Piper Autocontrol

This is the placeholder page for all the information that I have managed to assemble regarding the Piper Autocontrol I and II.  There is quite a bit of information so far including:

• Service manual for the Autocontrol I
• Service manual for the Autocontrol II
• Operating instructions for Autocontrol II
• Installation paperwork
• Reverse engineered schematic 
• Simulation file for LTSpice
• Sperry service manual for the gyros
• Various photographs of the components
• Cross reference of older autopilots

Another source of experience;  https://miscsolutions.wordpress.com/2011/02/16/troubleshooting-old-century-autopilots

An excellent write up on gyro instruments:  http://www.1journey.net/1candle/av/FL-107/FL-107.htm

The artificial horizon gyro is a modified AN 5736-1.  The part number is 52B9


The directional gyro is a modified AN 5735-1.  The part number is an 52D27E








Schematic Explanation:

The schematic of the Autocontrol I was reverse engineered from a non-functional example.  There is a considerable discussion of the the theory of operation in the two service manuals.  There are a few distinct sections that follow it works.

Oscillator:

There is an RF oscillator that is designed to operate near 10.7 MHz which is a frequency commonly used in FM radio.  This means that various components and circuitry would be reasonably available.  The frequency that it operates at is set by the resonance of a tank circuit.  There is an inductor in the attitude indicator in parallel with various capacitors in the control system.

Inductor "Lvar":

The "roll plate inductor" is installed in the attitude gyro.  Metal ferrous and aluminum plates mounted on the roll gimbal vary the inductance as the gymbal rolls.  This then varies the frequency at which the oscillator is resonating.

Capacitors "Cvar":

Air gap variable capacitors of between 3-30 pF are mounted in parallel (this means the capacitance simply adds together).  The roll (or trim) capacitor is connected to a knob on the amplifier.  The servo feedback capacitor is mounted at the servo.  A small trim capacitor is swapped back and forth with the heading gyro capacitor.  The sum of these capacitors reacts with the roll plate inductor to vary the frequency at which the oscillator is resonating.

Limiting Amplifier:

A transistor configured as an emitter-follower amplifies the output.  This provides a high impedance and prevents the discriminator coil from affecting the oscillator.

Discriminator Coil:

This is a lightly coupled coil with input and outputs that resonate at given frequency.  What it is doesn't really matter as differences are trimmed out during initial set up.  It is arranged with a rather unusual centre-tapped capacitor, rather than the more commonly illustrated centre-tapped secondary coil.  This provides two signals on the output that provide greater or lesser amplitude outputs depending on whether or not the frequency is above or below the centre frequency.

Foster-Seely Circuit:

The diode, resistor, and capacitor network rectify the AC signals and provide an output signal.  This signal is biased up or down to produce a small dead-band in the middle so that the motor isn't always running.  If the frequency is below the resonant frequency then the output on one line is at high voltage and the other is low.  If it is above the resonant frequency then the output on the other line is at high voltage.  If the oscillator is at or near the resonant frequency then the output on both lines is at a low voltage.

Switching Amplifiers:

The output of the Foster-Seely circuit cannot be used to drive an electric motor.  So it is fed through a pair of transistors that are either fully on or fully off.  If the output voltage connected to the base of the first transistor is about 0.3 volts below the supply voltage (for germanium) the transistor conducts.  This current then raises the voltage of the base of the second transistor high and turns it completely off.  Likewise if the first transistor is biased off then the ground connection at the base of the second pulls it low and it turns completely on.

Power Transistor H-Bridge:

This output can now sink the current from the base of the power transistor network.  If an output transistor is conducting it can pull the voltage of the resistor network down which will forward bias all of the pnp power transistors.  This allows current to flow through the motor.  Note that if at any time both sides get pulled low the system essentially shorts through the common collectors on either side of the bridge and that would blow the intrinsic 3 amp fuse.  There are also none of the various flyback diodes and so forth seen in a modern circuit.  Probably because the Ge transistors are leaky enough not to need it.

Silicon vs Germanium:

The old system was designed in the late 50's.  The transistors were germanium based.  The circuit is modeled using silicon transistor part numbers and there was only one resistor added to account for the extra gain.  The h-bridge should also have current fly-back diodes to deal with the various surges of current that will be produced by pushing one and two amp loads back and forth through the motor as it stops and reverses.

Component Values:

The component values were established via direct inspection, reading the Autocontrol I service manual, or back calculation from what was required to make it work.  For example, given the resonant frequency required and the value of the capacitors in the discriminator coil, the value of the inductors can be calculated.  The roll plate inductor and discriminator coils were the ones that needed to be calculated as all others could be directly measured.