent enable and load of CNT32 with decimal 3 and  
enable of the TMS0 shifter/buffer will initialize the shift opera-  
tion. Terminal count on CNT32 will complete the shift opera-  
tion. Since terminal count on CNT32 will cause the register  
selection to change within the shifter/buffer, the values la-  
beled as “x” will not be used and are treated as “don’t cares”.  
READING A PARTIAL BYTE FROM THE TDI SHIFTER/  
BUFFER. Data is shifted from the scan chain into each TDI  
register from most significant bit to least significant bit. Con-  
sequently, the valid (i.e., meaningful) bits in a partial byte  
shifted into a TDI register will reside in the upper significant  
bit locations. For example, if a scan operation involves shift-  
ing and evaluating 53 bits returning to TDI, TDI shifter/buffer  
must be read 7 times (i.e., 6 full bytes plus a partial byte con-  
taining 5 meaningful bits). If the last 5 bits shifted back to the  
TDI shifter/buffer are 11010, then upon completion of the  
shift operation (i.e., terminal count on CNT32), the shift reg-  
ister within the TDI shifter/buffer will contain the following  
partial byte:  
TDI SHIFTER/BUFFER  
The TDI Shifter/Buffer block diagram is shown in Figure 13.  
This block shifts in serial data from the TDI port and puts it in  
parallel form for read operations at the PPI. During normal  
shift modes, double-buffering is achieved by configuring the  
shifter/buffer as a 2 x 8 FIFO. This block can also be config-  
ured as a 16-bit Serial Signature Compactor (SSC). Write,  
read, and shift operations are controlled by a local state ma-  
chine that accepts stimulus from the PPI, Mode Registers,  
CNT32 and the TCK Control section. The TDI input always  
shifts in data on the rising edge of SCK. The order of shifting  
is least significant bit first. The TDI input includes a pull-up  
resistor to force a logic 1 when the test data signal returning  
from the scan chain is floating.  
MSB  
0
LSB  
TDI  
1
0
1
1
x
x
x
Following a read of a partial byte, the embedded test soft-  
ware must adjust the position of the valid bits read from the  
TDI shifter/buffer or the position of the expected data to as-  
sure that an accurate comparison is made (and the  
non-meaningful bits are masked).  
Read operations are completed if the shifter/buffer is not  
empty and SSC mode is not enabled. Otherwise they are ig-  
nored. Write operations are only possible while in SSC  
mode. Otherwise they are ignored.  
Shifting occurs when the following conditions are all true:  
www.national.com  
14  
Serial Scan Interface (SSI) (Continued)  
DS100325-8  
FIGURE 13. TDI Shifter/Buffer Block Diagram  
DS100325-13  
FIGURE 14. TDI SSC Block Diagram  
SSC MODE. By setting MODE1(3), the TDI Shifter/Buffer is  
reconfigured as a 16-bit SSC (Serial Signature Compactor)  
using the primitive polynomial:  
provide masking capabilities and, therefore, provides limited  
usefulness for boundary scan test operations. SSC mode  
may provide usefulness in other serial test or non-test re-  
lated implementations which contain predictable data return-  
ing into the TDI shifter/buffer.  
16  
+ X12 + X3 + X + 1  
=
F(X)  
X
Within a chain of 1149.1 compliant devices, there are typi-  
cally one or more input pins which are driven by uncontrolled  
signals (i.e., signals which are not driven to known logic lev-  
els during a boundary scan CAPTURE operation). These  
signals are masked during the evaluation of data returning  
from the scan chain. The SSC within the PSC100 does not  
Figure 14 shows a block diagram of the Linear Feedback  
Shift Register hookup.  
The SSC is loaded by two PPI writes to the TDI address.  
When the SSC enable bit is set, a pulse is issued internally  
that initializes the local parallel load logic such that the SSC  
15  
www.national.com  
dress or the CNT32 write control logic will be re-initialized.  
This re-initialization will result in a partially filled counter with  
an undesired value. CNT32 is reset each time the counter  
hits terminal count or by asserting the RST pin. A synchro-  
nous reset condition (setting Mode2(1)) does not reset the  
counter and a new value must be written to CNT32 to pro-  
vide the desired number of TCK cycles.  
Serial Scan Interface (SSI) (Continued)  
is loaded sequentially, most significant byte first, least signifi-  
cant byte last. When in SSC mode, writes can be completed  
at any time; however, shift operations will be disabled until  
the SSC is fully loaded. PPI reads from TDI are ignored while  
in SSC mode.  
Upon leaving SSC mode an internal pulse causes the TDI  
shifter/buffer to be full. Also, local read select logic is initial-  
ized such that the signature is read most-significant byte  
first.  
SINGLE STEP MODE: All four 8-bit registers are readable  
for testability; however, there are no update latches similar to  
the ones used for the status bits. To stabilize the counter for  
read operations during on-board test, the Single Step Mode  
has been added. This allows the user to place CNT32 in any  
state and then count for one SCK cycle (TCK will not toggle  
when in singled step mode). The result can then be read  
from the PPI. The counter can be tested by loading it with  
values at its boundary conditions, and then clocking for one  
cycle to see the results. For example, the counter could be  
loaded with the value:  
TEST LOOP-BACK MODES. This feature provides a means  
for testing ’PSC100 functionality by looping data appearing  
at the output of an outgoing shifter/buffer (i.e., TMS0, TMS1  
or TDO) back to the input of the TDI shifter/buffer. The loop  
back function is accomplished with a simple multiplexer (see  
Figure 13) whose path selection is determined by setting the  
mode bits in MODE1(1:0). Loop back does not disable TCK  
or prevent shifting of data in the shifter/buffers to the scan  
chain(s) connected to the PSC100. Therefore, the state and  
operation of the TAP controllers within the scan chain(s)  
must be considered when developing Loop-Back test vec-  
tors to prevent undesired shifting of data or TAP controller  
transitions within the scan chain.  
00000001 00000000 00000000 00000000  
The next step is to set the Single Step Mode bit so that the  
counter counts down to the next state and stops. The next  
value is:  
00000000 11111111 11111111 11111111  
Four read cycles using the PPI will reveal the results of the  
test.  
32-BIT COUNTER (CNT32)  
Note: CNT32 will not wrap from terminal count (i.e., 00000000h decremented  
by 1 will remain unchanged and will not wrap to FFFFFFFFh). There-  
fore, CNT32 should be loaded with a non-zero value prior to a Single  
Step Mode Operation.  
CNT32 is a 32-bit, count-down binary counter arranged in  
four 8-bit segments. CNT32 can be loaded independent of  
its enable/disable status. Loading requires four consecutive  
writes to its address (least significant byte first). These four  
writes must not be interleaved with writes to any other ad-  
TIMING WAVEFORMS  
DS100325-15  
FIGURE 15. Serial Scan Interface Timing  
www.national.com  
16  
Embedded Test Software Support  
A SCANPSC100 device driver is provided by National to  
supply functions for performing write, read and shift opera-  
tions. National also offers a suite of software tools (called  
SCAN EASE) which enables ATPG or custom generated test  
vectors to be embedded, applied and evaluated within an  
IEEE 1149.1 compatible system. SCAN EASE is written to  
run on a wide range of processor and memory architectures.  
SCAN EASE includes the source code (ANSI C) and is  
modular to allow user modification based on application spe-  
cific needs.  
17  
www.national.com  
Absolute Maximum Ratings (Note 7)  
ESD Last Passing Voltage (Min)  
4000V  
If Military/Aerospace specified devices are required,  
please contact the National Semiconductor Sales Office/  
Distributors for availability and specifications.  
Recommended Operating  
Conditions  
Supply Voltage (VCC  
)
−0.5V to +7.0V  
Supply Voltage (VCC  
)
DC Input Diode Current (IIK  
)
’PSC100F  
4.5V to 5.5V  
0V to VCC  
0V to VCC  
=
VI −0.5V  
−20 mA  
+20 mA  
Input Voltage (VI)  
=
VI VCC + 0.5V  
Output Voltage (VO  
)
DC Input Voltage (VI)  
−0.5V to VCC +0.5V  
Operating Temperature (TA)  
Military  
DC Output Diode Current (IOK  
)
−55˚C to +125˚C  
125 mV/ns  
=
VO −0.5V  
−20 mA  
+20 mA  
Minimum Input Edge Rate dV/dt  
SCAN “F” Series Devices  
VIN from 0.8V to 2.0V  
=
VO VCC + 0.5V  
DC Output Voltage (VO  
)
−0.5V to VCC + 0.5V  
±
±
DC Output Source/Sink Current (IO  
DC VCC or Ground Current  
per Output Pin  
)
50 mA  
50 mA  
@
VCC 4.5V, 5.5V  
Note 7: Absolute maximum ratings are those values beyond which damage  
to the device may occur. The databook specifications should be met, without  
exception, to ensure that the system design is reliable over its power supply,  
temperature, and output/input loading variables. National does not recom-  
mend operation of SCAN outside of recommended operation conditions.  
±
DC Latchup Source or Sink Current  
Junction Temperature  
Ceramic  
300 mA  
+175˚C  
Storage Temperature  
−65˚C to +150˚C  
DC Electrical Characteristics  
Military  
=
Symbol  
Parameter  
VCC  
(V)  
TA −55˚C to  
Units  
Conditions  
+125˚C  
Guaranteed Limits  
=
VIH  
VIL  
Minimum High  
Input Voltage  
Maximum Low  
Input Voltage  
Minimum High  
Output Voltage  
4.5  
5.5  
4.5  
5.5  
4.5  
5.5  
4.5  
5.5  
2.0  
2.0  
V
V
V
VOUT 0.1V or  
VCC − 0.1V  
=
0.8  
VOUT 0.1V or  
0.8  
VCC − 0.1V  
=
IOUT −50 µA  
VOH  
4.4  
5.4  
=
VIN VIL or VIH  
3.70  
4.70  
=
V
V
IOH −24 mA  
All Outputs Loaded  
=
IOUT −50 µA  
VOL  
Maximum Low  
Output Voltage  
4.5  
5.5  
4.5  
5.5  
0.1  
0.1  
=
VIN VIL or VIH  
0.50  
0.50  
=
V
IOL 24 mA  
All Outputs Loaded  
=
±
IIN  
Maximum Input  
Leakage Current  
5.5  
5.5  
5.5  
1.0  
µA  
VIN VCC for TDI, OE  
=
VIN VCC, GND for  
All Others  
=
IILR  
Maximum Input  
Leakage Current  
Minimum  
−385  
50  
µA  
VIN GND for  
TDI, OE Only  
=
IOLD  
VOLD 1.65V Max  
=
Dynamic  
mA  
Maximum Test Duration  
2.0 ms, One Output  
Loaded at a Time  
Output Current  
=
VOHD 3.85V Min  
IOHD  
Minimum  
=
Dynamic  
5.5  
−50  
mA  
Maximum Test Duration 2.0 ms,  
Output Current  
One Output Loaded at a Time  
www.national.com  
18  
DC Electrical Characteristics (Continued)  
Military  
=
Symbol  
IOZ  
Parameter  
VCC  
(V)  
TA −55˚C to  
Units  
µA  
Conditions  
+125˚C  
Guaranteed Limits  
=
±
Maximum  
10.0  
11.0  
VIN VCC, GND  
VIN (OE , R/W , CE , STB)  
VIL, VIH  
=
TRI-STATE  
Leakage Current  
Maximum I/O  
Leakage Current  
=
±
IOZT  
µA  
VIN VCC, GND  
=
VO VCC, GND  
IN(R/W , CE , STB) VIL, VIH  
=
V
ICC  
Maximum  
Quiescent  
5.5  
5.5  
5.5  
160  
920  
1.60  
µA  
µA  
TDI, OE Float  
Supply Current  
ICCmax  
Maximum  
Quiescent  
TDI, OE low  
Supply Current  
Maximum  
=
ICCT  
mA  
VIN VCC − 2.1V  
I
CC/Input  
Maximum  
CC/Input  
Float: TDI, OE  
=
ICCTR  
VIN VCC −2.1V  
I
5.5  
1.65  
mA  
TDI and OE Only  
Float Untested Pin  
Note 8: This parameter is not directly testable, but is derived for use in Total Power Supply calculations.  
AC Electrical Characteristics/Operating Requirements  
Symbol  
Parameter  
VCC  
Military  
Units  
Fig.  
No.  
(V)  
=
TA −55˚C to +125˚C  
(Note 9)  
=
CL 50 pF  
Min  
5.5  
5.5  
6.5  
7.0  
7.5  
9.0  
10.0  
2.0  
1.5  
1.5  
Max  
19.5  
21.0  
24.0  
24.0  
39.0  
34.0  
36.0  
13.0  
10.0  
16.5  
PARALLEL PROCESSOR INTERFACE (PPI)  
tpd1  
Prop Delay  
R/W to RDY  
Prop Delay  
A to RDY  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
2, 3, 4  
2, 3  
tpd2  
tpd3  
Prop Delay  
A to D  
3
tpd4  
Prop Delay  
STB to RDY  
Prop Delay  
SCK to D  
2–4, 6  
2–4, 6  
2–4, 6  
2–4, 6  
2, 3  
t pd10  
t pd11  
t pd12  
ten1  
Prop Delay  
SCK to INT  
Prop Delay  
SCK to RDY  
Enable Time  
CE to RDY  
Disable Time  
CE to RDY  
Enable Time  
CE to D  
tdis1  
2, 3  
ten2  
3
19  
www.national.com  
AC Electrical Characteristics/Operating Requirements (Continued)  
Symbol  
Parameter  
VCC  
(V)  
Military  
TA −55˚C to +125˚C  
Units  
Fig.  
No.  
=
(Note 9)  
=
CL 50 pF  
Min  
2.5  
3.0  
3.0  
3.0  
2.5  
0.5  
1.0  
4.5  
5.0  
4.5  
0.0  
1.0  
0.5  
7.5  
Max  
14.5  
17.5  
16.0  
16.0  
14.5  
PARALLEL PROCESSOR INTERFACE (PPI)  
tdis2  
ten3  
tdis3  
ten  
tdis  
th1  
ts1  
th2  
ts2  
th3  
ts3  
th  
Disable Time  
CE to D  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
3
3
Enable Time  
R/W to D  
Disable Time  
R/W to D  
3
Enable Time  
STB to D  
3
Disable Time  
STB to D  
3
Hold Time,  
R/W to STB  
Setup Time  
R/W to STB  
Hold Time,  
2–4, 6  
2–4, 6  
2, 3  
2, 3  
2
A to STB  
Setup Time,  
A to STB  
Hold Time,  
D to STB  
Setup Time,  
D to STB  
2
Hold Time,  
CE to STB  
Setup Time,  
CE to STB  
Setup Time,  
2–4, 6  
2–4, 6  
4–7  
ts  
ts4  
STB to  
SCK  
th4  
Hold Time,  
5.0  
0.0  
ns  
ns  
4–7  
4–7  
STB to  
SCK ↓  
th  
Hold Time,  
STB to RDY  
5.0  
5.0  
0.0  
tW  
Clock Pulse  
Width, SCK, H  
or L  
20.0  
ns  
ns  
4, 6  
4, 6  
tW1  
Pulse Width  
STB (H or L)  
Maximum  
5.0  
5.0  
6.0  
25  
fmax  
MHz  
Frequency  
Clock  
www.national.com  
20  
AC Electrical Characteristics/Operating Requirements (Continued)  
Symbol  
Parameter  
VCC  
(V)  
Military  
TA −55˚C to +125˚C  
Units  
Fig.  
No.  
=
(Note 9)  
=
CL 50 pF  
Min  
3.0  
5.5  
4.5  
3.0  
1.5  
2.0  
1.5  
0.5  
Max  
14.0  
19.5  
18.5  
13.5  
10.0  
13.0  
11.0  
SERIAL SCAN INTERFACE (SSI)  
tpd5  
tpd6  
tpd7  
tpd8  
tpd9  
ten4  
tdis4  
th5  
Prop Delay  
SCK to TCK  
Prop Delay  
SCK to TDO  
Prop Delay  
SCK to TMS  
Prop Delay  
FRZ to TCK  
Prop Delay  
TDI to TDO  
Enable Time  
OE to JTAG  
Disable Time  
OE to JTAG  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
15  
15  
15  
15  
15  
14  
14  
15  
Setup Time, H  
or L, TDI to  
SDK (Note  
10)  
ns  
ns  
ts5  
Hold Time, H  
or L, TDI to  
SDK (Note  
10)  
5.0  
7.5  
15  
RST RELATED TIMING  
tpd  
Prop Delay  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
5.0  
8.0  
8.0  
7.0  
5.5  
5.5  
6.5  
1.0  
29.5  
31.0  
28.0  
21.0  
20.0  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
RST to D  
tpd  
Prop Delay  
RST to RDY  
Prop Delay  
RST to INT  
Prop Delay  
RST to TDO  
Prop Delay  
RST to TMS  
Pulse Width  
RST (L)  
tphl  
tplh  
tplh  
tWR  
tREC  
Recovery  
Time  
SCK from  
RST  
±
Note 9: Voltage Range 5.0 is 5.0V 0.5V.  
Note 10: SSC Mode  
21  
www.national.com  
SHIFT-ENABLE signal which is generated internal to the  
’PSC100, based on the status of the TDI buffer and the  
Mode Registers.  
Application Note  
SCK MINIMUM PULSE WIDTH CALCULATION  
We now see the three major timing components which limit  
the duration of the SCK pulse width low. There are two minor  
additional delays which should be noted. The TCK signal  
from the ’PSC100 needs to arrive at the target device to be  
recognized, and this takes a finite amount of time depending  
on the signal trace length and impedance. Similarly, the TDO  
signal of the last target in the chain needs to reach the TDI  
pin of the ’PSC100, taking a finite amount of time as well.  
These two trace delays can be minimized by making the tar-  
get device closest to the ’PSC100 the last device in the  
chain. See Figure 17.  
The SCANPSC100 Parallel to Serial Converter is intended  
to act as the interface between a processor and an IEEE  
1149.1 boundary scan chain. When used in this configura-  
tion, there is a critical timing situation that is not obvious.  
This timing involves the system clock rate at which data from  
the scan ring is being read into the ’PSC100’s TDI pin (target  
TAP controllers in SHIFT-DR or SHIFT-IR states).  
To fully understand the events which are taking place during  
this critical period, it is useful to view the waveforms of inter-  
est as they relate in time. See Figure 16. The TCK is derived  
internally to the ’PSC100 based on the system clock (SCK)  
and clock gating control. The result is that when TCK is run-  
ning, it is at the same frequency as SCK but delayed in time  
by the SCK-TCK propagation delay.  
PROGRAMMING RESTRICTIONS  
Because certain mode bits enable shift operations for certain  
functions, these mode bits should not be changed when shift  
operations are in progress. The alignment of all registers  
during shift operations is controlled by a three bit counter in  
the TCK control block. Enabling or disabling a function in the  
middle of a shift operation may disrupt the logic necessary to  
keep all shifter/buffers byte-aligned. For example, if the TDO  
shifter/buffer (already loaded) is enabled while the three bit  
counter value is three, the shifter/buffer will only shift out 5  
bits of the first byte loaded.  
The TCK signal from the ’PSC100 drives all of the IEEE  
1149.1 target devices. On the rising edge of TCK, data  
present at each scan cell is clocked into it. On the falling  
edge, this data is presented at the output of the same scan  
cell for the next adjacent cell to read. With regards to the last  
cell in a particular target, the falling edge of TCK presents  
the data in the last scan cell to the TDO pin, a TCK-TDO  
propagation delay later.  
At the ’PSC100, data shifted in through the TDI pin is  
clocked in on the rising edge of SCK, not TCK. The reason  
for this is that TCK is generated internal to the ’PSC100 and  
intended to control the boundary scan targets. The ’PSC100  
is controlled by SCK, therefore the signal to be shifted into  
the TDI pin needs to be referenced to SCK not TCK. New  
TDI data must be present a TDI-SCK set-up time prior to the  
rising edge of SCK in order to guarantee validity. Although  
SCK is usually continuous, the TDI buffer is controlled by a  
The following bits should not be changed when shift opera-  
tions are in progress, i.e., when TCK is enabled (see TCK  
control section):  
MODE0(7:3)  
MODE1(4:3)  
MODE2(0)  
DS100325-16  
=
Pw(L) minimum Tpd1 + Td1 + Tpd2 + Td2 + Tsu  
FIGURE 16. System Clock Timing for Accurate TDI Data  
www.national.com  
22  
Application Note (Continued)  
DS100325-17  
Note:- - - - - Minimize the lengths of these two traces.  
FIGURE 17. SCANPSC100 Location Relative to Targets  
23  
www.national.com  
Physical Dimensions inches (millimeters) unless otherwise noted  
28-Pin Leadless Chip Carrier (LCC)  
NS Package Number E28A  
28-Pin Ceramic DIP  
NS Package Number J28A  
28-Lead Cerpack  
NS Package Number WA28D  
www.national.com  
24  
Notes  
LIFE SUPPORT POLICY  
NATIONAL’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT  
DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT AND GENERAL  
COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein:  
1. Life support devices or systems are devices or  
systems which, (a) are intended for surgical implant  
into the body, or (b) support or sustain life, and  
whose failure to perform when properly used in  
accordance with instructions for use provided in the  
labeling, can be reasonably expected to result in a  
significant injury to the user.  
2. A critical component is any component of a life  
support device or system whose failure to perform  
can be reasonably expected to cause the failure of  
the life support device or system, or to affect its  
safety or effectiveness.  
National Semiconductor  
Corporation  
Americas  
Tel: 1-800-272-9959  
Fax: 1-800-737-7018  
Email: support@nsc.com  
National Semiconductor  
Europe  
National Semiconductor  
Asia Pacific Customer  
Response Group  
Tel: 65-2544466  
Fax: 65-2504466  
National Semiconductor  
Japan Ltd.  
Tel: 81-3-5639-7560  
Fax: 81-3-5639-7507  
Fax: +49 (0) 1 80-530 85 86  
Email: europe.support@nsc.com  
Deutsch Tel: +49 (0) 1 80-530 85 85  
English Tel: +49 (0) 1 80-532 78 32  
Français Tel: +49 (0) 1 80-532 93 58  
Italiano Tel: +49 (0) 1 80-534 16 80  
Email: sea.support@nsc.com  
www.national.com  
National does not assume any responsibility for use of any circuitry described, no circuit patent licenses are implied and National reserves the right at any time without notice to change said circuitry and specifications.  
1 of 4  
National Semiconductor Company  
Design Purchasing Quality Company Jobs  
Products > Military/Aerospace > Logic > SCAN > SCANPSC100F  
Product Folder  
SCANPSC100F  
Embedded Boundary Scan Controller  
Generic P/N 100F  
Contents  
l
l
l
l
General Description  
Features  
Datasheet  
Package Availability, Models, Samples  
& Pricing  
l
Application Notes  
General Description  
The SCANPSC100F is designed to interface a generic parallel processor bus to a serial scan  
test bus. It is useful in improving scan throughput when applying serial vectors to system test  
circuitry and reduces the software overhead that is associated with applying serial patterns  
with a parallel processor. The 'PSC100F operates by serializing data from the parallel bus for  
shifting through the chain of 1149.1 compliant components (i.e., scan chain). Scan data  
returning from the scan chain is placed on the parallel port to be read by the host processor.  
Up to two scan chains can be directly controlled with the 'PSC100F via two independent  
TMS pins. Scan control is supplied with user specific patterns which makes the 'PSC100F  
protocol-independent. Overflow and underflow conditions are prevented by stopping the test  
2 of 4  
clock. A 32-bit counter is used to program the number of TCK cycles required to complete a  
scan operation within the boundary scan chain or to complete a 'PSC100F Built-In Self Test  
(BIST) operation. SCANPSC100F device drivers and 1149.1 embedded test application code  
are available with National's SCANEase software tools.  
Features  
l
l
l
Compatible with IEEE Std. 1149.1 (JTAG) Test Access Port and Boundary Scan  
Architecture  
Supported by National's SCAN Ease (Embedded Application Software Enabler)  
Software  
Uses generic, asynchronous processor interface; compatible with a wide range of  
processors and PCLK frequencies  
l
l
l
l
l
l
l
Directly supports up to two 1149.1 scan chains  
16-bit Serial Signature Compaction (SSC) at the Test Data In (TDI) port  
Automatically produces pseudo-random patterns at the Test Data Out (TDO) port  
Fabricated on FACT™ 1.5 µm CMOS process  
Supports 1149.1 test clock (TCK) frequencies up to 25 MHz  
TTL-compatible inputs; full-swing CMOS outputs with 24 mA source/sink capability  
Standard Microcircuit Drawing (SMD) 5962-9475001  
Datasheet  
Size  
(in  
Kbytes)  
Title  
Date  
Receive via  
Email  
View  
Online  
Download  
SCANPSC100F Embedded Boundary Scan Controller(IEEE 1149.1  
Support)  
16-Sep-  
99  
364 Kbytes  
View Online Download Receive via Email  
Please use Adobe Acrobat to view PDF file(s).  
If you have trouble printing, see Printing Problems.  
3 of 4  
Package Availability, Models, Samples & Pricing  
Samples  
Package  
Budgetary Pricing  
BSDL  
Model  
File  
Std  
Pack  
Size  
&
Package  
Marking  
Part Number  
Status  
#
$US  
Quantity  
Electronic  
Orders  
Type  
pins  
each  
tube  
of  
N/A  
[logo]¢Z¢S¢4¢A$E  
SCANSTA101  
ME  
SCANSTA101ME  
SSOP  
48  
Preliminary  
N/A  
N/A  
.
.
[logo]¢Z¢S¢4¢A  
SCANPSC  
100FLMQB/Q  
$E 5962-  
tray  
Full  
production  
5962-9475001Q3A  
LCC  
28  
250+ $38.5000 of  
25  
9475001Q3A  
tube  
[logo] ¢Z¢S¢4¢A$E  
Full  
production  
5962-9475001QXA  
SCANPSC100FMW  
Cerdip  
28  
28  
N/A  
N/A  
.
.
250+ $30.5000 of SCANPSC100FDMQB /Q  
12  
5962-9475001QXA  
SOIC  
WIDE  
[logo]¢Z¢S¢4¢A$E  
SCANPSC100FMW  
Preliminary  
N/A  
[logo] ¢Z¢S¢4¢A$E  
SCANPSC100FFMQB  
5962-9475001QYA  
/Q  
tube  
Full  
production  
5962-9475001QYA  
Cerpack  
Cerquad  
28  
48  
N/A  
N/A  
.
.
250+ $32.9000 of  
14  
tube  
of  
N/A  
[logo]¢Z¢S¢4¢A$E  
SCANSTA101W-  
QML  
SCANSTA101W-  
QML  
Preliminary  
Application Notes  
4 of 4  
Size  
(in Kbytes)  
Title  
Date  
Receive via Email  
Download  
View Online  
Please use Adobe Acrobat to view PDF file(s).  
If you have trouble printing, see Printing Problems.  
[Information as of 2-Sep-2000]  
Quick Search  
Parametric  
System  
Diagrams  
Product  
Tree  
Home  
Search  
About Languages . About the Site . About "Cookies"  
National is QS 9000 Certified . Privacy/Security  
Copyright © National Semiconductor Corporation  
Preferences . Feedback  

相关型号:

SCANPSC100FSC

Embedded Boundary Scan Controller (IEEE 1149.1 Support)
FAIRCHILD

SCANPSC100FSC

SPECIALTY MICROPROCESSOR CIRCUIT, PDSO28, 0.300 INCH, MS-013, SOIC-28
ROCHESTER

SCANPSC100FSCX

SPECIALTY MICROPROCESSOR CIRCUIT, PDSO28, 0.300 INCH, MS-013, SOIC-28
ROCHESTER

SCANPSC110

SCAN Bridge Hierarchical and Multidrop Addressable JTAG Port (IEEE1149.1 System Test Support)
NSC

SCANPSC110F

SCAN Bridge Hierarchical and Multidrop Addressable JTAG Port (IEEE1149.1 System Test Support)
NSC

SCANPSC110F

SCAN Bridge Hierarchical and Multidrop Addressable JTAG Port (IEEE1149.1 System Test Support)
FAIRCHILD

SCANPSC110FDM

暂无描述
NSC

SCANPSC110FDMQB

SCAN Bridge Hierarchical and Multidrop Addressable JTAG Port (IEEE1149.1 System Test Support)
NSC

SCANPSC110FE-QV

SPECIALTY MICROPROCESSOR CIRCUIT, CQCC28, LCC-28
TI

SCANPSC110FFMQB

SCAN Bridge Hierarchical and Multidrop Addressable JTAG Port (IEEE1149.1 System Test Support)
NSC

SCANPSC110FFMQB/NOPB

SPECIALTY TELECOM CIRCUIT, CDFP28
TI

SCANPSC110FLMQB

SCAN Bridge Hierarchical and Multidrop Addressable JTAG Port (IEEE1149.1 System Test Support)
NSC