Half a century before press accounts and a 1913 patent described Roy J. Meyers's "absorber" as drawing electricity from the air, the electrical state of the lower atmosphere had already been measured with calibrated instruments. A history paper by K.L. Aplin (University of Oxford) and R.G. Harrison (University of Reading), "Lord Kelvin's atmospheric electricity measurements", reconstructs that work from Kelvin's publications, letters and notebooks. It does not mention Meyers or any power-generating device. It is useful to this dossier as background: it sets out what calibrated instruments found in fair-weather air, and how Kelvin's water dropper came to share the potential of the air around it. The paper models equalisation and response time, not available power, so any comparison with the Meyers device remains an analogy.
What is in the air in fair weather
According to the paper, researchers had known since the late 1700s that fair-weather air sits at a positive potential relative to the ground. In the modern picture it gives, thunderstorms and shower clouds keep the ionosphere charged to about 300 kV positive relative to the ground, and this produces a vertical potential gradient (PG) of roughly 100 V per metre at the surface in fair weather.
Kelvin (William Thomson, 1824-1907) worked on the problem intensively from 1859 to 1861. According to the authors, his key step was to reference his electrometers to electrochemical cells, which made absolute rather than merely relative readings possible. The first calibrated measurement they cite was taken on the Links golf course at Aberdeen at 8am on 14th September 1859: a flame probe 3 feet above the ground stood at a potential equal to 115 Daniell cells. Taking 1.08 V per cell, a later calculation quoted in the paper converts this to 124.2 V over 0.91 m, a PG of 137 V per metre. Fair-weather values from the island of Arran the same year are given as 80-158 V per metre.
The instruments
The paper credits Kelvin with two new sensors, a recording method and a portable electrometer. Each is easiest to understand through the problem it solved.
- Winched can: one reading at a time. This sensor measured the effect of suddenly exposing a shielded electrode to the atmospheric field, a use of induction the authors liken to a modern field mill. The shield was an earthed can that could be raised and lowered by a pulley around an electrode standing about 60 cm above the roof of the Natural Philosophy Lecture Theatre at Glasgow University; with the earth connection broken, dropping the can about 46 cm gave a change in potential that Kelvin read on his divided ring electrometer. The authors estimate that the deflections he described correspond to at least 80 V. The device was reported on 8th March 1859, and the paper calls its operation cumbersome: each measurement required raising and lowering the can and taking an electrometer reading.
- Water dropper equaliser: a continuous reading. This sensor measured the potential of the air at a single point, the place where a stream of water flowing from an insulated tank broke up into drops, and its advantage, according to the paper, was that it could run without interruption. As long as the stream and the surrounding air are at different potentials, each departing drop takes charge with it, and the paper says this transfer continues until the difference has gone. Because the unbroken stream joins the tank electrically to the break-up point, an electrometer at the tank then shows the air's potential there, and the height of that point gives the potential gradient. The instrument was first described on 18th October 1859.
- Photographic recording: a trace without an observer. To get the "incessant recording" he wanted, Kelvin made the electrometer write its own record. Its mirror threw a spot of gaslight onto photographic paper that clockwork kept moving. The paper adds that this optical method put minimal load on the sensitive electrometer. Water dropper and recorder together became known as the Kelvin Electrograph, which Kelvin installed at Kew in January 1861.
- Portable electrometer: roaming measurements. Developed on Arran in the summer of 1859, this let Kelvin take a flame probe to the beach and onto Goatfell and measure the gradient there.
The water dropper lasted. The paper lists its use at Kew until 10th February 1932, on top of the Eiffel Tower in the 1890s, on an 1899 balloon ascent to about 4 km, at Eskdalemuir Observatory from 1908, and at Kakioka in Japan to the present.
How the water dropper equalises with the air
The water dropper is not a passive elevated conductor: it exchanges charge with the air through falling drops, and the paper's analysis concerns how that exchange brings the instrument to the air's potential. The authors model it with ordinary electrostatics: if drops of radius a leave n times per second, the current is 4πε₀·a·n multiplied by the potential difference between the water stream and the air, which gives an effective resistance between instrument and atmosphere of 1/(4πε₀·a·n). They note that Kelvin's actual geometry and flow rates are not recorded, and give two illustrative figures: about 10^16 ohms for drops of 1 mm radius falling once a second, and about 10^14 ohms for drops of 100 µm radius. The two figures are rough and do not sit easily with the formula or with each other, since in the formula resistance varies inversely with drop radius, so a tenfold change in radius would by itself change the resistance tenfold, not a hundredfold. They are best taken only as an indication that the resistance is very large.
The paper presents this resistance as a compromise: large enough that the instrument loads the atmospheric field negligibly, yet small enough that it still responds within 0.01-1 s. The resistance is not what brings the tank to the air's potential; in the paper's account that happens because the drops go on transferring charge until the potential difference is zero. The paper does not ask how much current such an instrument could deliver, and it states no current figure.
The same section touches on insulation. Charge leaked from the insulators at 5% per minute, a rate Kelvin believed could easily be improved to 5% per hour, and the paper says the dropper's rapid response reduced the demands on the insulation.
A footnote adds a distinction worth keeping: the equaliser is not Kelvin's later water dropper generator of 1867, which uses induction and positive feedback to build high voltages.
Variability
Kelvin was repeatedly surprised by how unsteady the readings were, the paper says, and the authors judge this the likely reason he called in 1859 for "incessant recording". He wrote that "even in fair weather, the intensity of the electric force in the air near the earth's surface is perpetually fluctuating". On Arran the PG rose markedly with east or north-east winds, which a later study cited by the authors attributes to smoke from the industrial mainland. With the winched can, Kelvin found the reading positive in fair weather but negative when a negative cloud, whether natural or of smoke, passed over. The authors also raise a link to the Carrington solar event. Kelvin's letter asking Joule for help with the portable electrometer is lost, but from the dates of Joule's letters and an assumed postal time of at least three days they infer it was written between 25th August and 2nd September 1859, a window of considerably enhanced solar activity. They consider it probable that Kelvin was trying to measure the PG during the disturbed flare period, and almost certain that the letter was written in strongly disturbed solar conditions. The connection to the readings themselves they treat as a possibility only: the solar disturbance, they say, cannot be excluded as a source of the anomalous measurements.
What this does and does not show about the absorber
The Meyers material summarised in Jerome Meyers and 'The Absorber': The source archive Energy Dossier (Provisional Record) makes two separate kinds of claim. British Patent GB191301098 asserts that a small apparatus 50-60 feet up delivers a constant current at commercially usable voltage and amperage. The press accounts report readings: a 1912 article says a model developed 8 volts, and a 1914 article says a meter test recorded 4.5 volts. These are claims by the patentee and by newspapers, not independently documented findings (see The Absorber (Meyers Dossier Device Claim) and Jerome Meyers).
The paper does not draw any comparison with such devices; read against it, two limited points stand. First, the paper documents a fair-weather potential gradient of about 100 V per metre and says nothing about devices of the Meyers type. Second, the one instrument it analyses is a water dropper that exchanges charge through falling drops, not a fixed elevated conductor. The paper models that instrument's equalisation and response time rather than any available power, puts its effective resistance to the air very high on rough estimates, and documents a field that is anything but constant. It offers no instance of atmospheric potential sustaining a load.
The limits run the other way too. The paper is a history of measuring instruments, not a test of magnets, zinc plates or mercury-vapour rectifiers, and its resistance figures are estimates for assumed drop sizes rather than measurements of Kelvin's apparatus. It does not quantify the current available to a large elevated conductor, and it notes that buildings and terrain distort the local field. It therefore cannot say what Meyers's meters registered or why, and it neither supports nor tests the claim of commercial current.
References
- K.L. Aplin and R.G. Harrison, "Lord Kelvin's atmospheric electricity measurements", arXiv:1305.5347, https://arxiv.org/pdf/1305.5347
Source notes & attribution
- https://arxiv.org/pdf/1305.5347