TMC3211G1C [FAIRCHILD]

Divider, 32-Bit, CMOS, CPGA120, CERAMIC, PGA-120;
TMC3211G1C
型号: TMC3211G1C
厂家: FAIRCHILD SEMICONDUCTOR    FAIRCHILD SEMICONDUCTOR
描述:

Divider, 32-Bit, CMOS, CPGA120, CERAMIC, PGA-120

时钟 外围集成电路
文件: 总10页 (文件大小:90K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
www.fairchildsemi.com  
TMC3 2 1 1  
In t e g e r Divid e r  
3 2 -Bit , 2 0 MOP S  
Data is input on separate busses, and quotients are available  
on a 32-bit output bus with synchronous three-state enable.  
All data inputs and outputs are registered and TTL compati-  
ble. All input and output signal timing is referenced to the  
rising edge of Clock.  
Features  
• 32-bit by 16-bit fixed-point integer division with 32-bit  
quotient  
• 20 MHz clock rate and pipelined throughput rate  
• Three-bus I/O architecture allows unrestricted throughput  
• Easy system interfacing  
• Status flags for divide-by-zero and inexact result  
• All inputs and outputs TTL compatible  
The TMC3211 has a single system clock and separate load  
enable controls for the dividend and divisor registers. This  
allows the device to be used in applications requiring divi-  
sion by a constant. Underflow automatically produces the  
expected zero quotient, and dividing by zero sets a divide-  
by-zero output flag.  
Applications  
• Graphics and image processors  
• Matrix operations and geometric transforms  
• Perspective extraction  
• Radar signal processing  
• Range scaling  
The internal architecture of the TMC3211 allows all 32-bit  
two’s complement integer dividends and nonzero 16-bit  
two’s complement integer divisors, without prenormaliza-  
tion. The output quotient format is 32-bit integer.  
Description  
The TMC3211 makes a full-precision, full-speed divide  
function available to designers of workstations, image  
processors, and radar systems who need to perform  
perspective extractions, matrix operations, range scaling, and  
other complex functions.  
The TMC3211 is a fast monolithic two’s complement integer  
divider which can divide a 32-bit dividend by a 16-bit divisor  
to produce a 32-bit quotient, with a maximum pipelined  
throughput of 20 MOPS (Million Operations Per Second).  
Block Diagram  
X
X
15-0  
16  
31-0  
32  
ENY  
ENX  
CLK  
Divisor  
16  
Dividend  
32  
1
16 - Stage  
2-18 Non-Restoring  
Divider  
Flags  
32  
19  
Quotients  
32  
20  
OE  
32  
2
Q31-0  
DZERO, INX  
REV. 1.0.1 5/9/00  
PRODUCT SPECIFICATION  
TMC3211  
The 32-bit parallel quotient output register includes  
Functional Description  
three-state output drivers with synchronous enable control,  
which permits multiple TMC3211s to be operated in parallel  
or connected directly to a system bus.  
General Information  
The TMC3211 consists of input registers, a pipelined array  
divider, and output (quotient) registers. The 16-bit divisor  
and 32-bit dividend input registers can each be loaded inde-  
pendently using the two synchronous load enable controls.  
The divider is a 16-stage pipelined non-restoring array which  
produces a 32-bit quotient and condition flags which indicate  
an attempted division by zero, or operations which yield a  
non-zero remainder or inexact result.  
The TMC3211 requires a total of 19 clock cycles to generate  
a full 32-bit quotient result. Once the internal pipeline is full,  
a new quotient is available at the output every clock cycle.  
Pin Assignments  
N M  
L K J H G F E D C B A  
Pin  
Name  
Pin  
Name  
Pin  
Name  
Pin  
Name  
1
2
A1  
GND  
C5  
C6  
C7  
C8  
C9  
C10  
Y
Y
Y
Y
G11  
G12  
G13  
H1  
Q
L10  
L11  
L12  
L13  
M1  
M2  
M3  
M4  
M5  
M6  
M7  
M8  
M9  
M10  
M11  
M12  
M13  
N1  
Q
27  
11  
8
17  
A2  
Y
Y
V
Y
Y
Y
Y
V
V
DD  
14  
13  
DD  
9
DD  
3
A3  
Q
GND  
4
16  
24  
25  
4
A4  
Y
Y
Q
0
24  
31  
0
5
A5  
Q
Q
H2  
Y
X
X
X
X
X
X
X
1
5
6
A6  
H3  
GND  
GND  
TOP VIEW  
CAVITY UP  
6
7
A7  
C11 GND  
C12 GND  
H11  
H12  
H13  
J1  
5
2
8
A8  
Q
Q
2
19  
18  
4
9
A9  
REM  
C13  
D1  
Q
9
6
10  
11  
12  
13  
A10  
A11  
A12  
A13  
B1  
Q
Q
Q
Y
Y
Y
Y
V
V
0
18  
17  
26  
9
D2  
J2  
3
27  
11  
14  
D3  
GND  
J3  
6
DD  
DD  
V
D11  
D12  
D13  
E1  
V
J11  
J12  
J13  
H1  
CLK  
DD  
DD  
YEN  
Q
Q
Q
Q
Q
30  
10  
11  
20  
19  
21  
20  
24  
25  
B2  
Y
V
Y
Y
Y
Y
Y
GND  
GND  
15  
DD  
12  
10  
7
B3  
Y
Y
Y
Y
B4  
E2  
H2  
Q
25  
B5  
E3  
GND  
GND  
H3  
GND  
GND  
GND  
B6  
E11  
E12  
E13  
F1  
H11  
H12  
H13  
L1  
N2  
X
X
X
X
X
X
X
21  
3
B7  
Q
Q
Q
N3  
3
12  
13  
19  
18  
30  
DD  
B8  
Q
N4  
1
5
B9  
DZ  
Y
Y
V
V
Y
V
N5  
22  
7
B10  
B11  
B12  
B13  
C1  
C2  
C3  
C4  
Q
Q
Q
Q
F2  
L2  
N6  
2
21  
10  
12  
13  
F3  
L3  
GND  
N7  
4
DD  
DD  
F11  
F12  
F13  
G1  
L4  
V
N8  
7
DD  
Q
Q
L5  
GND  
N9  
XEN  
OEQ  
8
14  
15  
23  
Y
V
V
L6  
X
V
X
N10  
N11  
N12  
N13  
16  
DD  
DD  
8
Y
L7  
Q
Q
Q
DD  
15  
29  
28  
26  
G2  
GND  
L8  
GND  
G3  
V
L9  
Q
31  
DD  
2
REV. 1.0.1 5/9/00  
TMC3211  
PRODUCT SPECIFICATION  
Pin Descriptions  
Signal Signal  
Pin Number  
Description  
Type  
Name  
Power  
VDD  
B3, A4, A13, D11, F11, G12, J11,  
K11, L11, L7, L4, L2, J3, G3, F3,  
C2, C3  
Supply Voltage, Ground. The TMC3211 operates on  
a single +5V supply. All power and ground lines must  
be connected.  
GND  
CLK  
A1, C4, C11, C12, E11, H11, L12,  
M12, M11, L5, L3, N1, K3, H3, G2,  
E3, D3  
Clock  
Inputs  
M9  
System Clock. The TMC3211 has a single Clock input.  
All input and output signal timing is referenced to the  
rising edge of Clock.  
Y31-0  
M1, L1, K2, K1, J2, J1, H2, H1, G1, Dividend Data. The 32-bit Dividend is presented  
F1, F2, E1, E2, D1, D2, C1, B2, A2, through the registered Y input port. Y31 is the sign bit.  
A3, B4, C5, B5, A5, C6, B6, A6, A7, The LSB is Y0.  
C7, B7, A8, B8, C8  
X15-0  
U8, M8, N8, N7, M7, N6, M6, L6,  
NS, MS, N4, M4, N3, M3, N2, M2  
Divisor Data. The 16-bit Divisor is presented through  
the registered X input port. X15 is the sign bit. The LSB  
is X0.  
Outputs Q31-0  
Controls YEN  
L9, M10, N11, N12, L10, N13, M13, Quotient Data. The current Quotient is available on the  
L13, K12, K13, J12, J13, H12, H13, registered Q output bus. Q31 is the sign bit. The LSB  
G11, G13, F13, F12, E13, E12,  
D13, D12, C13, B13, B12, A12,  
C10, B11, A11, B10, C9, A10  
is Q0.  
B1  
Dividend Write Enable. Data present at the Dividend  
input Y31-0 is latched into the input registers on the  
rising edge of clock when the enable control YEN is  
LOW.  
XEN  
OEQ  
N9  
Divisor Write Enable. Data present at the Divisor input  
X15-0 is latched into the input registers on the rising  
edge of clock when the enable control XEN is LOW.  
N10  
Quotient Output Enable. The quotient output bus Q31-0  
and flags DZ and REM are in the high-impedance state  
when the registered Output Enable OEQ is HIGH.  
When OEQ is LOW, they are enabled on the next clock  
cycle.  
Flags  
DZ  
B9  
A9  
Divide-By Zero Flag. When a zero divisor is input, the  
resulting invalid output quotient will be accompanied by  
a registered Divide-By-Zero Flag HIGH.  
REM  
Inexact Remainder Flag. Whenever a division  
operation leaves a nonzero remainder, the resulting  
qotient is accompanied by a registered nonzero  
Remainder Flag HIGH.  
No  
D4  
Index Pin (optional)  
Connect  
REV. 1.0.1 5/9/00  
3
PRODUCT SPECIFICATION  
TMC3211  
Data Formats  
Applications Discussion  
The TMC3211 supports fixed-point two’s complement data  
formats. By keeping track of the binary points of the input  
data, the user can then interpret the resulting quotient prop-  
erly. Two possible binary weightings of the input and output  
bits are as follows:  
Division Using A Constant  
By utilizing the separate input data register load enable con-  
trols, the TMC3211 can perform division by a constant. The  
data currently held remain in the input registers until updated  
by the user.  
Pin D31 D30 D29 D28 D27 D26 D25 D24 D23 D22 D21 D20 D19 D18 D17 D16 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0  
31  
30  
29  
28  
27  
26  
25  
24  
23  
22  
21  
20  
19  
18  
17  
16  
15  
14  
14  
14  
13  
13  
13  
12  
12  
12  
11  
11  
11  
10  
10  
10  
9
9
9
8
8
8
7
7
7
6
6
6
5
5
5
4
4
4
3
3
3
2
2
2
1
1
1
0
0
0
Y
X
Q
-2  
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
15  
-2  
31  
30  
29  
28  
27  
26  
25  
24  
23  
22  
21  
20  
19  
18  
17  
16  
15  
-2  
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Figure 1. Integer Data Format  
0
-1  
-2  
-3  
-4  
-5  
-6  
-7  
-8  
-9  
-10  
5
-11  
-12  
-13  
-14  
-15  
-16  
0
-17  
-1  
-18  
-2  
-19  
-3  
-20  
-4  
-21  
-5  
-22  
-6  
-23  
-7  
-24  
-8  
-25  
-9  
-26  
-10  
-11  
-27  
-11  
-12  
-28  
-12  
-13  
-29  
-13  
-14  
-30  
-14  
-15  
-31  
-15  
-16  
Y
-2  
.2  
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
X
Q
-2  
.2  
2
2
2
2
2
2
2
2
15  
14  
13  
12  
11  
10  
9
8
7
6
4
3
2
1
0
-1  
-2  
-3  
-4  
-5  
-6  
-7  
-8  
-9  
-10  
-2  
.2  
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Figure 2. Fractional Data Format  
where a leading minus sign indicates a sign bit.  
Inexact Results  
The flag REM is provided to indicate that the current quotient  
left a nonzero remainder and was truncated toward zero.  
Care must be taken when adopting fractional data formats.  
By observing the binary weighting applied to the input data  
in the dividend and divisor, the binary point of the quotient  
can then be correctly established. The difference lies only in  
constant scale factors, which must be considered in order to  
maintain a data format which is compatible with the bit  
weighting of the hardware system. The two most common  
choices are fractional and integer notation. If integer notation  
is used, the LSBs of the dividend, divisor, and quotient all  
have the same value. With fractional notation the MSBs are  
all of equal weight.  
Negative Full-Scale Overflow  
Due to a finite data word width, a two’s complement over-  
flow error occurs under the following unique condition:  
Divisor Y=80000000H (– Full-Scale)  
Dividend X=FFFFH (-1)  
Result:  
Quotient Q=80000000H (– Full-Scale)  
Divide by Zero  
The flag DZ indicates that the divisor input for the current  
calculation was a zero, independent of the dividend. Dividing  
by zero is an undefined operation yielding a meaningless  
quotient. Thus, this flag must be monitored to guard against  
possible errors.  
As stated above, this is due to a limitation in the number of bits  
available to indicate a positive full-scale quotient, and data  
overflows into the MSB position to indicate an incorrect sign.  
4
REV. 1.0.1 5/9/00  
TMC3211  
PRODUCT SPECIFICATION  
t
t
PWL  
PWH  
2
1
3
19  
20  
21  
CLK  
t
H
t
CY  
t
S
Y
Y
Y
2
Y
X
0
1
31-0  
Note 1  
X
X
1
15-0  
0
XEN  
YEN  
t
t
HO  
D
Q
Q
Q
2
2
0
1
Q
, DZ, REM  
31-0  
Notes:  
1. Demonstrates division by a constant, Q = Y /X .  
2
2
1
2. Assumes OEQ = Low.  
Figure 3. Timing Diagram  
V
V
DD  
DD  
n Substrate  
D1  
n Substrate  
p
p
D1  
p+  
p+  
Control  
Input  
Output  
1k  
n+  
D2  
n+  
D2  
n
n
p Well  
p Well  
GND  
Figure 5. Equivalent Output Circuit  
GND  
Figure 4. Equivalent Input Circuit  
REV. 1.0.1 5/9/00  
5
PRODUCT SPECIFICATION  
TMC3211  
Absolute Maximum Ratings (beyond which the device may be damaged)1  
Parameter  
Supply Voltage  
Input Voltage  
Output  
Min  
Max  
+7.0  
Units  
V
-0.5  
-0.5  
-0.5  
-3.0  
VDD + 0.5  
VDD + 0.5  
6.0  
V
Applied Voltage2  
Forced Current3,4  
V
mA  
sec  
°C  
°C  
°C  
°C  
Short-circuit duration (single output in HIGH state to ground)  
1
Temperature  
Operating, case  
Junction  
-60  
-65  
+130  
175  
Lead, soldering (10 seconds)  
Storage  
300  
+150  
Notes:  
1. Absolute maximum ratings are limiting values applied individually while all other parameters are within specified operating  
conditions. Functional operation under any of these conditions is NOT implied.  
2. Applied voltage must be current limited to specified range, and measured with respect to GND.  
3. Forcing voltage must be limited to specified range.  
4. Current is specified as conventional current flowing into the device.  
Operating Conditions  
Temperature Range  
Parameter  
Test Conditions  
Standard  
Nom.  
5.0  
Units  
Min.  
Max.  
5.25  
0.8  
VDD  
VIL  
VIH  
IOL  
IOH  
tCY  
tPWL  
tPWH  
tS  
Supply Voltage  
4.75  
V
V
Input Voltage, Logic LOW  
Input Voltage, Logic HIGH  
Output Current, Logic LOW  
Output Current, Logic HIGH  
Cycle Time  
2.0  
V
4.0  
-2.0  
50  
mA  
mA  
ns  
ns  
ns  
ns  
ns  
°C  
VDD = Min  
VDD = Min  
VDD = Min  
Clock Pulse Width, LOW  
Clock Pulse Width, HIGH  
Input Setup Time  
15  
15  
12  
6
tH  
Input Hold Time  
TA  
Ambient Temperature, Still Air  
0
70  
6
REV. 1.0.1 5/9/00  
TMC3211  
PRODUCT SPECIFICATION  
DC Characteristics within Specied Operating Conditions1  
Temperature Range  
Standard  
Parameter  
Test Conditions  
Units  
Min.  
Max.  
IDDQ  
IDDU  
IIL  
Supply Current, Quiescent  
VDD = Max, VIN = 0V  
5
mA  
mA  
µA  
µA  
V
Supply Current, Unloaded  
Input Current, Logic LOW  
Input Current, Logic HIGH  
Output Voltage, Logic LOW  
Output Voltage, Logic HIGH  
VDD = Max, OEQ = 5V, f = 20MHz  
VDD = Max, VIN = 0V  
150  
-10  
10  
IIH  
VDD = Max, VIN = VDD  
VDD = Min, lOL = Max  
VOL  
VOH  
IOZL  
0.4  
VDD = Min, lOH = Max  
2.4  
V
Hi-Z Output Leakage Current, VDD = Max, VIN = 0V  
Output LOW  
-40  
40  
µA  
IOZH  
IOS  
Hi-Z Output Leakage Current, VDD = Max, VIN = VDD  
Output HIGH  
µA  
Short-Circuit Output Current  
VDD = Max, Output HIGH, one pin to  
-150  
mA  
ground, one second duration max.  
Cl  
Input Capacitance  
Output Capacitance  
TA = 25°C, f = 1MHz  
10  
10  
pF  
pF  
CO  
TA = 25°C, f = 1MHz  
Note:  
1. Actual test conditions may vary from those shown, but guarantee operation as specified  
AC Characteristics within Specied Operating Conditions  
Temperature  
Standard  
Parameter  
Test Conditions  
Units  
Min  
Max  
tD  
Output Delay1  
Output Hold Time  
VDD = Min, CLOAD = 25pF  
VDD = Max, CLOAD = 25pF  
35  
ns  
ns  
tHO  
5
Note:  
1. Equivalent to t  
and t  
of the three-state outputs  
DIS  
ENA  
REV. 1.0.1 5/9/00  
7
PRODUCT SPECIFICATION  
TMC3211  
Mechanical Dimensions  
120-Pin Ceramic Pin Grid ArrayG1 Package  
Notes:  
Inches  
Millimeters  
Symbol  
Notes  
1. Pin #1 identifier shall be within shaded area shown.  
2. Pin diameter excludes solder dip finish.  
3. Dimension "M" defines matrix size.  
Min.  
Max.  
Min.  
Max.  
A
.080  
.040  
.125  
.016  
.160  
.060  
.215  
.020  
2.03  
1.01  
3.17  
0.40  
4.06  
1.53  
5.46  
0.51  
A1  
A2  
øB  
øB2  
D
4. Dimension "N" defines the maximum possible number of pins.  
5. Orientation pin is at supplier's option.  
2
2
6. Controlling dimension: inch.  
.050 NOM.  
1.27 NOM.  
1.340  
1.380  
33.27  
35.05  
SQ  
D1  
e
1.200 BSC  
.100 BSC  
30.48 BSC  
2.54 BSC  
L
.110  
.145  
.190  
2.79  
3.68  
4.83  
L1  
M
.170  
4.31  
13  
13  
3
4
N
120  
120  
.003  
.076  
P
A
A2  
A1  
øB  
øB2  
L
D
P
e
Top View  
Cavity Up  
D1  
Pin 1 Identifier  
8
REV. 1.0.1 5/9/00  
TMC3211  
PRODUCT SPECIFICATION  
Mechanical Dimensions  
120-Pin Plastic Pin Grid ArrayH5 Package  
Notes:  
Inches  
Millimeters  
Symbol  
Notes  
1. Pin #1 identifier shall be within shaded area shown.  
2. Dimension "M" defines matrix size.  
Min.  
Max.  
Min.  
Max.  
A
.080  
.040  
.105  
.017  
.125  
.060  
.180  
.020  
2.03  
1.02  
2.67  
0.43  
3.18  
1.52  
4.57  
0.51  
3. Dimension "N" defines the maximum possible number of pins.  
4. Controlling dimension: inch.  
A1  
A2  
øB  
øB2  
D
.050 NOM.  
1.27 NOM.  
1.340  
1.350  
34.04  
35.05  
D1  
e
1.200 BSC  
.100 BSC  
.120 .140  
30.48 BSC  
2.54 BSC  
3.05 3.56  
L
M
13  
13  
2
3
N
121  
121  
.003  
.076  
P
A
A2  
L
A1  
e
øB  
øB2  
D
P
Bottom View  
D1  
Top View  
Pin 1 Identifier  
Beveled Corner  
Vendor Option  
REV. 1.0.1 5/9/00  
9
PRODUCT SPECIFICATION  
TMC3211  
Ordering Information  
Temperature  
Product Number  
Screening  
Package  
Package Marking  
Range  
TMC3211H5C  
TMC3211G1C  
0 to 70°C  
0 to 70°C  
Commercial  
Commercial  
120 Pin Plastic Pin Grid Array  
120 Pin Ceramic Pin Grid Array  
3211H5C  
3211G1C  
LIFE SUPPORT POLICY  
FAIRCHILDS PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES  
OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT OF FAIRCHILD 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 (c) 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 of the user.  
2. A critical component in 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.  
www.fairchildsemi.com  
5/9/00 0.0m 002  
Stock#DS30003211  
2000 Fairchild Semiconductor Corporation  

相关型号:

TMC3211H5C

Divider, 32-Bit, CMOS, PPGA120, PLASTIC, PGA-120
ROCHESTER

TMC3211H5C

Divider, 32-Bit, CMOS, PPGA120,
RAYTHEON

TMC3281GT

SCREW MACHINE SOCKETS & TERMINAL STRIPS
ADAM-TECH

TMC3281TT

SCREW MACHINE SOCKETS & TERMINAL STRIPS
ADAM-TECH

TMC3282GT

SCREW MACHINE SOCKETS & TERMINAL STRIPS
ADAM-TECH

TMC3282TT

SCREW MACHINE SOCKETS & TERMINAL STRIPS
ADAM-TECH

TMC32NP

Complementary 30V MOS FET
TRINAMIC

TMC32NP-MLP

Complementary 30V MOS FET
TRINAMIC

TMC32NP2-MLP

Complementary 30V MOS FET
TRINAMIC

TMC32NPX-EVAL

Step/Direction to SPI converter
TRINAMIC

TMC332

Micro Step Sequencer for 2-phase and 3-phase Motors with Integrated PWM Units and Closed Loop Current Regulator
TRINAMIC

TMC332-BC

Micro Step Sequencer for 2-phase and 3-phase Motors with Integrated PWM Units and Closed Loop Current Regulator
TRINAMIC