LTC1042M データシートの表示（PDF） - Linear Technology

部品番号

コンポーネント説明

メーカー

LTC1042M

Window Comparator

Linear Technology 

LTC1042

APPLICATIO S I FOR ATIO

The LTC1042 uses sampled data techniques to achieve its

unique characteristics. It consists of two comparators,

each of which has two differential inputs (Figure 1). When

the sum of the voltages on a comparator’s inputs is

positive, the output is high; when the sum is negative, the

output is low. The inputs are interconnected such that

when (CENTER – WIDTH/2) ≤ VIN ≤ (CENTER + WIDTH/2)

both comparator outputs are low. In this condition VIN is

within the window and the WITHIN WINDOW output is

high. When VIN > CENTER + WIDTH/2, VIN is above the

window and the ABOVE WINDOW output is high.

An important feature of the LTC1042 is the non-interaction

of the inputs. This means the center and width of the

window can be changed without one affecting the other.

Also note that the width of the window is set by a ground

referred signal WlDTH/2).

Strobing

An internal oscillator allows the LTC1042 to strobe itself.

The frequency of oscillation sets the sampling rate and is

set with an external RC network (see typical curve, OSC

frequency vs REXT, CEXT). To assure oscillation, under all

conditions, REXT must be between 100kΩ and 10MΩ.

There is no limit to the size of CEXT.

A sampling cycle is initiated on the positive going transi-

tion of the voltage on the OSC pin. When this voltage is

near the positive supply, a Schmitt trigger trips and

initiates the sampling cycle. A sampling cycle consists of

applying power to both comparators, sampling the inputs,

storing the results in CMOS output latches and turning the

power off. This whole process takes approximately 80µs.

During the 80µs “active” time, the LTC1042 draws

typically 1.2mA (lS(ON)) at V+ = 5V. Because power is

consumed only during the “active” time, extremely low

average power consumption can be achieved at low sample

rates. For example, at a sample rate of 1 sample/second

the average power consumption is:

Power = (V+) (IS(AVG)) = 5V • 1.2mA • 80µs/1sec

= 0.48µW

At low sampling rates, REXT dominates the power con-

sumption. REXT consumes power continuously. The aver-

age voltage at the OSC pin is approximately V+/2. The

power consumed by REXT is:

P(REXT) = (V+/2)2REXT

Example: Assume REXT = 1MΩ and V+ = 5V. Then:

P(REXT) = (2.5)2/1MΩ = 6.25µW

This is more than ten times the typical power consumed by

the LTC1042 at V+ = 5V and 1 sample/second. Where

power is a premium, REXT should be made as large as

possible. Note that the power dissipated by REXT is not a

function of the sampling frequency or CEXT.

If high sampling rates are needed and power consumption

is of secondary importance, a convenient way to get the

maximum possible sampling rate is to make REXT = 100kΩ

and CEXT = 0. The sampling rate, set by the LTC1042’s

active time, will nominally be ≈ 10kHz.

WINDOW

CENTER

2

(VIN)

VIN

(WINDOW 3

CENTER)

WIDTH/2 5

GND 4

OSC 7

+

–

+ COMP A

–

4

+

–

+ COMP B

–

4

TIMING

GENERATOR

(A)

4

8 V+

1 WITHIN WINDOW

6

ABOVE WINDOW

(BELOW WINDOW)

WINDOW

CENTER

V+

WITHIN

WINDOW

–WIDTH/2 WIDTH/2

ABOVE

WINDOW

0V

POWER ON

80µs

POWER OFF

Figure 1. LTC1042 Block Diagram

VL

VU

INPUT VOLTAGE, VIN

(B)

LTC1042 • AI01

1042fa

Share Link: