Surrounded by History

by Dan Wiswell

Many years ago, I worked in a building in Boston that had a plaque on an outside wall that commemorated the invention of the telephone by Alexander Graham Bell.

Passing by it nearly every day, I used to think about how things must have been when those old experiments were going on. People may not realize that while Bell was working in his lab, there was an ambient chatter of telegraphic communication already going on and had been for decades.

By the beginning of the twentieth century, companies that catered to the needs of the growing electrical and communications industries employed thousands of people in the greater Boston area. Cambridge, Massachusetts became the birthplace of a multitude of companies that had their beginnings in or around the periphery of the radio and communications industries.

It must have been an absolute bonanza to be involved in metrology back then. With so many innovations occurring daily, the need for test and measuring instruments of all kinds must have been nearly inexhaustible. Back then, a local company that was involved in this trade was the Rawson Electrical Instrument Company, located at 110 Potter Street Cambridge, Massachusetts. Homer E. Rawson founded the company in 1918. Prior to the founding of his company, Mr. Rawson had been vice-president of the General Radio Company also located in Cambridge.

In his newly formed company, Mr. Rawson’s chief engineer was Arthur J. Lush. Mr. Lush had been previously employed by Paul Instruments of London, England. On January 19, 1920, Arthur Lush applied for a patent for an improved binding post with a captive cap. While still an English citizen and resident of Cambridge MA, he was granted patent number 1490336 by the United Stated Patent Office on April 15, 1924. Upon receipt, he assigned this patent to the Rawson Electrical Instrument Company.

In those days, beyond the commercialization of electricity, there was a tremendous amount of interest in the field of amateur radio. One publication of the time was QST magazine. It promoted itself as “A Magazine Devoted Exclusively to the Wireless Amateur.” The August 1919 edition contains an article written by H.E. Rawson titled “Measurement of Wavelength, Capacity and Inductance with Oscillating Vacuum Tube.” Towards the back of this issue of the magazine is an advertisement posted by the Rawson Electrical Instrument Company advertising itself as the “Sole U.S. agent for Robt. W. Paul, London, and Paul Instruments.” Makers of “Portable Galvanometers of Extreme Sensitivity and Measuring Instruments for all purposes on direct current and alternating current of any frequency.”

Robert Paul invented the unipivot meter movement in England during 1903. His meter movements can be seen in many of the earliest products in the Rawson Electrical Instrument product line. A classic example of a unipivot-based instrument is the Model 501 Galvanometer pictured below.

This instrument is equipped with a Paul Instruments unipivot meter movement. The top and bottom details of the meter movement are depicted to the right of the instrument. The bottom of the movement has a pivot and jewel assembly similar to those found in conventional d’Arsonval meter movements. However, at the top of the meter movement, instead of a corresponding pivot and jewel assembly, this design utilizes a tensioned helical coil to fix the at-rest meter indication at a desired point on the scale. As advertised, these instruments provided extreme sensitivity of measurements and were best used in a controlled laboratory environment.

To provide such accuracy, it was very important for the meter movement to be precisely balanced. Measurements were required to be made while the instrument was positioned on a flat surface and observed by looking down on the scale from above the instrument. The mirrored scale removed parallax errors when readings were viewed properly in this fashion.

There were a few notable deficiencies in this design. This type of movement was not as rugged as the improved d’Arsonval movement designed by Dr. Edward Weston in 1888. These instruments tended to drift off a set zero indication with a change in ambient temperature. A small but important feature of these products can be seen on the inside of the top cover. A small rubber button was positioned so that when the case cover was closed, it depressed a clamp on the front of the movement that locked the movement in place during transportation. This significantly added to the durability of the instrument. This clamp button was positioned in the “up” position when measurements were made and returned to the locked position during storage or transportation.

The Rawson Electrical Instrument Company made a variety of other types of measuring instruments including “thermal” and conventional multimeters. One problem that early instrument design engineers had to contend with was equilibrating an AC voltage with a corresponding DC voltage value. In those days, thermal transfer or reference standards were used for this purpose. Basically, a thermocouple was used to sense the value of an AC or DC voltage signal. The resultant meter deflection was plotted on a hand drawn scale. As a thermocouple is an inherently non-linear device, the resultant readings would also appear on a non-linear scale.

These scales (or dials) were made by a process called “hand pointing.” As each instrument reached its final assembly, it was fitted with a blank scale before a metrologist would put it through its calibration process. As readings were obtained, a small pencil mark was “pointed” on the scale identifying exactly where the pointer came to rest during each measurement. The marked-up blank was then sent to an art department where the scale was created from these cardinal points. The finished scale was then serialized and re-united with its host instrument before final calibration.

This technique for enhancing the accuracy of an analog meter was still being practiced in the first calibration laboratory I worked in during the late 1970s. I’m sure it is still going on somewhere.

Let’s look at a few examples of Rawson thermal multimeters. Depicted to the right is a Rawson Electrical Instrument Company Model 502 Thermal Multimeter, serial number 9239.

One general feature that stands out is that all ranges and functions are measured using various non-linear scales. This points to the fact that every measurement utilizes an internal thermo-electric sensor. Pictured below is an internal view of the product with the sensor in the foreground. Notice the heavily varnished, woven-fabric insulators that were in use before the days of circuit boards.

Another indicator of the age of this specimen is the font used on its scale. This font type was also used in the early days by the Boston Red Sox, a stylized version of which still adorns their team logo. The serial number is an important reference point as I have not found any date stamps or other notes created during the manufacturing process on any Rawson Electrical Instruments Company products, ever. So, without any other available information, we can make an informed guess as to its age using the following logic: If there are approximately two hundred and sixty workdays in a year and the company was able to produce roughly ten units per workday, this unit’s serial number could have been achieved within four years of production. As the supply chain from Paul Instruments appears to have been fully developed prior to 1920, this could possibly mean that this specific unit may be about one hundred years old, give or take a hand grenade. As of the date of this publication, that’s the best I can do.

Another version of a Rawson multimeter was the Rawson Electrical Instrument Company Model 5012 Twin Multimeter. This instrument was designed to measure AC and DC voltage and current across a fairly broad selection of ranges. A close look at its scale reveals that only the AC ranges have non-linear arcs. This means that only the AC functions pass through the internal thermal detector to produce a measured reading. An indication of the age of this instrument, beyond its archaic construction, is its ohms-per-volt rating. At one thousand ohms-per-volt on DC voltage ranges and one hundred ohms-per-volt on AC voltage ranges, these multimeters would load down a measured circuit significantly more than modern multimeters. As a point of reference, a few decades after the construction of this unit, the Simpson Electric Model 260-3 VOM offered an ohms-per-volt rating of twenty-thousand ohms-per-volt.

As the advertisement in QST magazine said, the Rawson Electrical Instrument Company offered a broad assortment of measuring instruments. One such instrument was the Model 503 Megohmmeter pictured below.

This megohmmeter was designed to measure high-resistance values with an applied test signal of one-hundred-and-fifty DC volts. Able to measure resistances of up to ten gigohms, its most prominent feature is the turret that protrudes above the rest of the front panel. I’m not sure why I have never taken the time to examine this instrument further. Now that I have, I must admit to some level of embarrassment at what I discovered when I did.

This causes me to think back on some of the stories that I was told by older metrologists that I worked with as a young man. Many of the older gentlemen that I worked with at the Mancib Company in the 1970s had been in World War II. One of them was a sailor that had volunteered for service in the submarine corp. He told many stories of being onboard when his submarine had been depth charged while on duty in the south Pacific. When that happened, he said that afterwards the sailors would find things that had shaken loose on the ship, like welding rods or flashlights that had been left in place during the ship’s construction. A close miss from a depth charge often caused damage to the pivot-and-jewel assemblies of d’Arsonval-movement-based meters on board. He told me that this problem was solved late in the war by retrofitting the old-style meters with newer, taut-band-based meters that could sustain and survive a much greater shock than a pivot-and-jewel-based meter movement could.

As I began researching other articles that I have had published in this magazine, I found numerous places online, including statements on the websites of some panel meter manufacturers, that taut-band meter technology gained wide acceptance only after advances in materials research made them strong enough to be a viable alternative solution to pivot-and-jewel based meters in the 1950s. Prior to this, spring-loaded jewel assemblies had been one solution for creating what were called “ruggedized” meters. Imagine my astonishment when I removed the outer shroud of this megohmmeter’s front-panel turret. The picture above clearly shows the leaf spring assembly of what the English manufacturer used to attach what they called a ligament wire to the coil assembly of the movement. Over here, we called it a taut band.

This meter movement is a hybrid of pivot-and-jewel and taut-band technology. It’s the only one I’ve ever seen. To me, it is proof that nothing beats direct research on any subject. It makes perfect sense to me that the creator of a single-pivot meter movement would also employ other techniques to solve a difficult engineering problem.

During my research of the Rawson Electrical Instrument Company, one thing that stands out is that the company must have had a very reliable supply chain of parts. Instruments separated by decades of time were built with the exact same switches, knobs, movements and Bakelite front panels. This is evident in the product shown as our next example, a Rawson-Lush Instrument Co. Inc. Model 963 AC/DC Thermal Voltmeter.

It features switches, knobs, meter movement and case-style that were ubiquitous throughout the entire life cycle of the Rawson-Lush legacy. For me, it underlies a somewhat melancholy aspect of this story. Although Homer Rawson founded his company in 1918, he passed away in 1923, and never saw the transfer of Arthur Lush’s patent to the company for his improved binding post in 1924. Arthur Lush went on to become the president of the company and passed away forty years after the passing of Homer Rawson in 1963. His son Morely J. Lush, born March 2, 1919, renamed the company the Rawson-Lush Instrument Company, Inc. and in 1963 moved the company to Acton, Massachusetts. Morely J. Lush brought his company into the modern era. He was the president and chief engineer of the company for over forty years and was a resident of Acton Massachusetts. He died on July 11, 2012, a prominent and well-respected member of his community. I wish that I had met this man during his lifetime. We would surely have had a lot to talk about.

---

Dan Wiswell, is a self-described Philosopher of Metrology and President/CEO of Amblyonix Industrial Instrument Company in North Billerica, Massachusetts.