DESIGN OF SAFE MEDICAL DEVICE PROTOTYPES - Design and Development of Medical Electronic Instrumentation - page 113

Figure 3.9 In this EEG amplifier, an extra optoisolator is used to introduce offset automatically to pass bipolar signals and to linearize the
response of the isolation amplifier. IC3B drives the LEDs of optoisolators IC4 and IC5. The phototransistor in IC4 is used to set the invert-
ing input of IC3B such that the LED is driven to a point that balances the signal at IC3's noninverting input. When the phototransistor in IC5
is not illuminated, its collector is pulled up to the nonisolated positive supply rail by R21. However, as signals cause IC3B to drive the LED,
the phototransistor pulls the collector towards the nonisolated negative supply rail.

the di ff erential ampli fi er implemented by IC1A. This di ff erential ampli fi er has unity gain

and is used to add an o ff set voltage that is set via potentiometer R3. The input signal is then

low-pass filtered through R6 and C6 to the dc-65 Hz modulation bandwidth, which can be

supported by the pulse-width modulator implemented around IC1B. The pulse-train output
from the pulse-width modulator drives an LED inside optocoupler IC2 via transistor Q1.
The Optek OPI1264 optoisolator used in this circuit was selected because it has a very
respectable 10-kV isolation rating (although its UL

fi

file rates it at 3500 V ac for 1 minute).
At the phototransistor end of IC2, the light pulses generated by the LED are received
and converted into a replica of the original pulse train through the RC high-pass

fi

filter
formed by R13 and C8. The circuit centered on IC3A acts as a pulse-train demodulator to
reconstruct the original modulating signal, which, depending on the characteristics of the
input signal, may have an o

fi

ff set. Finally, the low-pass fi filter implemented around IC3B

removes any remnants of the pulse train's carrier.