原创 Organ failure caused by organic liquid

You may already know the Hammond B-3 organ—it was a fixture in rock bands in the 1960s and 1970s. You may not be as familiar with the Hammond A-100, which is a console version of this classic organ.

My mom has an early 1960s manufactured Hammond A-100 that was not working. She moved into a smaller house with no room for this spinet style organ, so we loaded it into the back of a truck (it's heavy!) and took it to our house. It sat untouched for almost three years, until I had a large chunk of time to take a look inside and diagnose the problem. This chunk of time came along in summer of 2004—the organ was about 40 years old at the time.

We also had an old Wurlitzer organ I had repaired previously, and it was full of vacuum tubes, maybe 35 or 40, many for oscillators. When I opened the Hammond I was surprised to find only about a dozen tubes in an internal power amp and a big, long, heavy box suspended inside the organ enclosure.

Simply as a matter of principle, I first replaced all the tubes in the amplifier and in the preamp of the built-in mechanical reverb. Afterward, the organ did in fact make tones, only not when the vibrato function was enabled. Before we get too deep into this story, some background on the Hammond "tone wheel" organs is needed.

I've seen these organs described as "over-engineered," but my own description is "masterpiece." This machine is one of the most amazing electromechanical consumer products I've ever seen. The basic tones are generated by what's known as a tonewheel that starts with a synchronous electric motor whose rotational speed is set by the frequency of its AC power source, in this case, 60Hz. A shaft extends from the motor on one side, all the way across the width of the organ's internal cavity. A set of tone wheels are mounted on this shaft, which are not fully round, but instead, toothed wheels. As the shaft rotates, the teeth extend to almost touch a magnetic bar placed at the edge of the wheel. This whole motor and shaft assembly is mounted in that big, long, heavy, spring suspension shock isolated metal box.

The magnets have a coil around their ends, and as a wheel's teeth pass by a magnet, they modulate the magnetic field, which in turn generates a current in the coil. The frequency of the AC current is set by the rotational speed of the wheel and the number of teeth on it. As an analogy, consider it as a guitar pickup that, instead of a vibrating steel string nearby, you have a rotating toothed wheel. The AC current of one or more of the 48 tone wheels is summed into an amplifier to make the organ sound, depending on which keys are pressed. Each key has a fundamental and eight harmonics.

The standard signal path was working, it was the vibrato that was broken. I bought a copy of the service manual from Manual Manor and studied the schematic. There is also a service manual web page that discusses the operation of the vibrato. It describes a series of delays, via a cascade of low pass filters, that feeds into a rotating capacitive pickup.

There is a cylindrical chamber driven from the same synchronous motor that drives the tone wheels, only on the other side of the shaft coupler. It connects taps from the delay line, in forward and reverse, to the amplifier's summing point by coupling them through a commutator inside the chamber via rotating air gap capacitors created by meshing parallel plates on a rotor with fixed plates in the chamber. After locating the signal path in and out of the assembly, an oscilloscope showed the signal out was shunted to ground when the vibrato was engaged.

In other words, there are a set of static parallel plates through which a set of moving plates pass. The phase delayed signals are coupled into one of several sets of plates that create capacitors. The phase delayed signals are summed back into the primary tone to create a true frequency modulated vibrato (vs. tremolo which is amplitude modulation).

The chamber (upper left), the parallel plates, (lower left except two commutator plates centre right) and the phase delayed signal connectors (lower right). Also you can see the synchronous motor at centre left, which incidentally has a separate starter motor because the synchronous motor doesn't have enough torque to self-start.

After a thorough examination of the disassembled contraption and more measurements, I figured out that 40 years of wicked (both wik-ed and wikd) oil had crystallized into a low impedance carbon coating inside the rotation chamber. Out it the garage, a long session with a spray can of Gumout and some rags eliminated the short circuit. Another four or five hours of careful reassembly, including re-stringing cotton thread through some tiny crevices, and IT'S ALIIIIIVVVVVE!!

Brian Lowe submitted this article as part of Frankenstein's Fix, a design contest hosted by EE Times (US).