Smoke Deflector design and theory

I am curious about smoke deflectors as I think they can enhance or spoil an engine's look. Mostly spoil, though, I would say. I am struck by the contrast between, for instance, the Southern look, the "traditional" look of for instance the Britannias, the A1's and the 9F's, the "German" style, the total lack of them on the GWR, etc etc.

To my mind, Sir Lamiel and Lord Nelson look cute, Flying Scotsman with the German style just looks silly, the Brits etc look boring, the Duchesses look very nice, Princesses and all the GWR big engines look just right with no adornment, the way they should be, also the Black 5's etc.

So what determined whether a loco got steam deflectors? Was it just the chief engineer's foibles? Why do so many engines not have them at all, like the GWR? What made Southern go for the Dumbo ears, why does Scotsman have those strange things instead of Tornado-style, etc etc

I was a bit too young to worry about such things when I was growing into train-spotting, we fretted more about what colour the next "Western" would be outshopped in. But now I have time to ponder about such things :noidea:

And as a PS, why do the Castles and Kings have those beautiful curved steam pipes? Was that just a touch of design/publicity genius by Collett? Was there something that made it necessary? :crazy: It is extremely distinctive and good-looking but was it necessary or cosmetic?

 

Smoke deflectors were there for a purpose: they forced air down the side of the boiler thus creating a high pressure area each side of it to prevent steam being drawn down in front of the driver and so obscuring his view ahead. This was dangerous and led to some signals being passed at danger, sometimes disastrously.

The need for deflectors was dictated by several things, the height of the chimney and blast pressure being among them. A low chimney failed to lift the exhaust away from the smokebox so it would be drawn down, as would a soft blast, such as given by a double or multiple exhaust. High, single chimneyed engines tended not to need them.

The steam pipes transmit steam from the internal pipes inside the smokebox to the steam chest, so if the cylinders are mounted well to the rear, as in the GWR four cylinder engines and Stanier pacifics, they had to be curved simply to line up.

The steam loco was mostly a functional machine; aesthetics came second.

 

"Smoke deflectors became increasingly common on later steam locomotive designs as efficiency gains through improved smokebox design and reduced back pressure on the cylinders in turn reduced the uplift of smoke exiting the chimney"

http://en.wikipedia.org/wiki/Smoke_deflectors

"The German style smoke deflectors were a small price to pay for the rejuvenation of the A3s. The fitting of double chimneys in the late fifties gave the locos a second wind but BR then found they had the same problem as the LNER had had with 97. The soft exhaust caused smoke and steam to hang in drifting clouds around the boiler obscuring the driver's view ahead. Four of the class had small deflector plates fitted to the top of the smokebox, as had been tried on Humorist, but these were pretty ineffective. It was, we are told, Peter Townend (the boss at 34A) who suggested trying the German style deflectors. (They are called Witte deflectors in Germany - after their designer - they replaced the earlier Wagner style, which were like a huge version of the ones fitted to the Brits) These solved the problem. The books suggest that the Witte deflectors were the most efficient. The DB fitted all their pacifics with them as they were so much better than the original "elephant's ears"."

http://www.lner.info/forums/smoke-deflectors-4472-t3057.htm

Regards,

Neil

 

The views of Lms2968 and Njal are well representative, respect is due...

It also makes we wonder whether whatever technical college bulleid was using Knew sweet Fanny adams about aerodynamics as neither P2 or original Merchant navy profiles worked particularly well in this respect.

If the'Gill slit' design on the spamcans had been tapered (at least internally) it surely would have given the same effectiveness as the witte /trough style deflectors on the A3 i.e effective.

If the A1SLT do go on to do P2, if its a soft blast its hoped that they can get a few aeronautical boffins to tweak the profile inside the smoke deflectors to make them effective without having to bugger about with that divine external shape.
No doubt the curved mini fall plate on the GW/ Stanier designs has (by design or accident) increaed the airpressure / speed along the length of the lower boiler/running plate to dissipate steam drifting down
this would be stalled by the blocky steam pipes on the Princess Coranation (Duchess) Its intresting to note that of all the GW/LMS Types only the Scots/ Baby scots/ Princess coronations ended up with deflectors

 

Need to be careful when discussing P2 design; do you mean the original as fitted to 2001-2 or the later A4 Bugatti style fitted to 2003 - 2006 ?


In "Nigel Gresley Locomotive Engineer" F.A.S. Brown records that "in anticipation of a low-pressure exhaust and to guard against steam beating down over the cab spectacles, wind tunnel experiments were made at the City & Guilds College of Engineering to ascertain the most suitable shape for the top of the smokebox. The side deflector plates, on which the locomotive's name plates were mounted were also evolved from the wind tunnel experiments."

In December 1934 2001 was sent to Vitry [ France ] where Gresley and Bulleid tested 2001 and found it to be extremely efficient and easily produced 2800 hp; the front end of 2001 was changed to the A4 style in 1937 whilst 2002 was fitted earlier in October 1936 to standardise with the A4-style front fitted to the other class members from new.

If the work by the A1LST to date with 60163 is any guide I would expect the group to continue Gresley's policy of small but continuous improvements to his locomotives and - in that context - the problem of drifting steam will be one of the areas where improvements will be made. But in terms of modern day steam recent observations of 6233 in it black guise and 71000 during its last main line spell of activity show that exhaust still forms around the chimney lip and rolls some way along the boiler before being lifted clear of the cab.

It therefore seems a little graceless to criticise Bulleid when - it seems - all pre-war designers suffered the same problem of drifting exhaust in their locomotive designs - especially Pacifics.

 

from an asthetics POV, I'm not a fan of the southern style of deflectors, the seem a little half-arsed !

although I think a lot of it was down to Fowler's basic designs being ugly, the combination of smokebox and deflector on the rebuilt scots and baby scots were more attractive than on the orginals ... but also the bare princesses, rebuilt jubilees and GWR locos looks would not be improved by deflectors.


as for the standards ... the deflectors do a bit to disguise the oddness of the high running plate look.

air flow is a complex science and if all these different deflectors were effective solultions for the problems of smoke in the eyes with different combinations of loco shapes and exhausts, it just shows how complex it is. There is that story about how a thumbprint on the airflow model of the A4 was the solution to their smoke problems shows that accients sometimes work just as well.

 

The Southern style of deflector was the result of a lot of head-scratching throughout the early-mid 1920s on the part of Maunsell, Holcroft and Clayton. Credit must be given to this design team for taking the plunge in undertaking such a bold experiment (the Southern was the first British railway company to fit them), trialling 'German'-style deflectors, chimney wing-plates and everything in-between before arriving at the final design.

The Bulleid type was a compromise fitted retrospectively (the original air-smoothed casing design did not have smoke deflector plates). Bulleid had originally envisaged that a large rectangular hole in front of the chimney would be sufficient to lift the exhaust up, hence the so-called 'widow's peak' cowling of the early Merchant Navy locomotives. I imagine that the blast proved too soft for this alone to be effective in service, as the smoke tended to be pushed along the top of the casing before drifting downwards, leading to a complete redesign of the front-end. This resulted in the now familiar rounded cowling with larger, downward-facing hole to scoop the air hitting the smokebox face, and the addition of the side deflectors to help address the issue of the smoke clinging to the casing.

The problem with smoke deflection is that it nearly always comes down to a compromise, and aesthetics rightly plays second-string to operational practicalities. However, the story of the Bulleid deflector might have been very different had the Giesl ejector been available in Britain before the war...

For those interested, see D. L. Bradley's LSWR Locomotives: The Urie Classes for details of the smoke deflector experiments.

 

I think that what has been overlooked is that with a loco at speed it forces the air immediately in front of the smokebox to be displaced outwards, i.e., the air turns near 90 degrees. This causes a vacuum around the smokebox and boiler causing the air containing smoke to fill this vacuum.

The same effect can be seen with lorries fitted with 'curtainside' bodies at speed. The air hitting the front of the body is displaced outwards causing a lower air pressure (partial vacuum) at the forward end of the bodyside curtain. The atmospheric pressure inside the body is greater, thus causing the side curtains to bulge outwards, particularly at the front of the body.

Returning to steam locos, for smoke deflectors to be effective two things are necessary:-

1) They must extend forward enough to capture the displaced air, and,

2) The deflectors must be far enough away from the smokbox as possible to allow the air to pass down the length of the smokebox and boiler.

This explains why some, like the German Witte type as fitted to the Gresley A3s were effective, and other types (as fitted to the Thompson and Peppercorn Pacifics and Royal Scots, for example) were near useless.

Non-original they may be, but I cannot understand why 60163 wasn't fitted with the Witte type.

That said, personally I think the Witte type suited the A3s, viewed side on they fill in the gap forward of the smokebox. But then having lived most of my life by the GN main line I'll admit to being a teensy bit biased.

 

Yes i see, the fundamental difference with Bulleids style and most of the others is that his idea was to use the forward motion of the air to ramjet air around the blastpipe that would blow it up and away, but this as you say creates a vacuum along the sides of the boiler which draws it back down again, subsequent modifications to his 'air smootheds' allowed some of this pressured air to pass down the side of the boiler to mittigate this.
Most other smoke deflectors work only to pressurise and accelerate air down the side of the boiler. The Witte type creates a stream of higher pressure / velocity air a little higher up but again either side of the blast pipe...

Hopefully if 2007 gets the go ahead it will be with a poppet valve gear and have a sharp blast as i would hope that the original form (Yes actually a fan of Bulleids Aesthetics as well as most of his ideas which didnt always work as well as intended) can be left alone. But if not if a little slot inside the front end shroud gap at either side (in its original form) would introduce this higher speed/ pressure air stream along the boiler side...

Given that the outward apperance of a geisl ejector can be pretty well concealed in most 'double chimney' profile castings it surprises me we dont see more of them (well they do cost a bit i suppose) but as well as the strong blast they do work well with spark arrestor screens ...

 

When I was trainspotting in the late fifties through to the end of steam, the general opinion was that German - fitted A3s were the sexiest engines on the railway.

 

Beauty is in the eye of the beholder and I think deflectors improved the looks of some locomotive classes. In Germany I think the Witte type on the Prussian P8 4-6-0 looked plain daft but on the Pacifics, 01.5 and 03 classes in particular, they made the locos look mightily impressive IMHO. The A3s were graceful without them and pugnacious with them so I'm happy either way as far as they are concerned.

 

The original P2 design, i.e. 2001 was a failure in terms of safe exhaust deflection. The same style of design was tried on other designs, the Belgian class 1 for example, with the same outcome. The Bugatti front does work, it works very well. The original P2 front does not trap and channel sufficient air to be effective, the extra plates subsequently fitted trap more air but cause visibility problems in themselves. Bugatti all the way, unless you are an A3, then its the Witte type.

 

On GWR 4 cylinder engines the cylinders are well behind the smokebox, as the regulator is in the smokwbox you need to get the steam to the cylinders so you have these long pipes. The pipes are curved so that there is minimal restriction to the steam flow. If you look at the 2-cylinder engines you will see the pipes are straight down from the smokebox to the cylinders.

 

Thanks everyone for very interesting information.

 

Re the GWR curved steam pipes - their size and shape was chosen to minimise the pressure loss as the steam passed through them. But they do look good as well!

Regards,
John Webb

 

Thanks. It was interesting to observe yesterday that Princess Elizabeth has much smaller pipes, though of a similar shape.

 

But - as Stanier's first Pacific design - it has often been argued that the Princess Royal Class was the Pacific design that Swindon didn't think worth developing in view of the successful running of its Kings and Castles.

 

Interesting thing about deflectors is there were so many designs, yet all trying to solve the same problem; smoke drifting around the front end and obscuring the driver's view.

I saw photos once of a newish Britannia at Cardiff Canton and another of the same loco towards the end of it's life at Crewe. In the first view the smoke deflectors had hand rails bolted on, but the later one showed the rails had been replaced by cut outs in the deflector. Would the deflectors have been changed during it's 15 year life, or would the rails have been removed for some reason?

I agree with other posters, those German type smoke deflectors enhanced the appearance of the A3s.

 

The handrails were removed on some BR (W) Britannias and replaced with cut puts following an accident to I think Polar Star, that was blamed in part on poor forward visibilty. Not all Brittanias were so modified.

 

I can't recall the exact details, but there was an accident on the Western Region concerning a Britannia loco. Restricted vision due to the handrail was given as one of the causes. Consequently, they were removed.