Visible vs Actual Boiler Diameter

[MikeOxon]

A few years ago, when building a model of the GWR No.184, I looked up the boiler diameter in the RCTS booklets on GWR Locomotives (Part 3) as 4’ 2” and rolled my model boiler accordingly.

 

My write-up  led to an interesting discussion with @Buffalo on the subject of visible diameter vs. boiler diameter, once the boiler cladding is taken into account.

 

In the case of the GWR ‘157’ class, ‘Buffalo’ wrote : “As it happens, I have a copy of Swindon drawing No 11130 of 1893 sitting in front of me. This relates to alterations to the trailing end of a 517, but it shows the outlines of both the boiler and the cladding. The dimensions are unfortunately unreadable, but given the usual boiler diameter of 3' 6 7/8", a quick measurement and arithmetic suggest about 3' 11 1/2" for the outside of the cladding.” These figure show that the thickness of the cladding can make a very significant difference to the perceived diameter of a locomotive boiler!

 

More recently, I started to scratch-build a model of the Broad Gauge ‘Gooch Standard Goods’, which had a published boiler diameter of 4’ 6” (large for those days). I bought a length of 18 mm diameter brass tubing, to use for my model ‘boiler’. In correspondence, @Michael Edge reminded me again that the visible diameter would be 4” to 6” greater than this, depending on the thickness of the boiler cladding.

 

That started me thinking about how many modellers actually take account of this discrepancy. I’m sure that the true model-engineers, especially when working in larger scales, are well aware of the difference but how many 4mm-scale model boilers are of ‘true’ visible dimensions? It’s well known that many RTR models and several older plastic kits diverge considerably from prototype dimensions. One example is the Dapol ‘City of Truro’ kit, which has a boiler diameter somewhere between the Swindon Type 2 and Type 4 prototypes.

 

Out of curiosity, I had a look at a few well-known drawings, starting with the Broad Gauge ‘Waverley’ class, for which a drawing that I believe is from Gooch’s own notebook is in the NRM Collection.  Comparing the wheelbase of the drivers (quoted as 7’ 5”) with the boiler diameter as drawn, yielded a figure of 4’ 5.6”, which is actually less than the published 4’ 6” diameter!

 

I also checked the diameter of the boiler for the ‘Gooch Standard Goods’ in the Broad Gauge Society kit (FL02). The brass rectangle provided has a width of 58.1 mm which, when rolled, creates a boiler of 18.5 mm diameter (4’ 7½ “ prototype size), so very little allowance for any cladding.

 

Finally, I’ve noticed that several of the drawings in Russell’s GWR Locomotives show a dotted line inside the line marking the boiler diameter. The location of this dotted line seems somewhat variable but I assume it is to indicate the presence of cladding over the boiler itself.

 

I’d be interested to hear comments from other members of the forum on what they have measured on their own model locomotives and what allowances they make when scratch-building.

[Ian Simpson]

For early 1840s (standard gauge) locos I usually assume that the lagging and the wooden slats around the boiler added around 12 inches to the overall diameter.  It's one of those "facts" I seem to remember reading once, but heaven knows where!

[DCB]

The lagging changed from wooden slats to asbestos over the years and a number of boilers had thick lagging and or or oversized cladding to make them look more impressive, I have an idea the LNER Thompson L1 was one.

The GWR Churchward boiler cladding generally followed the boiler profile but some later ones and some stanier LMS boilers had short cone boilers with long cone cladding for appearance,   2251, 94XX and Stanier 5MT and 8F spring to mind.

 

[MikeOxon]

Table II in the 'Preliminary Survey' (Vol.1 of the RCTS series on GWR Locomotives) has sketches of clad and unclad boilers, which show that the external boiler bands often have little relationship the actual boiler rings!  Despite all the other details about boiler fittings, I've found nothing about the actual thickness of the cladding.  It seems curious that this vital information (for modellers to determine the visible diameter) is so hard to find.

Edited April 23, 2020 by MikeOxon
clarification

[Michael Edge]

GW weight diagrams usually have the boiler outline shown dotted. Most boilers are built with overlapping rings, the quoted diameter is normally the largest one. All boilers are lagged, early on this was often done with wood, leading to a planked appearance, later it changed to asbestos. The asbestos was either thrown on to a warm boiler by hand so that it stuck there (the dangers of asbestos were not really known until after the end of steam here), or it was incorporated in a sort of mattress. None of this is self supporting or paintable so there is an outer layer of thin steel sheet, the cladding, fitted outside it. This is the surface which is painted and the outside of the loco which we try to model. It has to be held on with steel bands round the boiler, these are what modellers call boiler bands and do not have to coincide with anything on the boiler - except where a taper boiler changes from parallel. The bands are fastened together with bolts through clips, mostly on the bottom but sometimes on the top of the boiler for accessibility. The cladding usually adds about 5" - 6" to the largest boiler diameter but this is not fixed and can vary quite a bit.

[Devo63]

In this scan (5600 class Diag A30) you can see a dotted line inside the outer cladding.

 

 

[Gibbo675]

Hi Mike,

 

The actual diameter of the boiler compared to the cladding sheets varies both depending upon the type of locomotive and the specific part of the boiler, also taper boilers can be quite deceptive.

 

The cladding over most fire boxes is usually about 1.5" to 2" from the shell of the boiler with the cladding over the barrel usually from 2" to 5" depending upon shape.

 

As the various sections of the boiler are riveted together you will have to take into account the thickness of the boiler plates and the head size of the rivets. Most boiler plates are between 5/8" and 3/4" depending upon boiler pressures and material types. At the joint between the fire box and the boiler barrel there is the thickness of both the throat plate and the wrapper sheet which could be 1.25"-1.5" plus the rivet heads standing proud of the the top surface of the boiler barrel when viewed in profile.

 

The cladding sheets are attached via crinolines which are frameworks running under the joints in the cladding sheets bolted to snubs riveted or welded to the boiler. The cladding sheets are screwed to these frameworks and boiler bands are used to both tighten down the sheets and also hide the joints in the sheets. Under the cladding would have been asbestos matting or asbestos paste rendered onto the boiler as plaster to a wall.

 

For an idea of scale the depth of the crinoline where the smoke box meets the boiler barrel is 2.125", as you may note the front ring of the boiler barrel is parallel and the depth goes to approx 4". The boiler is of an LMS black five.

 

Gibbo.

[Michael Edge]

The exact thickness of the insulation is of no interest to us as modellers, I was mainly pointing out that the actual boiler inside the insulation and cladding is irrelevant.

[stewartingram]

I always thought that the original Triang B12, with its obviously too small a diameter boiler, had been drawn up from the actual boiler (only) dimension. Just compare it to the recent Hornby new model.

 

Stewart

[MikeOxon]

44 minutes ago, Michael Edge said:

The exact thickness of the insulation is of no interest to us as modellers, I was mainly pointing out that the actual boiler inside the insulation and cladding is irrelevant.

True - but many reference books only tell us the boiler diameter, which is not much use to modellers, as you say.

[34theletterbetweenB&D]

On parallel boilers, the guidance I was given was 'for 4mm scale , add 1.5mm to the scale figure for the quoted diameter of the boiler. If that then looks too fat in proportion to the other dimensions, then the quoted diameter you have is probably over the cladding!' I am sure many of us recall when good information was often hard to come by and revel in the present rich resources.

 

51 minutes ago, stewartingram said:

I always thought that the original Triang B12, with its obviously too small a diameter boiler, had been drawn up from the actual boiler (only) dimension. Just compare it to the recent Hornby new model...

Or there might have been some dimensional 'hybridisation' with either the smaller diameter round top used on the B12/4, or the barrel of the original Stratford boiler. Unless it had an unusually large layer of insulation under the cladding, the old B12/3 model's boiler was I think undersize for the B12/3 over boiler plates. (And irresistibly; how fine the new model is...)

[Michael Edge]

1 hour ago, MikeOxon said:

True - but many reference books only tell us the boiler diameter, which is not much use to modellers, as you say.

Quite, best to ignore it and find more useful dimensions.

[MikeOxon]

I notice that Jim Champ, in his recent book 'An Introduction to GW Locomotive Development' writes: "Weight diagrams especially were neither intended nor used for construction or maintenance. ... The actual accuracy is ... certainly no better than the nearest scale inch and may sometimes be several inches out." 

 

Of his own drawings, he writes: "I also checked details against my own and published photographs and there have been occasions when I’ve got on the bike and ridden over to Didcot or elsewhere to clarify details."

 

As is always the case, the only way to be sure is to measure the thing yourself - but that's not possible when I model 19th century prototypes

Edited April 25, 2020 by MikeOxon

[Michael Edge]

Sometimes a weight diagram is all that you have but GW ones are much better drawn than most. No drawing can be  absolutely relied on, there are frequently errors in them and/or unmentioned alterations between drawing and building - and the process was sometimes the other way round, building before drawing. 

[rab]

13 hours ago, MikeOxon said:

I notice that Jim Champ, in his recent book 'An Introduction to GW Locomotive Development' writes: "Weight diagrams especially were neither intended nor used for construction or maintenance. ... The actual accuracy is ... certainly no better than the nearest scale inch and may sometimes be several inches out." 

 

Of his own drawings, he writes: "I also checked details against my own and published photographs and there have been occasions when I’ve got on the bike and ridden over to Didcot or elsewhere to clarify details."

 

As is always the case, the only way to be sure is to measure the thing yourself - but that's not possible when I model 19th century prototypes

I think I may have shared before, but as a draughtsman

I always wanted to add a note to my drawings:

 

Any similarity between this drawing

and the finished product

is purely coincedental. 

[Gibbo675]

17 minutes ago, Michael Edge said:

Sometimes a weight diagram is all that you have but GW ones are much better drawn than most. No drawing can be  absolutely relied on, there are frequently errors in them and/or unmentioned alterations between drawing and building - and the process was sometimes the other way round, building before drawing. 

Hi Mike,

 

I have used reproduction workshop drawings supplied by the NRM which were in the main excellent, however on occasion some have had either glaring errors or the locomotive seemed to have been since modified and no loner conformed to the specification of the drawing. Unexpected alterations were then the only way around the problem with a descriptive sketch placed in the locomotives job file.

 

Gibbo.

[Michael Edge]

4 hours ago, Gibbo675 said:

Hi Mike,

 

I have used reproduction workshop drawings supplied by the NRM which were in the main excellent, however on occasion some have had either glaring errors or the locomotive seemed to have been since modified and no loner conformed to the specification of the drawing. Unexpected alterations were then the only way around the problem with a descriptive sketch placed in the locomotives job file.

 

Gibbo.

Very true, I've found this sort of thing countless times. The best answer is to measure the loco yourself (if still possible), otherwise check all the time against photographs - if it doesn't look like the photos it's wrong. Sometimes there's a lack of photos as well and then the job gets really difficult - eventually it's a matter of doing the design work yourself but a fair bit of railway engineering knowledge is required for this and that tends to be in short supply amongst a lot of railway modellers.

[Gibbo675]

1 hour ago, Michael Edge said:

Very true, I've found this sort of thing countless times. The best answer is to measure the loco yourself (if still possible), otherwise check all the time against photographs - if it doesn't look like the photos it's wrong. Sometimes there's a lack of photos as well and then the job gets really difficult - eventually it's a matter of doing the design work yourself but a fair bit of railway engineering knowledge is required for this and that tends to be in short supply amongst a lot of railway modellers.

Hi Mike,

 

The trouble I had was that at the time I was chopping up a modified Hall and converting it into a Saint.................the chaps at Didcot were very happy with it once I had finished !

 

Gibbo.