The AD654 is a voltage-to-frequency converter that is often used to convert sensor-derived information in voltage form, such as the voltage obtained from a temperature sensor, into frequency, in order to reliably transmit the information from one point to another. In many situations, frequency transmission is more robust than simple voltage transmission. In this experiment we construct a simple circuit to flash a LED at a frequency that depends on the voltage applied to the AD654.
To build a simple LED flashing circuit using one IC and a few external components. To show a simple negative feedback amplifier, and show how to compensate for input bias currents. Following completion of this lab you should be able to describe the basic operation of the AD654 and its major application, explain what bias currents are and how to compensate for them, and describe the basic behavior of the two inputs of an op-amp that is connected in a negative feedback configuration.
The input voltage (Pin 4) range for single-supply operation is specified as 0 V to +VS – 4 V, which in this case is 0 V to 1 V. While this specification is for normal linear operation, the absolute maximum input limit is +VS on a single supply, so we are not in danger of damaging the chip with our input voltage that can run up to +5 V. The output frequency in the linear portion of the VIN range is defined as
For VIN = 1 V, we have f = 0.2 Hz.
The oscillation frequency is determined by RT and CT. In voltage-to-frequency conversion applications the oscillation frequency is typically set to be in the tens or hundreds of kilohertz, but for this lab it was set to flash the LED at a rate that was observable. The low rate requires a large RC time constant, which was obtained with the 0.1 μF capacitor and 5 MΩ resistor.
The op-amp requires a small input bias current on each input, and these currents are closely, but not perfectly matched. The current flowing through RT produces a voltage drop across RT, introducing an error into the frequency setting. Compensation resistor, RC, a resistor equal in value to RT, is added to the non-inverting input (Pin 4) to produce a nearly identical voltage drop is as is across RT. This technique minimizes the offset error between the two op-amp inputs, and produces the most accurate output frequency. This is an important consideration in circuits that are used as voltage-to-frequency converters, but not so important in simple LED flashing circuits. When observing the voltage on Pin 4, a jumper wire must be placed across RC (shown as dotted line in the schematic) in order to eliminate DC losses that occur due to the M1K input loading the 5 MΩ source resistance.
In a voltage-feedback op-amp circuit, negative feedback causes the feedback voltage on the inverting input to track the voltage on the non-inverting input. The voltage on Pin 3 is the feedback voltage, and this should track the voltage applied to Pin 4 very closely as long as the output of the emitter follower can follow it and the input common-mode range is not violated.