Locomotive Front End Designs

Re The Dean Goods I agree with JimC that the locos as originally built did not have the later chimney that was later used by Sam Ell in a number of tests. Neither did they have the original boilers, the later boilers being belpaire with superheating. They went through a number of boiler developments and changes.

I think it is a bit of a disservice to suggest Sam Ell based all his re-design of smokebox draughting on the Dean Goods. The Rugby test reports do simply not support this. I think he had a rather more empirical approach than is acknowledged, and his mathematical formulae (which was published at the time) was a bit off the mark and is probably best not examined in too great a detail and pulled to pieces. However it is a fact he transformed many well known designs. I knew many people who met Sam Ell, and the equally charasmatic Ernie Nutty.

He transformed the GWR Kings. He transformed the V2s. I could produce a long list. Some of the tests were never published. He was a very clever chap, and it is only in the last 15 years or so that this whole topic has undergone a complete re-assessment and greater understanding.

Cheers,
Julian

 

The trouble with Ell's formula is that it mixes 2/4 and 3 cylinder locomotives with the smaller exhaust of the latter. I wish I had
the original data for new regression formulae for both types.
Kind regards
Jos Koopmans

 

Quite agree with Julian...
The original form of the Dean Goods had 140psi round top firebox domeless boilers, 241 1 5/8" tubes and 17" * 24" cylinders and 13,313 lbs TE. In their ultimate form they had 180psi belpaire boilers with generous domes, 195 small tubes, 6 large ones, and 17½" * 24" cylinders and 18,138 lb TE. They must have gone through several iterations of front end design. Interesting that the blast pipe and chimney arrangement sketched above is the same as the 1940 GA drawing of a Dean Goods I have, but the dimensions appear to me rather different to those of the 5700 class, with similar culinders, valve gear etc and an unsuperheated version of the boiler. Whether they are critically different I cannot say.

 

Hi Jim,

I also have both the drawings you refer to, plus a few others. I will need to dig them out and have another look. The 57xx had different diameter wheels, but essentially the same boiler though not superheated.

Cheers,
Julian

 

Well, the evidence is against you. A1 frames cracked just as frequently, if not more so, than A3's. If anything the failure rate decreased in their old age, although that might well have been due to Cook's work in improving frame alignment etc.

It is odd that with a large class nothing was ever done specifically to correct the issue, although the redesigned A4 frames didn't have the problem. It seems that Doncaster had evolved cheap repair methods and it was just accepted as a matter of course, but equally it might have been felt that, as the problem was due to flexing, strengthening the frames in one location may have simply caused cracks elsewhere.

 

I'm still interested to know why you think he transformed the V2's. the people who had to operate them don't agree. Only a small number were ever fitted with the Swindon blastpipe arrangement and the others were not changed until eventually two were fitted with an LMS style double chimney - again no improvement - and then five were fitted with the Kylchap, which did improve their performance significantly.

I have no intention to denigrate Ell's work, I don't know enough about it regarding non-LNER classes, but the LNER were already pretty wise about blastpipes and high superheat.

 

Yes, that is odd, as the A4 frame was already there as an example. I understand, and am open to correction, that frame replacement was normal LNER policy on an age / mileage basis, and if so replacement by a more robust structure would be neither more complicated nor expensive. Is the frame replacement policy based in fact or another myth I've heard?

 

The story I heard, I think in Cox is that when the V2s were first fitted with self cleaning smokeboxes the performance was woeful, but after the Swindon blast pipe mods it ended up at a higher level. This paper appears to be the relevant BR report.

http://users.fini.net/~bersano/english-anglais/BR-tests/BR_P&E_No8_LNER_V2.pdf

According the the tests max steaming rate was 24,000lb per hour without the self cleaning system, 14,000lbs per hour with the original setup, and 30,000lbs per hour after the modifications and with the self cleaning plates.

One might speculate that maintaning a steaming rate much above 24,000lbs per hour may not have appealed to the average crew - especially the fireman.

 

Depends how you raise the power output, I would argue. While I haven't run numbers, I think the higher boiler pressure (and in preservation main steam pipe improvements) are more to blame than improvements to the exhaust and valve events. If anything, the heavy compression at the end of the stroke with short travel valves notched up is worse than the free exhaust with long travel valves; it puts the chassis (between cylinder and driving axle) under greater stress while doing negative work. A relatively small increase in back pressure (due to an inefficient exhaust) will also manifest itself in a larger increase in peak compression.

I believe the A3 frames later in life had a lot of the lightening holes from the original A1 design removed; it wouldn't have solved the horn stay issue but it may well have reduced cracking in other areas. I would go off and do some research to check that, but then I'd be yawning through tomorrow morning at work. I suspect that the long travel A1s had no more trouble than the short travel examples, and the Kylchap didn't hurt compared to a single chimney A3. I remain very doubtful about the wisdom of higher pressure boilers, overbored cylinders and improved main steam pipes as carried by 4472 for a while in preservation, without the benefits of A4-type horn stays. Like you, I don't understand why the LNER (or later BR) didn't gradually upgrade them as the frames were overhauled, but I guess that until the Kylchaps were fitted late in the day they were largely viewed as yesterday's design and not worth the effort by Gresley's various successors.

 

These early posts were very kindly removed and given a seperate thread to concentrate on smokebox draughting. Can we please keep on topic and not discuss frame defects please!

Cheers,
Julian

 

Does somebody has a timeline of the Dean Goods front-end development? The reason why I am asking is that a lot of locomotives, King as an example but also the Jubilee, had chimneys with 1:6 or 1:7 taper. In his 1933 booklet Young showed with his tests that 1:14 was optimal for a single orifice unit and I am wondering whether this was quietly adopted by Ell c.s. as a consequence.
Kind regards
Jos Koopmans

 

Hi Jos,

A quick answer without any in depth research is as follows.

Courier has attached the Swindon Mechanics Institute lectures showing that by 1908 Churchward and his team had carried out sufficient experiments of their own to formulate a number of standard smokebox draughting arrangements to suit his standard classes and new boilers. Holcroft confirms this also, and was there at the time.

Instead of an internal converging taper from choke to chimney top, with the top being 1" less in ID, they added 1" or 2" to produce a diverging internal taper.

This seems to be irrespective of choke to chimney top length. So the ratio or angle of the taper differed, but at least it was now diverging.

For the older classes, Swindon boiler design went through a number of stages and when overhauled in the early 1900s would receive a belpaire firebox boiler replacing the round top varieties. These also had extended and larger diameter smokeboxes and often new chimneys - which would have been pinched from the most suitable 'standard' arrangement. The chimney diameter is noticeably larger at this stage with those locos fitted with (externally) parallel chimneys.

Obviously, the longer the chimney the smaller the internal divergent taper.

I have a reference that the cast iron tapered chimney was introduced on the Dean Goods in 1919. This is the same chimney on the preserved Dean Goods, and the same as that 'used' later by Sam Ell and which was also tested Swindon against the Ivatt class 2.

So 1919 is the short answer.

Cheers,
Julian

 

I've never seen anything. Doesn't mean it doesn't exist. Surely some major studies of the Dean Goods have been written, and someone here has a copy and can comment?

Unfortunately I'm a long way from York, but it would be interesting to go through the drawing registers and see what drawings were made, and then look at what survives.

Looking at the list of drawings at York published on the NRM web site I can see a cast iron tapered chimney introduced in 1919. However you'd need to inspect the drawing to see if any changes were made later. It occurs to me the inside could be modified without changing the outside. I think this is where inspection is unavoidable. Drawings most certainly contained later amendments, so from the register I see a Dean Goods blast pipe introduced in 1908, but I can't see if any of the critical dimensions were changed later.

There look to have been variant boilers drawn at reasonably frequent intervals. RCTS records variant superheated P class boilers in 1911, 1915, 1917, 1921 and 1924. Whether there would have been front end changes I don't know.

There's an arrangement of smokebox drawing from 1921, which is listed as superceding one that must have been drawn about 1913 by the number.

There was, as noted, a new class GA drawing in 1940.

 

Always thought the Darlington pattern chimney for the 2mt was an excellent thing to look at. Very NER styled and far better than as originally done.

 

Well indeed. But had the Kylchap been fitted in 1952 instead of fiddling with Dean Goods arrangements, then the railway may indeed have profited.

The original motivation for change was not to improve the performance of the engines but to allow the employment of self-cleaning smokeboxes, desirable because of the increased ash content of post - WW2 coal. The whole story is quite complex and a general comment without giving specific engine numbers is no help.

I can't find any figures that support the Clay position, nor is it clear which version of the modification he is talking about - there were at least four variations. The original 'improvement' was aimed at improving the poor performance of the first few fitted with SC gear, and was only minimally effective , which is why 60845 was sent by Cook to Swindon to get it 'sorted' with Ell's expertise. So, the engine on which he made a dramatic improvement had already been modified both with the self-cleaning smokebox and with the first revised blastpipe arrangement, and was performing poorly compared to unmodified members of the class.

In the end only a further 20 engines had the Ell arrangement. making 21 in all, and then the fitting of SC apparatus was discontinued. 4 others had different variations on the GW proportions, 2 of which did not have SC apparatus, and about 18 others had been fitted with the same arrangement that 60845 had when it was sent to Swindon. None of these were modified back to standard, but I suspect the self-cleaning apparatus was gradually 'lost' from the engines that were so fitted and the original levels of performance restored.

The interesting thing would have been a proper test between an unmodified V2, and an an 'Ell' V2, but this didn't happen. It would then have been possible to determine if the intended objective had been achieved, which was to allow the fitting of SC gear with no loss in performance compared to an unmodified V2.

 

.
Very difficult to make like for like comparisons here. Stanier Pacific Boilers and Grates were larger than Gresleys ? , potentially able to generate more steam when draughted just as badly/well.
As you rightly point out the success of even the most ideal draughting geometry does depend on its proportions being suitable to the Boiler and Grate it serves, as does its relationship to the exhaust steam Profile ( no of cylinders, velve events etc)
The point that gets passed over with the more complicated /multiple draughting systems is that more efficient vacuum creation enables less constriction at the blast pipe,=less back pressure = more useful work from the cylinders.. its not all just about smokebox vacuum.

 

So true espeshaly about the exhaust pipe work and cylinder design. If that's wrong then your wasting your time. The laws about streamline and gas flow that applies to internal combustion engines also applies to steam.

 

One of the mysteries is how Doncaster came to forget the value of the Kylchap. They knew it in 1938, they knew it in 1948, but seem to have then forgotten it by the next year. I understand the lack of communication between engineers and operators predates K J Cook.....

While Gresley worked well enough with his district staff, under Thompson, with the strain of the wartime conditions upon every man, relations deteriorated. Arthur Peppercorn was in office for little more than a year before being caught up in the maelstrom of nationalisation. He in turn was succeeded by J F Harrison.... Unfortunately it seems that relations between Harrison and L P Parker could hardly have been worse and liason between Doncaster and the Districts became very poor.

(P J Coster - Essays in Steam)

It's odd that such important information was in people's heads rather than a report.

 

Just to be clear I was thinking of the difference in ratios (such as gas area through tubes/grate area) rather than size. Quoting Coster again ...the aberrations in boiler design at Doncaster had been effectively masked by the use of Kylchap double blastpipes.

 

The piece I quoted was published in SLS journal in the 1960s - so there were plenty of people capable of contradicting his statement if it hadn't been true. I believe the Swindon tests showed the V2 w/o self cleaning (but with the non std reduced dia blastpipe) steamed almost as well as the Ell arrgt with self cleaning - so it is probably true that a std V2 w/o SC was the equal of the Swindonised one with SC - so probably they removed the grills and everyone was happy apart from the person who had to shovel out the smokebox.