Hostname: page-component-848d4c4894-5nwft Total loading time: 0 Render date: 2024-05-26T13:30:57.560Z Has data issue: false hasContentIssue false

The Regenerator Principle in the Stirling and Ericsson Hot Air Engines

Published online by Cambridge University Press:  05 January 2009

Edward E. Daub
Department of General Engineering, University of Wisconsin, Madison, Wisconsin 53706, U.S.A.


During the first half of the nineteenth century the Stirling brothers and John Ericsson made significant attempts to design hot air engines which could provide an economical alternative to high-pressure steam. They believed that it would be possible to gain fuel efficiencies superior to those obtained with steam by employing a heat exchanger within the air engine cycle, erroneously supposing that such a heat exchanger could eliminate the large loss of heat incurred in the condenser of the steam engine. The Stirling brothers called their heat exchanger an economizer because it prevented the waste of heat, but Ericsson coined the term regenerator because he imagined that the power of the heat (caloric) could actually be regenerated, i.e. re-used in the engine to generate mechanical work. Ericsson's term became commonly accepted, so that Stirling's economizer is now also referred to as a generator. These attempts to attain high horsepower outputs with hot air engines were abandoned after the dramatic failure of Ericsson's notorious paddle steamer in the early 1850s. Further ventures were aimed at the more limited goal of low horsepowers and generally ignored the regenerator principle.

Research Article
Copyright © British Society for the History of Science 1974

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)


1 For a discussion of the history of air engines, see Finkelstein, T., ‘Air engines’, The engineer, ccvii (1959), 492–7 522–7. 568–71, 720–3.Google Scholar

2 Church, W. C., The life of John Ericsson (2 vols., New York, 1890), i. 198.Google Scholar

3 ‘The centenary of the heat regenerator and the Stirling air engine’, The engineer, cxxiv (1917), 516–17.Google Scholar The document was found by pure chance on the occasion of the hundredth anniversary of the patent by someone rummaging through the Stirling family papers. Although it had not been enrolled in London in 1816, the details of the engine were known because ‘copies of the Scotch specification were produced in evidence at the celebrated trials of the Neilson Hot-blast patent’; see ibid., p. 516. For the strange circumstances which account for the failure to enroll the patent in London, see Finkelstein, , op. cit. (1), p. 523Google Scholar, or The engineer, cxxiv (1917), 523, 537, 567.Google Scholar

4 ‘Mr. Braithwaite recalled the circumstances attending the first trials of Ericsson's caloric engine, in England; there was not any regenerator.’

5 ‘Julius Jeffreys’, Dictionary of national biography, x. 723.Google Scholar Reference to Jeffreys's Respirator first occurred when James Stirling appeared before the Institution of Civil Engineers in 1845 to discuss their engine. ‘In answer to a question from Mr. Lowe, Mr. Stirling agreed, that Jeffreys’ Respirator, used by consumptive patients, was based upon the same principle of the alternate absorption and giving out of caloric to air traversing the capillary passages' see Stirling, James, ‘Stirling's air engine’, Mechanics' magazine, xlv (1846), 559–66 (563).Google Scholar

6 Church, op. cit. (2), p. 72.Google Scholar

7 For an excellent review of attempts to use hot combustion gases directly during the nineteenth century, see Finkelstein, , op. cit. (1), especially pp. 493–5.Google Scholar ‘The dust and ashes carried over into the cylinder from the fire, combined with the difficulty of maintaining some lubrication in the power cylinder, destroyed the engine fairly rapidly in nearly all designs’ (ibid., p. 495).

8 See Stirling, , op. cit. (5), p. 561.Google Scholar Stirling pointed out that, although his engine used air at pressures as high as sixteen atmospheres, safety was guaranteed by leakage of air from the cylinder when the engine was at rest. (An air compressor supplied air to maintain the pressure while the engine was operating.) Steam, he said, is most dangerous when the engine is at rest since the furnace heat can cause a sudden rise in pressure.

9 Eugene Ferguson has called attention to an excellent review of nineteenth-century attempts to find alternatives to steam: Babcock, G. H., ‘Substitutes for steam’. Transactions of the American Society of Mechanical Engineers vii (1886), 680741.Google Scholar Carbon bisulphide was tried in America as late as 1872, when an explosion nearly cost the life of a manufacturer and hastened its departure from the industrial scene; see ibid., pp. 680–6.

10 ‘The Operation of the engine is explained as follows: the part of the cylinder surrounded by the flues is heated to a temperature of 480° higher than the part AB. In the position represented … the plunger is in contact with the piston, by which means the included air is brought to the warm part of the cyclinder and has its elasticity increased and presses upon the piston with a force greater than that of the atmosphere. The piston is thus forced downwards … till the pressure of the included air and that of the atmosphere become equal. The impulse communicated to the fly wheel acts then … to raise the plunger from the piston. The included air is thus made to descend between the plunger and cylinder and brought to the cold part; it is cooled in its descent, has its elasticity diminished, and its pressure becomes less than that of the atmosphere, the piston is forced upwards … and the revolution of the fly and crank again bring the plunger toward the piston, the air ascends through the same passage by which it descended, is heated in its ascent and forces the piston downwards … and so on alternately’; The engineer, op. cit. (3), P. 516.Google Scholar

13 The only respect in which my reconstruction may fail to do justice to the content of the original patent is that I ignore the fact that the full first half of the patent is devoted to Stirling's claims for the invention of a heat enconomizer in which currents of hot and cold air flow alternately and counter-currently along narrow passages between thin metal plates. Stirling presupposed that discussion when he described his engine. However, no patent purports to give the actual historical progression of its ideas, and my reconstruction in reverse order seems more credible than to believe that he first thought of his heat economizer for more general purposes and then invented an engine that included one.

14 Law, R. J., James Watt and the separate condenser (London, 1969), p. 13.Google Scholar

15 ibid., Figures 32, 39.

16 ibid., p. 15.

17 ‘Ericsson's caloric engine’, Mechanics magazine, xx (1833), 81–3.Google Scholar

18 Mauel, Kurt, Die Rivalität zwischen Heissluftmaschine und Verbrennungsmotor als Kleingewerbemaschinen zwischen 1860 und 1890 (Düsseldorf, 1967), p. 73.Google Scholar

19 Zeuner, G., Grundzüge der mechanischen Wärmetheorie (Leipzig, 1877), p: 212Google Scholar: ‘Auf die Wirkungslosigkeit des Regenerators bei richtig arbeitenden calorischen Maschinen hat zuerst Hirn aufmerksam gemacht.’

20 Hirn, G. A., Exposition de la théorie mécanique de la chaleur (2 vols., Paris, 1876), ii. 101–15.Google Scholar

21 ibid., p. 109.

23 ibid., p. 103.

24 ibid., pp. 110–15. In a steam engine, the boiler and condenser operate at constant temperature so that all the heat supplied enters at the maximum available temperature To and all the heat rejected leaves at the minimum available temperature T1. Since the Ericsson air engine received heat at constant pressure and rejected heat likewise at constant pressure, it was only at the very end of the former and beginning of the latter that the air was operating at the maximum and minimum temperatures available.

25 Zeuner, , op. cit. (19), pp. 205–17.Google Scholar The quotation is from p. 212.

26 ibid., pp. 217–18.

27 Kurt Mauel has commented on the fact that the Stirling engine was very little known on the Continent. See Mauel, op. cit. (18).

28 Rankine's first treatment of the Stirling cycle was completely theoretical; see his paper ‘On the geometrical representation of the expansive action of heat, and the theory of thermo-dynamic engines’, Philosophical transactions of the Royal Society (1854), reprinted in Miscellaneous scientific papers of W. J. M. Rankine, ed. Tait, P. G. (London, 1881) pp. 338409Google Scholar (especially pp. 372–4). He also presented an analysis at the meeting of the British Association for the Advancement of Science in 1854, and this is the paper most frequently cited; see Rankine, , ‘On the means of realizing the advantages of the air engine’, Report of the British Association for the Advancement of Science, 1854 (London, 1855), pp. 159–60.Google Scholar This was merely a brief abstract of his paper. The most readily available source for Rankine's thinking about the Stirling and Ericsson cycles and the role of the regenerator was his book, A manual of the steam engine (London, 1859)Google Scholar, and all subsequent editions; see pp. 345–70 of the first edition.

29 Zeuner may have been misled by a cursory reading of Rankine, for Rankine referred to the Ericsson cycle as a constant-pressure cycle and the Stirling cycle as constant-volume, meaning in both cases the process of heat exchange in the regenerator. Since Zeuner attributed the constant-pressure processes in the Ericsson cycle to the exchanges with the heater and cooler, he could easily have assumed that Rankine intended the same. Rankine, however, had a different image of the Ericsson cycle. He believed that the exchanges with the heater and cooler were isothermal for the Ericsson cycle. This was not an error on Rankine's part, for he was analysing the 1851 Ericsson engine in which Ericsson had adopted the Stirling arrangement where the furnace was in direct contact with the air expanding in the working cylinder, so making the expansion isothermal.

30 Rankine, , ‘On the geometrical representation …’, op. cit. (28), pp. 372–4.Google Scholar

31 Mauel, , op. cit. (18), p. 101Google Scholar: ‘Es ist doch geradezu auffallend, dass die erfolgreichste Heissluftmaschine des Kontinents keinen Regenerator hatte, und dass die in so beträchtlichen Stückzahlen gebaute kleine Heissluftmaschine von Ericsson eine offene Maschine war und ebenfalls keinen Regenerator hatte’. For a description of Ericsson's successful and popular hot air engine designed for small power output, see Finkelstein, , op. cit. (1), p. 496.Google Scholar

32 Edelmann, H. C., ‘In search of Stirling’, The announcer, xxiii (1969), 612 (12)Google Scholar: ‘The “History of the Galston Parish Church”, records the following: The close of the year 1848 and the beginning of 1849 brought a terrible calamity to Galston and the country. The scourge of cholera came. It was then that Dr. Stirling showed the stuff of which he was made. Then did he manifest the fidelity and Christian courage of a true minister of Christ. Fearlessly he moved among the plague-stricken homes in the parish; faithfully did he minister to the physical and spiritual wants of the suffers. He toiled among them night and day; he tended them; he prayed with them; he buried them. And I am only speaking the literal truth when I say that his conduct at that time won something nobler than the presentation which was given to him for his faithful and devoted service, and that was the lasting gratitude of many a soul in this parish. When I think of these days, and when I think of that work so faithfully and fearfully performed, I have no hesitation in characterizing this country minister as a Christian hero.’

33 James Leslie reported on the fate of the Stirling engine at Dundee to the Institution of Civil Engineers during an extended discussion of the new open-cycle caloric engines of Ericsson in 1853:

‘As it has been asserted, that Stirling's engine did work effectively and economically for a number of years, it will naturally be asked, for what reason was it abandoned, at the Dundee foundry? From information furnished by Mr. David Mudie, now one of the lessees of the Foundry, it appears, that the motion of the engine was not perfectly smooth and uniform, which was the only mechanical objection, and that not an important one; but the real cause of the engine having been set aside, or reconverted into a steam engine was, that the bottoms of the air vessels could not be made to withstand the heat to which they were exposed.

The engine of 45 horsepower was started in March 1843. In December 1845, (viz., two years and nine months after starting), one air vessel gave way; in May 1846, another failed, and in January 1847, a third, when the parties carrying the foundry, Mr. Stirling having left the management and gone to Edinburgh, got discouraged by these repeated failures, and removed the engine.’; Leslie, James, ‘On the principle of the caloric, or heated air engine’, Minutes of the proceedings of the Institution of Civil Engineers, xii (1853), 563–71 (570).CrossRefGoogle Scholar

34 ‘Mr. Stirling has since turned his attention to the best mode of remedying the defects, and there can be little doubt that he could satisfactorily accomplish his object, if he were to meet with support and encouragement, in carrying out and completing his experiments. At present, however, the only active movement being made towards turning this engine to account, seems to be in New York, where certain parties having collected all the information they can procure, as to the patents, and as to the engines constructed at Dundee, are endeavouring to improve Stirling's air engine with the intention of bringing it out in preference to Ericsson's.’; ibid.

35 Rankine, , ‘On the means of realizing …’, op. cit. (28), p. 160.Google Scholar

36 Rankine, , Manual of the steam engine, op. cit. (28), pp. 362–71.Google Scholar

37 ibid., p. 371.

38 Rankine, ‘On the means of realizing …’, op. cit. (28), p. 160.Google Scholar He listed the advantages: ‘compact in its dimensions, easily worked, not liable to get out of order, and consumed less oil, and required fewer repairs than any steam engine.’ He made no mention of his calculation which demonstrated superior fuel efficiency, perhaps because that had been treated earlier. One wonders whether Rankine had never heard of the failure of the air vessels under severe heating as reported by Leslie (see note 33).

40 From the Boston evening transcript, reported in ‘The caloric ship “Ericsson”’, Mechanics' magazine, Iviii (1853), 469.Google Scholar For an excellent study of the Ericsson engines, see Ferguson, E., ‘John Ericsson and the age of caloric’, Contributions from the Museum of History and Technology, Bulletin 228 (1960), pp. 4260.Google Scholar For a discussion of the stimulus Ericsson's new engines gave to thermodynamic debate, see Bryant, Lynwood, ‘The role of thermodynamics in the evolution of heat engines’, Technology and culture, xiv (1973), 152–65CrossRefGoogle Scholar, especially 152–7.

41 ‘Ericsson's caloric engine’, op. cit. (17), p. 82.Google Scholar

42 Sargent, J. O., A lecture on the late improvements in steam navigation and the arts of naval warfare with a brief notice of Ericsson's caloric engine (New York, 1844); see especially pp. 5964.Google Scholar

43 ibid., pp. 60–1.

44 ibid., p. 62.

45 ibid., pp. 63–4.

46 Ferguson, , op. cit. (40), p. 46.Google Scholar Ferguson's account of the trials is most interesting; see p. 45.

47 Reported from the Journal of the Franklin Institute, in ‘The caloric engine’, Mechanics' magazine, lxii (1855), 7880 (78).Google Scholar The editor of the Journal replied: ‘We do not accuse Mr. Ericsson of having ever asserted that his engine was, in principle, a perpetual motion. But this claim was decidedly and frequently made in the various newspaper articles which reported his banquets, and for which we cannot but consider him responsible, since they were published under his auspices (on his account as it were) and without public remonstrances on his part, so far as we have ever heard’; ibid., p. 80.

48 ‘Dunn's patent caloric engine’, Mechanics' magazine, lv (1851), 42–5 (42).Google Scholar The patent was in the name of Dunn.

49 Ericsson, J., ‘Observations on Major Barnard's calculations relative to the theoretical power of the caloric engine’, Appleton's mechanics' magazine and engineers' journal, iii (1853), 121–3 (123).Google Scholar

50 ‘The caloric ship “Ericsson”’, Mechanics' magazine, lviii (1853), 469.Google Scholar

51 Barnard, J. G., ‘Theoretical investigation of the caloric engine’, Appleton's mechanics' magazine and engineers'journal, iii (1853), 152–8, 241–5 (154).Google Scholar

52 ibid., p. 241.

53 ‘Captain Ericsson on the caloric engine’, Mechanics' magazine, lxiii (1855), 56 (6).Google Scholar

54 Quoted in Church, op. cit. (2), p. 198.Google Scholar

55 ‘The caloric ship Ericsson’, Mechanics' magazine, lviii (1853), 62–4 (64).Google Scholar

56 Finkelstein, , op. cit. (1), p. 525.Google Scholar

57 Edelmann, . op. cit. (32), p. 6.Google Scholar