Excerpted from Email sent by Ed Lyon:
Subject: Re: Brown chopper amp
Date: Sat, 29 Dec 2007 08:42:19
The Brown amplifier you so nicely drew up is quite familiar to me, except for the vibrator power supply and the prime power of 28 Vdc [24Vdc ed.]. Most of the ones I worked with were in labs (Chemistry, and various biological labs), and ran directly from the 120-Vac line. Otherwise, the circuit is standard. The load, connected from the "amber short" centertap on the plate winding of the power transformer, to ground, is one winding of a two-phase a-c motor, with the other winding fed from the filament line (or in the lab models I used, from the a-c line). Then the motor turns CW with positive input d-c voltages and CCW for negative. Geared to the motor was a cable drum that dragged a ink-pen across a chart and an answer potentiometer. This pot had a d-c voltage applied that allowed its wiper to be fed back in series with the input voltage, and so the motor turning would drag the pen across the page to a point where the answer voltage would match the input, and the motor would stop. Sort of a big microvoltmeter. Very useful in measuring (and getting a paper record of) pH in the lab, as well as many other chemical reaction effects, like microphotometry, spectroscopy, etc. The chart recorder ordinarily pulled the chart linearly with time, but could be geared to a moving slit and photomultiplier tube in a closed spectroscope, and thus the spectral lines of an unknown specimen as it was burned in a carbon arc could be recorded. We used this setup in police forensic work with the 1.5-meter spectroscope in college. (Found that the glass particles in a suspects trouser cuffs were identical to glass in a jewelry store front window, placing him at the scene, whereas his alibi and so-called witnesses claimed he was in another city at the time.
You see how the amplifier works, I assume? The chopper puts the d-c input into the amplifier as either an in- phase a-c wave or an out-of- phase wave, WRT [With Respect To ed.] to the a-c line (power transformer primary). This is amplified, as you measured, and produces half-wave-rectified a-c current pulses of either + or - phase through the motor winding. The other motor winding is at fixed phase, fed from the power transformer with a 90-degree phasing capacitor. The d-c pulses fed to the motor act as self-damping currents, keeping it from coasting, and therefore, hunting. In fancier servos, they used drag-cup generators to produce a rate-of-turning voltage that was fed back to the amplifier to do the damping.
This very interesting email got me thinking further about the amplifier and I sent this response:
Date: Mon, 31 Dec 2007 01:24:44
Ed, thank you very much for your comprehensive comments to the Brown chopper amp. I will be posting your comments in the Philbrick Archive.
Your description of the how this amplifier was used in real life will be of interest to a wider audience, than just the purely technical aspects.
I forgot about the required 90o phase shift between the fixed winding drive and the variable winding drive. A cap is ideal to get this phase shift if the source AC is sinusoidal. Wouldn't an inductor be better for the 80% duty cycle square wave that the vibrator supplies?
But if the cap is put on the control winding, the peak currents will be limited by the triode drive. Perhaps there is enough inductance in the motor to limit peak currents with a square wave driving the cap, so the cap could be put in series with the fixed winding, or in shunt with the control winding.
One item that seemed a bit odd is the large DC component that the output supplies. Hmm, maybe having a large DC current driving the winding when the AC signal output is zero will make a strong DC electromagnet to assist with eddy current breaking to prevent coasting. Any increases in AC drive are taken directly from the DC component.
I learned about this basic type of servo loop from my Digitec 201 digital voltmeter. It uses an Airpax chopper, a solid state 60Hz IF amp and a small two phase induction motor driving a precision pot with a turns counter. The operation is pretty similar to the pen recorder you describe.
I just took a quick look at the schematic for the Digitec and it also has a class A drive to the motor, thus giving it a strong DC field when the AC signal goes to zero. Just this past week, a friend [Mark Stevens] gave me two specimens of a Kintel model 112a amplifier. It that looks like another servo loop amp. I attached a few photos. I have no documentation. Perhaps someday I will trace it out too. I wonder if you are familiar with this particular amp. All signals and power come in and out through a special connector in the back. The case is fashioned as a plugin.