Steve,

I think it is accurate to conclude that cap stacks served a decorative function, but I see no evidence that decoration was the only function; or evidence to conclude that no documentation exists that indicates a function other than ornamentation. Generally, I find almost no reference sources covering cap stacks, while there is a lot of published information about the various shaped spark arrestor stacks. That is why I am curious about cap stacks.

Regarding the question of whether cap stacks had cast iron barrels or were made as rolled sheet metal: I agree that a sheet metal barrel would have a riveted seam. So, I have looked for that detail in all photos, and actually do find several photos that clearly show such a seam running the length of the stack barrel.

I would say that most cap stack photos do not show such a seam, however, this is a fine detail, and may not show in many photos which lack sufficient clarity. It barely shows when you can clearly see it. Also, photos show only a part of the stack surface, so a seam might be there on the surface not shown in a photo. I note that the visible seams vary in their location around the stack barrel. And also, since aesthetics was an objective, measures may have been taken to hide the seam. They may have hid the seam by making a butt strap seam with the strap on the inside, and peened the rivets flush on the outside.

In any case, I do clearly see a visible seam, including rivets and the metal edge of a lap, on cap stacks in the following photos:

1) The photo I mentioned earlier here showing New Jersey & Railroad Transportation Co. #36 with the Hudson’s coal-burning stack (a cap stack).

2) In the book, Minneapolis and the Age of Railways by Don Hofsommer, page 42, photo shows M&StL 4-4-0 #47 built by Manchester 11/1881, in operation and rebuilt with an extended smokebox, probably circa 1890.

3) In RL&HS Railroad History #154, the M&StL locomotive roster by William Edson, photo shows M&StL 4-4-0 #6, built by Baldwin 8/1877, in operation and rebuilt with an extended smokebox, probably circa 1890.

4) In the book GREAT RAILROAD PHOTOGRAPHS by John Winthrop Adams, page 34, photo shows St. LOUIS & SAN FRANCISCO 4-4-0 #96 built by Cooke 1884, as builder’s photo with an extended smokebox.

5) In the book, A LOCOMOTIVE ENGINEER’S ALBUM by George Abdill, page 36, photo shows Delaware, Lackawanna, & Western 4-6-0 #85, built Danforth & Cooke 3/1885, in operation at unknown date. Design elements appear to be earlier than 1885.

6) In the same book, page 98, photo shows Mt. Gretna Narrow Gauge Ry. 4-4-0 #11.

7) In the book, RAILS WEST by George Abdill, page 104, photo shows MKT 4-4-0 #82 built by Baldwin 1877, in operation and rebuilt with an extended smokebox, probably circa 1890.

Other evidence of the use of rolled metal for cap stacks is the drawing of a cap stack posted above and the one on the previous page, which both indicate a relatively thin wall such as .25 inch. The drawing on page 5 refers to the cap assembly as being castings, and showed the parts with section cross hatch symbolism. The base, which is always a casting, is also shown with the same cross hatch. Yet, the stack barrel is not cross hatched. To me, this strongly suggests that the barrel is not cast iron.

The drawing on page 4 shows both the cap and the base in section view with a solid hatch fill, whereas the stack barrel is shown as a single line where the section cuts each wall, again indicating that barrel is rolled rather than cast iron.

So I see this evidence as indicating that cap stack barrels were frequently rolled sheet iron with a longitudinal, riveted seam, and no taper. However, I would not conclude that cast iron barrels were never used with cap stacks.

I would estimate that 95% of cap stacks in perhaps 100 photos I have examined show a straight (non-tapered) barrel. A lack of taper at least suggests that the stack barrel was rolled sheet since casting would permit taper without any complications. And also, most cast iron shotgun stacks do have some taper increasing the diameter toward the top. While others have some taper to a smaller diameter toward the top. Opinions seemed to have varied as to how much taper was needed, and the amount was generally quite small. So with this recognized need for taper; and with cap stacks generally having no taper; it suggests that cap stacks were not tapered because it simplified the sheet metal pattern, and the telescoping fit to the cap assembly.

The few cap stacks with a taper that I have found in photos show a prominent taper increasing the diameter toward the top of the stack. The taper extends right up to the cap. In my opinion, this is not good looking design. It is surprising how much different the tapered barrel makes a cap stack look. Actually, I think a cap stack with a taper to a smaller diameter at the top of the barrel would look better. But generally, the straight barrel with the prominent “flying saucer” cap and an echo of that shape in the stack base seems to give the best appearance.

I too am fascinated by cap stacks, and the era of railroading when they were prominent. And again, what you write above goes more to decoration than any other function.

You are correct that much has been written about the function of balloon and diamond stacks (among others), explaining the inner workings of the stacks necessary for spark reduction. If there was some function for the cap stack other than decorative, why don't we see any mention of this whatsoever, even in the little we have in writing on these stacks? Why don't we see a single short sentence in Meyer, for example, stating something to the effect that, "the weight of the cap adds necessary stability to the stack itself?" if that were indeed the case? After all, Meyer went so far to include two drawings of such stacks, and refers to them in the text.

The only books you mention that I have are the Abdill books, and I'm afraid I don't see the seams that you seem to see. That being said, I do suppose there may have been instances of rolled sheet metal stacks being made. Even so, I suspect that there were easier, cheaper means of stabilizing the tube than a fairly elaborate casting or set of castings.

Actually, from a molder's view, no taper is needed. The stack is way too long to pull the pattern off green sand, therefore there must be a core, in which case, perfectly cylindrical is no more difficult than tapered.

If the stacks do have taper, it is for aesthetic reasons, not ease of production.

Or, to optimize vacuum. The advent of the Master mechanic's front end, extended smokebox, etc moved some spark arresting capabilities form the stack into the smokebox, and people learned a lot about increasing exhaust vacuum through better nozzles and stack arrangements after a while. Expanding the stack from base to top lowers pressure of entrained gasses and pulls more vacuum in very general terms. I got a close look at TEXAS a couple weeks ago, and was very interested in how the smokebox and front end was put together compared to something from 1890 or 1920. Steam locomotives in 50 years starting in 1850 was doing in its time what computers have done in the past 50 - and, are much more interesting and fun to play with.

When I said, "Opinions seemed to have varied as to how much taper was needed...", I did not mean that as referring to a need for taper as mold draft, or for aesthetics. I was assuming that there was stack gas flow theory behind the purpose of the taper and assumed there were differing theories and opinions about that.

For the plain cast iron shotgun stack, I most often see outward taper rising; and less common are the ones that taper inward rising to about half the stack height, and then from there, taper outward to the top of the stack. Earlier I see straight stacks without caps that are a cylinder with no taper.

Most of those theories would be considered fairy tales in our days. Using standard text book theory of our days momentum of the steam ( its mass times velocity) would be shared with the combustion products. Assuming a) a uniform stack exit velocity and b) for the sake of the argument no vacuum, a parallel stack with 3 times the orifice diameter would give triple the mass at 1/3 of the velocity. With taper(inverted cone) the mass flow would increase with a decreased velocity. The massflow from a decreased exit, wrong taper, area would decrease which is not really what we would want.
Kind regards
Jos Koopmans

Looking at the book, MASON STEAM LOCOMOTIVES by Arthur Wallace, page 118, Fig. 159, photo shows 0-4-4T named “Peter Stuyvesant,” of New York & Manhattan Beach RR, with a cap stack. Caption mentions that the engine was equipped for spark abatement, and a note by Mason regarding the stack says, “straight with perforated cone.”

Like the earlier mentioned cap stack referred to as the Hudson’s coal-burning” stack, this Mason stack appears to be a larger outside diameter than what would seem to fit the base casting. Both stacks appear to be maybe 4” larger O.D. than the actual smoke barrel. I wonder if the cinders caught by the “perforated cone” were thrown into the 2” space between the inner and outer stack wall for collection, and then later discharged from that space.

Here is a link showing a stack that looks like other cap stacks, but with a couple other appendages. It does have a function of spark arrestor and cinder storage. Like the description by Mason above, this stack has a perforated cone in the stack barrel.

This stack is called out as being cast iron, and unlike the design of typical cap stacks, this stack “cap” is cast as part of the cast iron stack barrel, so it is a doughnut shaped chamber that is part of the entire stack interior. The chamber is to catch the cinders riding up the outside of the perforated cone. The cinders strike the upper inside surface of the doughnut, and bounce back down to the bottom of the doughnut. Apparently, the cinders bounce around inside of the doughnut until they fall into the discharge port leading down into the collector pipe.

Refer to 5th post down in this discussion:
https://www.smokstak.com/forum/showthread.php?t=92122