Faraday to James Cosmo Melvill   7 June 1845¹

Royal Institution | 7 June 1845

Sir,

I have received your note of the 30th May 1845² accompanied by the following papers - Extract from the Minutes Pol[itica]l. and Mil[itar]y. Committee 29th May 1845³ - Military Letter from Bengal 14 Feby 1843 [sic] No 23⁴ - No 5806⁵ Letter to Sir Henry Hardinge⁶ 31 Decr. 1844 signed W.H.L. Frith⁷, R Benson⁸, W. Burlton⁹ - No 134¹⁰ from Captain Pillans¹¹ and Dr. O'Shaughnessy¹² to Captn Green¹³ 24 Augt 1844¹⁴ - No 196¹⁵ from Major Sturt¹⁶ to the Military Board 7th Feby 1845 - From Professor Daniell 10 May 1841¹⁷ - from the same 15 June 1841¹⁸ - from myself 9 June 1841 ¹⁹ - certain extracts from Military letters to Bengal²⁰ - and a letter from Professor Daniell 23 July 1844²¹:- and your object is that I should "furnish you with a report of my opinions on the views and sentiments which have been expressed on the subject in India as contained in the collection".

In reply I will quote you the exact words of my former report i.e. the one referred to above and contained in the papers you have now sent me for "these with the other papers I have carefully considered, and, as the general result beg to say I see no reason to alter even a word of my former report²².

In saying so much perhaps I say every thing that the Honorable Court of Directors desire of me, for it is probably more the judgment of the individual than the reasons for the judgment that is required. The latter it is almost impossible for any man to give fully in a case where there are arguments and reasons on both sides of the question for the bearings of these become almost infinite in number by reason of the variation in the degree of force which they possess under varying circumstances and at last it is the discretion, experience, tact, and caution of the person which gives to his opinion any worth it may in reality possess."

The copy of my report on the 9th June 1841 I will beg you again to read as if it were now an original document. I have paged it in red ink and corrected the references; and I have lined nearly the whole of it down the side with red ink and the present date of 7 June 1845. Every part so marked I present to you now as the result of my deliberate consideration and cautious judgement and I have nothing at all to alter of the conclusions therein drawn:- and therefore I will ask you to accept that report in conjunction with the sentence I am writing as the expression of my opinion on the views and sentiments contained in the collection of papers you have sent me.

My report contains at page 7 and the following pages an account of certain experiments on the division of a charge and the production of what is often called the lateral spark, the causes of which and the consequences that flow from it I have stated in general terms. On the other hand Dr. O'Shaughnessy in the papers you have just sent me letter dated 24 August 1844, whilst objecting to the kind of lightning protector which I and the late Professor Daniell recommended, has himself advised another which both in principle and practice he considers absolutely safe. I would have avoided if possible for the reasons expressed at page 2 of my former report commenting on this plan; but as the very experiments I have just referred to shew that proposition cannot fulfil the expectation of the proposer and that it might place the magazines in much greater danger than the plan I approved of; and, as you call on me for an opinion in continuation, I feel that I cannot in justice to the Honorable board refrain from drawing its attention to certain points in the proposed arrangement and its effects.

It is said by the proposer that the most powerful batteries may be discharged harmless "on a metal powder flask, without the slightest danger to its contents"; that "shells of the thinnest copper charged with percussion powder may be placed with equal safety in the current of the most violent artificial explosions"- that "it is impossible to kill a bird in a wire cage by the electric fluid" that "within such envelopes of conducting materials no subdivision of charge can take place" - "all inductive influence being destroyed" - and "it is accordingly thus that our powder should be protected".

There is evidently here and a little further on, the confusion of two distinct series of electrical effects essentially different, namely the static and the dynamic and it is no doubt, this confusion which leads to the erroneous practical result which the proposer arrives at. He then states, & truly, that expence is the only objection which can be urged against the obvious mode of protecting magazines by enveloping them altogether in metallic sheathing - but immediately after says "the subjoined plan, while equally efficient will reduce the expence to a fair practical amount and will guarantee security". The plan is then proposed of carrying a copper strap upwards from the ground against one side of the magazine leading it over the top and down to the ground on the other side; this is to be done at every 6 or 10 feet of distance. Similar band[s] are to run at intervals of 6 feet along the roof and across the former bands, being fastened to them by rivetting or soldering. A similar band is to run round the bottom of the building at the level of the ground and connect all the ends together; and, lastly from each corner of the building, copper rods with branches are to go into the ground. The copper straps or bands are to be two inches wide but their thickness is not given. However it is said that hoop iron might be used if it were not liable to rust - and it is also said that the mass of the copper (whether of each strap or of all the vertical straps is not expressed) is much greater than that employed in Mr. Harris' ship conductors²³.

It is then said "within a magazine thus arranged it is impossible - according to the present state of our knowledge that an accident from lightning can occur. The system can provoke no discharge upon itself &c - and then the philosophy of static effect i.e. of quiescent electricity and not of discharge (which when it occurs is dynamic in its character) is referred to.

I had no doubt from the result described in my former report and now recalled to your attention by the red ink lines at p 7, 8, 9, 10, 11 &c that such a system as that proposed would easily give lateral sparks within; and that also under circumstances very similar to those likely to happen with a powder magazine. But to make it more evident to the Honorable Board I arranged a system of copper wires exactly in the manner proposed so as to form a metallic cage (as in the figure at the next page) of which the intervals between the vertical wires were only from 18 to 22, - 30 and 48 inches, instead of six feet or 72 inches; and the vertical interval no more than 57 inches. The wires were from 1/12th to 1/16th of an inch in diameter well connected together, and, finally, for the sake of a perfect discharge connected by a thick wire to the London water pipes.

Being thus arranged an electrical machine was set in action so as to throw simple sparks (without any Leyden Jar) from the conductor by a jointed rod on to any part of the top of the frame, as for instance where the balls are represented. The sparks were not more than one inch & six tenths long and the wire of the frame or any single wire was sufficient to carry several hundred thousand times this quantity at one discharge without in the least suffering from it. Yet when a wire was connected with the lower rim, as at B or otherwise well connected with the earth as by the pipes p6 and being taken up inside the frame was brought near the under part of any of the upper wire a spark passed between that wire and the frame, at the occur[r]ence of every spark from the machine - nay more:- whatever part of the lower edge of the frame this wire was made fast to, its upper end could take a spark from any place of the upper part of the frame at every spark from the machine. Still more, if the wire represented by the red line was only held near the frame at its top and bottom ends, then, a spark from the machine on the top of the frame caused two sparks on the experimental wire, one at each end. If instead of one experimental or inner wire, two, three, four or more were used, being connected below with the bottom of the frame or with the water pipes p.6 then one spark from the machine could produce a secondary spark on each of these wires i.e four internal or lateral discharges could be produced at different places within by one external spark.

I covered a part of the top of the wire frame with sheets of wire gauze the wires being only the 1/8th of an inch apart and therefore far closer than the wires of a bird cage; and I could produce the lateral discharge as easily from the inner surface of the gauze as from the former wires of the frame. I hung a cylinder of wire gauze inside the frame and then had abundant lateral or secondary sparks from the inside of the cylinder. I took a large strong copper pan 20 inches in diameter and turning it upside down upon the frame made it a metallic dome; when the machine discharged a small spark only on to it or any part of the wire frame, a lateral spark could be obtained with the utmost facility from any part of the inside.

My model was only about five feet high but the powder magazine is 23 feet in height; and, increase in distance from the place where the electricity strikes to the place of its discharge (i.e the earth) tends to render the effect more striking. I had only wires of copper instead of copper bands or iron hoop; but my wires were far larger in proportion to the electric spark which I used than the bands of copper (judging from the weight given) or iron would be to the lightning flash. Such a system of bands because of its infinitely better conducting power than stone or brick even when the latter is wet will tend to make the lightning strike upon a magazine which without the bands might not have been struck; and considering the metallic nature of the contents of a magazine, that each powder barrel is lined with copper; and that these and shells are often piled on each other from the ground upwards towards the roof, so as even sometimes to touch it, I think it very probable that a strong flash of lightning striking the proposed framework might divide after the manner of a lateral spark as in the experiment described and part of it take its [course] through the contents of the magazine and thus cause its entire destruction.

In order to shew that mere surface has nothing at all to do with these, or any dynamic effects of electricity, I constructed a similar frame with copper bands nearly two inches wide and obtained the lateral sparks just as easily on the inside surface of these bands as with the wires. It could not be otherwise after the results of the experiments described with the wire gauze and the copper pan.

I have no doubt that if the magazine were entirely enclosed in a case of copper and that copper were thick enough to conduct most freely the electric fluid, it would be protected:- but I doubt whether a cage of copper straps such as those proposed would be sufficient to remove entirely the effects I speak of even though the straps were so near each other as to have no space exposed larger than a square foot:- further, supposing that these straps were in some places to come apart, or else overlie not being in actual or soldered metallic contact; there would almost certainly be a spark at each of these places when the lightning fell upon any part of the frame.

I have paged in ink the copy of Dr. O'Shaughnessy's letter or report of the 24th August 1844 which you have sent me and which I return with this paper of mine. You will find that from page 37 to page 67 it speaks of the power of a lightning conductor through which a discharge is passing, to affect neighbouring bodies and cause sparks to pass between them; and it gives many cases of the production of such sparks as at p.p. 41, 43 to and including 49 being the Strasburg Cathedral, also 52, 53, 59, 60, 62 and 63; and these are considered as reasons against such a lightning conductor as I have recommended, Now, all these effects (being dynamic) take place as easily within the frame work of straps proposed as they do near a lightning rod and upon exactly the same principles. They are within that frame work just as surely "increased by the proximity of the conductor (i.e the straps)" and the quantity of the primary or lightning discharged p64 "as they are in any other condition of the lightning conductor:- and because the frame work is in much closer approximation to the contents of the magazine than the rod I recommend, are more productive of danger to these contents.

For these, and for all the reasons formerly and again adduced, my conclusion is the same as before, namely, that the conductor should not be put in contact with the magazine; but, as I have said in the close of my last report page 20, to which I again beg to refer you, should be about 3 feet distance from it.

I have the honor to be | Sir | Your Obedient humble Servant | Signed / W. [sic] Faraday

James C. Melvill Esqre | Secretary | &c &c &c | East India House