©2008 Fairchild Semiconductor Corporation www.fairchildsemi.com
FOD8001 Rev. 1.0.3 8
FOD8001 — High Noise Immunity, 3.3V/5V Logic Gate Optocoupler
Application Information
Noise is defined as any unwanted signal that degrades
or interferes with the operation of a system or circuit.
Input-output noise rejection is a key characteristic of an
optocoupler, and the performance specification for this
noise rejection is called, “Common Mode Transient
Immunity or Common Mode Rejection, CMR”. The CMR
test configuration is presented in high speed optocoupler
datasheets, which tests the optocoupler to a specified
rate of interfering signal (dv/dt), at a specified peak volt-
age (Vcm).
This defined noise signal is applied to the test device
while the coupler is a stable logic high or logic low state.
This test procedure evaluates the interface device in a
constant or static logic state. This type of CMR can be
referred to as “Static CMR”. Fairchild’s high speed opto-
couplers, which use an optically transparent, electrically
conductive shield, and offer active totem pole logic out-
put have static CMR in excess of 50KV/us at peak ampli-
tudes of 1.5kV to 2.0kV.
Dynamic Common Mode Rejection
The noise susceptibility of an interface while it is actively
transferring data is a common requirement in serial data
communication. However, the static CMR specification is
not adequate in quantifying the electrical noise suscepti-
bility for optocouplers used in isolating high speed data
transfer.
A serial data communication network’s noise perfor-
mance is usually quantified as the number of bit errors
per second or as a ratio of the number of bits transmitted
in a specified time frame. This describes Bit Error Rate,
BER. Test equipment that evaluates BER is called a Bit
Error Rate Tester, BERT. When a BERT system is com-
bined with a CMR tester, the active or dynamic noise
rejection of an isolated interface can then be quantified.
This type of CMR is thus defined as “Dynamic CMR”.
Therefore, evaluating the common mode rejection while
the optocoupler is switching at high speed represents a
realistic approach to understand noise interference.
Test circuit functions were built to interface a commercial
pseudo-random bit sequence (PRBS) generator and
error detector with a pair of high speed optocouplers,
FOD8001, connected in a loop-back configuration. With
a 10MBaud PRBS serial data stream, no error was
detected until the common mode voltage rose above
2.5kV with a dv/dt of 45kV/us. And increasing the data
rate beyond 10Mbaud, the test was conducted at
20MBaud, and no error was detected at dv/dt of 25kV/us
at common mode voltage of 1.5kV.
The test data for the dynamic CMR is comparable or
better than the static CMR specifications found in the
datasheet. These excellent noise rejection performances
are results of the innovative circuit design and the
Power Supply Noise Rejection
High levels of electrical noise can cause the optocoupler
to register the incorrect logic state. The most commonly
discussed noise signal is the common mode noise found
between the input and output of the optocoupler. How-
ever, common mode noise is not the only path of noise
into the input or output of the optocoupler. Due to the
high gain and wide bandwidth of the transimpedance
amplifier used for the photo detector circuits, power sup-
ply noise can cause the optocoupler to change state
independent of the LED operation. Power supply noise is
typically characterized as either random or periodic
pulses with varying amplitudes and rates of rise and fall.
The necessary tests have been conducted to understand
the influence of the power supply noise and its effect of
the proper operation of the FOD8001. The optocoupler
under test offered power supply noise rejection in excess
of 10% of the supply voltage for a frequency ranging
from 100kHz to 35MHz, for logic high and logic low
states.
proprietary coplanar assembly process.