Some Steam Engine Prehistory

Case Model

Mr. Ball's Case Model at Blaker's Reunion in 1948. Courtesy of Ernest Hoffer, 444 Starr Ave., Toledo, Ohio 43600

Ernest Hoffer

Content Tools

178 Emerson Place, Brooklyn, New York 11205.

After reading the May-June issue of I.M.A. I especially enjoyed the article, 'My First Engine' by George Eves. Hopefully Mr. Eves can be persuaded to give us some more of his experiences.

Many history books and encyclopedias have taken to copying earlier material without checking out all the facts and have thus been guilty of perpetrating errors. For example, how many history books state that Robert Fulton 'invented' the steamboat? Similarly Watt is commonly known as the inventor of the steam engine when his credit should actually be an improver of an already existing machine, for the Newcomen steam engine was at work for over 60 years before Watt put his hand to it.

Numerous attempts were made during the late 1600's to harness the force of steam but almost all proved impractical. Ironically, the man who came closest to 'discovering' the principle of the steam engine did not apparently realize what he had and didn't follow up his experiments. This was Denys Papin who was one of the large number of amateur scientists of the 17th century who together did so much of the work which laid the foundations of 'hard science', which was to follow.

As one of his experiments, Papin fitted a piston in a small vertical cylinder and attached a cord to the piston which ran up over pulleys and was attached to some weights. Water was placed into the cylinder and a fire (probably an alcohol or similar type lamp) placed under it. When the water boiled the resulting steam allowed the weights to pull the piston up in the cylinder. The source of heat was removed and when the cylinder had cooled the steam condensed and formed a vacuum. Atmospheric pressure then forced the piston down which, by means of the cords, pulled the weights up. Here was the Newomen type engine in principle but Papin saw it as nothing more than a toy and of no practical use. To add to the irony, Papin used a type of device which had been developed by Von Guericke about 1672 in the course of his experiments on atmosphere pressure. The difference between the two was that Von Guericke used an an pump to create his vacuum while papin used the condensation of steam. Papin did not invent the piston and cylinder concept, for not only Von Guericke, but the ancient Romans knew of piston type water pumps. Plunger type pumps were also known at the time of Christ, but without packing glands, which made them of little use. It was not until Sir Samuel Morland invented the packed Plunger pump in the late 1600's that this type became practical. There are some writers who do not even credit Papin with pioneering the use of steam under a piston to develop a vacuum and as early as 1678 experiments were being made to create a vacuum in closed vessels by exploding gun powder. By 1698 Thomas Savery had developed his pumping ' engine', which involved creating a vacuum by means of steam condensation and the stage was now set for the reciprocating steam engine. Savery's device consisted of two closed copper chambers connected at the top to a steam supply from a boiler and at the bottom to the water suction and discharge lines. The attendant started the apparatus by opening a cock, or valve, in the steam line and filling one of the chambers with steam. The steam supply was shut off and cold water from another line was allowed to run over the outside thus condensing the steam inside. As a vacuum formed, atmospheric pressure forced the water to be pumped up the suction pipe into the chamber. When the steam cock was once again opened, steam (which at one point was reported to be as high as 120-150 psi pressure) was admitted into the top of the chamber forcing the water out. Check valves in the suction and discharge lines directed the flow thru the proper lines and with two chambers in use, one was filling while the other was discharging. All valve operations were performed by hand. The practical suction lift of this device was about 25' or so and as Savery had developed his pump for use in mines deeper than this, he planned to install both pump and boiler in the mine itself. Available steam pressure limited the discharge head, however, and this device still left much to be desired, especially insofar as economy of operation. The same theory interestingly enough was used for 'Pulsometer' type sinking pumps for mine work which remained in service until fairly recently especially in Europe. One of Papins other developments without which steam boilers would be rather dangerous, was the safety valve. This was of the simple lever and weight type and here again we find that although a person (in this case Papin) is credited with discovering an idea, he actually took an existing concept and made it practical. Very primitive types of safety valves consisting merely of tapered plugs driven into a hole had been used for hundreds of years on stills and retorts. Papin's, however, was first of all capable of fairly fine pressure adjustments and, secondly, was self resetting when the steam pressure dropped below the blow off point. And on this basis he gets and deserves credit.

Papin developed this device for use on the pressure cooker he invented in 1680. Here again, one writer, Thomas Ewbaun, on page 393 of 'Descriptive and Historical Account of Hydraulic and other Machines for Raising Water', published in 1856, quotes several Biblical sources which he interprets as referring to a type of pressure cooker, thus indicating that the item in question was known previously.

To return to the steam engine, we find an interesting reference in a diary kept by one Roger North to a device which he saw in about 1680. His description and crude drawing includes a boiler, two vertical cylinders, racks, pawls, ratchet wheels and 'wheelwork', the latter of which operated stop cocks to control the steam flow to the cylinders. There is strong evidence to link this engine to Sir Samuel Morland and if this be the case he has the distinction of building the world's first reciprocating steam engine with self-acting valve gear. Unfortunately, no solid evidence has come to light to prove or disprove this point nor are any details know. And it was 1712 before the first recorded full size reciprocating engine was built. Savery's pumps were, however, in use draining mines by this period but it is not known for certain if Newcomen was at first familiar with them or not. What is known is that Newcomen and Savery became partners due to the fact that Savery held a patent issued in 1698 for raising water by fire, the wording of which, of course, covers any type of engine powered by a boiler. It is believed by some writers that a semi-experimental Newcomen engine may have been in service as early as 1705 and it is interesting to note that this is also the date at which Savery abandoned his efforts on behalf of his own engine, and it is also believed to be the date of the Savery-Newcomen partnership. It seems that we shall never know for certain.

The Newcomen engine was of a beam type but it did not use a slide valve. The steam admission valve was located inside the boiler and consisted of a flat fan shaped sector plate mounted on a stem which passed thru the boiler shell. A slight rotation of the stem caused the plate to pass over the steam opening, like a shutter to open or close the supply to the cylinder. Pressure were so low, 1/2 to 1-1/2 psi, that this mechanism worked quite well. The slide valve was unkown at this time for it (and also the eccentric) was developed by William Murdoch, who was a foreman, and later manager of Boulton and Watt, some time during the 1780's. Slide valves did not really become popular till after 1800 and many types of pumping engines used poppet valves and plug rods similar to those of Watts day till long after 1900.

The pistons of Newcomen engines were not pushed during the non-condensing portion of the engine cycle for two reasons. First, the piston was connected to the end of the beam by chains of a type similar to today's link-belt chains, and this arrangement would not have transmitted any compressive force from the piston to the beam as a solid piston rod would. The engine was so constructed that the pump rods at the other end of the beam outweighed the engine parts thus pulling the piston up when the vacuum was broken by the opening of the steam valve. Secondly, the steam pressure was so low that there was not enough 'push' to have done much good and in fact the piston often pulled the steam out of the boiler! This condition led to a series of events which eventually came down to us as the never-to-die-out story of the boy, Humphrey Potter, and his cords.

Newcomen's first experimental engines were undoubtedly hand valved but by the time they were being offered on a commercial basis the self-acting valve gear had been developed. This consisted of a 'plug rod', which was a long timber, suspended vertically by chains from an arch head on the beam and fitted with pins or 'stops' which struck the handles of the valve gear as the beam rose and fell, thus opening or closing the appropriate valves. Obviously, great care was needed in setting the stops for there was nothing to prevent the piston going out of the top of the cylinder or hitting the bottom if they were incorrectly adjusted. In later engines spring beams were fitted so that they would be struck by the engine beam before the piston traveled too far in either direction.

A boy was involved in the valve gear story, as was Humphrey Potter, but it was to properly team it a boy!

It was soon discovered that the primitive boilers then employed could not supply steam fast enough for the engine, and of course lack of steam meant loss of vacuum, bringing the engine to a halt. Newcomen, therefore, regulated his engine speed by boiler pressure so that the condensing, or power stroke, did not occur until enough steam was available in the boiler to refill the cylinder for the next stroke. To accomplish this, the valve gear was so arranged that as the pump rods pulled the piston to the top of its cylinder, the weight operated cold water injection valve was 'cocked' but held from being opened by a catch. A large vertical tube passed thru the top of the boiler into the water space and a float was fitted inside this tube and connected to the valve gear catch. As the boiler steam pressure recovered, the water in the tube was forced up slightly, thus raising the float which, in turn, tripped the injection water gear. The steam in the cylinder was condensed, the piston made its power stroke as it moved down and was again pulled up as the steam admission valve opened and broke the vacuum. It is important to note at this point that it was necessary for the steam pressure, and hence the water level and float, to drop each stroke to allow the injection gear to latch before it could be tripped again. As boilers were made larger and better, it was evident that under slow rates of pumping the boiler steam pressure did not fall enough during each stroke to allow the valve gear to function properly. Under these circumstances, the procedure was to make the bouy inoperative by disconnecting or jamming it in its tube. The injection valve gear was then tripped by hand as the piston reached the top of its stroke and here we come to our old friend, Humphrey Potter. The Potter family had been involved with mines and Newcomen engines and so it is not surprising to find one of its younger members, Humphrey, helping to operate an engine. It was he who added a cord from the injection gear to the plug rod, thus causing the engine to trip the valve and eliminating the need for manual attention to this chore. However, his contribution was merely an additional refinement and he in no way can be credited with inventing the self-acting valve gear. A much modified version of the Potter cord remained an integral part of Newcomen engines from then onwards although engines were built for several years with both 'bouys' and 'Potter cords' so that they could be operated at will with either. By 1718 Henry Beighton built a Newcomen engine and is said to have been the first to eliminate both the cord and bouy substituting instead a linkage operated solely by the plug rod.

Removing the bouy pipe from boilers created a new problem, that of preventing excess pressure build up. The open topped pipe acted as a vent, for an overpressure condition would blow some of the water up the pipe until the steam could relieve itself to the atmosphere. It was about 1718, therefore, that the safety valve appeared on boilers to soon become a standard fixture. Some early boilers were, by the way, fitted with two safety valves. They were so weak they needed one to prevent explosions and one to prevent the boilers from collapsing due to formation of a vacuum when they cooled down.

Although Watt's early engines operated at only several pounds of steam pressure, he was not only aware of the expansive force of steam, but was the first to use a cut off. He had thought about doing this as early as 1769 but it was not until 1776 that the engine driving 'Boulton and Watts' Works was altered to operate expansively. There is, of course, no difference insofar as operating an engine with a steam cut off if the engine exhausts into a vacuum or a back pressure, other than varying the actual point of cut off itself to compensate for the loss of power due to the back pressure. In 1778 Watt constructed an engine for the Shadwell Water Works in London that had adjustable pins in the plug rod to allow a cut off at 2/3 of the stroke. Steam pressure at this time was 10-14 psi. This expansive principle was patented in 1782 by Watt who also at about this time invented the indicator, another device which is still being used today. It was in a much cruder form than those with which we are familiar but its principles were the same.

One of the main factors that contributed to the acceptance of Watt's engines over those of Newcomen's was the lesser cost of fuel, for the former were using less than half as much for a given horse power. This was due not only to the adopting of an early cut off but also to the separate condenser which by itself may well rank as the most important single improvement in the history of the steam engine.

The foregoing is of necessity a very condensed version of the steam engine's pre-history and many factors, as well as many early pioneers have been skipped over.