LT1720IS8 [Linear]

Dual/Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs; 双/四路, 4.5ns ,单电源3V / 5V比较具有轨至轨输出
LT1720IS8
型号: LT1720IS8
厂家: Linear    Linear
描述:

Dual/Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs
双/四路, 4.5ns ,单电源3V / 5V比较具有轨至轨输出

放大器 光电二极管
文件: 总28页 (文件大小:320K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
LT1720/LT1721  
Dual/Quad,  
4.5ns, Single Supply  
3V/5V Comparators  
with Rail-to-Rail Outputs  
U
FEATURES  
DESCRIPTIO  
The LT®1720/LT1721 are UltraFastTM dual/quad compara-  
tors optimized for single supply operation, with a supply  
voltage range of 2.7V to 6V. The input voltage range extends  
from100mVbelowgroundto1.2Vbelowthesupplyvoltage.  
Internal hysteresis makes the LT1720/LT1721 easy to use  
even with slow moving input signals. The rail-to-rail outputs  
directly interface to TTL and CMOS. Alternatively, the sym-  
metric output drive can be harnessed for analog applications  
or for easy translation to other single supply logic levels.  
UltraFast: 4.5ns at 20mV Overdrive  
7ns at 5mV Overdrive  
Low Power: 4mA per Comparator  
Optimized for 3V and 5V Operation  
Pinout Optimized for High Speed Ease of Use  
Input Voltage Range Extends 100mV  
Below Negative Rail  
TTL/CMOS Compatible Rail-to-Rail Outputs  
Internal Hysteresis with Specified Limits  
Low Dynamic Current Drain; 15µA/(V-MHz),  
The LT1720 is available in the 8-pin MSOP and SO packages;  
three pins per comparator plus power and ground. The  
LT1721 is available in the 16-pin SSOP and S packages.  
Dominated by Load In Most Circuits  
U
APPLICATIO S  
ThepinoutsoftheLT1720/LT1721minimizeparasiticeffects  
byplacingthemostsensitiveinputs(inverting)awayfromthe  
outputs, shieldedbythepowerrails. TheLT1720/LT1721are  
ideal for systems where small size and low power are  
paramount.  
High Speed Differential Line Receiver  
Crystal Oscillator Circuits  
Window Comparators  
Threshold Detectors/Discriminators  
Pulse Stretchers  
Zero-Crossing Detectors  
High Speed Sampling Circuits  
, LTC and LT are registered trademarks of Linear Technology Corporation.  
UltraFast is a trademark of Linear Technology Corporation.  
U
TYPICAL APPLICATIO  
2.7V to 6V Crystal Oscillator with TTL/CMOS Output  
Propagation Delay vs Overdrive  
8
2.7V TO 6V  
25°C  
1MHz TO 10MHz  
7
6
5
4
3
2
1
0
V
V
C
= 100mV  
= 10pF  
STEP  
CC  
LOAD  
2k  
CRYSTAL (AT-CUT)  
= 5V  
RISING EDGE  
(t  
220  
)
PDLH  
620Ω  
GROUND  
CASE  
+
FALLING EDGE  
(t  
C1  
)
PDHL  
OUTPUT  
1/2 LT1720  
2k  
1.8k  
1720/21 TA01  
0.01µF  
10  
20  
OVERDRIVE (mV)  
40  
0
50  
30  
1720/21 TA02  
1
LT1720/LT1721  
W W  
U W  
ABSOLUTE MAXIMUM RATINGS (Note 1)  
Storage Temperature Range ................. 65°C to 150°C  
Operating Temperature Range  
Supply Voltage, VCC to GND ...................................... 7V  
Input Current ...................................................... ±10mA  
Output Current (Continuous) ............................. ±20mA  
Junction Temperature........................................... 150°C  
Lead Temperature (Soldering, 10 sec).................. 300°C  
C Grade .................................................. 0°C to 70°C  
I Grade .............................................. 40°C to 85°C  
U
W U  
PACKAGE/ORDER INFORMATION  
TOP VIEW  
–IN A  
+IN A  
GND  
1
2
3
4
5
6
7
8
16  
15  
14  
13  
12  
11  
10  
9
–IN D  
+IN D  
TOP VIEW  
TOP VIEW  
V
CC  
+IN A  
–IN A  
–IN B  
+IN B  
1
2
3
4
8
7
6
5
V
CC  
+IN A  
–IN A  
–IN B  
+IN B  
1
2
3
4
8 V  
CC  
OUT A  
OUT B  
GND  
OUT D  
OUT C  
7 OUT A  
6 OUT B  
5 GND  
OUT A  
OUT B  
GND  
V
CC  
MS8 PACKAGE  
8-LEAD PLASTIC MSOP  
+IN B  
–IN B  
+IN C  
–IN C  
S8 PACKAGE  
8-LEAD PLASTIC SO  
TJMAX = 150°C, θJA = 230°C/ W  
GN PACKAGE  
16-LEAD NARROW  
PLASTIC SSOP  
S PACKAGE  
16-LEAD PLASTIC SO  
TJMAX = 150°C, θJA = 200°C/ W  
TJMAX = 150°C, θJA = 135°C/ W (GN)  
TJMAX = 150°C, θJA = 115°C/ W (S)  
ORDER PART  
NUMBER  
MS8  
PART MARKING  
ORDER PART  
NUMBER  
S8  
ORDER PART  
NUMBER  
GN  
PART MARKING  
PART MARKING  
LT1720CMS8  
LTDS  
LT1720CS8  
LT1720IS8  
1720  
1720I  
LT1721CGN  
LT1721CS  
LT1721IGN  
LT1721IS  
1721  
1721I  
Consult factory for Military grade parts.  
The denotes specifications that apply over the full operating temperature  
ELECTRICAL CHARACTERISTICS  
range, otherwise specifications are at TA = 25°C. VCC = 5V, VCM = 1V, COUT = 10pF, VOVERDRIVE = 20mV, unless otherwise specified.  
SYMBOL  
PARAMETER  
CONDITIONS  
MIN  
2.7  
TYP  
MAX  
UNITS  
V
V
V
Supply Voltage  
6
V
V
CC  
Common Mode Voltage Range  
Input Trip Points  
(Note 2)  
(Note 3)  
0.1  
V
– 1.2  
CC  
CMR  
TRIP  
+
2.0  
3.0  
5.5  
6.5  
mV  
mV  
V
V
V
Input Trip Points  
(Note 3)  
(Note 3)  
(Note 3)  
5.5  
6.5  
2.0  
3.0  
mV  
mV  
TRIP  
OS  
Input Offset Voltage  
1.0  
3.0  
4.5  
mV  
mV  
Input Hysteresis Voltage  
Input Offset Voltage Drift  
2.0  
3.5  
10  
5.0  
mV  
HYST  
V /T  
µV/°C  
OS  
2
LT1720/LT1721  
The denotes specifications that apply over the full operating temperature  
ELECTRICAL CHARACTERISTICS  
range, otherwise specifications are at TA = 25°C. VCC = 5V, VCM = 1V, COUT = 10pF, VOVERDRIVE = 20mV, unless otherwise specified.  
SYMBOL  
PARAMETER  
CONDITIONS  
MIN  
TYP  
MAX  
0
UNITS  
µA  
I
I
Input Bias Current  
–6  
B
Input Offset Current  
0.6  
µA  
OS  
CMRR  
PSRR  
Common Mode Rejection Ratio  
Power Supply Rejection Ratio  
Voltage Gain  
(Note 4)  
(Note 5)  
(Note 6)  
55  
65  
70  
80  
dB  
dB  
A
V
V
V
+
Output High Voltage  
Output Low Voltage  
I
I
= 4mA, V = V  
+ 10mV  
V – 0.4  
CC  
V
V
OH  
OL  
SOURCE  
IN  
TRIP  
= 10mA, V = V  
– 10mV  
0.4  
SINK  
IN  
TRIP  
I
t
t
Supply Current (Per Comparator)  
V
V
= 5V  
= 3V  
4
3.5  
7
6
mA  
mA  
CC  
CC  
CC  
Propagation Delay  
Propagation Delay  
V
= 20mV (Note 7)  
4.5  
7
6.5  
8.0  
ns  
ns  
PD20  
PD5  
OVERDRIVE  
V
= 5mV (Notes 7, 8)  
10  
13  
ns  
ns  
OVERDRIVE  
t  
Differential Propagation Delay  
Propagation Delay Skew  
Output Rise Time  
(Note 9) Between Channels  
0.3  
0.5  
2.5  
2.2  
1.0  
1.5  
ns  
ns  
ns  
ns  
PD  
+
t
t
t
t
(Note 10) Between t /t  
SKEW  
PD PD  
10% to 90%  
90% to 10%  
r
Output Fall Time  
f
+
Output Timing Jitter  
V
V
= 1.2V (6dBm), Z = 50Ω  
t
t
15  
11  
ps  
ps  
JITTER  
IN  
P-P  
IN  
PD  
RMS  
RMS  
= 2V, f = 20MHz  
CM  
PD  
f
Maximum Toggle Frequency  
V
V
= 50mV, V = 3V  
70.0  
62.5  
MHz  
MHz  
MAX  
OVERDRIVE  
OVERDRIVE  
CC  
= 50mV, V = 5V  
CC  
Note 1: Absolute Maximum Ratings are those values beyond which the  
life of a device may be impaired.  
Note 7: Propagation delay measurements made with 100mV steps.  
Overdrive is measured relative to VTRIP .  
±
Note 2: If one input is within these common mode limits, the other  
input can go outside the common mode limits and the output will be  
valid.  
Note 3: The LT1720/LT1721 comparators include internal hysteresis.  
The trip points are the input voltage needed to change the output state  
Note 8: tPD cannot be measured in automatic handling equipment with  
low values of overdrive. The LT1720/LT1721 are 100% tested with a  
100mV step and 20mV overdrive. Correlation tests have shown that  
tPD limits can be guaranteed with this test, if additional DC tests are  
performed to guarantee that all internal bias conditions are correct.  
Note 9: Differential propagation delay is defined as the larger of the  
two:  
+
in each direction. The offset voltage is defined as the average of VTRIP  
and VTRIP, while the hysteresis voltage is the difference of these two.  
Note 4: The common mode rejection ratio is measured with VCC = 5V  
tPDLH = tPDLH(MAX) – tPDLH(MIN)  
tPDHL = tPDHL(MAX) – tPDHL(MIN)  
where (MAX) and (MIN) denote the maximum and minimum values of  
a given measurement across the different comparator channels.  
and is defined as the change in offset voltage measured from VCM  
0.1V to VCM = 3.8V, divided by 3.9V.  
=
Note 5: The power supply rejection ratio is measured with VCM = 1V  
and is defined as the change in offset voltage measured from VCC  
2.7V to VCC = 6V, divided by 3.3V.  
Note 6: Because of internal hysteresis, there is no small-signal region  
in which to measure gain. Proper operation of internal circuity is  
ensured by measuring VOH and VOL with only 10mV of overdrive.  
=
Note 10: Propagation Delay Skew is defined as:  
tSKEW = |tPDLH – tPDHL  
|
3
LT1720/LT1721  
TYPICAL PERFORMANCE CHARACTERISTICS  
W
U
Input Offset and Trip Voltages  
vs Supply Voltage  
Input Offset and Trip Voltages  
vs Temperature  
Input Common Mode Limits  
vs Temperature  
3
2
3
2
4.2  
4.0  
V
CC  
= 5V  
+
+
V
TRIP  
V
TRIP  
3.8  
3.6  
0.2  
0
1
0
1
V
V
OS  
OS  
0
–1  
–2  
–3  
–1  
–2  
–3  
V
V
TRIP  
TRIP  
0.2  
25°C  
CM  
V
= 1V  
0.4  
–50 25  
0
25  
50  
75 100 125  
4.5  
SUPPLY VOLTAGE (V)  
5.5 6.0  
50  
TEMPERATURE (°C)  
100 125  
2.5 3.0  
3.5 4.0  
5.0  
50  
25  
75  
25  
0
TEMPERATURE (°C)  
1720/21 G02  
1720/21 G01  
1720/21 G03  
Input Current  
vs Differential Input Voltage  
Quiescent Supply Current  
vs Temperature  
Quiescent Supply Current  
vs Supply Voltage  
7
6
5
4
3
2
1
0
2
1
6.0  
25°C  
CC  
V
= 5V  
125°C  
5.5  
5.0  
0
–1  
–2  
–3  
–4  
–5  
–6  
–7  
4.5  
4.0  
3.5  
3.0  
2.5  
25°C  
V
CC  
= 5V  
55°C  
V
CC  
= 3V  
2.0  
0
2
3
4
5
6
7
25  
0
50  
75 100 125  
1
50  
25  
–5 –4 –3 –2 1  
0
5
1
2
3
4
SUPPLY VOLTAGE (V)  
DIFFERENTIAL INPUT VOLTAGE (V)  
TEMPERATURE (˚C)  
1720/21 G06  
1720/21 G05  
1720/21 G04  
Propagation Delay  
vs Load Capacitance  
Propagation Delay  
vs Temperature  
Propagation Delay  
vs Supply Voltage  
9
8.0  
5.0  
4.5  
4.0  
t
25°C  
STEP  
PDLH  
CM  
STEP  
LOAD  
25°C  
STEP  
RISING EDGE  
(t  
V
V
C
= 1V  
= 100mV  
= 10pF  
8
7
6
5
4
3
2
1
0
V
= 100mV  
V
= 100mV  
7.5  
7.0  
)
PDLH  
OVERDRIVE = 20mV  
= 5V  
OVERDRIVE = 20mV  
= 10pF  
V
= 3V  
CC  
V
CC  
C
LOAD  
V
= 5V  
6.5  
6.0  
5.5  
5.0  
4.5  
CC  
FALLING EDGE  
(t  
RISING EDGE  
(t  
)
PDHL  
)
OVERDRIVE = 5mV  
OVERDRIVE = 20mV  
PDLH  
V
CC  
= 5V  
FALLING EDGE  
(t  
PDHL  
)
V
CC  
= 3V  
4.0  
10  
20  
40  
0
50  
30  
4.5  
5.5  
6.0  
25  
0
50  
75 100 125  
2.5 3.0  
3.5 4.0  
5.0  
50  
25  
OUTPUT LOAD CAPACITANCE (pF)  
TEMPERATURE (°C)  
SUPPLY VOLTAGE (V)  
1720/21 G07  
1720/21 G08  
1720/21 G09  
4
LT1720/LT1721  
W
U
TYPICAL PERFORMANCE CHARACTERISTICS  
Output High Voltage  
vs Load Current  
Output Low Voltage  
vs Load Current  
Supply Current vs Frequency  
0.5  
0.4  
0.3  
0.2  
0.1  
0.0  
0.2  
0.4  
0.6  
0.8  
–1.0  
10  
9
25°C  
CC  
V
V
V
= 5V  
= 1V  
V
V
V
= 5V  
= 1V  
CC  
CM  
IN  
CC  
CM  
IN  
V
= 5V  
125°C  
= 15mV  
= 15mV  
125°C  
125°C  
= 2.7V  
8
V
CC  
55°C  
C
= 20pF  
LOAD  
25°C  
7
6
5
4
3
25°C  
55°C  
NO LOAD  
25°C  
= 2.7V  
V
CC  
4
8
12  
16  
20  
10  
20  
FREQUENCY (MHz)  
40  
0
4
8
12  
16  
20  
0
0
30  
OUTPUT SINK CURRENT (mA)  
OUTPUT SOURCE CURRENT (mA)  
1720/21 G10  
1720/21 G11  
1720/21 G12  
U
U
U
PIN FUNCTIONS  
LT1720  
LT1721  
–IN A (Pin 1): Inverting Input of Comparator A.  
+IN A (Pin 1): Noninverting Input of Comparator A.  
–IN A (Pin 2): Inverting Input of Comparator A.  
–IN B (Pin 3): Inverting Input of Comparator B.  
+IN B (Pin 4): Noninverting Input of Comparator B.  
GND (Pin 5): Ground.  
+IN A (Pin 2): Noninverting Input of Comparator A.  
GND (Pins 3, 6): Ground.  
OUT A (Pin 4): Output of Comparator A.  
OUT B (Pin 5): Output of Comparator B.  
OUT B (Pin 6): Output of Comparator B.  
OUT A (Pin 7): Output of Comparator A.  
VCC (Pin 8): Positive Supply Voltage.  
+IN B (Pin 7): Noninverting Input of Comparator B.  
–IN B (Pin 8): Inverting Input of Comparator B.  
–IN C (Pin 9): Inverting Input of Comparator C.  
+IN C (Pin 10): Noninverting Input of Comparator C.  
VCC (Pins 11, 14): Positive Supply Voltage.  
OUT C (Pin 12): Output of Comparator C.  
OUT D (Pin 13): Output of Comparator D.  
+IN D (Pin 15): Noninverting Input of Comparator D.  
–IN D (Pin 16): Inverting Input of Comparator D.  
5
LT1720/LT1721  
TEST CIRCUITS  
±VTRIP Test Circuit  
15V  
P-P  
LTC203  
BANDWIDTH-LIMITED  
TRIANGLE WAVE  
~
1kHz  
14  
15  
3
2
V
CC  
0.1µF  
+
1000 × V  
TRIP  
50k  
10nF  
1µF  
10k  
+
16  
9
1
8
50Ω  
50Ω  
1/2 LT1112  
200k  
+
1000 × V  
HYST  
DUT  
1/2 LT1720 OR  
1/4 LT1721  
V
CM  
11  
10  
6
7
1000 × V  
OS  
10k  
LTC203  
3
2
14  
15  
1/2 LT1638  
+
1000 × V  
TRIP  
100k  
100k  
1µF  
10nF  
1
8
16  
9
+
2.4k  
100k  
100k  
1/2 LT1638  
1/2 LT1112  
+
0.15µF  
6
7
11  
10  
1720/21 TC01  
NOTES: LT1638, LT1112, LTC203s ARE POWERED FROM ±15V.  
200kPULL-DOWN PROTECTS LTC203 LOGIC INPUTS  
WHEN DUT IS NOT POWERED  
Response Time Test Circuit  
+V – V  
CC  
CM  
0V  
DUT  
1/2 LT1720 OR  
1/4 LT1721  
–100mV  
0.01µF  
+
25Ω  
10 × SCOPE PROBE  
(C 10pF)  
25Ω  
50k  
IN  
0.01µF  
50Ω  
V1*  
0.1µF  
130Ω  
400Ω  
PULSE  
IN  
2N3866  
0V  
1N5711  
–V  
+
CM  
–3V  
50Ω  
750Ω  
*V1 = –1000 • (OVERDRIVE + V  
TRIP  
)
NOTE: RISING EDGE TEST SHOWN.  
FOR FALLING EDGE, REVERSE LT1720 INPUTS  
1720/21 TC02  
–5V  
6
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
Input Voltage Considerations  
Input Protection  
The LT1720/LT1721 are specified for a common mode  
range of –100mV to 3.8V when used with a single 5V  
supply. In general the common mode range is 100mV  
below ground to 1.2V below VCC. The criterion for this  
common mode limit is that the output still responds  
correctly to a small differential input signal. Also, if one  
input is within the common mode limit, the other input  
signal can go outside the common mode limits, up to the  
absolute maximum limits (a diode drop past either rail at  
10mA input current) and the output will retain the correct  
polarity.  
The input stage is protected against damage from large  
differential signals, up to and beyond a differential voltage  
equal to the supply voltage, limited only by the absolute  
maximum currents noted. External input protection cir-  
cuitry is only needed if currents would otherwise exceed  
these absolute maximums. The internal catch diodes can  
conduct current up to these rated maximums without  
latchup, even when the supply voltage is at the absolute  
maximum rating.  
The LT1720/LT1721 input stage has general purpose  
internalESDprotectionforthehumanbodymodel.Foruse  
as a line receiver, additional external protection may be  
required. As with most integrated circuits, the level of  
immunity to ESD is much greater when residing on a  
printed circuit board where the power supply decoupling  
capacitance will limit the voltage rise caused by an ESD  
pulse.  
When either input signal falls below the negative common  
mode limit, the internal PN diode formed with the sub-  
strate can turn on, resulting in significant current flow  
through the die. An external Schottky clamp diode  
between the input and the negative rail can speed up  
recovery from negative overdrive by preventing the sub-  
strate diode from turning on.  
Input Bias Current  
When both input signals are below the negative common  
mode limit, phase reversal protection circuitry prevents  
false output inversion to at least 400mV common mode.  
However, the offset and hysteresis in this mode will  
increasedramatically,toasmuchas15mVeach.Theinput  
bias currents will also increase.  
Inputbiascurrentismeasuredwithbothinputsheldat1V.  
As with any PNP differential input stage, the LT1720/  
LT1721 bias current flows out of the device. With a  
differential input voltage of even just 100mV or so, there  
will be zero bias current into the higher of the two inputs,  
while the current flowing out of the lower input will be  
twice the measured bias current. With more than two  
diode drops of differential input voltage, the LT1720/  
LT1721’s input protection circuitry activates, and current  
out of the lower input will increase an additional 30% and  
there will be a small bias current into the higher of the two  
input pins, of 4µA or less. See the Typical Performance  
curve “Input Current vs Differential Input Voltage.”  
When both input signals are above the positive common  
mode limit, the input stage will become debiased and the  
output polarity will be random. However, the internal  
hysteresis will hold the output to a valid logic level, and  
because the biasing of each comparator is completely  
independent, there will be no impact on any other com-  
parator. When at least one of the inputs returns to within  
thecommonmodelimits,recoveryfromthisstatewilltake  
as long as 1µs.  
High Speed Design Considerations  
The propagation delay does not increase significantly  
whendrivenwithlargedifferentialvoltages.However,with  
low levels of overdrive, an apparent increase may be seen  
with large source resistances due to an RC delay caused  
by the 2pF typical input capacitance.  
Application of high speed comparators is often plagued  
by oscillations. The LT1720/LT1721 have 4mV of internal  
hysteresis, which will prevent oscillations as long as  
parasitic output to input feedback is kept below 4mV.  
7
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
However, with the 2V/ns slew rate of the LT1720/LT1721  
outputs, a 4mV step can be created at a 100input  
source with only 0.02pF of output to input coupling. The  
pinouts of the LT1720/LT1721 have been arranged to  
minimize problems by placing the most sensitive inputs  
(inverting) away from the outputs, shielded by the power  
rails. The input and output traces of the circuit board  
should also be separated, and the requisite level of  
isolationisreadilyachievedifatopsidegroundplaneruns  
betweentheoutputsandtheinputs. Formultilayerboards  
where the ground plane is internal, a topside ground or  
supply trace should be run between the inputs and  
outputs, as illustrated in Figure 1.  
The supply bypass should include an adjacent  
10nF ceramic capacitor and a 2.2µF tantalum capacitor no  
farther than 5cm away; use more capacitance if driving  
more than 4mA loads. To prevent oscillations, it is helpful  
tobalancetheimpedanceattheinvertingandnoninverting  
inputs; source impedances should be kept low, preferably  
1kor less.  
The outputs of the LT1720/LT1721 are capable of very  
high slew rates. To prevent overshoot, ringing and other  
problems with transmission line effects, keep the output  
traces shorter than 10cm, or be sure to terminate the lines  
to maintain signal integrity. The LT1720/LT1721 can drive  
DCterminationsof250ormore,butlowercharacteristic  
impedance traces can be driven with series termination or  
AC termination topologies.  
Hysteresis  
The LT1720/LT1721 include internal hysteresis, which  
makes them easier to use than many other comparable  
speed comparators.  
(a)  
(b)  
1720/21 F01  
The input-output transfer characteristic is illustrated in  
Figure 2 showing the definitions of VOS and VHYST based  
upon the two measurable trip points. The hysteresis band  
makestheLT1720/LT1721wellbehaved, evenwithslowly  
moving inputs.  
Figure 1. Typical Topside Metal for Multilayer PCB Layouts  
Figure 1a shows a typical topside layout of the LT1720 on  
such a multilayer board. Shown is the topside metal etch  
including traces, pin escape vias, and the land pads for an  
SO-8 LT1720 and its adjacent X7R 10nF bypass capacitor  
in a 1206 case.  
The ground trace from Pin 5 runs under the device up to  
the bypass capacitor, shielding the inputs from the  
outputs. Note the use of a common via for the LT1720 and  
the bypass capacitor, which minimizes interference from  
highfrequencyenergyrunningaroundthegroundplaneor  
power distribution traces.  
V
OH  
V
HYST  
+
(= V  
– V  
)
TRIP  
TRIP  
V
/2  
HYST  
Figure 1b shows a typical topside layout of the LT1721 on  
a multilayer board. In this case, the power and ground  
traces have been extended to the bottom of the device  
solely to act as high frequency shields between input and  
output traces.  
V
OL  
+
V = V – V  
IN  
IN  
IN  
0
+
V
V
TRIP  
TRIP  
+
V
+ V  
2
TRIP  
TRIP  
V
=
OS  
1720/21 F02  
Although both VCC pins are electrically shorted internal to  
the LT1721, they must be shorted together externally as  
well in order for both to function as shields. The same is  
true for the two GND pins.  
Figure 2. Hysteresis I/O Characteristics  
8
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
The exact amount of hysteresis will vary from part to part  
asindicatedinthespecificationstable.Thehysteresislevel  
will also vary slightly with changes in supply voltage and  
common mode voltage. A key advantage of the LT1720/  
LT1721 is the significant reduction in these effects, which  
is important whenever an LT1720/LT1721 is used to de-  
tect a threshold crossing in one direction only. In such a  
case, the relevant trip point will be all that matters, and a  
stable offset voltage with an unpredictable level of hyster-  
esis, as seen in competing comparators, is of little value.  
TheLT1720/LT1721aremanytimesbetterthanpriorcom-  
parators in these regards. In fact, the CMRR and PSRR  
tests are performed by checking for changes in either trip  
point to the limits indicated in the specifications table.  
Because the offset voltage is the average of the trip points,  
the CMRR and PSRR of the offset voltage is therefore  
guaranteed to be at least as good as those limits. This  
more stringent test also puts a limit on the common mode  
and power supply dependence of the hysteresis voltage.  
voltage on the right side of R3 is 300mV or VCC – 300mV,  
for a total voltage swing of (VCC – 300mV) – 300mV =  
VCC – 600mV.  
Withthisinmind, calculationoftheresistorvaluesneeded  
isatwo-stepprocess.First,calculatethevalueofR3based  
on the additional hysteresis desired, the output voltage  
swing, and the impedance of the primary bias string:  
R3 = (R1 R2)(VCC – 0.6V)/(additional hysteresis)  
Additional hysteresis is the desired overall hysteresis less  
the internal 3.5mV hysteresis.  
The second step is to recalculate R2 to set the same  
averagethresholdasbefore. Theaveragethresholdbefore  
was set at VTH = (VREF)(R1)/(R1 + R2). The new R2 is  
calculated based on the average output voltage (VCC/2)  
and the simplified circuit model in Figure 4. To assure that  
the comparator’s noninverting input is, on average, the  
same VTH as before:  
R2= (VREF – VTH)/(VTH/R1 + (VTH – VCC/2)/R3)  
Additionalhysteresismaybeaddedexternally. Therail-to-  
rail outputs of the LT1720/LT1721 make this more pre-  
dictable than with TTL output comparators due to the  
LT1720/LT1721’s small variability of VOH (output high  
voltage).  
For additional hysteresis of 10mV or less, it is not uncom-  
mon for R2to be the same as R2 within 1% resistor  
tolerances.  
This method will work for additional hysteresis of up to a  
few hundred millivolts. Beyond that, the impedance of R3  
is low enough to effect the bias string, and adjustment of  
R1 may also be required. Note that the currents through  
the R1/R2 bias string should be many times the input  
currents of the LT1720/LT1721. For 5% accuracy, the  
current must be at least 120µA(6µA IB ÷ 0.05); more for  
higher accuracy.  
To add additional hysteresis, set up positive feedback by  
adding additional external resistor R3 as shown in Figure  
3. ResistorR3addsaportionoftheoutputtothethreshold  
set by the resistor string. The LT1720/LT1721 pulls the  
outputs to the supply rail and ground to within 200mV of  
the rails with light loads, and to within 400mV with heavy  
loads. For the load of most circuits, a good model for the  
V
REF  
R2′  
V
REF  
V
TH  
R3  
R3  
V
2
CC  
V
=
AVERAGE  
R2  
R1  
+
+
1/2 LT1720  
1/2 LT1720  
R1  
1720/21 F04  
INPUT  
1720/21 F03  
Figure 3. Additional External Hysteresis  
Figure 4. Model for Additional Hysteresis Calculations  
9
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
Interfacing the LT1720/LT1721 to ECL  
100µA of forward current. R4 prevents this with the  
minimum additional power dissipation.  
The LT1720/LT1721 comparators can be used in high  
speed applications where Emitter-Coupled Logic (ECL) is  
deployed. To interface the outputs of the LT1720/LT1721  
to ECL logic inputs, standard TTL/CMOS to ECL level  
translators such as the 10H124, 10H424 and 100124 can  
be used. These components come at a cost of a few  
nanosecondsadditionaldelayaswellassupplycurrentsof  
50mA or more, and are only available in quads. A faster,  
simpler and lower power translator can be constructed  
with resistors as shown in Figure 5.  
Finally, Figure 5d shows the case of driving standard,  
negative-rail, ECL with the LT1720/LT1721. Resistor val-  
ues are given for both ECL interface types and for both a  
5V and 3V LT1720/LT1721 supply rail. Again, a fourth  
resistor,R4isneededtopreventthelowstatecurrentfrom  
flowing out of the LT1720/LT1721, turning on the internal  
ESD/substrate diodes. Not only can the output stage func-  
tionality and speed suffer, but in this case the substrate is  
common to all the comparators in the LT1720/LT1721, so  
operation of the other comparator(s) in the same package  
could also be affected. Resistor R4 again prevents this  
with the minimum additional power dissipation.  
Figure 5a shows the standard TTL to Positive ECL (PECL)  
resistive level translator. This translator cannot be used  
for the LT1720/LT1721, or with CMOS logic, because it  
depends on the 820resistor to limit the output swing  
(VOH)oftheall-NPNTTLgatewithitsso-calledtotem-pole  
output. The LT1720/LT1721 are fabricated in a comple-  
mentary bipolar process and their output stage has a PNP  
driver that pulls the output nearly all the way to the supply  
rail, even when sourcing 10mA.  
For all the dividers shown, the output impedance is about  
110. This makes these fast, less than a nanosecond,  
with most layouts. Avoid the temptation to use speedup  
capacitors. Not only can they foul up the operation of the  
ECL gate because of overshoots, they can damage the  
ECL inputs, particularly during power-up of separate  
supply configurations.  
Figure 5b shows a three resistor level translator for inter-  
facing the LT1720/LT1721 to ECL running off the same  
supply rail. No pull-down on the output of the LT1720/  
LT1721 is needed, but pull-down R3 limits the VIH seen by  
the PECL gate. This is needed because ECL inputs have  
both a minimum and maximum VIH specification for  
proper operation. Resistor values are given for both ECL  
interface types; in both cases it is assumed that the  
LT1720/LT1721 operates from the same supply rail.  
The level translator designs assume one gate load. Mul-  
tiple gates can have significant IIH loading, and the trans-  
missionlineroutingandterminationissuesalsomakethis  
case difficult.  
ECL,andparticularlyPECL,isvaluabletechnologyforhigh  
speed system design, but it must be used with care. With  
less than a volt of swing, the noise margins need to be  
evaluated carefully. Note that there is some degradation of  
noise margin due to the ±5% resistor selections shown.  
With10KH/E,thereisnotemperaturecompensationofthe  
logic levels, whereas the LT1720/LT1721 and the circuits  
shown give levels that are stable with temperature. This  
will degrade the noise margin over temperature. In some  
configurations it is possible to add compensation with  
diode or transistor junctions in series with the resistors of  
these networks.  
Figure 5c shows the case of translating to PECL from an  
LT1720/LT1721 powered by a 3V supply rail. Again,  
resistorvaluesaregivenforbothECLinterfacetypes.This  
time four resistors are needed, although with 10KH/E, R3  
is not needed. In that case, the circuit resembles the  
standard TTL translator of Figure 5a, but the function of  
the new resistor, R4, is much different. R4 loads the  
LT1720/LT1721 output when high so that the current  
flowing through R1 doesn’t forward bias the LT1720/  
LT1721’s internal ESD clamp diode. Although this diode  
can handle 20mA without damage, normal operation and  
performance of the output stage can be impaired above  
For more information on ECL design, refer to the ECLiPS  
data book (DL140), the 10KH system design handbook  
(HB205) and PECL design (AN1406), all from Motorola.  
10  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
5V  
5V  
180Ω  
270Ω  
820Ω  
DO NOT USE FOR LT1720/LT1721  
LEVEL TRANSLATION. SEE TEXT  
LSTTL  
10KH/E  
(a) STANDARD TTL TO PECL TRANSLATOR  
V
CC  
R2  
R1  
V
R1  
10KH/E 5V OR 5.2V 510180750Ω  
100K/E 4.5V 620180510Ω  
R2  
R3  
CC  
1/2 LT1720  
R3  
(b) LT1720/LT1721 OUTPUT TO PECL TRANSLATOR  
V
CC  
3V  
R2  
R1  
R4  
V
R1  
R2  
R3  
R4  
CC  
1/2 LT1720  
10KH/E 5V OR 5.2V 300180OMIT 560Ω  
100K/E  
4.5V  
33018015001000Ω  
R3  
(c) 3V LT1720/LT1721 OUTPUT TO PECL TRANSLATOR  
V
CC  
R4  
ECL FAMILY  
V
V
R1  
R2  
R3  
R4  
EE  
CC  
R1  
5V  
3V  
5V  
3V  
5602703301200Ω  
270510300330Ω  
6802703001500Ω  
330390270430Ω  
1720/21 F05  
1/2 LT1720  
10KH/E  
5.2V  
4.5V  
R2  
R3  
100K/E  
V
EE  
(d) LT1720/LT1721 OUTPUT TO STANDARD ECL TRANSLATOR  
Figure 5  
11  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
Circuit Description  
Technology’s rail-to-rail amplifiers and other products.  
But the output of a comparator is digital, and this output  
stagecandriveTTLorCMOSdirectly.ItcanalsodriveECL,  
as described earlier, or analog loads as demonstrated in  
the applications to follow.  
TheblockdiagramofonecomparatorintheLT1720/LT1721  
is shown in Figure 6. There are differential inputs  
(+IN/IN), an output (OUT), a single positive supply (VCC)  
and ground (GND). All comparators are completely inde-  
pendent, sharing only the power and ground pins. The  
circuit topology consists of a differential input stage, a  
gain stage with hysteresis and a complementary  
common-emitter output stage. All of the internal signal  
pathsutilizelowvoltageswingsforhighspeedatlowpower.  
Thebiasconditionsandsignalswingsintheoutputstages  
are designed to turn their respective output transistors off  
faster than on. This nearly eliminates the surge of current  
from VCC to ground that occurs at transitions, keeping the  
power consumption low even with high output-toggle  
frequencies.  
The input stage topology maximizes the input dynamic  
range available without requiring the power, complexity  
and die area of two complete input stages such as are  
foundinrail-to-railinputcomparators.Witha2.7Vsupply,  
the LT1720/LT1721 still have a respectable 1.6V of input  
common mode range. The differential input voltage range  
is rail-to-rail, without the large input currents found in  
competing devices. The input stage also features phase  
reversal protection to prevent false outputs when the  
inputs are driven below the –100mV common mode  
voltage limit.  
The low surge current is what keeps the power consump-  
tion low at high output-toggle frequencies. The frequency  
dependence of the supply current is shown in the Typical  
Performance Characteristics. Just 20pF of capacitive load  
on the output more than triples the frequency dependent  
rise. Theslopeoftheno-loadcurveisjust32µA/MHz. With  
a 5V supply, this current is the equivalent of charging and  
discharging just 6.5pF. The slope of the 20pF load curve is  
133µA/MHz, an addition of 101µA/MHz, or 20µA/MHz-V,  
units that are equivalent to picoFarads.  
Theinternalhysteresisisimplementedbypositive,nonlin-  
ear feedback around a second gain stage. Until this point,  
the signal path has been entirely differential. The signal  
path is then split into two drive signals for the upper and  
lower output transistors. The output transistors are con-  
nected common emitter for rail-to-rail output operation.  
The Schottky clamps limit the output voltages at about  
300mVfromtherail, notquitethe50mVor15mVofLinear  
The LT1720/LT1721 dynamic current can be estimated by  
addingtheexternalcapacitiveloadingtoaninternalequiva-  
lent capacitance of 5pF to 15pF, multiplied by the toggle  
frequency and the supply voltage. Because the capaci-  
tance of routing traces can easily approach these values,  
the dynamic current is dominated by the load in most  
circuits.  
V
NONLINEAR STAGE  
CC  
+
+
Σ
+IN  
+
+
A
A
OUT  
V1  
V2  
+
Σ
–IN  
+
GND  
1720/21 F06  
Figure 6. LT1720/LT1721 Block Diagram  
12  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
Speed Limits  
The second output speed limit is the clamp turnaround.  
The LT1720/LT1721 output is optimized for fast initial  
response, with some loss of turnaround speed, limiting  
the toggle frequency. The output transistors are idled in a  
low power state once VOH or VOL is reached by detecting  
the Schottky clamp action. It is only when the output has  
slewed from the old voltage to the new voltage, and the  
clamp circuitry has settled, that the idle state is reached  
andtheoutputisfullyreadytotransitionagain. Thisclamp  
turnaround time is typically 8ns for each direction, result-  
ing in a maximum toggle frequency of 62.5MHz, or a  
125MB data rate. With higher frequencies, dropout and  
runt pulses can occur. Increases in capacitive load will  
increase the time needed for slewing due to the limited  
slew currents and the maximum toggle frequency will  
decrease further. For higher toggle frequency applica-  
tions, consider the LT1394, whose linear output stage can  
toggle at 100MHz typical.  
The LT1720/LT1721 comparators are intended for high  
speed applications, where it is important to understand a  
few limitations. These limitations can roughly be divided  
into three categories: input speed limits, output speed  
limits, and internal speed limits.  
There are no significant input speed limits except the  
shunt capacitance of the input nodes. If the 2pF typical  
input nodes are driven, the LT1720/LT1721 will respond.  
The output speed is constrained by two mechanisms, the  
firstofwhichistheslewcurrentsavailablefromtheoutput  
transistors. To maintain low power quiescent operation,  
the LT1720/LT1721 output transistors are sized to deliver  
25mA to 45mA typical slew currents. This is sufficient to  
drive small capacitive loads and logic gate inputs at  
extremelyhighspeeds.Buttheslewratewillslowdramati-  
callywithheavycapacitiveloads.Becausethepropagation  
delay (tPD) definition ends at the time the output voltage is  
halfway between the supplies, the fixed slew current  
actually makes the LT1720/LT1721 faster at 3V than 5V  
with 20mV of input overdrive.  
The internal speed limits manifest themselves as disper-  
sion. All comparators have some degree of dispersion,  
defined as a change in propagation delay versus input  
overdrive. The propagation delay of the LT1720/LT1721  
will vary with overdrive, from a typical of 4.5ns at 20mV  
overdrive to 7ns at 5mV overdrive (typical). The LT1720/  
LT1721’s primary source of dispersion is the hysteresis  
stage. As a change of polarity arrives at the gain stage, the  
positive feedback of the hysteresis stage subtracts from  
the overdrive available. Only when enough time has  
elapsedforasignaltopropagateforwardthroughthegain  
stage, backwards through the hysteresis stage and for-  
ward through the gain stage again, will the output stage  
receive the same level of overdrive that it would have  
received in the absence of hysteresis.  
Another manifestation of this output speed limit is skew,  
the difference between tPD+ and tPD. The slew currents of  
theLT1720/LT1721varywiththeprocessvariationsofthe  
PNP and NPN transistors, for rising edges and falling  
edges respectively. The typical 0.5ns skew can have either  
polarity, rising edge or falling edge faster. Again, the skew  
will increase dramatically with heavy capacitive loads.  
The skews of comparators in a single package are corre-  
lated, but not identical. Besides some random variability,  
there is a small (100ps to 200ps) systematic skew due to  
physical parasitics of the packages. For the LT1720 SO-8,  
comparatorA, whoseoutputisadjacenttotheVCC pin, will  
have a relatively faster rising edge than comparator B.  
Likewise, comparator B, by virtue of an output adjacent to  
the ground pin will have a relatively faster falling edge.  
Similar dependencies occur in the LT1721 S16, while the  
systemic skews in the smaller MSOP and SSOP packages  
arehalfagainassmall. Ofcourse, ifthecapacitiveloadson  
the two comparators of a single package are not identical,  
the differential timing will degrade further.  
With5mVofoverdrive, theLT1720/LT1721arefasterwith  
a 5V supply than with a 3V supply, the opposite of what is  
true with 20mV overdrive. This is due to the internal speed  
limit, because the gain stage is faster at 5V than 3V due  
primarily to the reduced junction capacitances with higher  
reverse voltage bias.  
Inmanyapplications, asshowninthefollowingexamples,  
there is plenty of input overdrive. Even in applications  
providing low levels of overdrive, the LT1720/LT1721 are  
13  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
fast enough that the absolute dispersion of 2.5ns  
noninvertinginput.ThecircuitwilloperatewithanyAT-cut  
crystal from 1MHz to 10MHz over a 2.7V to 6V supply  
range. As the power is applied, the circuit remains off until  
the LT1720/LT1721 bias circuits activate, at a typical VCC  
of 2V to 2.2V (25°C), at which point the desired frequency  
output is generated.  
(= 7 – 4.5) is often small enough to ignore.  
The gain and hysteresis stage of the LT1720/LT1721 is  
simple, short and high speed to help prevent parasitic  
oscillations while adding minimum dispersion. This in-  
ternal “self-latch” can be usefully exploited in many  
applicationsbecauseitoccursearlyinthesignalchain, in  
a low power, fully differential stage. It is therefore highly  
immune to disturbances from other parts of the circuit,  
either in the same comparator, on the supply lines, or  
from the other comparator(s) in the same package. Once  
a high speed signal trips the hysteresis, the output will  
respond, after a fixed propagation delay, without regard  
to these external influences that can cause trouble in  
nonhysteretic comparators.  
The output duty cycle for this circuit is roughly 50%, but  
it is affected by resistor tolerances and, to a lesser extent,  
by comparator offsets and timings. If a 50% duty cycle is  
required, the circuit of Figure 7 creates a pair of comple-  
mentary outputs with a forced 50% duty cycle. Crystals  
are narrow-band elements, so the feedback to the nonin-  
vertinginputisalteredanalogversionofthesquarewave  
output.Changingthenoninvertingreferencelevelcanthere-  
fore vary the duty cycle. C1 operates as in the previous  
example, whereas C2 creates a complementary output by  
comparing the same two nodes with the opposite input  
polarity. A1 compares band-limited versions of the out-  
puts and biases C1’s negative input. C1’s only degree of  
freedom to respond is variation of pulse width; hence the  
outputs are forced to 50% duty cycle. Again, the circuit  
operatesfrom2.7Vto6V, andtheskewbetweentheedges  
of the two outputs are shown in Figure 8. There is a slight  
duty cycle dependence on comparator loading, so equal  
capacitive and resistive loading should be used in critical  
±VTRIP Test Circuit  
The input trip points are tested using the circuit shown in  
the Test Circuits section that precedes this Applications  
Information section. The test circuit uses a 1kHz triangle  
wave to repeatedly trip the comparator being tested. The  
LT1720/LT1721outputisusedtotriggerswitchedcapaci-  
tor sampling of the triangle wave, with a sampler for each  
direction. Because the triangle wave is attenuated 1000:1  
and fed to the LT1720/LT1721’s differential input, the  
sampled voltages are therefore 1000 times the input trip  
voltages. The hysteresis and offset are computed from the  
trip points as shown.  
V
CC  
2.7V TO 6V  
1MHz TO 10MHz  
CRYSTAL (AT-CUT)  
2k  
220  
Crystal Oscillators  
620Ω  
GROUND  
CASE  
A simple crystal oscillator using one comparator of an  
LT1720/LT1721 is shown on the first page of this data  
sheet. The 2k-620resistor pair set a bias point at the  
comparator’s noninverting input. The 2k-1.8k-0.1µF path  
sets the inverting input node at an appropriate DC average  
level based on the output. The crystal’s path provides  
resonant positive feedback and stable oscillation occurs.  
Although the LT1720/LT1721 will give the correct logic  
output when one input is outside the common mode  
range, additional delays may occur when it is so operated,  
opening the possibility of spurious operating modes.  
Therefore, the DC bias voltages at the inputs are set near  
the center of the LT1720/LT1721’s common mode range  
and the 220resistor attenuates the feedback to the  
+
C1  
OUTPUT  
100k  
1/2 LT1720  
2k  
+
A1  
LT1636  
0.1µF  
0.1µF  
1.8k  
0.1µF  
1k  
100k  
+
C2  
OUTPUT  
1/2 LT1720  
1720 F07  
Figure 7. Crystal Oscillator with Complementary  
Outputs and 50% Duty Cycle  
14  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
1000  
56% low duty cycle, sufficient to allow 2ns between the  
high pulses. Figure 10 shows the two outputs.  
800  
600  
400  
200  
The optional A1 feedback network shown can be used to  
force identical output duty cycles. The steady state duty  
cycles of both outputs will be 44%. Note, though, that the  
addition of this network only adjusts the percentage of  
time each output is high to be the same, which can be  
important in switching circuits requiring identical settling  
times. Itcannotadjusttherelativephasesbetweenthetwo  
outputs to be exactly 180° apart, because the signal at the  
input node driven by the crystal is not a pure sinusoid.  
0
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0  
SUPPLY VOLTAGE (V)  
1720/21 F08  
Q0  
2V/DIV  
Figure 8. Timing Skew of Figure 7’s Circuit  
applications. This circuit works well because of the two  
matcheddelaysandrail-to-railstyleoutputsoftheLT1720.  
Q1  
2V/DIV  
ThecircuitinFigure9showsacrystaloscillatorcircuitthat  
generatestwononoverlappingclocksbymakingfulluseof  
the two independent comparators of the LT1720.  
C1 oscillates as before, but with a lower reference level,  
C2’s output will toggle at different times. The resistors set  
thedegreeofseparationbetweentheoutput’shighpulses.  
With the values shown, each output has a 44% high and  
20ns/DIV  
Figure 10. Nonoverlapping Outputs of Figure 9's Circuit  
V
CC  
2.7V TO 6V  
10MHz  
CRYSTAL (AT-CUT)  
2k  
220Ω  
620Ω  
GROUND  
CASE  
+
C1  
OUTPUT 0  
1/2 LT1720  
OPTIONAL—  
SEE TEXT  
100k  
2k  
+
A1  
LT1636  
0.1µF  
1.3k  
0.1µF  
0.1µF  
1k  
2.2k  
100k  
+
C2  
OUTPUT 1  
1/2 LT1720  
1720/21 F09  
Figure 9. Crystal-Based Nonoverlapping 10MHz Clock Generator  
15  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
Timing Skews  
high speed instrumentation. The circuit in Figure 12 is a  
delay detector which will output a pulse when signals X  
and Y are out of sync (specifically, when X is high and Y is  
low). Note that the addition of an identical circuit to detect  
the opposite situation (X low and Y high) allows for full  
skew detection.  
For a number of reasons, the LT1720/LT1721’s superior  
timing specifications make them an excellent choice for  
applications requiring accurate differential timing skew.  
The comparators in a single package are inherently well  
matched, with just 300ps tPD typical. Monolithic con-  
struction keeps the delays well matched vs supply voltage  
and temperature. Crosstalk between the comparators,  
usually a disadvantage in monolithic duals and quads, has  
minimal effect on the LT1720/LT1721 timing due to the  
internal hysteresis, as described in the Speed Limits  
section.  
Comparators U1A and U1B clean up the incoming signals  
and render the circuit less sensitive to input levels and  
slew rates. The resistive divider network provides level  
shifting for the downstream comparator’s common mode  
input range, as well as offset to keep the output low except  
during a decisive event. When the upstream comparator’s  
outputs can overcome the resistively generated offset  
(and hysteresis), comparator U1C performs a Boolean  
“X*_Y” function and produces an output pulse (see Fig-  
ure 13). The circuit will give full output response with  
input delays down to 3ns and partial output response with  
inputdelaysdownto1.8ns.CapacitorC1helpsensurethat  
an imbalance of parasitic capacitances in the layout will  
not cause common mode excursions to result in differen-  
tial mode signal and false outputs.1  
The circuits of Figure 11 show basic building blocks for  
differential timing skews. The 2.5k resistance interacts  
with the 2pF typical input capacitance to create at least  
±4ns delay, controlled by the potentiometer setting.  
A differential and a single-ended version are shown. In the  
differentialconfiguration,theoutputedgescanbesmoothly  
scrolled through t = 0 with negligible interaction.  
3ns Delay Detector  
1 Make sure the input levels at X and Y are not too close to the 0.5V threshold set by the R8–R9  
divider. If you are still getting false outputs, try increasing C1 to 10pF or more. You can also look  
fortheproblemintheimpedancebalance(R5||R6=R7)attheinputsofU1C. Increasingtheoffset  
by lowering R5 will help reject false outputs, but R7 should also be lowered to maintain impedance  
balance. For ease of design and parasitic matching, R7 can be replaced by two parallel resistors  
equal to R5 and R6.  
It is often necessary to measure comparative timing of  
pulseedgesinordertodeterminethetruesynchronicityof  
clock and control signals, whether in digital circuitry or in  
LT1720  
LT1720  
C
IN  
C
IN  
+
+
INPUT  
2.5k  
C
IN  
C
IN  
0ns TO 4ns  
SINGLE-ENDED  
DELAY  
DIFFERENTIAL ±4ns  
RELATIVE SKEW  
INPUT  
2.5k  
C
IN  
C
IN  
+
+
C
IN  
C
IN  
V
V
REF  
REF  
1720/21 F11  
Figure 11. Building Blocks for Timing Skew Generation with the LT1720  
16  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
DELAY DETECTOR  
5V  
R5  
1.82k*  
OPTIONAL LOGARITHMIC PULSE STRETCHER (SEE TEXT)  
+
X
301*  
R6  
301*  
U1A  
5V  
1/4 LT1721  
51*  
CAPTURE  
0.33µF  
R1  
1N5711  
499*  
Z
U1C  
R8*  
4.53k  
C1  
5.6pF  
301*  
+
1/4 LT1721  
V
V
U1D  
IN  
C
R9  
487*  
5V  
0.1µF  
L
+
1/4 LT1721  
C2  
540pF  
**  
5V  
301*  
U1B  
1/4 LT1721  
475*  
Y
+
+
R7  
261*  
51*  
R2  
1k*  
R3  
1*  
V
OFF  
1V  
R4  
DELAY  
X
Y
Z
30*  
0V  
1V  
0V  
1720/21 F12  
* 1% METAL FILM RESISTOR  
** 270pF ×2 FOR REDUCED LEAD INDUCTANCE  
5V  
0V  
RESULT OF X AND NOT Y  
Figure 12. 3ns Delay Detector with Logarithmic Pulse Stretcher  
X
Y
Z
Figure 13. Output Pulse Due to Delay of Y Input Pulse  
17  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
Optional Logarithmic Pulse Stretcher  
For simplicity, with tP < τ1, and neglecting the very slight  
delay in turn-on due to offset and hysteresis, the equation  
can be approximated by:  
The fourth comparator of the quad LT1721 can be put to  
work as a logarithmic pulse stretcher. This simple circuit  
can help tremendously if you don’t have a fast enough  
oscilloscope(orcontrolcircuit)toeasilycapture3nspulse  
widths (or faster). When an input pulse occurs, C2 is  
charged up with a 180ns capture2 time constant. The  
hysteresis and 10mV offset across R3 are overcome  
within the first nanosecond3, switching the comparator  
output high. When the input pulse subsides, C2 dis-  
chargeswitha540nstimeconstant,keepingthecompara-  
tor on until the decay overrides the 10mV offset across R3  
minus hysteresis. Because of this exponential decay, the  
output pulse width will be proportional to the logarithm of  
theinputpulsewidth. Itisimportanttobypassthecircuit’s  
VCC well to avoid coupling into the resistive divider. R4  
keepsthequiescentinputvoltageinarangewhereforward  
leakage of the diode due to the 0.4V VOL of the driving  
comparator is not a problem.  
tOUT = τ2 ln [(VCH • tP/τ1)/(VOFF – VH/2)]  
(2)  
For example, an 8ns input pulse gives a 1.67µs output  
pulse. Doubling the input pulse to 16ns lengthens the  
output pulse by 0.37µs. Doubling the input pulse again to  
32ns adds another 0.37µs to the output pulse, and so on.  
The rate of 0.37µs per octave falls out of the above  
equation as:  
tOUT/octave = τ2 ln(2)  
(3)  
There is ±0.01µs jitter5 in the output pulse which gives an  
uncertainty referred to the input pulse of less than 2%  
(60ps resolution on a 3ns pulse with a 60MHz oscillo-  
scope—not bad!). The beauty of this circuit is that it gives  
resolution precisely where it’s hardest to get. The jitter is  
due to a combination of the slow decay of the last few  
millivolts on C2 and the 4nV/Hz noise and 400MHz  
bandwidthoftheLT1721inputstage.Increasingtheoffset  
across R3 or decreasing τ2 will decrease this jitter at the  
expense of dynamic range.  
Neglectingsomeeffects4, theoutputpulseisrelatedtothe  
input pulse as:  
tOUT = τ2 ln {VCH • [1 – exp (–tP/τ1)]/(VOFF – VH/2)}  
τ1 ln [VCH/(VCH – VOFF – VH/2)]  
The circuit topology itself is extremely fast, limited theo-  
retically only by the speed of the diode, the capture time  
constant τ1 and the pulse source impedance. Figure 14  
shows results achieved with the implementation shown,  
compared to a plot of equation (1). The low end is limited  
by the delivery time of the upstream comparators. As the  
input pulse width is increased, the log function is con-  
strained by the asymptotic RC response but, rather than  
becoming clamped, becomes time linear. Thus, for very  
long input pulses the third term of equation (1) dominates  
and the circuit becomes a 3µs pulse stretcher.  
+ tP  
(1)  
where  
tP = input pulse width  
tOUT = output pulse width  
τ1 = R1 || R2 • C2  
τ2 = R2 • C2  
the capture time constant  
the decay time constant  
the voltage drop across R1  
LT1721 hysteresis  
VOFF = 10mV  
2 So called because the very fast input pulse is “captured,” for later examination, as a charge on the  
capacitor.  
VH = 3.5mV  
3 Assumingtheinputpulseslewrateatthediodeisinfinite.Thiseffectivedelayconstant,about0.4%  
of τ1 or 0.8ns, is the second term of equation 1, below. Driven by the 2.5ns slew-limited LT1721,  
this effective delay will be 2ns.  
4 VC is dependent on the LT1721 output voltage and nonlinear diode characteristics. Also, the  
Thevenin equivalent charge voltage seen by C2 is boosted slightly by R2 being terminated above  
ground.  
5 Output jitter increases with inputs pulse widths below ~3ns.  
VC = VIN – VFDIODE  
the input pulse voltage after  
the diode drop  
VCH = VC • R2/(R1 + R2) the effective source voltage  
for the charge  
18  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
14  
easily measured 1.70µs output pulses. A 12 foot cable  
length difference will result in ~18.4ns delay and 2.07µs  
output pulses. The difference in the two output pulse  
widths is the per-octave response of your circuit (see  
equation (3)). Shorter cable length differences can be  
used to get a plot of circuit performance down to 1.5ns (if  
any), which can then later be used as a lookup reference  
whenyouhavemovedfromquantifyingthecircuittousing  
the circuit. (Note there is a slight aberration in perfor-  
mance below 10ns. See Figure 14.) As a final check, feed  
the circuit with identical cable lengths and check that it is  
not producing any output pulses.  
12  
10  
8
6
MEASURED  
4
EQUATION 1  
2
0
1
10  
100  
(ns)  
1000  
10000  
t
PULSE  
1720/21 F14  
Figure 14. Log Pulse Stretcher Output Pulse vs Input Pulse  
10ns Triple Overlap Generator  
The circuit of Figure 16 utilizes an LT1721 to generate  
three overlapping outputs whose pulse edges are sepa-  
ratedby10nsasshown. ThetimeconstantissetbytheRC  
network on the output of comparator A. Comparator B and  
Dtripatfixedpercentagesoftheexponentialvoltagedecay  
across the capacitor. The 4.22kfeed-forward to the C  
comparator’s inverting input keeps the delay differences  
the same in each direction despite the exponential nature  
of the RC network’s voltage.  
NANOSECOND  
INPUT RANGE  
MICROSECOND  
OUTPUT RANGE  
X
Y
1 FOOT CABLE  
L
t
OUT  
(SEE TEXT)  
CIRCUIT OF  
FIGURE 12  
2V  
0V  
There is a 15ns delay to the first edge in both directions,  
due to the 4.5ns delay of two LT1721 comparators, plus  
6ns delay in the RC network. This starting delay is short-  
ened somewhat if the pulse was shorter than 40ns be-  
cause the RC network will not have fully settled; however,  
the 10ns edge separations stay constant.  
SPLITTER  
n FOOT CABLE  
1720/21 F15  
Thevaluesshownutilizeonlythelowest75%ofthesupply  
voltagespan, whichallowsittoworkdownto2.7Vsupply.  
Thedelaydifferencesgrowacouplenanosecondsfrom5V  
to 2.7V supply due to the fixed VOL/VOH drops which grow  
as a percentage at low supply voltage. To keep this effect  
to a minimum, the 1kpull-up on comparator A provides  
equal loading in either state.  
Figure 15. RG-58 Cable with Velocity of Propogation = 66%;  
Delay at Y = (n – 1) • 1.54ns  
You don’t need expensive equipment to confirm the actual  
overall performance of this circuit. All you need is a  
respectablewaveformgenerator(capableof>~100kHz), a  
splitter, a variety of cable lengths and a 20MHz or 60MHz  
oscilloscope. Split a single pulse source into different  
cable lengths and then into the delay detector, feeding the  
longer cable into the Y input (see Figure 15). A 6 foot cable  
length difference will create a ~9.2ns delay (using 66%  
propagation speed RG-58 cable), and should result in  
Fast Waveform Sampler  
Figure 17 uses a diode-bridge-type switch for clean, fast  
waveform sampling. The diode bridge, because of its  
inherent symmetry, provides lower AC errors than other  
semiconductor-based switching technologies. This cir-  
cuit features 20dB of gain, 10MHz full power bandwidth  
19  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
V
CC  
V
CC  
OUTPUTS  
+
U1B  
1k  
1/4 LT1721  
V
CC  
10ns  
10ns  
V
CC  
INPUT  
750Ω  
909Ω  
+
1.37k  
215Ω  
681Ω  
U1A  
+
1/4 LT1721  
U1C  
100pF  
V
REF  
1/4 LT1721  
10ns  
10ns  
681Ω  
4.22k  
+
U1D  
1/4 LT1721  
453Ω  
1720/21 F16  
Figure 16. 10ns Triple Overlap Generator  
5V  
2.2k  
2.2k  
INPUT  
±100mV FULL SCALE  
+
OUTPUT  
±1V FULL SCALE  
LT1227  
1k  
909Ω  
= 1N5711  
AC BALANCE  
3pF  
= CA3039 DIODE ARRAY  
(SUBSTRATE TO 5V)  
100Ω  
5V  
1.5k  
3.6k  
1.1k  
1.1k  
0.1µF  
1.1k  
+
C
IN  
1/2 LT1720  
1.1k  
MRF501  
MRF501  
SKEW  
COMP  
2k  
10pF  
SAMPLE  
COMMAND  
DC BALANCE  
2.5k  
+
500Ω  
680Ω  
11  
8
1/2 LT1720  
2k  
820Ω  
820Ω  
6
9
C
IN  
LM3045  
13  
10  
51Ω  
7
51Ω  
1720/21 F17  
–5V  
Figure 17. Fast Waveform Sampler Using the LT1720 for Timing-Skew Compensation  
20  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
and 100µV/°C baseline uncertainty. Switching delay is  
less than 15ns and the minimum sampling window width  
for full power response is 30ns.  
is adjusted for minimal bridge ON vs OFF variation at the  
output. The skew compensation and AC balance adjust-  
ments are then optimized for minimum AC disturbance in  
the output. Finally, unground the input and the circuit is  
ready for use.  
The input waveform is presented to the diode bridge  
switch, the output of which feeds the LT1227 wideband  
amplifier.TheLT1720comparators,triggeredbythesample  
command, generate phase-opposed outputs. These sig-  
nals are level shifted by the transistors, providing comple-  
mentary bipolar drive to switch the bridge. A skew  
compensation trim ensures bridge-drive signal simulta-  
neity within 1ns. The AC balance corrects for parasitic  
capacitive bridge imbalances. A DC balance adjustment  
trims bridge offset.  
Voltage-Controlled Clock Skew Generator  
It is sometimes necessary to generate pairs of identical  
clock signals that are phase skewed in time. Further, it is  
desirable to be able to set the amount of time skew via a  
tuning voltage. Figure 18’s circuit does this by utilizing  
theLT1720todigitizephaseinformationfromavaractor-  
tuned time domain bridge. A 0V to 2V control signal  
provides ±10ns of output skew. This circuit operates  
from a 2.7V to 6V supply.  
The trim sequence involves grounding the input via 50Ω  
and applying a 100kHz sample command. The DC balance  
CLOCK  
INPUT  
V
CC  
2.7V TO 6V  
+
C1  
FIXED  
OUTPUT  
Q
1/2 LT1720  
V
CC  
2k  
2.5k  
2.5k  
2.5k*  
10ns  
TRIM  
14k  
2k*  
“FIXED”  
“SKEWED”  
+
C2  
SKEWED  
OUTPUT  
36pF  
Q′  
1/2 LT1720  
12pF  
MV-209  
VARACTOR  
DIODE  
1M  
1M  
0.1µF  
0.005µF  
47µF  
INPUT  
L1**  
0V TO 2V ≈  
±10ns  
+
V
CC  
A1  
LT1077  
+
SKEW  
1.1M  
100k  
2.2µF  
V
SW  
IN  
= 1N4148  
= 74HC04  
6.2M*  
LT1317 FB  
GND  
V
C
200pF  
* 1% FILM RESISTOR  
** SUMIDA CD43-100  
1.82M*  
POLYSTYRENE, 5%  
1720/21 F18  
Figure 18. Voltage-Controlled Clock Skew  
21  
LT1720/LT1721  
U
W U U  
APPLICATIONS INFORMATION  
Coincidence Detector  
AND gate could instead be used, but would add consider-  
ably more delay than the 300ps contributed by this dis-  
crete stage.  
High speed comparators are especially suited for interfac-  
ing pulse-output transducers, such as particle detectors,  
to logic circuitry. The matched delays of a monolithic dual  
are well suited for those cases where the coincidence of  
two pulses needs to be detected. The circuit of Figure 19  
isacoincidencedetectorthatusesanLT1720anddiscrete  
components as a fast AND gate.  
This circuit can detect coincident pulses as narrow as 3ns.  
For narrower pulses, the output will degrade gracefully,  
responding, but with narrow pulses that don’t rise all the  
way to “high” before starting to fall. The decision delay is  
4.5ns with input signals 50mV or more above the refer-  
encelevel.ThiscircuitcreatesaTTLcompatibleoutputbut  
it can typically drive CMOS as well.  
Thereferencelevelissetto1V,anarbitrarythreshold.Only  
when both input signals exceed this will a coincidence be  
detected. The Schottky diodes from the comparator out-  
puts to the base of the MRF-501 form the AND gate, while  
the other two Schottkys provide for fast turn-off. A logic  
For a more detailed description of the operation of this  
circuit, see Application Note 75, pages 10 and 11.  
5V  
5V  
GROUND  
CASE LEAD  
300Ω  
+
MRF501  
OUTPUT  
1/2 LT1720  
51Ω  
5V  
3.9k  
1k  
0.1µF  
1/2 LT1720  
+
300Ω  
4× 1N5711  
51Ω  
1720/21 F19  
300ps AND GATE  
COINCIDENCE COMPARATORS  
Figure 19. A 3ns Coincidence Detector  
22  
LT1720/LT1721  
W
W
SI PLIFIED SCHE ATIC  
23  
LT1720/LT1721  
U
PACKAGE DESCRIPTION Dimensions in inches (millimeters) unless otherwise noted.  
S8 Package  
8-Lead Plastic Small Outline (Narrow 0.150)  
(LTC DWG # 05-08-1610)  
0.189 – 0.197*  
(4.801 – 5.004)  
7
5
8
6
0.150 – 0.157**  
(3.810 – 3.988)  
0.228 – 0.244  
(5.791 – 6.197)  
1
3
4
2
0.010 – 0.020  
(0.254 – 0.508)  
× 45°  
0.053 – 0.069  
(1.346 – 1.752)  
0.004 – 0.010  
(0.101 – 0.254)  
0.008 – 0.010  
(0.203 – 0.254)  
0°– 8° TYP  
0.016 – 0.050  
(0.406 – 1.270)  
0.050  
(1.270)  
BSC  
0.014 – 0.019  
(0.355 – 0.483)  
TYP  
*DIMENSION DOES NOT INCLUDE MOLD FLASH. MOLD FLASH  
SHALL NOT EXCEED 0.006" (0.152mm) PER SIDE  
**DIMENSION DOES NOT INCLUDE INTERLEAD FLASH. INTERLEAD  
FLASH SHALL NOT EXCEED 0.010" (0.254mm) PER SIDE  
SO8 1298  
24  
LT1720/LT1721  
U
PACKAGE DESCRIPTION Dimensions in inches (millimeters) unless otherwise noted.  
MS8 Package  
8-Lead Plastic MSOP  
(LTC DWG # 05-08-1660)  
0.118 ± 0.004*  
(3.00 ± 0.102)  
8
7
6
5
0.118 ± 0.004**  
(3.00 ± 0.102)  
0.193 ± 0.006  
(4.90 ± 0.15)  
1
2
3
4
0.040 ± 0.006  
(1.02 ± 0.15)  
0.034 ± 0.004  
(0.86 ± 0.102)  
0.007  
(0.18)  
0° – 6° TYP  
SEATING  
PLANE  
0.012  
(0.30)  
REF  
0.021 ± 0.006  
(0.53 ± 0.015)  
0.006 ± 0.004  
(0.15 ± 0.102)  
0.0256  
(0.65)  
BSC  
MSOP (MS8) 1098  
* DIMENSION DOES NOT INCLUDE MOLD FLASH, PROTRUSIONS OR GATE BURRS. MOLD FLASH,  
PROTRUSIONS OR GATE BURRS SHALL NOT EXCEED 0.006" (0.152mm) PER SIDE  
** DIMENSION DOES NOT INCLUDE INTERLEAD FLASH OR PROTRUSIONS.  
INTERLEAD FLASH OR PROTRUSIONS SHALL NOT EXCEED 0.006" (0.152mm) PER SIDE  
25  
LT1720/LT1721  
U
PACKAGE DESCRIPTION Dimensions in inches (millimeters) unless otherwise noted.  
S Package  
16-Lead Plastic Small Outline (Narrow 0.150)  
(LTC DWG # 05-08-1610)  
0.386 – 0.394*  
(9.804 – 10.008)  
16  
15  
14  
13  
12  
11  
10  
9
0.150 – 0.157**  
(3.810 – 3.988)  
0.228 – 0.244  
(5.791 – 6.197)  
5
7
8
1
2
3
4
6
0.010 – 0.020  
(0.254 – 0.508)  
× 45°  
0.053 – 0.069  
(1.346 – 1.752)  
0.004 – 0.010  
(0.101 – 0.254)  
0.008 – 0.010  
(0.203 – 0.254)  
0° – 8° TYP  
0.050  
(1.270)  
BSC  
0.014 – 0.019  
(0.355 – 0.483)  
TYP  
0.016 – 0.050  
(0.406 – 1.270)  
S16 1098  
*DIMENSION DOES NOT INCLUDE MOLD FLASH. MOLD FLASH  
SHALL NOT EXCEED 0.006" (0.152mm) PER SIDE  
**DIMENSION DOES NOT INCLUDE INTERLEAD FLASH. INTERLEAD  
FLASH SHALL NOT EXCEED 0.010" (0.254mm) PER SIDE  
26  
LT1720/LT1721  
U
PACKAGE DESCRIPTION Dimensions in inches (millimeters) unless otherwise noted.  
GN Package  
16-Lead Plastic SSOP (Narrow 0.150)  
(LTC DWG # 05-08-1641)  
0.189 – 0.196*  
(4.801 – 4.978)  
0.009  
(0.229)  
REF  
16 15 14 13 12 11 10 9  
0.229 – 0.244  
(5.817 – 6.198)  
0.150 – 0.157**  
(3.810 – 3.988)  
1
2
3
4
5
6
7
8
0.015 ± 0.004  
(0.38 ± 0.10)  
× 45°  
0.053 – 0.068  
(1.351 – 1.727)  
0.004 – 0.0098  
(0.102 – 0.249)  
0.007 – 0.0098  
(0.178 – 0.249)  
0° – 8° TYP  
0.016 – 0.050  
(0.406 – 1.270)  
0.0250  
(0.635)  
BSC  
0.008 – 0.012  
(0.203 – 0.305)  
* DIMENSION DOES NOT INCLUDE MOLD FLASH. MOLD FLASH  
SHALL NOT EXCEED 0.006" (0.152mm) PER SIDE  
** DIMENSION DOES NOT INCLUDE INTERLEAD FLASH. INTERLEAD  
FLASH SHALL NOT EXCEED 0.010" (0.254mm) PER SIDE  
GN16 (SSOP) 1098  
Information furnished by Linear Technology Corporation is believed to be accurate and reliable.  
However, no responsibility is assumed for its use. Linear Technology Corporation makes no represen-  
tationthattheinterconnectionofitscircuitsasdescribedhereinwillnotinfringeonexistingpatentrights.  
27  
LT1720/LT1721  
U
TYPICAL APPLICATION  
Pulse Stretcher  
capacitor C now begins charging through R and, at the  
end of 100ns, C2 resets low. The output of C1 also goes  
low, latching both outputs low. A new pulse at the input  
of C1 can now restart the process. Timing capacitor C can  
be increased without limit for longer output pulses.  
For detecting short pulses from a single sensor, a pulse  
stretcher is often required. The circuit of Figure 20 acts as  
a one-shot, stretching the width of an incoming pulse to a  
consistent 100ns. Unlike a logic one-shot, this LT1720-  
based circuit requires only 100pV-s of stimulus to trigger.  
This circuit has an ultimate sensitivity of better than 14mV  
with 5ns to 10ns input pulses. It can even detect an  
avalanche generated test pulse of just 1ns duration with  
sensitivity better than 100mV.6 It can detect short events  
better than the coincidence detector of Figure 14 because  
the one-shot is configured to catch just 100mV of upward  
movement from C1’s VOL, whereas the coincidence  
detector’s 3ns specification is based on a full, legitimate  
logic high, without the help of a regenerative one-shot.  
The circuit works as follows: Comparator C1 functions as  
a threshold detector, whereas comparator C2 is config-  
uredasaone-shot. Therstcomparatorisprebiasedwith  
a threshold of 8mV to overcome comparator and system  
offsets and establish a low output in the absence of an  
input signal. An input pulse sends the output of C1 high,  
which in turn latches C2’s output high. The output of C2  
is fed back to the input of the first comparator, causing  
regeneration and latching both outputs high. Timing  
6 See Linear Technology Application Note 47, Appendix B. This circuit can detect the output of the  
pulse generator described after 40dB attenuation.  
5V  
0.01µF  
15k  
OUTPUT  
C1  
PULSE SOURCE  
1/2 LT1720  
50Ω  
100ns  
+
51Ω  
24Ω  
R
1k  
6.8k  
1N5711  
C
100pF  
C2  
1/2 LT1720  
+
2k  
2k  
1720/21 F20  
2k  
Figure 20. A 1ns Pulse Stretcher  
RELATED PARTS  
PART NUMBER  
DESCRIPTION  
COMMENTS  
LT1016  
UltraFast Precision Comparator  
Industry Standard 10ns Comparator  
Single Supply Version of LT1016  
6mA Single Supply Comparator  
450µA Single Supply Comparator  
Single Comparator Similar to the LT1720/LT1721  
LT1116  
12ns Single Supply Ground-Sensing Comparator  
7ns, UltraFast, Single Supply Comparator  
60ns, Low Power, Single Supply Comparator  
4.5ns Single Supply 3V/5V Comparator  
LT1394  
LT1671  
LT1719  
17201f LT/TP 1099 4K • PRINTED IN USA  
28 LinearTechnology Corporation  
1630 McCarthy Blvd., Milpitas, CA 95035-7417  
LINEAR TECHNOLOGY CORPORATION 1998  
(408)432-1900 FAX:(408)434-0507 www.linear-tech.com  

相关型号:

LT1720IS8#PBF

LT1720 - Dual, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs; Package: SO; Pins: 8; Temperature Range: -40&deg;C to 85&deg;C
Linear

LT1720IS8#TR

LT1720 - Dual, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs; Package: SO; Pins: 8; Temperature Range: -40&deg;C to 85&deg;C
Linear

LT1720IS8#TRPBF

LT1720 - Dual, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs; Package: SO; Pins: 8; Temperature Range: -40&deg;C to 85&deg;C
Linear

LT1720_15

Dual/Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs
Linear

LT1721

Dual/Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs
Linear

LT1721CGN

Dual/Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs
Linear

LT1721CGN#PBF

LT1721 - Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs; Package: SSOP; Pins: 16; Temperature Range: 0&deg;C to 70&deg;C
Linear

LT1721CGN#TR

LT1721 - Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs; Package: SSOP; Pins: 16; Temperature Range: 0&deg;C to 70&deg;C
Linear

LT1721CGN#TRPBF

LT1721 - Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs; Package: SSOP; Pins: 16; Temperature Range: 0&deg;C to 70&deg;C
Linear

LT1721CS

Dual/Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs
Linear

LT1721CS#TRPBF

LT1721 - Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs; Package: SO; Pins: 16; Temperature Range: 0&deg;C to 70&deg;C
Linear

LT1721IGN

Dual/Quad, 4.5ns, Single Supply 3V/5V Comparators with Rail-to-Rail Outputs
Linear