Lima "Double Belpaire"?

I have been reading for years (decades) about how Lima, in the late 1940s, tried to market steam locomotives with a "Double Belpaire" boiler, but have never known exactly what that means. The usual image -- the one we've all seen, the re-touched photo of a C&O 4-8-4 -- shows Belpaire "shoulders" over the fire-box, but I've never seen how the design differs from an ordinary Belpaire boiler.

So... Can anyone tell me what Lima's idea was? And what made it "Double"?

The other photo we've all seen is a shot of the inside of the firebox of a model boiler. Was this model made by Lima for experimental purposes? Model tests are useful in lots of areas of engineering, but I would think the operation of a locomotive boiler would depend on exactly how water circulates in it, and hydrodynamics is plagued with scale-dependent complexities that I think would make it difficult to use results from a model test... So, second question, what was Lima's research for their new boiler design like and how did the model figure in it?
--
Water circulation... Locomotive boilers are made out of opaque materials. On the other hand, water circulation was an important feature of operation. (Think about the intended function of thermic syphons.) So, third question, did engineers in the steam locomotive era have equipment that allowed them to observe water circulation inside an operating locomotive boiler, and if so, what was it like?

Allen, could you post these photos.. or links? I have seen neither picture that you mention.

I found a picture of the scale model here:
http://www.leclairerail.com/MiscSubjects.html
http://www.leclairerail.com/Additions-2 ... erLima.jpg

the "photo" of the C&O J3a and a brief discussion can be found in "Perfecting the American Steam Locomotive"
By J. Parker Lamb
scanned here: http://i894.photobucket.com/albums/ac14 ... /COJ3a.jpg

It appears that the 'double part refers to the fact that the Belpaire squarish shape of the furnace roof is carried through into the barrel, both top and bottom, in a way to form a combustion chamber of roughly square cross section between the furnace proper and the rear tube sheet.
I believe the advantage claimed is that the Belpaire shape has more radiant heat absorption surface than a cylindrical shape in the same envelope. hence a higher steaming capacity for a given boiler size/weight.

As for your other question about flow measurement inside boilers, one method is to mount a number of threaded bushings in various locations , flush with the surface on the water side, and insert 'averaging pitot' flow elements. the deltaP between the two sides of the element is proportional to fluid velocity. since the dP is only a few inches of water the attached indicator can be a simple U-tube manometer. the elements can be very thin tubing in a faired sheath to minimize the device's impact on the flow ! they can be inserted to various depths and rotated to find direction of flow as well as velocity.
In a confined channel like a tube this can produce very accurate mass flow measurement. In less restricted geometries such as a water leg around a radiant furnace with arrays of staybolts and two phase flow, the problem would be very complex, but precision would probobly not be needed, so much as assurance of a certain minimum velocity under varying conditions.

Based on the picture of the model boiler, I also would believe that the second part of the belpaire is for the combustion chamber, possibly also offering some water-tube capability in the form of additional water-legs inside the combustion chamber?

I would think the obvious advantage would be the same as the firebox portion.. the ease of aligning the staybolts between the firebox/combustion chamber sheets and the boiler courses.

Did the belpaire firebox have any further advantages? I would think the obvious disadvantage would be the stresses of corners in the during the heating & firing.

Thanks, both of you!
Re: the photos.
The ones from "leclairerail.com" are new to me: the only photo I am familiar with of the model is inside the firebox; I've never seen the external appearance. (Or any hint that it might still be in existence, on display!) The retouched image illustrating the proposed locomotive is the one I know.
---
Particularly the model photo makes it plausible that the "Double Belpaire" feature involves squaring-off at the lower sides of the combustion chamber. (Conventional Belpaire boilers with combustion chambers ((as on the PRR M1 4-8-2)) continue the "flat top" effect from the firebox forward over the combustion chamber, so a cross section of the outer shell through the combustion chamber would be round below, squared at the top: the "Double Belpaire" looks as if it would have four square corners in such a cross section.) Perhaps this is visible in the picture of the locomotive as well: there is a light area visible below the running board below the Belpaire combustion chamber which might represent this lower squaring. Thanks! I think this explains the terminology!
---
Re:
Flow measurement.
Thanks again. That sounds like a sort of instrumentation that would have been available in the steam era. Actually USING it with a (full-scale or model) boiler would be ... difficult... particularly for things like thermic siphons! (So: I still wonder to what degree designers of locomotive boilers were guided by actual measurement and to what extent their ideas about the functions of siphons, etc., was derived from theory.
---
Re: Advantages.
Well, my recollection is that heat transfer trough the firebox and combustion chamber walls was MUCH more efficient than through tubes/flues, and this arrangement would -- slightly -- increase the wall area.... But whether this would actually be useful, it seems to me, would depend on other things, like... water flow around the combustion chamber. So... The answer is probably available somewhere in the literature on the Lima Locomotive Works (or maybe in the yellowing pages of issues of "Railway Mechanical Engineer" from the 1940s): did Lima use the model for testing how the idea would actually work?

Allen Hazen, I think you are right that alot of boiler design was and is based on theory, imagination, or even wishfull thinking, and let us not forget trial and error, rather than empirical measurement, though there was and is plenty of that done too.

you are also right that radiant surface has a much higher heat absorption value than convection surface.
This is codified in the ASME code as follows in section A-44 ( PG-70 prior to 1992)
minimum lb/hr safety valve relieving capacity per sqft of heating surface:
.......................firetube...... water tube
convection surface
hand fired..................5.......... 6
stoker fired............... 7.......... 8
oil or gas fired............8.......... 10

radiant surface
hand fired................ 8............8
stoker fired..............10...........12
oil or gas fired...........14..........16

this table , dutifuly memorized by every power engineering license candidate in the US for 85 years was finally relegated to the "Alternate rules" by the ASME in 1992. today PG-70 says "it shall be the responsability of the boiler manufacturer to determine the required minimum relieving capacity" . This acknowledges that both theoretical and empirical methods have improved enormously in recent decades, as have materials.

Parker Lamb unfortunately has little to say about the why and how of the "double bubble" boiler, and only mentions the model to say that Lima built it "to confirm their analysis of the anticipated improvements."

(Thanks, Eliphaz, for your typically informative comments!)

I hound an extract from J. Parker Lamb's book on Google Books. Apparently Lima (perhaps after failing to interest any railroad in investing in an all-new locomotive?) tried to interest the New York Central in a project to re-boiler a Hudson with a double Belpaire. This would have been ... interesting ... visually! Victorian Railways (in Australia) had a class of Hudsons with (conventional, not double) Belpaire boilers, built in Britain in the early 1950s. To my, American, eye a Hudson with a Belpaire looks incongruous: as if the Penn Central merger had somehow been back-dated to the 1930s!