Scientific American Supplement, No. 388, June 9, 1883

In a preceding article, we have described a ventilator which is in use at the Decazeville coal mines, and which is capable of furnishing, per second, 20 cubic meters of air whose pressure must be able to vary between 30 and 80 millimeters.
In order to actuate such an apparatus, it was necessary to have a motor that was possessed of great elasticity, and that nevertheless presented no complications incompatible with the application that was to be made of it.
In the ventilation of mines it has been demonstrated that the theoretic power in kilogrammes necessary to displace a certain number of cubic meters of air, at a pressure expressed in millimeters of water, is obtained by multiplying one number by the other. Applying this rule to the case of 20 cubic meters under a hydrostatic pressure of 30 millimeters, we find:
20 × 30 = 600 kilogrammeters.
In the case of a pressure of 80 millimeters, we have:
20 × 80 = 1,600 kilogrammeters.
If we admit a product of 50 per cent., we shall have in the two cases, for the power actually necessary:
600 / 0.05 = 1,200 kilogrammeters, or 16 H.P.
1,600 / 0.05 = 3,200 kilogrammeters, or 43 H.P.
Such are the limits within which the power of the motor should be able to vary.
After successively examining all the different systems of engines now in existence, and finding none which, in a plain form, was capable of fulfilling the conditions imposed, Mr. E.D. Farcot decided to study out one for himself. Almost from the very beginning of his researches in this direction, he adopted the Woolf system, which is one that permits of great variation in the expansion, and one in which the steam under full pressure acts only upon the small piston. There are many types of this engine in use, all of which present marked defects. In one of them, the large cylinder is arranged directly over the small one so as to have but a single rod for the two pistons; and the two cylinders have then one bottom in common, which is furnished with a stuffing-box in which the rod moves. With this arrangement we have but a single connecting rod and a single crank for the shaft; but, the stuffing-box not being accessible so that it can be kept in a clean state, there occur after a time both leakages of steam and entrances of air.