University of Virginia Library

Appendix S
Charles Bonnycastle Plan for Curing Smoking Chimneys
[ca 5 October 1828]

The imperfect action of Chimneys in carrying off smoke appears to arise from two distinct
causes, the want of sufficient permanent draught, & the liability to be affected by currents of
air; the first can only be effectually removed by increasing the height of the chimney, but
where this is not practicable the evil may still be lessened; the second may I beleive be
wholly cured.

The effect of the wind on the action of a chimney is not produced by the wind entering the
chimney & driving the smoke before it, but results chiefly from its blowing irregularly, or
being heaped over the chimney by the resistance experienced from some neighbouring
object that rises higher than the chimney itself. I will consider these cases seperately.

All fluids are found by experiment to have a very powerful lateral action upon their own
particles--a stream of water passing over a pipe which communicates with a vessel of water
below, will raise the latter in the pipe; & a similar effect is found to take place with a current
of air. Hanksber caused a stream of air ab to pass through a vessel into which introduced a
barometer cd,[873] with the open end plunged into a cup of mercury, as at c:

[drawing]

The current instead of compressing the air in the vessel, & that causing the mercury to rise
in the tube, rarified it by carrying, or rather drawing by its lateral attraction, part of the air in
the vessel along with it, & the mercury descended in the tube below the height of the
barometer at the time of the experiment.

From this cause when a gust of air passes by a house it acts as an air pump in attenuating the
air within; upon the side on which the wind blows of course the effect will be the reverse,
but from every other part of the house, & especially from the chimney, the air will be
rushing out to join the current which is passing. When therefore the gust ceases the air
within the house is less condensed than that without, for if a partial vacuum existed without
it will be filled instantly from the surrounding medium, but the deficiency within can only be
supplied through the passages by which the air escaped, & chiefly by the chimney; down
which the air from without rushes, carrying the smoke before it, until there is an equilibrium
within & without the house. This action is repeated with every gust of wind, & as only a
small part of the smoke which is driven down into the room when the gust has passed, is
drawn to the chimney to be carried up again whilst a new gust is passing, the house must
after a short time be filled with smoke.

The remedy which immediately suggests itself is to place a light valve on the top of the
chimney which will freely allow the air to ascend, but will close immediately it attempts to
descend. It is true that whilst the valve is closed the smoke must be accumulating in the
Chimney, but this will only continue until the vacuum below has been supplied through the
crevices of the doors & windows. But if instead of trusting to these channels of ventilation,
we close them as completely as possible, & supply their places by pipes containing light
valves opening inward, the supply will be readily obtained from within, whilst there will be
no draught by which the room can be exhausted but that up the chimney.

In the figure on the right I have drawn

[two drawings]

a room with a valve of this kind at B, & another in the chimney at A. The construction of the
valves is shewn in fig 3; where AB is a circular box of tin, which for the chimney may be 10
inches diameter, but for a lateral valve, as at B, need not exceed 3 or 4. This box is pierced
completely through by an opening pqrs, more than half its diameter. At i & k a wire about
1/8 dia, passes through a small collar worked in a wire which passes over the opening; this
axis ik carries a circular plate mn, of the lightest tin, whose diameter is greater than the
opening, but less than the box. A small stop at r prevents the valve when it is driven upward
from closing the opening pq; but when driven downward it falls on vs, & closes the opening
there. The lateral valve is of the same construction but smaller, & worked in the thickness of
the wall, it is chiefly intending for small rooms with low chimneys, where were such a valve
placed in each of the four walls they would probably much assist the action of that in the
chimney.

The second case, where there is an obstacle that causes the air to be heaped over the
chimney when the wind blows, differs from the first in this, that even when the wind does
not act by gusts, the increased density of the air above the chimney, to which is pressed
against the neighbouring obstacle by the current, faster than it can escape, will have a
tendency to drive the smoke down the chimney; & might render the valve less efficacious by
keeping it closed for a considerable time, In such cases an other outlet on the opposite side
of the obstacle against which the air was heaped would always afford a remedy; & in the
Rotunda we have this remedy within reach; a pipe of tin or sheet iron a few inches in
diameter carried round the parapet wall from one chimney to the other, would enable the
smoke of one chimney, when the valve was closed there, to flow into the other, & it is
obvious that when the air is heaped by the wind on one side of the roof it is attenuated by
the same current on the other, so that both valves cannot be closed at once.

[drawing]

AD, ViU:PP, 4p, with ASB docket "Prof Bonnycastles plan for curing smoking Chimneys."
This plan was enclosed in John Hartwell Cocke's letter to ASB of 5 October. It is separated
from that letter and can be found in the undated material for 1828.