Faraday to Jacob Herbert   1 February 1847

Royal Institution | 1 Feby 1847

My Dear Sir

On the 26th of last month I went to the South Foreland lights, according to an intent referred to in preceding letters¹, and remained there on that and the following nights for the purpose of examining particularly into the cause of the occasional occurrence of dampness on the interior of the glass of the upper lanthorn, and testing the efficacy of the system of tubes by which the lamps in both lighthouses are ventilated. The weather was changeable being fair in the day - showery at night, not frosty, and the wind from the west south west, generally strong & often in sharp squalls.

Confining the attention at present to the upper house, I may briefly state, that, the lamp there, which is a central four wicked lamp, has the ventilating tubes attached to it; & that Mr Finnis², the chief keeper, says, no desposition of moisture occurs on the inside of the glass except when the wind is North or South East; &, then, not always but in cold weather. He has an impression, also, though a doubtful one, that something goes from the lamp and condenses on the glass on these occasions; and that this is more abundant with Rape oil than with sperm oil; but has no idea why this should happen only with North or South East winds[.]

The tower & lanthorn were new three years ago & the tower is not inhabited, nor has it any chimney in any part of the walls. The feeling to me whenever I have gone into it is that of a damp uninhabited place. At the bottom of the tower is the oil cellar the whole of which is below the level of the ground & in contact with it, & when the door is shut, is entirely closed from the air. Above that is an air chamber, occupying the whole horizontal extent of the tower; for the entrance of air at the level of the ground, to a central stove & ventilating apparatus. Above the air chamber comes the general space of the tower containing the stair case; over which is the Watch chamber, and over that again the lanthorn. The lanthorn is almost entirely of metal or of glass. Beneath the glass it is double & of Iron. The glass part is 14 feet in diameter & ten feet high - exposing above 413 square feet of glass. The roof above this is conical & consists of a single thickness of copper. When condensation occurs within it takes place both on the glass & on the copper roof above.

The present arrangement for warming & ventilating the lanthorn is as follows. Three large opening[s] exist on the East, South & West sides of the tower about 2 or 3 feet from the ground which give access of external air to the air chamber above the oil cellar already referred to. Weather boards are put up at the outside of these openings, & within they are boxed up so that by using a string the lids of the boxes may be opened & shut, with the intention of regulating the entrance of the air. A large thin metal prismatic tube probably 18 inches in width rises up in the centre of the tower from this chamber & enters a shallow air chamber below the floor of the lanthorn. This floor itself is slate but in it are (I think) seven ventilators equidistant from the center & the sides of the lanthorn by which the air passes into the latter and finally out by the cowl & thus it is ventilated. But to assist in the ventilation & to effect the warming, a stove is placed in the tower air chamber below & its chimney, which is a copper tube 6 inches in diameter and about 80 feet in length passes up the central air tube, then through the upper air chamber under the floor of the lanthorn, next through the floor & up the inside of the lanthorn & roof & finally through the roof into the air where it is furnished with its own cowl.

This stove when alight has a powerful effect in warming the air about its chimney in the large ventilating tube; & if there were no wind outside would give a good ascending power to the air, so warmed, & cause a large quantity to pass into the lanthorn through the regulators, they being open. It is in fact a very excellent arrangement for an abundant supply of ventilating & warming air; and I thought so at the time when Mr. Walker³, before its erection, did me the honor to ask my opinion of it, but from circumstances the effect of which could not have been anticipated at the time it is perhaps not the best arrangement for this tower.

In the first place when the wind is in any direction except North, so much air is drawn in at the weather or outer openings, that the keeper, to reduce it, is obliged to close all the lids of the air boxes; & one of them, belonging to the chief or south opening, he has been under the necessity of nailing down, because of the wind by its force actually blew it up though of considerable weight & then rushed through the central tube & lanthorn. Even this was not precaution enough for general circumstances, for so much air still enters by the crevices of the wood work as to require that six of the seven ventilators should be usually shut; & when the wind is strong, that the seventh should be shut, & all covered with cloths. At the time that I was there & the air boxes were all closed if one of the seven ventilators in the lanthorn were only in a small degree open & then covered with a large cloth, the cloth was instantly flown up into a great dome, many feet in circumference, which indeed grew until the edges were lifted from the floor & the air escaped into the lanthorn. When the ventilator was quite shut, & a large card laid over the small hole in the center, at which the turning key is applied, the card was blown up & sustained floating in the air over the aperture[.] Hence, when there is any wind other than the North, there is a powerful tendency of air to pass through the great central tube & the lanthorn; &, if there be no fire in the stove & the air cold, it exerts great cooling agency on the tower & lanthorn which have no extra or constant sources of warmth.

In the next place Mr Finnis states that he very rarely lights the fire; because, he finds, that, if there be any tendency to dampness on the windows the use of the fire increases it:- and this is not improbable or unlikely; for the stove, being in the lower air chamber, will always when alight, tend to dry it, i.e. will evaporate the water from its walls & floor & will give it to the air which is passing up. In such cases the air will go in a moister state into the lanthorn, than if it were sent in absolutely as it is on the outside; & then this dampness will condense on the glass of the lanthorn cooled by the external temperature. This moisture is not inconsiderable; for, the floor of the air chamber is the roof of the oil chamber, & that is found by observation to be always damp;- indeed the general dampness of this part of the tower may be judged of by an observation respecting the oil chamber which I will give presently. On the other hand, the warmth which is given by the stove to the air at the same time with the moisture, is not enough when the air enters the lanthorn to ensure its retaining this extra water dissolved. I had a good fire made in the stove for the purpose of ascertaining what the temperature of the air sent by it through the ventilators into the lanthorn could be ordinarily raised to. After about one hours firing this temperature was only 58˚ the external air being at 44˚ or a difference of 14˚ & this was the highest point I attained to. One reason why the air does not attain a higher temperature, is, that the air tube being of metal, much of the heat is communicated to it, & from it to the air of the interior of the tower (as I found was the case,) & from it again to the walls & so a consequence results which could hardly have been anticipated that this warmth causes the air in the tower to take moisture from the walls & such air going by the doors first into the watch room & then into the lanthorn is more loaded with water, & will sooner deposit on the glass, than air taken in its natural state directly from the outside[.]

If an attempt be made to urge the fire & so by a large consumption of coals to keep the tower, watch room & lanthorn all so warm that their temperature shall continually & effectually counteract the tendency of the air passing through them to deposit moisture; then, the present stove does not seem able for that purpose. I found it with a draught down into the tower; Mr. Finnis says the fire burns badly & smokes much. A blower has been put up covering in all the opening except the bars & yet I was not able to make such a constant fire as would prevent by its draught the smoke from creeping out. Then it is to be remembered that, the lanthorn, which is the part most needing elevation of temperature, is furthest from the source of heat; and I believe that the 24 or 25 feet of the upper part of the flue which passes through the lanthorn, can do more to warm the lanthorn than the warm air which the ventilators can deliver into it. I hope I shall be excused for these details but I am anxious to give them clearly both because of the good opinion I had of the plan of the stove generally before it was executed, & because of the conclusions I may be emboldened at last to offer.

There was no fire in the stove on the nights of the 26th & 27th when I was there, & I will now speak of the lanthorn & lighting apparatus as I found it under Mr. Finnis' care. The ventilating tube was right in all its parts, & was continued well up (three feet) into the cowl. The draught at all the opening[s] of the tube was carefully examined, both by day & night, & found to be right both in direction & sufficiency. Besides the air which enters into the lanthorn from the ventilators & their chamber, I found that there was a draught of air from the tower & watch room by the stairs & doors. There was some deposition of moisture on the glass outside, both by day & night, & this was chiefly to leeward. I examined it & found it saline; & its cause is as follows. Particles of salt spray fly up & lodge on the glass & then if the air be near the dew point, i.e near the temperature at which it will deposit moisture, then these saline particles being considerably hygrometric, cause deposition, & hence the moisture. The reason why they appear to leeward & not to windward is this, (or at least was the reason when I was there,) namely, the showers or even the mists which drive against the glass on the windward side wash off these saline particles & then the glass has no more tendency to condense on the outside than on the clear inside; whereas to leeward where the particles are not washed off, that tendency continues. At all times this inclination to deposition on the outside indicates that the air which passes the lighthouse is nearly saturated with moisture[.]

On both nights when I was in the lighthouse there was no deposition on the inside of the glass, but the circumstances were such as to bring the air within very near the depositing points. I found that if I stood for five minutes near the glass not breathing towards it but inwards towards the lamp a little damp began to appear which disappeared again after a while when I had changed my place. This shewed that the air was almost saturated with moisture since the slight evaporation from my body so placed brought it up to the saturated & condensing point.

I am thoroughly convinced that the lamp is perfectly ventilated & that no moisture or any thing condensible or noncondensible passed from it into the lanthorn: nevertheless to prove this, & also to shew what real & good effect the ventilation had, I made the following experiment. By means of plates of Iron which I had prepared & taken down with me, I stopped off the second joint of the tube upwards, so as to throw the lamp ventilators out of action but without making the least alteration of the burning of the lamp or the general ventilating opening from the lanthorn into the cowl; I simply threw the ventilating tube out of action & let the results of the combustion enter into & diffuse themselves through the air of the lanthorn. In less than one minute all the glass in the middle & upper tier of panes became dim with moisture, and by the time that I had removed the obstructions from the tube which time was less than two minutes some of the middle panes of glass were on the point of running with condensed water. On the other hand, the instant the obstructions were taken away, the dew began to disappear from those panes which were the least dimmed; and by the time that five minutes had elapsed all was bright again. There is no doubt, but, that the state of the lanthorn & of the tower was favourable to such an experiment; because the conditions were such as to bring the air before hand nearly up to the depositing point. But the effect does not the less indicate the action of the lamp ventilator; & the quickness of the appearance & disappearance shews the power & amount of that ventilation. Hence we may I think surely conclude that when deposition does occur it is not now due to the lamp; and, also, that if the lamp ventilation were away, it would occur far more frequently than is the case at present.

<_><_><_>

I will now give you the best judgement in my power upon the cause of the deposition of water on the glass inside when it does occur; and that I may make the reasoning more apparent will state at the outset the conclusion I have come to. I believe in the first place that the air which has access to the lanthorn at such times, is rendered more damp than the air outside, by the moisture which it gathers in passing through the tower; and, in the next, that the temperature which the external air communicates to the glass (& roof) of the lanthorn is below the condensing point of such extra moist air, & causes a continual deposition from it as long as such air continues to pass through the lanthorn.

In order to make the points clear I will state simply a few of the principles of action concerned. Suppose a portion of air shut up in a room with wet walls, all being at the temperature of 50˚; the air will take up a certain amount of moisture & no more, & no further change will ensue. Consider this air as removed into a room with metal or glass walls but still at the temperature of 50˚; no change will occur, nor any deposition take place. If the room & its included air be now raised to 55˚ in temperature, the air becomes in a considerable degree dry air, to use common language, - not because it contains less water, but because it could not by reason of the higher temperature dissolve more water:- but if such air be returned to 50˚, then it becomes as damp as before, & if reduced to 49˚ or even to 49˚1/2, it will deposit moisture as dew upon the glass walls. All this is true for any other given temperature; and no matter how dry an air may seem to be or how warm, the moment it arrives by cooling at that temperature, at which it is saturated with moisture then it is ready to deposit by the least additional cold. Even the flame of a lamp or candle, which may well be considered as hot dry air, will deposit moisture on the inside of a cool glass tumbler, when by contact, it is itself, reduced to the same temperature as the glass[.]

It is very easy therefore to comprehend how the air which plays against the outside of the glass of the lighthouse window, without causing any condensation against it, may, after passing through the tower of the lighthouse, produce condensation inside. The effect only happens when the wind is in the North, or East, or South East, i.e. in cold quarters. Suppose the wind and temperature were to change from such a condition as I found it in (about 44˚ outside at 9 oclk in the Evening) into the north, & lower the temperature to 35˚, the glass panes would then also be lowered in temperature & perhaps be at 39˚, (for I found a difference between the inside of the glass & the outside air when I was there of four degrees). Then this or similar air would enter from below into the lanthorn, partly by the ventilating chambers, and partly as I found to be the case by the tower. It would enter both these below & find them at higher temperatures than itself; yet, both damp, and, as it entered it would send the air already there, forward into the lanthorn and whatever the state of this tower air as to temperature & moisture might be, it would itself assume nearly if not quite the same state whilst in the tower to be urged forward in its turn. Now I found the dew point of the air coming up the ventilator to be 42˚ and that of the air in the bottom & body of the tower to be 41.5 that in the watch room was 41˚ and that in the lanthorn 42˚ and therefore it is manifest that all these airs when they went into the lanthorn with glass & roof at the temperature of 39˚ would deposit water on them, & continue to do so as long as the circumstances remained unchanged.

That this kind of distillation should continue to go on may seem at first strange, but is indeed very natural. When cold weather occurs after warmer weather it will be the windows & roof of the lanthorn that is first cooled & most cooled. The part of the tower below, consisting of thick walls & partly bedded in the earth, will at no time be so cold nor will they be so changeable:- I found the temperature in the oil cellar 42˚; I understand it never freezes there if the door be shut, and I dare say that in summer its temperature does not rise much above 50˚. So at the times of condensation on the windows, the whole lighthouse may be regarded as a great distilling apparatus, of which the lower part of the tower is the warm or distilling end, & the lanthorn, the cold or condensing end; & the air which does pass through, helps, actually, to carry the aqueous vapour from the place where it first is raised by the warmth to the place where it ought not to go. And, though if the cold continue many nights the evaporation of the water below will tend gradually to diminish its quantity & to lower the warmth which raises it into vapour; yet, the very contact & condition of that end of the tower in relation to the earth, tends to the continual supply of more water in place of that which has dried off, & more warmth to cause its successive vaporization.

To these results & this reasoning I will now add a few confirmatory & illustrative facts[.]

I found in the watch room i.e. the room below the lanthorn some water lying at the bottom of the glass, within each window. Mr. Finnis informed me it was water that had ran down from condensation on the inside of the glass during the previous night. This proves that during the night the external air had cooled the windows so much as to bring them below the dew point of the air within; and that during that time of condensation the air in the watch room, which afterwards goes into the lanthorn, was damper than that without.

In the late frosty weather, when the moisture actually froze on the inside of the lanthorn windows, water was condensed in the tower beneath upon the stove air pipe & stove chamber, & ran off the latter on to the floor. Now it will be remembered that the external air passes up the inside of this chamber & pipe; in doing so, it had cooled them so low, that the moisture which the air in the tower had taken from the walls & floor condensed in this central pipe; so that the water which the air had received from the warmer parts of the tower was actually condensed out again before the air had left the place. In fact the cold air which entered the pipe did in the tower, what the cold air that came against the glass & roof did in the lanthorn; it condensed the extra moisture in the air & from the same cause.

On putting out the lamp at sun rise there is often a little deposition of moisture afterwards on the windows. This shews how near circumstances are before hand to a condensation water, since the cessation of the ventilation caused by the lamp & lamp tube only, & the absence of the little heat which it could communicate laterally to the lanthorn, is enough, in such cases, to make condensation come on[.]

That the air in the tower is damp, is shewn by this, that, often, in damp weather, the presence of one or two persons in the day time in the lanthorn will make the windows dim more or less. The following fact proves the same thing. There is a Lobby between the tower & the cottages, on the North side of the tower, which is always open to the tower by a large opening without a door. Opposite this opening & facing the North is a large window; right & left are two doors leading to the cottages which are generally shut; & two other doors with glass windows, lead to short passages, at the end of which are outer doors of wood. When the wind is in the North (which is when the lanthorn windows condense water inside) this lobby window also condenses water inside abundantly; & when the weather is sharp & continuous, the two inner glass doors of the passages will have water condensed on them inside i.e from the air of the tower & the lobby.

In fact it appears that any thing, window, or door, or stove, or pipe that is in contact with the air of the tower has water condensed on it from the air of the tower when cooled, as they often are sufficiently, by the external temperature.

As a further illustration of the continual tendency of water to rise in vapour from the basement of the building & condense on the colder parts, I will describe an effect in the oil cellar. The door ports & lintel consist of three pieces of stone, each of which has two faces bedded in the brick work, a third face exposed to the air inside of the door & chamber, & a fourth face outside of the door & therefore in the lobby. On opening the door I found the under face of the lintel studed with many large drops of pure water ready to fall & many others had already fallen; also the faces of the stone side posts were covered with water as dew & in several parts running. Mr Finnis told me that in cold weather these faces dropped & ran abundantly, & that the effect continued more or less all the cold months. The faces of the stones which were outside i.e. in the lobby were damp but not in this wet state. The cause of this effect is as follows, & well illustrates the relative condition of the bottom of the tower. When the door is closed the air in the cellar becomes saturated with moisture; & in fact I found at my visit the temperature of the cellar 42˚ and also the dew point of the air in it at 42˚. Then, when cold weather comes on, the lobby falls in temperature and the exposed faces of the three stones becomes cooled; this cold being conducted by the body of the stone lowers the other three faces within beneath the condensing point; & then a continual distillation goes on in the cellar, water rising from one part & condensing on another i.e. on these stones. The effect manifestly proves how damp a state the air in the lower part of the tower is & how continual are the causes which tend to this dampness[.]

From the deliberate consideration of all the circumstances of this lighthouse; and of those which take place at the Saint Catherines lighthouse Isle of Wight; and also of the general principles which are concerned in these and similar arrangements, I have come to a conclusion, which, I beg respectfully yet earnestly to submit to the Deputy Master⁴ & Elder Brethren of the Trinity House. It is, that, where the tower of a lighthouse is uninhabited having neither house fires in it nor flues in use, it is a bad thing to be aerially connected with the lanthorn:- that therefore, the lanthorn ought to be separated by double doors from it, and that the warming and ventilation of the lanthorn ought to be independent & irrespective of the tower or of the air in it; and each separately under command[.]

Respecting ventilation; as far as I can judge, the ventilators which are, in many cases, put round a lanthorn opening from the outside, and which can be opened or closed at pleasure, seem very well fitted for the purpose of admitting air[.] They should, of course, be so placed that rain could not drive into them; & if the openings are in stone or brick work, the passages should be lined with metal tube, to prevent the communication of moisture to the entering air. No anxiety need be entertained about the supply of air to the lamp; for, if enough air (which is easily supplied) enters into the lanthorn the lamp is sure to be supplied & its ventilation will take care of itself.

As respects warming;- a stove in the lanthorn with a copper flue appears to me to be the best means of raising the temperature of that which, only, requires warming, namely the general air of that space; and, as the air near the windows is the part which most requires to be kept warm, for the purpose of preventing deposition, so any arrangement of this stove & its chimney, which best effects that purpose, will, I anticipate, be found the best in practice. To illustrate my meaning I will make the following supposition as to arrangement without being so bold as to assert that it is the best. The platform under the glass at the Upper Foreland light is iron, without perforation, & at such a height, that the head of those standing below does not come within a foot or more of it. If a sheet of iron or copper were fastened to the inner edge of this platform and descended 6 or 8 inches or more as might be found convenient, and if the platform itself were open all round for the space of 2 inches from the glass, such an arrangement would make a warm air trough for the delivery of air to the inside of the windows. Then, if a stove were fixed in the lanthorn, & its copper flue placed under the platform about 2 inches from it & from the side of the lanthorn, passing horizontally for about 10 or 20 feet under the part most liable to deposition or most important to be kept clear according as the draught would allow, before it ascended vertically; I think, that, such a stove would then do its best in warming the air of the lanthorn, & that that warm air would be in the most favourable situation to produce its good effect. I do not think that the ventilators should open into this warm air though but lower down; & if, in intense cold, the windows should be at so low a temperature as to have a descending current down them inside, still the warm air, which would then flow into the lanthorn from under the edge of the platform, & the warmth of the platform, would both be in the most favourable condition which they could have to counteract its effect.

In concluding these observations upon the state of the Upper South Foreland lighthouse I beg leave to refer to my letter respecting it of the 14th of September⁵ last every word of which I approve of at this time & would join to the present account[.]

I may also refer here to a letter which I received from you on the 12th of August last⁶ respecting the Lundy light. I have not seen that lighthouse; nor do I know whether it is a new or old tower, or one uninhabited or not. Still my impression inclines me to agree with the extract you sent me from the report of the 16 July 1846⁷; though as regards the ventilating tube of the lamp, I have no doubt it is sufficient for all it has to do or can do if it is as properly put up as that at the South Foreland.

There is one cause of occasional slight deposition at a lighthouse, which I witnessed & ought to mention, that it may not be confounded with more general & more important causes. It is so difficult to make the joints of the glass tight with the frames, that, whenever the wind is strong with rain the water is actually blown through, & runs down the glass inside in streams; sometimes in such quantity as to run over the platform & fall on to the floor. On the first night that I was there about as much as a pint came through in different places. As this water flows down the glass, cooled by the rain & the wind at the same moment on the outside, it causes a little vapour & deposition on it; as soon as the rain ceases or the wind falls, this effect goes off; & generally it is of very little consequence[.]

<_><_><_>

Lower lighthouse. I examined the lower lighthouse both by day & night. Mr Nott⁸ the principal keeper reports that he always had dew upon the glass inside, before the lamp ventilating apparatus was up, the wind being either North or South East;- but now, that, he never has it after the lamps are lighted:- for though deposition will form on the inside in the day time before the lamps are burning, from the moisture contained in the air which enters from the stair case of the old house; yet, on lighting the lamps that moisture in a few minutes disappears. There is no doubt a considerable degree of warmth on the glass from the effect of the lamps & reflectors, & this causes the dissipation of the dampness; but there was the same warming effect before the ventilating tubes were put up; & then the moisture went on increasing instead of decreasing, notwithstanding the warmth.

The reflectors do not become dull & tarnished at their upper surfaces as they did before; & the place altogether keeps clean instead of becoming smoked.

At night I found the lamps burning well & the draught currents all right, both at the ventilating pipes, against the glass, & at the cowl;- even though a current set down the inside of the copper dome because of the cold there, yet the dome itself was perfectly dry, shewing that none of the products delivered into the cowl by the ventilating tube returned into the lanthorn.

I could not well shut off the ventilating apparatus here not having arranged for that purpose: but if I had, I expect from the appearances, that an effect like that I had in the upper house would have followed; for the water from the lamps was at this time condensing in the globe in the middle of the lanthorn.

The water which thus collects in the condensing globe, occurs in all the winter months; and in cold weather, above a pint & a half of water is drawn out of the globe every morning, which is to be considered as over & above that which the tubes have conveyed as steam into the cowl & so into the air. The condensed water sometimes drops from the edges of the tubes, but this is a consequence merely of the form of the parts. This apparatus was one of those early constructed & before minute particulars were learned by experience. It is easy so to arrange these tubes that all the water shall be conducted inwards & so pass away or on into the globe[.]

I brought some of this water from the globe which was condensed on the night that I was there, away with me for examination. I find it to contain a little copper in solution & a little salt. I refer this to the particles of spray which are carried in the air when the wind is strong. It gives however a little power of corrosion especially upon the iron of the lower joint at its upper end. This may be easily remedied by a little more thickness of metal there, as recommended in a recent letter⁹. There is also a small appearance of blue deposit at some of the copper joints; it is a consequence of their construction suffering the inner water to run out at them; the copper tubes themselves which have been up three years are in very good & excellent condition[.]

The keeper Nott still says that he burns 100 gallons of oil less, per year, than he did; and this difference I have traced to the following cause. He is instructed to keep the flame of the lamp between 1 1/4 and 1 1/2 inches high. If he raises it up to 1 1/2 inches, it is liable by any increase of flame to smoke, and so foul the inside of the ventilating tube with soot. As he has to clean this out with a small brush, he endeavours so to regulate his flame that it shall be nearer to 1 1/4 inches than 1 1/2; and by this care, I believe, the oil is saved. It follows that in old times he had a taller flame than he has now, & if such a tall flame were durable & again required, he must have tube[s] a little larger in diameter. As a taller flame does not increase the horizontal divergence or brightness but only the vertical divergence (See my letter of the 9th Octr. 1845¹⁰) so there seems to be no reason to regret the care that the keeper has been obliged to take[.]

In concluding this long letter I can only say for excuse that I was anxious the Deputy Master & Elder Brethren should be able themselves to judge of my conclusions & reasons. I am happy in the thought that both forms of the lamp ventilating principle as established at the South Foreland & elsewhere are worthy of the favour & confidence which the Board has given to them¹¹[.]

I am My dear Sir | Your Very faithful Servant | M. Faraday

Jacob Herbert Esq | &c &c &c