10.5.1.1 Transition edge

The output of a phototransistor (in a tranmissive sensor) does not have sharp vertical edges when the sensor transitions (from light to dark or vice versa). This is because of physical properties of a transmissive sensor.

First, the IrED (infrared emitting diode) is not a point source of infrared. Instead, from the perspective of the phototransistor, it looks more like a disc of infrared. This means that during the transition from opague to clear (from the encoder disc), a phototransistor experiences a continuous change from full IrED disc to no IrED disc. This continuous change of IrED disc exposure translates to a continuous change of the output voltage.

For a normal CMOS/TTL input, this is a problem. This is a problem because the amount of time to transition from full exposure to the IrED disc to no exposure is longer than the response time of a CMOS/TTL gate. Most CMOS/TTL input have specifications similar to the following:

• High-level minimum input voltage: $\mathrm{V}_\mathrm{IH} = 3.15\mathrm{V}$
• Low-level maximum input voltage: $\mathrm{V}_\mathrm{IL} = 1.35\mathrm{V}$

This means that the device guarantees to register a one when the input voltage is 3.15V or more, and guarantees to register a zero when the input voltage is 1.35V or less. But what if the input voltage is, say, 2.5V? Is that a one, or is that a zero?

Many CMOS devices bounce the interpretation between 0 and 1 when the input voltage is between the high-level minimum and the low-level maximum. This leads to false transition readings.