Keeping time
Joseph Mansfield

Clarence Jones was an astronomer and architect, among other things. He combined both fields when the Chattanooga city school board engaged him to plan and build the observatory that would eventually bear his name. If I am not mistaken, he also designed the majestic Brainerd Junior High School, the city's showcase junior high, which stands at the foot of the hill which the observatory crowns. Other schools bear his mark as well.

One of Jones's lesser contracts was for the building of Dunlap Elementary School (the “grammar school”) in 1949. Both pairs of my grandparents lived on Rankin Avenue just over a hundred yards from the construction. With all of three years on me, I rode my trike up and down Rankin as I contemplated the new school which, I was assured, I would soon attend. The thought comforted me, for I was lonely: most of the neighborhood children were a bit older than me and already walked those privileged corridors.

Did I say this was a lesser contract? Well, Brainerd it was not. But it was by far the finest schoolhouse Sequatchie County had ever seen. In addition to twelve classrooms, there were a library, a lunchroom equipped to prepare and serve hot meals (the first in the county), and an auditorium with a real stage, with draw curtain, colored footlights, piano, the works. And those privileged corridors: they were paved in terrazzo, something few of us had even heard of.

Nothing, however, in Mr Jones's creation captured my attention so securely as the clocks and bells. Jones had specified that the building be fitted with an IBM self-synchronizing clock system. Let's look at it.

The idea was to have clocks throughout the building which would always display the correct time. Or at least correct themselves quickly if they failed to display correct time.

The clock in the principal's office stayed on time so long as the electric grid was working. The grid delivered current alternating at 60 Hertz. The clockwork was driven by a synchronous electric motor, so called because it was synchronized with the grid. So long as the grid delivered 60 Hz – which it did if it delivered current at all – the clock stayed on time. If the grid went down, a spring-driven motor kicked in to keep the clock on time. When grid power returned, a little electric motor would rewind the spring.

So far, so good. The office clock now displayed correct time all the time. But what about the other clocks scattered throughout the building?

With reason, these clocks were called slave clocks, and the main clock in the office was called the master clock. The slave clocks had simple stepper motors driven by pulses from the master clock. Each time a pulse went out from the master, the slaves advanced one minute.

The slave clocks did not operate through an outage. Instead, the spring-driven part of the master clock kept a count of minutes elapsed. After the return of power, the necessary number of pulses would be sent out to bring all slave clocks to the correct time.

Neat, eh? And simple (even if not quite as simple as I have made it here). But there are two more details to cover.

First detail: the master clock rang the school bells according to a preset program.

Second: the master clock broke down a lot, and it cost a lot to repair.

In fact the repair cost in an average year exceeded the $75 (in 1950s dollars) that Mr Johnson, the principal, paid for a little Montgomery program timer which he bought to replace the IBM master clock when the IBM broke down one time too many. The Montgomery did not drive the slave clocks. But it did ring the bells just fine, and it never broke down. The bells rang merrily and the slave clocks were quickly forgotten.

Until, that is, I returned to Dunlap after having finished Brainerd Junior High. On my first visit to the grammar school, I found the slave clocks abandoned. I also found that the old master had been given to Raymond Spangler, who had incorporated some of its organs into a mechanical man he was making for a science fair.

What a shock! But I soon recovered as the ramifications of this turn became clear. Specifically, I saw an opportunity to build a clock. A master clock. A clock that would outperform the IBM which it replaced. I approached Mr Johnson and asked if I might undertake the project. His only question was, What would it cost. I said, Let me study that a while.

There was not much study involved. I took one of the slave clocks home with me to study its workings and determine how to build a master that would drive it. I priced a few components in the Allied Radio mail order catalog, priced some more at Curle Radio in Chattanooga, prepared a bid, and presented it to Mr Johnson, who summarily advanced me the money I needed to get going.

The master clock I built remained in service at the grammar school for many years. The clock was simpler than the IBM it replaced, which is one big reason why it worked better. It did not attempt to recover from power outages; it had a button which was pushed once for every minute of an outage to get itself, and the slaves, back on time. And, as we noted above, another device had taken over the job of ringing the bells.

It was during a time when I was living away from Dunlap that my clock died. I had used a synchronous motor geared down to 1 RPM to generate the pulses that drove the slave clocks. That motor gave out. And by the time of my next sojourn in Dunlap, the grammar school had closed. I did manage to recover my clock, which stayed for a long time in an outbuilding of the Mansfield house.

In early 2011, probably in January, just after the death of my father, I had a conversation with Carson Camp, the prime mover of the local historical museum. He told me that my father had told him the story of the clock some time back and invited him to take the clock for the museum, but Carson could not find the clock. I found it for him. Carson graciously accepted it, and asked me to submit notes on the clock. This essay, two years later, is my response.



Copyright © 2013 by Joseph Mansfield. All rights reserved.
2013.01.15 First draft
2013.01.22 Revisions
2013.02.05 Posted





[home]



Site Meter