As long as the signal is within these limits, all three tran-

sistors conduct and a low-resistance path is maintained

from the IN to OUT pin. (Note that, since the device is

symmetrical, IN and OUT pins can be interchanged.)

When the signal is beyond the gate threshold of either

Q2 or Q1/Q3, the path resistance rises dramatically.

When power is off, none of the transistors have gate

bias, so the circuit from IN to OUT is open.

Normal Operation

In normal operation, the protector is placed in series

with the signal line and the power supplies are con-

nected to V+ and V- (see Figure 2). V- is ground when

operating with a single supply. When power is applied,

each protector acts as a resistor in the signal path.

Any voltage source on the “input” side of the switch will

be conducted through the protector to the output. (Note

that, since the protector is symmetrical, IN and OUT

pins can be interchanged.)

If the output load is resistive, it will draw current, and a

voltage divider will be formed with the internal resistance

so the output voltage will be lower than the input voltage.

Since the internal resistance is typically less than 100

Ω,

high-impedance loads will be relatively unaffected by the

presence of the protector. The protector’s path resis-

tance is a function of the supply voltage and the signal

voltage (see

Typical Operating Characteristics).

Power Off

When power is off (i.e., V+ = V- = 0V), the protector is a

virtual open circuit, and all voltages on each side are

isolated from each other up to ±40V. With ±40V applied

to the input pin, the output pin will be 0V, regardless of

its resistance to ground.

Fault Conditions

A fault condition exists when the voltage on either sig-

nal pin is within about 1.5V of either supply rail or

exceeds either supply rail. This definition is valid when

power is applied and when it is off, as well as during all

the states as power ramps up or down.

During a fault, the protector acts as a variable resistor,

conducting only enough to sustain the other side of the

switch within about 1.5V of the supply rail. This voltage

is known as the “fault knee voltage,” and is not symmet-

rical. It is approximately 1.3V down from the positive

supply (V+ pin) or approximately 2.0V up from the neg-

ative supply (V- pin). Each fault knee voltage varies

slightly with supply voltage, with output current, and

from device to device.

During a fault condition, all the fault current flows

from one signal pin through the protector and out

the other signal pin. No fault current flows through

either supply pin. (There will be a few pico-amps of

leakage current from each signal pin to each supply

pin, but this is independent of fault current.)

During the fault condition, enough current will flow to

maintain the output voltage at the fault knee voltage, so

the fault current is a function of the output resistance

and the supply voltage. The output voltage and cur-

rent have the same polarity as the fault.

The maximum input fault voltage is 40V from the “oppo-

site-polarity supply rail.” This means the input can go

to ±35V with ±5V supplies or to ±25V with ±15V sup-

plies. The fault voltage is highest (±40V) when the sup-

plies are off (V+ = V- = 0V).

Using the circuit of Figure 2, the approximate fault cur-

rents are as follows:

1) For positive faults:

2) For negative faults:

MAX366

OUT1

1

4

IN1

7

V+

V-

V+

V-

8

Figure 2. Application Circuit

Signal-Line Circuit Protectors

6

_______________________________________________________________________________________