Dials Way Over Engineered.

[Dan/Panther]

I've finally taken the plunge and completely disassembled a dial I recently got ripped off by a seller on Ebay.
The main shaft was bent. So I took it apart and put the shaft into my lathe, and straigtened it out. Now the dial works fine, and the finger wheel rotates at a flat plane once again.

HOWEVER, in looking at how dials are made, I came to the conclusion, that the same idea could have been accomplished wiuth a 10 postion cam, like those found  in a distributor for a car, and a set of points. Hardly any moving parts, and very cheap to make. Who designed the clock work mechanisms in phone dials ? Seems to me, he, or she was an unemployed Swiss watch maker. WAY OVER ENGINEERED.
Has anyone ever thought that same thing ?

D/P

[BDM]

Dan, which model dial?

[Dan/Panther]

BDM;

5J.
It also has marking of I-40, 57L, and I-57.

D/P

[McHeath]

I've not yet taken a dial apart but it would not surprise me if it's overdone.  I did pull a hookswitch apart in a 500 last week, and Holy Cow Batman is it complicated!  Like someone with a fondness for arcane little parts all squeezed in a little unit designed it.  I have a 2554 by Radio Shack from 89' that is an interesting comparison, it's hookswitch is nothing more than a single little clicker switch with only two wires, and it still works.  However, perhaps the complex WE switches will work longer and are more reliable in the long run, I don't know.  But it seems that all those contact points in the WE switch would be a liability as they age.

[Dan/Panther]

I wish I had the time. I would like to try to make a simple dial to see if it would work. Maybe someday.
D/P

[benhutcherson]

I've been giving this some thought over the past couple of days, and here's the conclusion I've reached:

There's really not a whole lot to a dial. At its core, you really just have a spring and some sort of cam to open and close the contacts. On older dials(#5 and earlier), the cam is actuated by notches on a center wheel that make it flap back and forth. On #6 and later dials, a cam is geared to the main shaft, and makes one rotation for every digit dialed.

The issue that comes up, though, is that the amount of force applied by the spring is not directly proportional to the amount that the spring is wound(Hooke's law is only true over small displacements). Furthermore, springs can and do weaken over time, especially if they're subjected to temperature extremes. Thus, it would be impossible to get a consistent and repeatable pulse rate out of a dial that had no means of regulating its speed.

This is where the other main part of the dial, the governor, comes in to play. By designing the dial such that the spring alone would make it run too fast, the governor is able to slow the dial down to give a consistently and repeatable rate of rotation, both within a digit and from digit to digit.

Thus, I don't think that a simple dial would be able able to dial as consistently and reliably as a simple one such as described. Bear in mind, too, that the basic dial designs in use(WE #2/4/5, WE #6/7/8/9, AE) were developed in a time when step switches were widely in use. My understanding of step switches is that they're very particular about how the pulses are spaced. In particular, the sheer mass and distance traveled by the moving parts wouldn't allow them to work faster than about 13 pulses per second. Crossbar and panel, since there's very little movement, could handle 20+ pps, but then both of these weren't really used outside of major cities until the '50s. My town of Frankfort, KY got a crossbar switch in the mid-50s, but some of the county switching needs, and the switching in all of the smaller towns, still used step switching until the mid-60s.

[Dan/Panther]

Somewhere here I read that a dial is properly calibrated at 10 clicks per second, or one second for one revolution of the dial, then someplace else I read between 9 and 12 clicks per second is correct. Either way don't tp precdision, or seem to hard to accomplish with a good spring, and a cam with 10 lobes and a points set up with the proper amount of contacts, All that could accomplished with two moving parts, the mainshaft, and the points.

D/P