Electrokinetica The Electro-mechanical Museum

DC meters

Whilst modern electrical distribution systems supply their consumers with AC (alternating current), DC (direct current) was commonly used from the earliest days up to the middle of the 20th century. Both types of current offer certain advantages for specific applications, although for economy of distribution and large-scale generation AC is the clear winner. Amongst the disadvantages of DC while it was still in widespread use was the difficulty of recording customers' energy consumption using the available technology. Smaller customers were at first often billed using a tariff based on the number and size of lamps installed, or some other fixed criteria, regardless of their actual energy usage. Meters were devised which monitored only the current, from which the energy consumption could be calculated so long as the system voltage remained constant, but the quest was always for an instrument capable of measuring the energy itself, accurately and with a minimum of maintenance. Although such meters existed and some types worked well, DC versions remained complicated and troublesome relative to their AC equivalents.

The Aron clock meter

Meter dials

Meter dials

The Aron Clock Meter was indeed complicated but achieved good accuracy and was a true watt-hour meter. Based on a principle devised in 1882 by electrical engineers Ayrton and Perry, the mechanism was developed in Germany by Dr. Hermann Aron, specialist in electrical horology and founder of the Aron-Werke, a significant manufacturer of electrically driven clocks and time control equipment. The meters were available from the 1890s and were the first to be widely used in the UK, made under license by GEC. Early versions were somewhat simpler, using one clock instead of two and lacking the electric winding motor; the type shown here includes a number of refinements to improve accuracy developed around the turn of the century. The method of measurement used by the clock meter has a number of advantages over many competing DC meter technologies; it is reasonably linear even during overload, has nominally zero starting current, requires no topping-up or resetting and is stable over time. Its principal disadvantage was the high cost of manufacture.

Pendulum coils

Pendulum coils

Side of meter

Side of meter

Aron meter front view

Aron meter front view

Principle of operation

Left view of movement

Left view of movement

The Aron Clock Meter contains two pendulum clocks, each of which keeps its own time. Both are driven from a single small mainspring by way of a differential gear (the driving differential), which allows them to run at different rates while receiving the same driving torque. The spring is wound frequently by an impulse motor, typically every 30 seconds, this being a more efficient method than running the motor continuously. Each clock has an ordinary escapement and pendulum, although a coil of many turns of fine wire is attached to the underside of each pendulum bob fed via wire pigtails which run down the pendulum from near its pivot. These coils are fed with a current proportional to the circuit voltage, i.e. they are shunt coils. Below each pendulum is a series coil of heavy wire carrying a fraction of the load current, over which the shunt coil swings. As it does so, the interacting magnetic fields of the shunt and series coils either augment or oppose the gravitational force acting on the pendulum, and hence increase or decrease the rate of the clock proportionally to both voltage and current over the duration of a full swing. The coil polarities are arranged such that when one clock gains, the other loses time, so the instantaneous difference between the two clock rates is effectively proportional to the power. Another differential gear (the recording differential) is connected between the two clock geartrains, arranged in this case to output the difference rather than the sum, and it is this which operates the recording geartrain carrying the dial pointers, integrating the power to show total energy.

Right view of movement

Right view of movement

Two subtleties are introduced to overcome practical limitations in the clockwork. Firstly the clocks are regulated to run at different rates in the absence of load. This minimises the tendency for them to fall into synchronism at low load (and hence fail to register) by transmission of torque pulsations through the differential gears which tend to retard the faster clock and accelerate the slower. The difference in rates is accounted for by a difference in gear ratios between the going trains and the recording differential sun wheels. Secondly a device is included to cancel any static rate error: The polarity of the shunt coils is reversed frequently by an electrical commutator after a fixed total number of pendulum swings, the direction of the recording train being simultaneously reversed relative to the recording differential. Thus as each clock movement serves equally in the roles of 'gaining' and 'losing', any static rate error accrued during one period is automatically subtracted during the next, while the integration of power consumption continues unabated. The direction reversal occurs approximately every ten minutes under the control of an escapement connected to the driving differential, simultaneously providing a half-turn indexing of the commutator and switching an extra wheel into, or out of, the recording train drive.

Our Aron meter

This meter is dated 1904, and was supplied for 220V 75A DC service. The current rating was later increased by a factor of four, by the addition of an external shunt. In this configuration, it served to record the output of the Belliss and Morcom generating set featured elsewhere on this site. It is in working order and is a delight to watch.

The Reason electrolytic meter

Mercury tubes

Mercury tubes

Resetting meter

Resetting meter

Front view

Front view

Amongst the meters which measured charge (rather than energy) using an electrolytic cell was that devised by Mr. A Wright and made by the Reason Manufacturing Co. of Brighton. This has a completely sealed glass tube construction which needs no topping-up or cleaning; the cell contains mercurous nitrate as the electrolyte and metallic mercury as the indicating fluid, which is transferred from a reservoir to the measuring tubes by electrolytic action. When the meter is in use, fresh mercury is dispensed by gravity as required into a circular trough using an arrangement like a bird-feeder to keep the level constant. The surface of the mercury in the trough forms the anode of the cell and is completely submerged in the electrolyte. The cathode is a button shaped iridium electrode supported in the middle of the cell above a funnel leading to the measuring tubes, positioned so that the metallic mercury deposited on it drips off into the tube. The small U-tube in which it collects is scaled to indicate units of consumption (calculated on the designated system voltage). When this tube is full of mercury the contents syphon into the outer tube scaled in thousands of units and the U-tube begins to fill again. This provides for better readability than would a single tube. When most of the mercury has been transferred from the reservoir to the outer tube, typically after one year of use, it is necessary to reset the meter. The meter-reader releases a catch and swings the entire sealed tube assembly out of the case and upside-down, causing the mercury to flow back into the reservoir.

Meter shunt

Meter shunt

It is important to note that nothing in this cell decomposes (in contrast to the Bastian and similar electrolytic meters which decomposed water into hydrogen and oxygen) the mercury simply being transferred from one part of the cell to another. A yet more important advantage lies in its low cell EMF of one tenth of a millivolt; from the external connections the cell presents a very linear resistance, enabling it to be used for any range of current by connection of a metallic shunt resistance with up to one volt drop at full load. This is not possible with the Bastian meter in which the cell EMF swamps the resistive drop and makes the use of a shunt impossible, requring the entire current to flow through the cell. The electrolyte in the Wright cell has a negative temperature coefficient and a compensating positive-coefficient resistance is therefore fitted in series, which is physically attached to the cell for which it is calibrated.

Our Reason meter No. 413982

Made for 100A service at 200V, our meter is in working order and retains satisfactory calibration. The glassware is mounted in an iron case with a compartment in the bottom for the shunt; a window in the case allows the tube to be read without removing the security seal. The handwritten scale has faded making reading somewhat difficult but it is fascinating to see the mercury appear in the tube 'out of nowhere' over the course of days or weeks, although we have never yet managed to witness the rare moments when the small tube syphons into the large one.

-----

All content © copyright Electrokinetica 2007-2019 except where otherwise stated • Valid XHTML 1.0Valid CSS