RETURN TO INDEX

DONCASTER WORKS

 

LOCOMOTIVE DRAWING OFFICE 

 STIRLING DESIGNS

Stirling’s Arrival at Doncaster

Following Sturrock's notification to the Board of his impending retirement, Patrick Stirling, Locomotive Engineer of the Glasgow & South Western Railway (1853-66), was interviewed at a GNR Board meeting (1st December 1865) for the post of Works Manager.  At that time the Works was only repairing locomotives, so it would appear to be a curious downward move. During the next Board meeting (9th January 1866) the Chairman of the Locomotive Committee stated that “neither of the short listed candidates for Parker’s job matched the requirements” but he believed Stirling was “qualified by his attainments and references to fill the role of assistant locomotive engineer and to succeed Sturrock on his retirement”.  Stirling was to “report to Sturrock and run the workshops” thus filling both posts (from about March 1866).  At a further Board meeting (10th July) there was a surprise announcement that Sturrock would now retire on 30th September leaving Stirling to quickly find someone to succeed him as Works Manager.  John Shotton took over that post on 7th August and Stirling finally succeeded Sturrock on 1st October 1866.  Curiously later both Shotton and Stirling died in the same year, 1895.

Drawing Office

The usual practice, certainly in place in LNER days, was for draughtsmen to be allocated specific design features in which they then specialised for perhaps forty years.  In later years the draughtsmen recorded their own drawings in the registers adding their initials but originally it was probably the Chief Draughtsmen who vetted a drawing before registering.  There are the usual stories going the rounds of a component being designed and finished and finally the draughtsman came along and produced the drawing.  I can certainly vouch for this in one instance.  I once took the Pipe & Rod drawing (Q133) for the Peppercorn A1 Pacifics to Farnley Shed where some were in store and tried to correlate the pipe work in the cab with the drawing.  It was impossible.  The pipe work weaved around and then out of sight and bore no relation to what I assume had been intended according to the drawing.  I asked the Chief Draughtsman about it and he said threading the pipe work was a job for the experts.  When the first engine of a class was completed a young draughtsman would be then sent down to examine it and do his best to produce the Pipe & Rod drawing.  He agreed they were a fabrication but it kept the bosses happy.  An engine could not be built based solely on the LDO drawings and the Works Manager would have his own design staff to take care of minor matters as they cropped up in the works and did not warrant LDO intervention.  There was certainly a rivalry between the two distinct camps.  In particular the Pattern Shop would prepare patterns for the cast-iron cylinders which required complicated castings to be made with hollow parts for piston and valves and had to allow for shrinkage on cooling.  The castings were then machined to conform to the LDO drawing.  Strangely, the LDO prepared only one drawing for cylinder, valves, valve spindles, etc., between drawing M42 (25 May 1872) and M44 (28 August 1883), which was M43 (10 February 1876) the puny cylinders for Stirling’s smallest design, 0-4-2ST Nos. 501/02.  This gives the impression that for almost eleven years it had been more convenient to leave all the design work to the man who had to prepare the drawings for the patterns, perhaps even a draughtsman transferred from the LDO  There were other somewhat lengthy gaps in the LDO drawing lists such as from I36 (29 January 1884) to I37 (8 May 1889) in the series for smokeboxes, blast pipes and chimneys.  There is nothing concrete here but you have to be on your guard when possibly looking in vain for a drawing for a particular design feature.  It just may not exist.

Engine classifications

The well-known wheel notation devised by Frederick Methvan Whyte around 1900 (known as the Whyte system) will be used hereinafter.

Stirling inherited a variety of wheel arrangements, two of which disappeared in his time: 0-4-0 (1872) and 0-8-0T (1880) and he introduced no new ones.  His engines were segregated into goods and passenger according to their chief usage with a grey area called “luggage engines” or “Universal Class” which roughly corresponds to latter-day “mixed traffic”.  These were the 0-4-2s, which started off in the goods category and later finished up in the passenger category.  Eventually the relevant general Arrangement drawings P24, P31 in the goods engine series became Q71 and Q72 in the passenger engine series in 1907.  A better idea of their segregation is in their running numbers.

Running Numbers

Stirling inherited the segregation between passenger and goods engines as follows:
Passenger: 1-100, 200-99
Goods: 101-199, 300-99, 400-99

Stirling continued with this system giving “replacement” engines the numbers of the engines they replaced with costs being charged to revenue account.  “Replacement” usually meant engines entirely brand new but there were four engines, Nos. 10, 20, 42, 43, which whilst being official “rebuilds” must have used so much new material as to warrant them being given works numbers.  On occasion the Sturrock engine being replaced was too good to scrap so it was placed on the duplicate list and its number was given the suffix A.  The intention was for such engines to be scrapped as soon as convenient though in the case of No. 112A this meant almost twenty-one years later (and a tender used by it at the end of its life exists to this day).  The record was held by No. 245A from June 1879 to June 1905.  Those brand new engines that could be legitimately charged to capital account took vacant numbers in the fleet in the appropriate range for the category of goods or passenger, e.g. 280-99 for new 2-4-0s and 474-500 for new 0-6-0s and 0-6-0Ts.  Then a further range was added for goods engines: 501-620.  Thereafter blocks of numbers were allocated regardless of category, chiefly so as not to leave numbers blank when it was important not to exceed a certain number of authorised engines in capital stock.

There appears to have been a practice in the LDO whereby engines on order would be given numbers 1000 and upwards, on paper only, to assist in segregating drawings according to their intended use.  For example around 1884 there were two types of 2-2-2 being developed, the basic difference being the size of boiler.  It seems Stirling was being cautious and ordered two of each type to be built, intending later to decide which version to proceed with.  The two frame drawings for example were O32 (31/10/1884) for Nos. 1000 and 1001 and O33 (17/8/1885) for Nos. 1002 and 1003.  There was no intention of the engines bearing these numbers, the actual numbers would be allocated at the last minute depending on which Sturrock engines they were intended to replace.  In the event “1000 and 1001” became Nos. 238/32 and “1002 and 1003” became Nos. 234/29.  

0-4-2 Nos. 67 and 70 (built 1891) replaced two ancient 0-4-2s originally 2-2-2s ordered from Hawthorn’s by Benjamin Cubitt, the first Locomotive Superintendent, which then became 67A and 70A.  Later it was later discovered that the number of duplicate stock engines was in excess of the authorised number.  The new engines were renumbered 951/52 and 67A and 70A returned to capital stock.  This choice of a block starting at 951 (rather than 871) suggests the renumbering took place around January 1893 though Norman Groves in his “GREAT NORTHERN LOCOMOTIVE HISTORY” (Volume 2) quotes a date of January 1892.  Later it was found that 0-4-2 No. 359 (new 7/1893) duplicated a Sturrock 0-6-0 bearing that number and still in capital stock.  Presumably the duplicate stock already had its full quota so the new 0-4-2 was renumbered 957.  Norman Groves quotes the date of renumbering as January 1894 though at that date Nos. 955/56 had not yet been built (new 10/1894) and would have been available instead, assuming Groves’ date is correct.  The allocation of the 1001-10 block of numbers to 4-2-2s was out of date sequence.  It is possible that the allocation of 991-1000 (2-4-0s not built until 1894-95) had already been fixed before agreement was reached for the two new 4-2-2 Nos. 264/65 built in 1893 on the renewal account to be charged instead to capital in the 1894 building programme and re-numbered 1001/02.  This all suggests that capital list numbers were allocated well in advance and not deviated from.  This may well have been because certain parts, axles, etc., were stamped with running numbers during construction.

Drawing Registers

Drawings were registered sequentially within separate lists with a prefix letter to segregate them according to their need, latterly these were for example M cylinders, N boilers, O frames, P general arrangement (goods), Q general arrangement (passenger), R general arrangement (tenders), U axle-boxes, etc.  Almost certainly this particular system had been introduced about 1874.

Engine General Arrangement Drawings

It appears the Stirling carried on with the old system until about 1873, when a new series commenced in which P was for Goods and Q for Passenger engine general arrangements. These new categories continued through to the end, though the distinction between P and Q latterly became somewhat blurred.  Earlier arrangement drawings that were still needed (as at say 1873) were renumbered P1 or Q1 upwards as appropriate in date order.  Entries in the drawing register prior to the mid 1870s suggest they were written at the same time by the same hand.  In the course of time as a drawing was successively re-copied evidence of the earlier number was gradually lost.  But the earlier tinted general arrangement drawings on cartridge paper were too good to throw away and were kept.  Most were still valid but had been replaced by modern copies, the exception being 15A (dated 1858, see APPENDIX AA).  Another obsolete drawing would have been that for Sturrock's 4-2-2 No. 215.  This engine had been scrapped in 1869, so its general plan never acquired a Q number.  This drawing too should have survived for some time, but had disappeared by 1960.  It was the absence of an entry for this engine in the drawing list that was the clue to dating the register.  Although older drawings would disappear in the course of time, at least their references would remain in the registers.

General arrangement drawings were prepared for most of the classes, though not necessarily for all variations within a class.  The list is therefore incomplete as far as following the development of Stirling engines.  Drawings were though also produced for rebuilds of Sturrock engines, in particular around 1873-74, when by fitting them with modern boilers their life would be extended and be useful machines alongside his own engines.

 P series Stirling General Arrangement Drawings

 

P16 (3/1867) 0-6-0 Nos. 474-93, etc.
P17 (5/1867) 0-6-0ST Nos. 124/62, 392
P18 (12/1868) 0-6-0ST Nos. 396, etc.
P19 (9/1870) 0-6-0ST Nos. 395/98, etc.
P20 (11/1871) 0-6-0 Nos. 174, etc.
P21 (6/1872) Sturrock 0-6-0 Nos. 400-69 altered
P22 (7/1872) 0-6-0 Nos. 171, etc.
P23 (12/1873) 0-6-0ST Nos. 494-99, etc.,  plus 606-15 (with short wheelbase)

P- (23/7/1874) Sturrock 0-6-0ST No. 397 altered [not in drawing register]
P24 (7/1875) 0-4-2 Nos. 74, etc.  Later amended to Q71 in the passenger engine series
P25 (7/1875) 0-4-2 ditto, side elevation

P26 (5/1875 4/1878) Sturrock 0-6-0ST Nos. 111/34/39/40/44/49/55 altered
P27 (10/1879) 0-6-0 Nos. 640, etc.
P28 (3/1880) 0-6-0ST Nos. 634-38, etc.
P29 (12/1881) 0-6-0 Nos. 716-50, etc.
P30 (6/1882) 0-6-0ST Nos. 684-87, etc.
P31 (9/1882) 0-4-2 Nos. 105-10/12-15.  Later amended to Q79 in the passenger engine series

P32 (12/1882) 0-6-0 374, etc.

P33 (1/1883) 0-6-0ST Nos. 688-93, etc.

P34 (9/1891) 0-6-0ST Nos. 921-30, etc., outside elevation

P35 (1/1892) 0-6-0ST Nos. 921-30, etc.

P36 (1/1895) 0-6-0 Nos. 1021-30
P37 (11/1895) 0-6-0 Nos. 1031-45

 

 Q Series Stirling General Arrangement Drawings

 

Q15 (6/4/1867) Neilson 0-4-2T Nos. 270-74

Q16 (6/4/1867) Avonside 0-4-2T Nos. 275-79

Q17 (10/5/1867) 2-4-0 Nos. 280-99 mountings

Q18 (8/1867) 0-4-2 Nos. 18, etc.

Q19 (1/1868) 2-2-2 Nos. 4, etc.

Q20 (9/1868) 0-4-2WT Nos. 125-27, etc.

Q21 (2/1869) 2-4-0 Nos. 280-99

Q22 (3/1870) Hawthorn 0-4-2 Nos. 67/70 altered

Q23 (3/1870) 2-2-2 No. 92

Q24 (3/1870) 4-2-2 Nos. 1, etc.

Q25 (9/1871) Sharp 0-4-2WT No. 12 altered

Q26 (1/1872) 0-4-4WT Nos. 120, etc.

Q27 (5/1872) 0-4-2WT Nos. 10, 20, 42/43

Q28 (10/1872) 2-4-0

Q29 (7/1873) Hawthorn 2-2-2 Nos. 51, etc. altered

Q30 (6/8/1873) Longridge 2-2-2 Nos. 91, etc altered

Q31 (23/7/1873) Hawthorn 2-4-0 Nos. 223-28, altered

Q32 [blank] 2-2-2 Kitson, Stewart & Stephenson Nos. 229-40, altered

Q33 (13/10/1874) Fowler & YEC 2-2-2 Ns. 264-69, altered

Q34 (10/7/1874) Hawthorn Nos. 204-14, altered

Q35 (22/9/1874) Sharp Stewart 2-4-0 Nos. 251-60, altered    

Q36 (16/12/1874) Wilson 2-4-0 No. 79, altered

Q37 (8/1874) 4-2-2 Nos. 1, etc

Q38 [blank] 4-4-2 Nos. 1, etc, cross-section

Q39 [blank in drawing register]

Q40 (2/1876) 0-4-2ST Nos. 501, etc

Q1 (8/3/1878) 0-4-4T Nos. 621, etc

Q2 (3/5/1879) Neilson 0-4-2WT Nos. 270-74, altered

Q3 (10/2/1880) 2-4-0

Q4 (22/6/1880) 0-4-4WT Nos. 652, etc

Q5 (7/1880) 4-2-2 Nos. 662, etc, outside elevation

Q6 (30/7/1880) 4-2-2 Nos. 662, etc

Q7 (3/1881) 0-4-4T Nos. 658, etc

Q12 (2/1881) 0-4-4T Nos. 658, etc, outside elevation

Q41 (12/1881) 2-4-0 outside elevation

Q42 (9/1/1883) 0-4-4T Nos. 694, etc

Q43 (31/10/1884) 2-2-2 Nos. 232/38

Q44 (1/1886 1/7/1885) 2-2-2 Nos. 232/38, outside elevation

Q45 (1/1886) 2-2-2 Nos. 229, etc

Q46 (10/1887) 2-4-0

Q47 (1/1888) 4-2-2 Nos. 771, etc

Q48 (20/2/1888) 4-4-2 Nos. 771, etc, outside elevation

Q49 (20/5/1889) 0-4-4T Nos. 766, etc

Q50 (4/1892) 0-4-4T Nos. 931-40

Q51 (12/1894) 4-2-2 Nos. 1003-08

Q52 (9/1895) Avonside 0-4-4T No. 245, altered

Doncaster Works References

Geo. Fredk. Bird in his monumental work “The Locomotives of the Great Northern Railway” quoted works reference numbers.  Almost certainly these were references to the drawers where the relevant drawings were kept for Doncaster-built engines, perhaps in the Works Manager's Drawing Office (WMDO).  As a parallel, Kitson & Co Ltd used the term Drawer No. on its drawings.  At Doncaster, drawings for contract engines were kept separate and Bird ignored these.  These works references (with rare exceptions) did not appear on the drawings prepared in the LDO.  One such example was in the Boiler Order book, where the boilers in BO500 were "available for 6' 6" H or S".  In addition I found a mention of a works reference in a book kept in the LDO: from memory the Materials Ordered book.  The works references (A, B, C, etc.) were allocated on receipt of drawings for new batches of engines being built at Doncaster.  It is likely the references and thus the drawers themselves were allocated and maintained in order of receipt.  There could have been a key to the references in book form with brief descriptions of contents.  It is unlikely the drawers themselves would have been arranged in this order as it could have meant a certain amount of rearranging each time a new sub-division (e.g. E2, F2, G2 and G3) was introduced.  In fact it appears that Bird wrote about the Doncaster-built engines in the order of the allocation of these works references as if he was working from a list of these drawers.  He was thus conveniently dealing with the engines in approximate order of introduction of each new class or sub-division.

Stirling Doncaster Works References

 

Reference

Introduced

Type

Wheels

First Engine

Total

A

1868

0-4-2

5ft-7in

18

46

A2

1874

ditto

ditto

74

25

A3

1882

ditto

ditto

103

12

A4

1887

ditto

ditto

10

21

B

1868

2-2-2

7ft-0in

6

12

C

1868

0-6-0ST

various

392

8

D

1868

0-4-2WT

5ft-7in

126

13

E

1869

0-6-0

5ft-2in

369

17

E2

1874

ditto

ditto

372

36

E3

1886

ditto

ditto

791

72

F

1870

2-2-2

7ft-6in

92

1

G

1870

4-2-2

8ft-1in

1

37

G2

1884

ditto

ditto

771

10

G3

1894

ditto

ditto

1003

6

H

1871

2-4-0

6ft-6in

261

2

H2

1874

ditto

ditto

86

19

H3

1881

ditto

ditto

208

9

H4

1884

ditto

ditto

751

18

H5

1888

ditto

ditto

210

56

I

1872

0-6-0

5ft-2in

174

6

J

1872

0-6-0ST

4ft-0in

471

2

J2

1874

ditto

ditto

136

6

J3

1882

ditto

ditto

684

4

J4

1890

ditto

ditto

134

4

K

1872

0-4-4BT

5ft-6in

120

22

K2

1878

ditto

various

621

26

L

1873

0-4-2WT

5ft-6in

43

4

M

1874

0-6-0ST

4ft-8in

494

35

M2

1881

ditto

ditto

672

43

M3

1891

ditto

ditto

854

7

M4

1892

ditto

ditto

921

10

M5

1893

ditto

ditto

961

20

N

1876

0-4-2ST

5ft-8in

501

4

N2

1878

ditto

ditto

631

2

O

1881

0-4-4T

5ft-0in

658

16

P

1883

0-6-0

4ft-8in

374

8

P2

(1896)

ditto

ditto

1021

10

Q

1885

2-2-2

7ft-6in

238

2

Q2

1886

ditto

ditto

234

10

Q3

1892

ditto

ditto

871

11

R

1889

0-4-4T

5ft-6in

766

25

R2

1895

ditto

Do

941

4

 The list was added to under Ivatt.

Engine Frame

The frame plate thickness was variously 11/8in, 113/16in and 1¼in but the lack of reliable source material has made it impossible to follow the changes with any degree of reliability.  The frames were originally made of wrought iron though because of manufacturing methods of the period these dimensions were only nominal.  As a result plates were probably specified thicker than they needed to have been, to ensure they would not fracture. But this also meant they could have been thicker in places where clearance tolerances were critical.  As far as the 4-2-2s were concerned the frames were originally 1¼in thick.  Then commencing with No. 1 (rebuilt December 1880) and 662 (new February 1881) their frames tapered in at the front from 4ft-0½in to 3ft-8¾in at the buffer-beam, to give more clearance for the leading bogie wheels on curves.  Drawing Q47 (January 1888), also redrawn for “The Engineer” and used for the illustration on p. 79 of “The Stirlngles” gives this dimension as 3ft-11¼in.  This is in fact the distance across (not between) the frames at this pointing Si.  Following the introduction of mild steel and its application by Stirling for boilers and frames commencing with No. 773 (August 1885), the thickness was perhaps reduced to 13/16in.  Drawing Q47 showed the thickness as 1¼in originally but the figure appears to have been altered later to 13/16in.  This was a general arrangement drawing, so may not have been accurate with all figures.  It also seems likely that the frames thickness was reduced still further to 11/8in for the 1003 Class though admittedly this dimension is shown was 1¼in on their general arrangement drawing, Q51 (December 1894).  According to this drawing their frames were parallel throughout. Perhaps it was considered that the present quality of materials had eliminated the need for extra clearance at the front.

Most main frames were spaced 4ft-1½in apart although those on the 4-2-2s were 1in closer together to provide extra clearance for the large driving wheels.  On the 2-2-2s, 2-4-0s and 0-4-2WTs outside frames were also provided with the multiple function of supporting the running plate (which ran the full length of the engine), the cab footsteps and the axle boxes for the smaller carrying wheels.  This frame was attached to the running plate by an angle iron which ran for the full length.  On all other Stirling types including the 8-foot Singles this outer frame was much shallower, supporting only the running plate and the footsteps.  However it too was affixed to an angle iron for the complete length.  The outside frame system employed by Sturrock (a former Swindon man) was different as he followed early G.W.R. practice with “proper sandwich” frames which also held the outside axle boxes on the driving and coupled wheels. This meant that the side (or coupling) rods were outside this frame whereas with the Stirling system the rods were much nearer the wheels, not least due to the inside axle boxes. 

The boiler was rigidly anchored to the frames at the front by the cylinders, whilst at the rear the firebox was supported on expansion brackets which rested freely on the frames and were able to move.  Thus as the boiler heated and expanded the firebox was able to compensate for this by moving slightly.  The clearance problem cannot have been too serious for action not to have been taken sooner as the class had been in service for ten years and for the remedy not to have been applied to the 1003 Class for their frames were parallel throughout.  The dates the frames were replaced on the earlier 4-2-2s (which took just over ten years to complete) are the dates they were recorded as having been rebuilt.

Boilers

By 1866 boiler design had settled down though with the multiplicity of contractor-built designs on the GNR  Stirling had to select what he considered to be the best options which can be summarised as follows.  The boiler barrel comprised three concentric rings, narrowest at the front.  Shaped like a dome, a mud-drum (“mud-collector”) was provided at the bottom of the middle ring, shaped like a dome, with an inspection cover for the removal sludge and loose scale.  From about 1880 to 1887, the injector water feed for the Single-wheelers only was fed into the side of the mud-drum.  Steam was collected in the upper part of the boiler through a large number of holes in the upper surface of the collecting pipe running from the firebox backplate to the smokebox tubeplate with the regulator valve in the smokebox.  The safety valves on the firebox top were encased in a polished brass “trumpet” casing.  At the first these were two Salter spring valves, later changed after 1875 to the Ramsbottom type which was more tamper-proof but needed a larger “trumpet”.  Ramsbottom valves continued into the Gresley regime but the brass “trumpet” was long gone.

The diameter over the largest (rear) ring of his original boilers was 4ft-0½in and the barrel plates were ½in thick.  Originally boilers were made of wrought iron but commencing with 4-2-2 No. 773 (built August 1885) all new boilers were made of mild steel.  The boiler diameter was increased to 4ft-2in (drawing N41 dated 11th July 1879) for the 4-2-2s, its size being limited by the distance between the driving wheels.  For general use the diameter was 4ft-2½in (N43, dated September 1881) and finally 4ft-5in (N55, 1886) for the 120 Class 0-4-4BTs.  The N43 version was gradually fitted to the earlier tender engines (including those of Sturrock) with just three Stirling engines still to be dealt with at the time of his death (0-6-0 Nos. 481/82/83).  Sturrock 0-6-0 No. 450, one of the batch originally ordered with steam tenders, did not acquire one of these boilers until 1894.  Actually the 4ft-5in diam. of N55 was not new, a similar diameter boiler (N36, 26 June 1872) having been built for the six 174 Class mineral 0-6-0.  Another type (N40, 21 January 1878) had been used to reboiler the Sturrock 0-4-2Ts and earlier still Sturrock had used this diameter for his 400-19 Series 0-6-0s.  Interestingly a similar diameter boiler (N42, 18 August 1879) had been proposed for the 2-4-0s but the plans came to naught.  Finally, Stirling was working on two new 4ft-5in diam. boiler designs.  The first (N65, 15th January 1895) was for new 0-6-0s (allocated Nos. 1021-26, amended later to 1021-30) having 4ft-8in diam. wheels.  The second (N66, November 1895) was intended for new “standard” 0-6-0s (numbers not yet allocated) having 5ft diam. wheels.

As mentioned earlier the boiler could only be attached to the main frames at the front end, because of unequal expansions of boiler and frames.  An expansion bracket was riveted to the firebox side and this rested, or rather slid, on the frame tops.  Some boiler types could be fitted to several engines classes, though according to the level of the frames and the pitch of the boiler the position of the expansion bracket had to be raised or lowered.  In addition, because of the 1in closer setting of the main frames on the 4-2-2s there were two widths of firebox for the Singles boilers; narrow width for 4-2-2s (No. 1 Class) and standard width for the otherwise equivalent 2-2-2s (234 Class).  As boilers were not interchanged in Stirling’s day this difference created no problem.  Under Ivatt however, a No. 1 Class boiler could be fitted to a 234 Class engine but not vice versa.  In the former case it is probable that wider expansion brackets had to be used to bridge the ½in gap on each side.  No drawing was provided by the LDO and it would be simply left to the Works to sort it out for themselves.

Safety Valves

Engines had Salter spring-balance safety valves, with a horizontal lever that passed into the cab for the driver to operate at his discretion.  The valve was out of sight under a brass “trumpet”.  From 1875 (drawing W13, 5/6/1875) Ramsbottom “duplex” safety-valves (which were more tamper-proof) were fitted instead.  These had two valves in line pitched a little higher so that the driver’s lever had a reverse curve in it (known as a “cow’s-tail”).  The brass trumpet was noticeably larger.

Cabs

The crew needed to keep a watchful eye to the rear to ensure the train was still there and look out for signals from the guard, hence they needed a clear view.  For protection against rain and wind, before Stirling, there was a vertical plate surrounding the back of the firebox, bent over at the top and with two small “spectacle” windows at each side.  This was Spartan protection but it ensured the crew remained alert at all times.  Even this protection was not always provided if thought unnecessary, such as on the two 0-8-0 Tank engines, Nos. 472/73, intended for the London “underground”. 

When Stirling took over in 1866 his first three designs incorporated better crew protection.  These were 2-4-0 Nos. 280-99 and 0-6-0 Nos. 474-93 from contractors delivered in 1867-68 as well as the first six 7ft 2-2-2 built at Doncaster which had substantial “cabs”, with 18in. diam. windows at the side resembling portholes on ships.  Stirling had previously designed in 1865 the 45 Class 2-2-2 for the G&SWR., also with 7ft wheels and cabs having porthole windows.  Doncaster Works started construction towards the end of 1867 and their first three engines were 0-4-2 Nos. 18, 23, 40, also with “porthole” cabs.  As the “porthole” cab first appeared in the month of October it would be expected be appreciated, though during the summer months they could have become uncomfortable hot.  Instead they were unpopular with crews, perhaps because it was difficult looking back along the side of the train, requiring them to stand precariously in the swaying cab entrance.  The answer was the wrap-over cab with cut-out sides allowing the crew to lean out safely.  The shape of which is believed to have been suggested by Stirling’s sister.  These were provided for all subsequent Stirling tender engines with no exceptions and the earlier engines were soon brought into line.  The length of the cut-out varied according to the space available, i.e. overall cab length.  Some time during the 1880s the vacuum ejector exhaust pipe was relocated in view outside the boiler and slightly wider cabs were provided.  The running plate alongside the cab had restricted width so after a while a horizontal handrail was provided.  Stirling’s final 4-2-2s, Nos. 1003-08, had their cab roofs extended back a small amount whilst preserving the cut-out at waist height.  This gave a little extra protection from the rain.  Tank engines were a different matter.  A simple wrap-over was provided at first but better protection was provided afterwards for those on passenger duties with the cabs almost entirely enclosed.  Later shunting engines built from 1892 had enclosed cabs.

Cylinders and Valves

The general practice under Patrick Stirling was to arrange two cylinders between the frames, with connecting rods to the cranks on the driving axle.  Adjacent to each cylinder was a steam chest (actually in a single casting), with a vertical flat slide-valve admitting steam to the cylinder and exhausting it at the end of the piston stroke.  The slide-valve was kept in close contact with the ports by steam pressure surrounding the valve, though this inevitably resulted in the gradual wear of the surfaces in contact.  The valve was driven by rodding connected to two eccentrics on the driving axle, one each for forward and reverse motion.  Thus each driving axle had two cranks and four eccentrics.  The cranks were 90 degrees apart on the axle, so that when the piston in one cylinder was at the end of its stroke (twice per axle revolution) the piston in the other cylinder was at mid-point (again twice per axle revolution).  Stephenson link motion was employed.

The first exception to the general rule was Stirling’s first 4-2-2, No. 1, new in April 1870, with two outside cylinders.  Each cylinder and steam chest was cast together, with the frame straddling the casting.  This entailed lengthy live and exhaust steam passages to connect with the cylinder, hitherto adjacent.  The slide-valves were made to drawing M32 (23 December 1869), similar to those intended to be fitted around this time to Sturrock 0-6-0s Nos. 400-19, though there is no record of which engines were actually fitted other than No. 411 most likely in late 1867.  The valve was the balanced type, presumably held in place against the port faces by springs instead of steam pressure.  Its application was not a success and as early as 4th August 1870 a fresh drawing was made for ordinary slide-valves.  This was marked off for Nos. 8 and 33, after which No. 1 was brought into line and then Stirling never again fitted balanced slide-valves.  Eventually there were forty-seven 4-2-2s in the No. 1 class constructed over a period of twenty-three years and having the same basic arrangement of cylinders and valves (though there were other significant differences, mentioned later).  The 1003 Class which appeared in 1894-95 had a new arrangement of cylinders and valves. 

Maximum valve travel on the No. 1 Class was 4½in. though the late Kenneth Leech noted that “short travel was a practical necessity with unbalanced flat slide-valves.  Even with the travel cut to a minimum quite a large amount of the engine horsepower was absorbed in driving them, especially with the regulator wide open” and consequently “most engines with flat slide-valves tended to be driven with only partially opened regulators, thereby reducing appreciably the steam chest pressure and with it the friction of the slide-valves”.  The steam chests on the final batch of Stirling 2-2-2s Nos. 229-40, 871-80, 981 had larger steam ports 16in wide instead of 14in and larger exhaust ports 4in. wide instead of 3½in. when compared with the 4-2-2s.  This probably made the 2-2-2s superior to the 4-2-2s and as Hitching driver W Freestone once commented to Kenneth Leech: “The 879 Class are certainly the fastest engines the Great Northern ever had”.  The steam chests on the final batch of Stirling 4-2-2s Nos. 1003-08 differed in having their flat slide-valves placed 16 degrees from the vertical, so that they fell away from the port faces when coasting.  This also meant that the connecting live and exhaust steam passages were shorter.  The valve ports were the same larger sizes as in the latest 2-2-2s.  The maximum valve travel was increased slightly to 4¾in. which presumably helped when starting away with the heavier loads of the day.  In 1930 Gresley commented that the 1003 Class had no boiler to speak of but steamed alright.  Darlington Drawing Office accordingly remodelled the blast arrangement on Class W1 No. 10000 on similar lines, including the 4¾in. diam. blast-pipe top.  The late O S Nock in his “The Locomotives of Sir Nigel Gresley” stated that No. 10000 performed best in this modified condition.

Axle box side play

As an engine travels over rough track, or sways from side to side, stresses are produced in the frames which can be compensated for to a certain extent by allowing lateral side-play of the axle journals within the axle boxes (no more than say ½in. each way) and vertical-play by allowing the axle boxes to ride up and down in the horn guides with spring control.  Lateral side-play has to be minimised in the driving wheels because of its effect on the valve gear and at the extreme front end there is the closeness of the frames to consider.  But where more lateral side-play was needed at the front end the answer was a separate 2-wheel pony truck or 4-wheel bogie truck so that it was not just the journals within the axle boxes that moved, it was the whole truck.  If there were trailing wheels under the cab, extra lateral side-play was unnecessary because of the steadying effect of the tender.  Vertical-play was never a problem, the suspension springs could be adjusted as necessary.  Stirling’s 2-2-2s had 7/16in. lateral side-play for the leading wheel axle boxes but none for the trailing wheels.  Their absence from the trailing wheels ensured the rear four wheels formed a rigid wheelbase, avoiding unnecessary side-way movement on straight track   His 4-2-2s had some lateral side-play for the trailing wheels but none at all for the bogie, other than its pivotal action on curves.  In their case the leading six wheels effectively acted as a rigid wheelbase on straight track.

Crank axles

The majority of Stirling engines had two cylinders with the connecting rods operating the two cranks on the driving axle, thus if the piston stroke in the cylinder was 24in., then the “big ends” of the connecting rods encircled the centre of axle with a radius of 12in.  The “crank-axle” was thus built up of a number of pieces and there were frequent fractures in the early days.  The two cranks were arranged 90 degrees apart to ensure that when one piston was at the end of its stroke in the cylinder the other piston was at mid-point.  Because of this 90 degree shift the driving axles were not reversible so care had to be taken when erecting an engine and after periodic tyre turning.  As an aid, axles were stamped L at one end and R at the other (in addition to the engine number which identified the axle when in Works).  In addition to the cranks there were four eccentrics on the driving axle for the Stephenson valve gear, two for each steam chest and one each for forward running and reverse.  The 2-2-2s had the one driving axle with cranks and eccentrics.  On other inside-cylinder engines there was one crank-axle plus eccentrics and either one or two plain axles according to the wheel arrangement.  However the 4-2-2s had outside cylinders and therefore outside cranks on their driving wheels.  Only the four eccentrics were on the axle between the wheels.  As the connecting rods are in view, it can be seen on No. 1 today that the right side crank leads the left side crank by 90 degrees in forward gear.  Also, No. 1’s present axle is stamped 778, denoting that it started life with that engine.

Driving wheels

Coupled and driving wheels originally had laminated bearing springs.  The 2-2-2s and 4-2-2s from 1886, had Timmis helical springs for their driving wheels, intended to avoid the internal friction with laminated springs.  It was left to Ivatt to extend their use to the driving wheels of other classes.  The translation of reciprocal motion in the cylinders to rotary motion at the driving-axle by means of the connecting rods necessitated balancing the mass of the cranks and eccentrics.  The need for balancing was well-known, for example when Sturrock ordered auxiliary steam tenders for 0-6-0 Nos. 400-19, he stipulated that the wheels were “to be properly balanced with weights”.  In the event, only the tender driving wheels were fitted with balance weights.  As far as the Stirling engines were concerned, on Single-wheelers with inside cylinders (i.e. 2-2-2s) this meant balancing the crank-axle and inside eccentrics.  On Single-wheelers with outside cylinders (i.e. 4-2-2s) this meant balancing the outside connecting-rod crank and inside eccentrics, though in practice the mass of the inside eccentrics was apparently ignored and only the outside crank was balanced.  On coupled-wheel engines with inside cylinders (e.g. 2-4-0s) this meant balancing the mass of the crank-axle, inside eccentrics and coupling rods.  There were no coupled-wheel engines with outside cylinders.  Where necessary the balance weights were fitted to the wheels just inside the rim.  The actual position had to be calculated and could be different on opposite sides of the engine as the connecting rod cranks were 90 degrees out of phase.  Also the eccentrics were 90 degrees out of phase with their associated connecting rod crank.  On coupled-wheel engines (which all had inside cylinders) the weights were usually different on each axle, the compensation for the inside eccentrics being concentrated on the driving axle only.  On some engines balancing was ignored where there seemed little point, for example with shunting engines.

The construction of wrought-iron wheels was arduous, labour-intensive and expensive, so much so that when Sturrock’s 4-2-2 No. 215 was taken out of service in October 1869 its flangeless 7ft.-6in. diam. wheels were too good to scrap.  New flanged tyres were fitted to the wheel centres and the wheels were used as the basis for a new 2-2-2, given the number 92.  Surprisingly few drawings were needed: frames O14 (January 1870), general arrangement Q23 (March 1870), connecting rod S12 (1869).  In particular, there was no new drawing (K series) for the valve motion.  Later a fresh drawing T61 (17 February 1883) was prepared for new wheels for No. 92.  The date is somewhat indistinct but it is certainly not the December 1882 given in “The Stirling Singles”.  It is doubtful if the replacement wheels had a new steel crank-axle, as claimed in “GREAT NORTHERN LCOMOTIVE HISTORY” Volume 2.  The word “steel” is not mentioned against T61 in the drawing index.  The first drawing specifically noted as “steel” was T66 (24 October 1885) for 4ft. diam. steel wheels for 3000 gallon tenders. This was immediately followed by T67 (November 1885) for 8ft diam. steel wheels and axle for passenger engines.

No. 775 (new January 1887) was the first 8-footer (and almost certainly the first GNR engine of any class) with each wheel complete with spokes and balance weight in one steel casting together with a steel axle.  Incidentally the diameter of the driving wheels on the 7ft. Singles was 7ft.-1in., on the 7ft.-6in' Singles it was 7ft.-7in', whilst on the 8-footers it was 8ft.-1in..  These measurements are theoretical, because it all depended on actual tyre thickness and eventual wear.

Bogies

Stirling provided his 4-2-2s with bogies to support the weight at the front end and guide the engine around curves.  The bogie drawing was O13 (24 December 1869) replaced (for the frame anyway) by O17 (April 1872).  The diameter of the wheels was a nominal 3ft.-11in.  The centre of the bogie was in line with the chimney but the bogie pivot was placed 3in behind this centre line which Stirling said had the effect of “laying down the road for the driving wheels” though it also meant that the lateral swing of the front wheels was greater than that of the rear pair.  The leading pair of wheels could have scraped the main frames so in many cases these latter tapered in slightly ahead of the cylinders (see earlier under “Frames”).  The thickness of the bogie frame plates was gradually reduced, eventually to 1in. on the 1003 Class 4-2-2s.  The wheels had laminated springs above the axle boxes but no spring side control.  There were other minor changes such as spring length reduced by 2in. from 1884. Surprisingly Stirling did not fit helical springs to his bogies and this improvement was left to his successor to introduce.  The 120 Class 0-4-4T engines introduced in October 1872 also had bogies, with 3ft nominal diameter wheels spaced 5ft. apart.  The frame drawing (O16) is dated 5th December 1872 in the drawing register but the year is believed to be a copying error as the next drawing in the list is O17 (April 1872).  Entries on this page are in the same handwriting and give the impression of having been brought forward from an earlier book.  It is almost certain that O16 was drawn in December 1871, just preceding the general arrangement drawing for these engines Q26 (January 1872).

Trailing Carrying Wheels

Several Stirling classes had trailing carrying wheels and the method of springing varied according to the space that was available.  The first engines to appear from Doncaster Works in 1868 were his 18 Class 0-4-2 with 3ft.-6in. diam. trailing wheels.  Two wrought-iron troughs were arranged transversely below the cab.  The lower trough was seated at each end to the top of an axle box.  The top trough was inverted and secured to the frames and drag-box. Between the troughs were four volute springs, two at each side.  The springing was adequate to deal with track irregularities, but probable made for rough riding.  Next appeared the No. 6 Class 2-2-2, with laminated springs above the axle boxes out of sight between the sandwich frames.  Then followed the 126 Class 0-4-2WT, with outside axle boxes and laminated springs over them in full view above the running plate, in an excellent position for maintenance. 

After a short interval the No. 1 Class 4-2-2 appeared in 1870, with their 4ft. diam. trailing carrying wheels.  By then you would imagine that the merits of the different methods of springing would have been assessed.  Their arrangement followed that of the 0-4-2s, except for having three volute springs at each side instead of two.

In 1876 a change was made to the 4-2-2 design commencing with No. 221. The trailing wheels were 6in. larger in diameter to accommodate laminated springs below the axle boxes.  All the earlier engines of this class were brought into line.  The 103 Series 0-4-2s which appeared in 1882 also had 6in. larger diameter wheels and laminated springs below their axle boxes.  The improvement was not considered sufficient to justify the earlier engines (in excess of 100) being brought into line.

Reversing gear

The reversing lever in the cab was used to change direction, pushing it to go forward and pulling it to reverse.  The lever could be left in any intermediate point (i.e. notch) in the quadrant thus altering the point in the cylinder piston cycle at which steam supply was cut-off, hence the term “notching up” to (say) 25 per-cent cut-off.  Whilst it would be usual to start the train from rest in full gear (usually limited to say 65 per-cent cut-off) it would be quickly notched up to (say) 45 per cent cut-off.  65 per-cent seems to have been the ideal figure and it was not until the heavy loadings encountered during the 1939-45 War that it became a noticeable limitation.  From 1946 the Gresley Pacifics gradually had the limit increased to 75 per-cent.  Screw-operated reversing gear had limited use, in particular for passenger tank engines turning around at destinations but on long distance working it would be rarely used. 

According to the LDO drawing indexes, Stirling fitted this gear to the following engines:

0-6-0 with 19in by 28in cylinders (174 Class).  Drawing K23 (15 August 1871) referred.
0-4-2WT Nos. 10, 20, 42, 43, ostensibly rebuilds of old “Sharpies”.  Drawing K27 (13 November 1872) was made six months before these engines started to appear.

0-4-2BT Nos. 504-07 and presumably later engines down to 657.  Drawing K28 (16 June 1873) referred.

0-4-4T Nos. 658-61 and presumably later engines 682/83.  Drawing K37 (22 February 1881) referred.

0-6-0 with 4ft-6in diam. wheels (374 Series).  Drawing K43 (January 1883) referred.

0-4-4T Nos. 694-98 with condensing gear, and presumably later engines 761-65.  Drawing K46 (29 August 1883) referred

4-2-2 Nos. 1003-08, with their higher boiler pressure and larger slide-valves.  Drawing K54 (7 May 1894) referred.

0-6-0 with 4ft-8in diam. wheels (Nos. 1021-26).  Drawing K57 (20 December 1894) was made almost a year before Stirling’s death.  This particular order (Nos. 1027-30 added later) was completed early in 1896 but they may be considered as being Stirling products.

Brakes

Tender engines had no engine brakes, only the hand brake on the tender and it was not until 1875 that the matter was taken seriously by the GNR when 7ft Single-wheeler No. 55 was fitted with Smith's system of vacuum brake.  This was operated by the driver wishing to reduce speed or stop creating a vacuum to the train pipe which applied the engine and carriage brakes.  It was a "fail unsafe" system because if the train became divided the brakes could not be applied.  It was perhaps better than nothing and the subsequent tests on the Midland line at Newark were satisfactory.  Nine suburban tank engines were fitted with vacuum brakes later that year including four new engines (Nos. 528-31) and five existing engines (Nos. 55, 120, 505/11/15) for suburban in the London area.  By 1879 the majority of passenger engines had been fitted and work carried on steadily fitting the remaining 0-4-2s.  The fitting of vacuum brakes when new to the 0-6-0 goods engines commenced in 1880.  Further existing engines (mostly 0-4-2s) continued to be fitted.  The remaining 0-4-2s were fitted during 1881-83, also one Sturrock 0-4-2T and six Stirling passenger 0-4-2STs thus completing the fitting of the simple brake system to passenger and mixed traffic engines.

During 1883 a start was made fitting vacuum brakes to older 0-6-0 goods engines.  Several existing Stirling 0-6-0s were fitted in 1883-87 but work ceased upon the introduction of the automatic type.  Thus the majority were never fitted with the simple type.  The automatic brake was devised as a "fail safe" system.  Brakes were applied to engine, tender and carriages in the event of the driver or guard applying the vacuum brake, a passenger operating the communication cord, or a train becoming divided.  The system was introduced in 1884 (according to the Repairs Volumes) on 2-4-0 No. 51 (ex-works 30/8/1884) presumably for trial purposes.  Two new 2-4-0s built shortly afterwards (late 1884), Nos. 206 and 209, had the equipment when new.  Several passenger engines built in 1885 also had the equipment when new: 2-2-2 (Nos. 238 and 232), 4-2-2 (Nos. 772/73/74), 2-4-0 (Nos. 753/54).  Further passenger engines built in 1886 were similarly fitted: 2-2-2 (Nos. 234/90), 2-4-0 (Nos. 211/17/24/28/16/25, 755/56).

It had been intended to similarly equip the ten new 0-6-0s built in 1886 (Nos. 791-800), for which purpose the frames at the rear were lengthened to accommodate the 21in diam. brake cylinder but no doubt it was realised that the new equipment would be better utilised for the time being on passenger engines.  At the very least the simple vacuum brake system provided power brakes for the engine and tender.  Automatic brakes were eventually fitted to these particular engines from 1889.  Further new passenger tender engines built in 1887 were similarly fitted: 2-2-2 (Nos. 237/33/36/39), 2-4-0 (Nos. 757-60), 4-2-2 (Nos. 775-78), 0-4-2 (possibly Nos. 10, 12, though their histories make no mention of brakes).  Also new 0-6-0 goods engines (Nos. 322/07, 199, 320, 176/83, 389, 147), though the histories of Nos. 322/07 simply say "vacuum brakes new" without specifying the type but there is no subsequent reference to a brake type change.

 

Further new passenger engines (tender and tank) built from 1888 were similarly fitted, as were 0-6-0s and 0-4-2s.  However 2-2-2 No. 231 was shown as having the simple system when new (9/2/1888) being converted to automatic a year later (7/1889).  New 0-6-0STs were fitted from 1890.  From mid-1887 the system was fitted as soon as practical to existing passenger then goods engines.  Not all engines were fitted, e.g. existing shunting engines were not all equipped with some fitted for the first time in Gresley's time.  Not all dates have been recorded in the Repairs Volumes but in some cases it is not clear if this was a clerical omission or the engine concerned was never fitted.  Where there is a date for the simple system but not for the automatic system this is clearly a clerical omission.  There is also the problem with the engines withdrawn prior to about 1901 for which there are no surviving repairs records. 

Vacuum Brake Cylinders – Automatic System

The situation regarding the position of the brake cylinders was complicated. On tender engines the standard depth 21in diam. cylinder could be located under the cab, though sometimes set at an angle. Difficulties were encountered on tenders and in general the methods employed were as follows:

On tenders with flat bottoms and 3ft.-6in. wheels.  Existing tenders which still had their draw-bar springs at the back of the tender had these repositioned to the front in the manner. of those tenders with well tanks.  A shallow 21in. brake cylinder was located under the sole bar immediately behind the draw-bar spring.  The vertical piston of the brake cylinder just cleared the leading axle.

On tenders with well tanks and 4ft. wheels whose axles were therefore 3in. higher.  A shallow 21in. brake cylinder was tried out, with reduced clearance, so it then became the practice to fit smaller diameter twin-cylinders instead, side by side. 

Engine – Tender Intermediate Coupling

The surviving evidence is based on general arrangement drawings which are notorious for not being entirely accurate, containing obsolete features and may have been updated over the years to show a more up-to-date arrangement.

The earliest tender drawing available (R8, October 1866) shows a screw coupling, held in tension at the tender end by a 3ft. long horizontal laminated spring, and at the engine end secured with an eye-bolt under the “fall-plate”.  Alongside the coupling were two safety single-link chains, secured at each end under the tender and engine fall-plates, with a little slack permitted for the eye-bolts.  The centres of the safety-links were 2ft.-6in. apart.  In addition there was a pair of sprung buffers, placed at 3ft.-10in. centres, i.e. exactly lining up with the centre line through the inner oak beam of the tender frames, where the shock could be absorbed.  The buffers pressed against the engine buffer plate, strengthened at these points by extra plating. 

A later drawing (R12, August 1872) showed a thinner inner oak beam.  The centres of the buffers were still 3ft.-10in. apart and central with the beams.  The safety chain now had three links, the middle one being short, which was an innovation.

Another drawing (R13, February 1879) showed the latest arrangement of frame with thinner inner oak beam, yet archaic single-link couplings and single-link safety chains were depicted.  The purpose of this drawing was to draw the vacuum brake layout, so other details may have been incidental, but it does raise the unanswered question were earlier coupling layouts brought into line with current practice.

Tender Drawings

For detailed information the reader is recommended to consult “Great Northern Railway Locomotive Tenders” (late Malcolm Crawley, 2013) which is well illustrated with drawings taken from those general arrangement drawings still available in the 21st Century.

One of those minor mysteries is the lack of detail drawings for tenders throughout the Stirling period.  The evidence suggests that drawings for purely tender parts, e.g. frames and tanks, were prepared by a draughtsman in the Works Manager’s Drawing Office (WMDO), or perhaps even located in the Tender Shop.  The Works also prepared tender general arrangement drawings too, copies of which were sent to the LDO for their information and that of the Locomotive Engineer. The LDO dealt with couplings, wheels, axle boxes, springs, brake gear and any other item of a general nature and not specific to a tender.

The first 12 entries (R1 to R12) in the drawing register appear to have been brought forward about 1875 from an earlier drawing register.  They were mostly general arrangements, some for obsolete tenders, suggesting it could have been a complete list of all the drawings the LDO held, and nothing had been discarded.  From 1875 the LDO then started producing general arrangement drawings, commencing at R13, though confusingly the Works continued producing such drawings also commencing at R13.  Fortunately only R13 and R14 were duplicated, after which general arrangement drawings were only drawn out in the LDO.

Because of the lack of detailed tender drawings during the Stirling period, tracing their development relies largely on the available general arrangements, and so the dates quoted are of necessity approximate.  For the purpose of this survey only, the relevant general arrangement drawings are identified by their drawing number followed by water capacity in gallons, then an indication of whether Flat-Bottom or Well provided, and then the wheel diameter (where known) expressed as 3 for 3ft.-6in. and 4 as 4ft. nominal diameter, thus (R8-1500 FB-3) represents drawing R8 for a flat-bottom 1500 gallons tender, having 3ft.-6in. diam. wheels.  Drawings R1 to R7 were for the pre-Stirling period and it will also be noted that Stirling’s first tender general arrangement (R8) appeared in the month he took office.

General points of interest were the following.  The intermediate coupling between engine and tender was fixed (with a vertical eye-bolt) at the engine end and sprung at the tender end.  The original springing was by means of a horizontal laminated spring (3ft.-6in. across) at the rear of the tender connected by long tie bars to the back of the coupling.  The introduction of the well left no room for the tie bars, and the laminated spring was then located just behind the “drag beam”, with short swing links at the outer ends to provide a little side-control.  There were two sizes of wheel, 3ft.-6in. diam. and 4ft.-0in.  The first specific reference to steel being used was on drawing T66 (24 October 1885) for 4ft.-0in. wheels. The bearing springs were positioned out of sight behind the outside frames, immediately above each axle box, and no doubt were awkward to examine for broken plates, and even more awkward to change.  Following the introduction of iron (later steel) frames in 1883, the springs had to be located in full view outside the frames.  With the introduction of the vacuum brake, this was applied to the tender also, with the brake cylinder(s) located just behind the “draw bar” spring, and head of the well tank where fitted.

Tender Frames

R8-1600 FB-4.  In October 1866 Stirling drew out a new tender design which had a modified frame arrangement.  The outside frames were made of 7/8in. iron plate placed 6ft.-7½in. apart transversely; this figure was shortly afterwards altered to 6ft.-8in.  Next to the inner faces of the plate frames were timber beams, 11in. deep and 4½in. wide.  Inside frames were used and these were timber beams too, 11in. deep and 4in. wide.  The inner and outer timber frames were 10¼in. apart.  For additional strength, transverse timber beams 11in. deep and 4in. wide separated the inner frames at two places.  At the ends of the frames, two 1ft.-4in. deep by 6in. wide timbers provided the buffer-beams.  The whole wooden framework was “well morticed and tanned and put together with white lead”.  For his timbers, Stirling specified “well seasoned Quebec white oak or pitch pine”.  Each laminated bearing spring was located in the space between the axle box below and outer oak beam above, hemmed in at the sides by the iron frame plate and wheel respectively, clearly a difficult position for maintenance purposes.  The spring hangers just fouled the oak beam and small pieces were cut out of the timber, ¾in. deep by 3in. long for clearance.  These tenders were intended for goods engines and so had flat bottom tanks.

R9-2500 FB-4.  In February 1867 a larger capacity version was drawn out, noted as specifically intended for Sturrock 0-6-0 Nos. 420-29.  The timber framework was basically the same as for R8, but whereas the earlier design had employed 3ft.-6in. diam. wheels, in this new design they were 4ft.-0in.  As the axle boxes were appreciably higher there was even less room for the springs above the axle boxes.  This was solved by cutting away a section from the outer oak beam above each axle box, providing a recess 3¾in. deep by 3ft.-3in. long.  Both R8 and R9 designs, having flat bottom tanks, were essentially for use attached to goods engines.  Stirling soon found that with the slower speeds at which these tenders ran, there was no marked superiority of the larger wheel over the smaller one.  Thereafter he only used the cheaper 3ft.-6in. wheels for his goods engines.  On the other hand, he found that with passenger tenders larger wheels were less liable to run hot at the higher speeds, this advantage outweighing the higher cost.  Henceforth all his passenger tenders had 4ft-0in wheels, and also wells – again adding to the cost; but high capacity tenders were more essential for passenger engines than goods.

R8-2000 FB-4.  An alteration (shown in red on the drawing) was made in March 1867 increasing the capacity.  To support the extra weight the slots in the iron frames were made slightly smaller.  This would suggest the design was intended for new construction only, as replacement larger tanks would have been too heavy for the original frames.

R10-2470 W-4.  The well version of the R8-2000 type mentioned earlier followed in August 1867.  The transverse timbers had to be dispensed with, the well being in the way.  At this point, the reader is reminded that whilst Doncaster may have been building tenders, the first new engine did not appear until January 1868.

R11-1200 FB.  A 4-wheel flat bottom version was drawn out in July 1868, for 0-4-2 Nos. 11, 31, 218/20, intended for specialised work requiring a short-wheelbase tender engine.  The frames had the usual roughly triangular slot just behind the steps, and a long slot between the wheels.  The diameter of the wheels was not recorded.  So far up to the present day no copy of this drawing has been found.

R12-2000 FB-3.  A slight change in the shape of the front frame slot (near the step) appeared in August 1872, with the leading edge now vertical instead of slanting.  The same arrangement drawing shows the large wooden brake blocks replaced by the cast-iron pattern.  The outside-frame plates are now noted as being 6ft.-9in. apart

R13-2700 W-4.  The LDO produced a general arrangement drawing in February 1879, specifically noted along with the later R14 and R15 drawings as being “original”, suggesting that the earlier drawings were all copies of drawings received from the Works.  The drawing showed twin-vacuum brake cylinders fitted.  No coal rails shown. 

R13-2800 W-4.  This was the final wood frame tender design, drawn out in October 1881 (presumably in the Works) bearing the reference R13 which had already been used by the LDO.  This Works drawing turned up in the LDO some time between 1883 and 1889 and its title was entered in their drawing register following the entry for R15.  This design had 6½in wide outside timber beams, and 4in wide inside ones.  These beams were now only 8¾in. apart, whist the outside frames were still of 7/8in. wrought-iron.  (The timber was now of “well seasoned Quebec white oak” only, pitch pine having evidently fallen out of favour.

R15 was a detail drawing for the iron frame drawn out by the LDO in May 1883. This showed 4ft.-0in. wheels matching the general arrangement drawing R14 (though there was no separate detail drawing for the tank).  The use of timber was discarded except for the oak buffer-beams.  An inner iron frame was introduced, 1ft.-1in. deep and ¾in. thick.  The inner and outer frames were now 79/16in apart, the outside frame plates having been brought inwards nearer to the wheels, by dispensing with the timber beams.  The outside frames were now only 5ft.-6¾in. apart and this enabled the axle box guides to be bolted now to the outer faces of the outside frames.  Thus the axle boxes themselves were positioned outside the frames and consequently the springs too – in a more accessible position.

R14-3200 W-4.  The general arrangement drawn out in the LDO in May 1883 was for the first “iron frame” tender, with a total water capacity of 3200 gallons.  The drawing shows the twin-cylinder vacuum brake arrangement. It is likely this proposal was rejected as the 7ft.-6in. wide tank would presumably have been unacceptable.

R14-2900 W-4.  Another general arrangement drawing turned up later, bearing the number R14 and so was presumably drawn out in the Works who were unaware the LDO had already used this number.  This time its title was not entered in the drawing register, and the surviving copy gives no indication of its date.  It shows that the main tank held 2400 gallons of water, and the well tank a further 500 gallons.  Also shown is the single-cylinder vacuum brake arrangement, but otherwise the design resembles the LDO version of R14.  It was probably the Works’ response to the LDO wide tank proposal.

R16-2100 FB-4.  A general arrangement drawing was prepared August 1889 for the old tender from the recently withdrawn Hawthorn No. 213 to be apparently refurbished for further use.

R18 was a detail drawing prepared 13th March 1890 for the frames of a 2600 gallon tender.  Following close after R17, this suggests the two drawings may have been related.  This was immediately followed by drawing T73 (April 1890) for the 3ft.-6in. wheel for a 2600 gallon tender, the diameter confirming it related to a flat-bottom tender.  So far up to the present day no copy of this drawing has been found.

R19-3500 W-4.  A general arrangement drawing was prepared 5th March 1891.  These tenders had wells, and may be those tenders described in W B Yeadon’s list as 3553 gallons tenders.  So far up to the present day no copy of this drawing has been found.

R20-3850 W-4.  A general arrangement drawing was prepared July 1894 for the last Stirling 8ft. Singles, Nos. 1003 to 1008.  This was the highest capacity Stirling tender.

R21-2654 FB-3.  Stirling’s final general arrangement drawing was prepared November 1895. This was for the proposed 5ft.-0in. goods engines, and therefore required no well.  So far up to the present day no copy of this drawing has been found.

It will be appreciated that the majority of the aforementioned drawings were general arrangements.  No information is available concerning detail drawings prepared by the Works, needed in connection with frames and tanks.  During the Stirling period a small number of detail drawings for tenders were prepared by the LDO, see table below for known examples.

                                    Table 6 – Stirling Tender Detail Drawings 

Drawing

Date

Details

T28

25/10/1875

Tender wheels & axles

T57

4/4/1878

3’–6” tender wheels & axle for new [axle] box U-55

Z20

3/10/1879

Details for tender brake

Z21

15/12/1881

Angle plates for tender frame

Y109

-/4/1883

Improved buffer for engines and tender

T63

21/6/1884

4’–0” wheel for 3000 gallon tender

T66

24/10/1885

4’–0” wheel for 3000 gallon tender (Steel)

Z25

12/4/1886

Details for tenders (including intermediate coupling)

Z26

-/1887

Details for tenders (including drag)

T73

4/1890

3’-6” wheels for 2600 gallon tender

Z27

7/1895

Coupling for end of tender

In the preceding table, T list entries were taken from the drawing register, Y list entry from the NRM (York) catalogue and Z list entries from a “modern” transcript.

For a few years after Stirling’s death the same divided method of preparing drawings for tenders still applied as late as 1904.  Eventually all work was done by the LDO.

Tender Tanks

The tank followed the same pattern throughout Stirlings time with one exception (R20-3850 W-4 described later).  The tank extended as far forward as practical leaving plenty of easily accessible coal space in front of it.  (Less accessible was the extra coal stacked on top of the tender.)   The “horizontal” top of the tank sloped down slightly towards the front, usually by ½in.  The tank terminated in the vicinity of the front wheels, as far forward as possible whilst allowing clearance for the feed pipe at the bottom of the tank of the flat-bottom tenders and stopping short at the splashers in the case of the tenders with 4ft wheels.  This allowed coal to collect well forward of the tank, within easy reach of the fireman.  The upper part of the front of tank sloped back at an angle of about 50 deg. from the horizontal, to provide extended coal space.  Variations in water carrying capacity were usually provided by altering the depth of the tank.  The “shovelling plate” was at footplate level so the fireman had to raise the coal about 15in. while firing.  Some tenders had supplementary wells, located between the frames and above the axles.

R8-1600 FB-4.  In October 1866 Stirling drew out a new tender design. The tank was 2ft-11in high and capacity 1600 gallons. The figure of 1500 appears in the drawing register but as this entry had been copied from an earlier register it was probably a clerical error.

R9-2500 FB-4.  In February 1867 a 2500 gallons version was drawn out with the tank now 4ft.-4in. high, noted as specifically intended for Sturrock 0-6-0 Nos. 420-29.  A feature of the larger wheels was the need for an 8½in. wide curved recess, reminiscent of a wheel “splasher”, above each wheel within the tanks to clear the tyres.  Both R8 and R9 designs, having flat bottom tanks, were essentially for use attached to goods engines.

R8-2000 FB-4.  An alteration (shown in red on the drawing) was made in March 1867 increasing the capacity to 2000 gallons increasing the tank height to 3ft.-8in.

R10-2470 W-4.  The well version of the R8-2000 type followed in August 1867, increasing the capacity to 2470 gallons.  The shallower tank was balanced by a 16ft.-5in. long well between the inside frames.  The depth of the well was limited by the proximity of the tender axles and the width by the inside timber frame.  It is difficult to see what advantage this design had over R9-2500 which actually held more water.  Perhaps lowering the centre of gravity by employing a well was thought advisable when contemplating high speeds on a rough track, with the curves not canted as they are today.  Still, the well left scope for increasing the capacity in the future, and in fact the height was later varied a little.  At this point, the reader is again reminded that whilst Doncaster may have been building tenders, the first new engine did not appear until January 1868.

R11-1200 FB.  A 4-wheel flat bottom version was drawn out in July 1868, for 0-4-2 Nos. 11, 31, 218/20, intended for specialised work requiring a short-wheelbase tender engine.  The tank capacity was 1200 gallons.

R12-2000 FB-3.  The drawing shows a 2000 gallons tender, with a red ink undated amendment, showed a 4¼in. deeper tank, increasing the tank capacity to 2300 gallons. Deeper tanks were being introduced around 1881, so this amendment could have been made about then, see R12-2300 FB-3 later.

R10-2420 W-4.  Around 1876 a start was made fitting simple vacuum-brake to new passenger engines, but the well tank was in the way of the intended location for the  brake cylinder.  New wells were needed, 1ft-8in shorter at 14ft.-9in., set 1ft.-6in. further back, reducing their capacity by about 10 per-cent, with the total capacity reduced to, say, 2420 gallons.  The necessary drawings would have been prepared in the Works, not the LDO, but there would have been no need for a fresh general arrangement drawing.  The tank from one of these tenders finished up behind No. 1 and measurements taken in July 1938 showed the well to be 14ft.-3in long.  During 1876-78 about 45 R10 tenders were similarly altered for existing 2-2-2, 4-2-2 and 2-4-0 engines with the possible exception of the last one in 1878, 2-4-0 No. 294, see next entry.

R10-2700 W-4.  The drawing was amended in October 1878 to increase the overall capacity to 2700 gallons.  On the drawing was noted “tanks 5 inch longer, 2 inch deeper and 2 inch wider”.  The new size well was installed, as had been the practice anyway over the previous two years, though the alterations to the main tank dimensions may have been a new feature.  The last existing passenger engine to receive the simple vacuum-brake was No. 294, ex-works in October 1878, and it is thought its tender was the first one altered to the new arrangement, which was the reason for the preparation of this drawing.  Tenders continued to be built for some time to this drawing.

R13-2700 W-4.  The LDO produced a general arrangement drawing for a 2700 gallons tender in February 1879, specifically noted along with the later R14 and R15 drawings as being “original”, suggesting that the earlier drawings were all copies of drawings received from the Works.

R13-2800 W-4.  This was the final wood frame tender design, drawn out in October 1881 (presumably in the Works) bearing the reference R13 which had already been used by the LDO.  This Works drawing turned up in the LDO some time between 1883 and 1889 and its title was entered in their drawing register below the entry for R15.  The tank was made 4¼ in deeper and held 2800 gallons.  Three coal-guards were now added to the top of the coping plate, another Stirling feature come to stay, except for a variation in the numbers of rails at a later date.

R12-2300 FB-3.  The basis was the R12-2000 FB-3 design of August 1872, but with a red ink amendment showing a 4¼in deeper tank and capacity 2300 gallons, suggesting its appearance closely followed the aforementioned R13-2800 W-4.

R14-3200 W-4.  The general arrangement drawn out in the LDO in May 1883 was for the first “iron frame” tender, with a total water capacity of 3200 gallons, including 480 gallons in the well.  It is likely this proposal was rejected as the 7ft-6in wide tank would presumably have been unacceptable.

R14-2900 W-4.  Another general arrangement drawing turned up later, bearing the number R14 and so was presumably drawn out in the Works who were unaware the LDO had already used this number.  This time its title was not entered in the drawing register, and the surviving copy gives no indication of its date.  It shows that the main tank held 2400 gallons of water, and the well tank a further 500 gallons.  The tank design resembles the LDO version of R14.  It was probably the Works’ response to the LDO wide tank proposal.

There were two drawings prepared in the LDO for 4ft-0in wheels for 3000 gallon tenders, T63 (June 1884, wrought iron) and T66 (October 1885, steel), which cannot be related to a known tender type.

R16-2100 FB-4.  A general arrangement drawing was prepared August 1889 for the old 2100 gallons tender from the recently withdrawn Hawthorn No. 213 to be apparently refurbished for further use.

R17 was a detail drawing prepared February 1890 for the tender tank for Sturrock 0-6-0 No. 406.  Nothing further is known.  So far up to the present day no copy of this drawing has been found.

R18 was a detail drawing prepared 13th March 1890 for the frames of a 2600 gallon tender.  Following close after R17, this suggests the two drawings may have been related.  This was immediately followed by drawing T73 (April 1890) for the 3ft-6in wheel for a 2600 gallon tender, the diameter confirming it related to a flat-bottom tender.  So far up to the present day no copy of this drawing has been found.

R19-3500 W-4.  A general arrangement drawing was prepared for a 3500 gallons tender 5th March 1891.  These tenders had wells, and may be those tenders described in W B Yeadon’s list as 3553 gallons tenders.  So far up to the present day no copy of this drawing has been found.

R20-3850 W-4.  A general arrangement drawing was prepared July 1894 for the last Stirling 8ft-Singles, Nos. 1003 to 1008.  The tank design was a change from Doncaster practice.  The main tank now straddled all six wheel splashers and not just the rear four.  Where the front of the tank had previously sloped down there was a gentle “S” curve with the lower part sloping down at an angle of about 12 deg from the horizontal to the front of the tender, where it was 1ft-6in deep.  A separate sloping shovelling plate was fitted over this last section, which was no doubt a boon to firemen, being at about the same level as the fire-hole door.  It is unlikely the alteration had been made as the result of complaints from firemen, who were a hardy breed, but if any had suffered bad backs and thereby caused train delays then this would have been of concern to the Directors.    This was the highest capacity Stirling tender and also differed from earlier tenders in having only two coal guards instead of three.

R21-2654 FB-3.  Stirling’s final general arrangement drawing was prepared November 1895. This was for the proposed 5ft-0in goods engines, and therefore required no well.  So far up to the present day no copy of this drawing has been found.

It will be appreciated that the majority of the aforementioned drawings were general arrangements.  No information is available concerning detail drawings prepared by the Works, needed in connection with frames and tanks.  During the Stirling period a small number of detail drawings for tenders were prepared by the LDO, see table below for known examples.

The difference in dimensions of the tanks of those types for which drawings are available are summarised below.

Table 7 – Stirling Tank Dimensions


Type

Tank length

Tank
depth

Tank
width

Well
length

Well
depth

Well
width

R8-1500

14ft-9in

2ft-11in

6ft-1in

 

 

 

R8-2000

14ft-9in

3ft-8in

6ft-1in

 

 

 

R9-2500

15ft-4½in

4ft-4in

6ft-6in

 

 

 

R10-2470

15ft-4½in

3ft-8in

6ft-0½in

16ft-5½in

1ft-9in

3ft-3in

R10-2350

15ft-4½in

3ft-8in

6ft-0½in

14ft-9in

1ft-9in

3ft-3in

R10-2700

14ft-9in

4ft-2in

6ft-21/8in

14ft-9in

1ft-9in

3ft-3in

R12-2000

14ft.9in

3ft-9½in

6ft-0½in

 

 

 

R12-2300

14ft-9in

4ft-1¾in

6ft-0½in

 

 

 

R13-2700

15ft-1½in

4ft-2in

6ft-2in

14ft-9in

1ft-9in

3ft-3in

R14-3200

15ft-1¼in

4ft-2in

7ft-6in

14ft-9in

1ft-8in

3ft-3in

R14-2900

15ft-1½in

4ft-2in

6ft-8in

14ft-5½in

1ft-8in

3ft-3in

R16-2100

15ft-1½in

4ft-2in

6ft-0in

 

 

 

R20-3850

18ft-97/8in

4ft-8¼in

7ft-1½in

14ft-4in

1ft-8½in

3ft-6in

Epilogue

Stirling died “in office” in November 1895 without the chance of handing over to his successor.  His close friend and assistant John Shotton (Works Manager) had died not long beforehand, and we are left to speculate what Stirling’s intentions would have been over the following months.  Bearing in mind up to two years could be spent between planning and completion, allowing for changing priorities in the various departments, and the need to ensure there was always work available at all times in the Works and in the Drawing Office, it is essential to study the drawings that appeared in 1894 and 1895, looking for clues.

The entries in the drawing registers were as follows.  Those for the 8ft Singles may be ignored as these engines appeared before Stirling’s death.  Some details have been taken from later undated typed transcripts (I-M, S, W and Z drawings) but others (N-T drawings) are exactly as entered in the drawing register.  I have ignored register entries for U, V, X and Y drawings (axle boxes, springs, ashpans and miscellaneous).

I series (smoke boxes, chimneys and blast pipes).  No drawings produced between 16/3/1893 and 10/4/1896.

J series (injectors)
J43 (13/10/94) feed pumps
J44 (       2/94 sic) No. 9 m/m Gresham & Craven’s injector, Delivery.
J45 (       3/96) Injector and ejector steam cock feed and delivery valve.

K series (valve gear)
K52 to K56 (/94) various drawings for 8ft-0in engines Nos. 1003 & 1004
K57 (20/101894) Screw Reversing Gear for Nos. 1021 to 1026
K58       (1/1896) Steel Cross Stay & Valve Spindle for - [left blank]

L series  (slide bars and crossheads)
L16  (22/2/94) Crosshead and Slidebars for Nos. 1003 & 1004
L17 (4/1896) Crosshead & Slideblock for Goods & Pass. Engs.

M series (cylinders and valves)
M59/61/62 ( /94) various
drawings for Nos. 1003 & 1004
M58  (   16/1/94) Slide Valve and Buckle (18½“ x 26” Cyls) for 7’-6” Pass. Engs.
M60  (        2/94) 18” Piston and Cylinder Cover for Met. Tank Engs
M63  (      12/95) 17½“ x 26” Cylinders for 5’-0” Goods Engs.
M64  (      12/95) 18” x  26” Cylinders  for 4’-6” Tank Eng.

N series (boilers and fireboxes)
N64 (April 19th/94) Boiler for Nos
. 1003 & 1004
N65 (Jany 15th /95) Boiler for Goods Engs with 4’-6” Wheels – Nos 1021 to 1026 (b)
N66 (       Novr /95)    “        “       “          “    “   5’-0”      “   (Standd)    (a)
N67 (        Feby/96) Arrangement of Tubes for Tank Engs Nos 1031to1045. Firebox end (c)
N68 (        Feby/96)    "                      "               "       "       "       "        "       Smokebox end (c)

O series (frames)
O42/43 (         /94) drawings for 8ft singles Nos 1003 & 1004
O44   (   Novr /94) Frame for Metn Side Tank Engines Nos 941 to 944
O45   (      “       “ )    “       “    4’-6” Six Wheels Coupled Engs Nos 1021 to 1026 (b)
O46   (May 18/96)    "       "     6'-6" passenger Engines Cylrs 17½ x 26   Nos 1061 to 1070 added later

P series (goods engine plan drawings)
P36 (Jany 12th/95) 6 Wheels Coupd Goods Eng. 4’-0” Wheels
P37 (Novr       /95)  “      “          “          “         “      5’-0”      “

Q series (passenger engine plan drawings)
Q51 (Decr /94) Express passr engines  8’-0” 19½” x 28” Cyls Nos 1003,1004,1005,1006 &c
Q52 (Sepr /95) Alteration to Passr. Tank Eng. No 245

R series (tenders)
R20 (   Jul /94) Tender for 8ft engs Nos 1003, 1004 &c            “(3850 Gallons)” added later
R21 (Novr /95) Tender   “  5’-0” Goods Engs (Flat bottom)     “
[blank]   gallons” added later

S series (coupling and connecting rods)
S30 (3/94) Connecting rod for 8’ Pass. Engs.
S31 (4/94) Connecting rod

T series (wheels and axles)
T77 (Feby 8th /94) 8’-0” Wheel & Axle (Steel) for Nos. 1003, 1004
T78 (May /95)       7’-0”     “           “          “      “ 
   [blank]
T79 (Aug. /95)       3’-0”                                   for Metn Tank Engines

W series (brass fittings)
W40  (27/4/94) Regulator for 8’ pass engines 1003 & 1004
W41 (20/6/94) Brass work No. 6 G[resham] & C[raven] Injectors
W42 (8/94)       “        “        steam sanding details

Z series (tender details)
Z27 (7/95) [Incidentally the first drawing since 1887] Coupling for end of tender

 

Notes:

(a) N66 for the 5ft Standard engines was the usual Stirling 4ft-2½in boiler, an indication that his later standard goods engines would continue to have this size of boiler.  The final series comprised Nos. 1031-45 built by Dubs in 1896 to General Plan drawing P37 with this boiler.  These engines had 13in wider cabs than the earlier 5ft engines (excepting Nos. 1011/12).  Had Doncaster inadvertently provided the wrong cab drawing? 

 

(b) N65 and O45 are clear indications that Stirling intended these new mineral engines to have 4ft-5in boilers, needing a new frame design.  The numbers 1021-26 were allocated in advance, leaving a gap for further 1003 Class 8ft. singles to appear after 1008 (which they never did).  Therefore the ten mineral engines Nos. 1021-30 that appeared in 1896 were Stirling engines, the material for them having been ordered about November 1895.  More than that is conjecture.

Then followed a short Interregnum before Ivatt took office in March 1896.

 

(c) N67 Error went uncorrected. Not obvious if intended for tender engines 1031-45 or tank engines 1046-60.
Roy Burton apparently never located this drawing, and simply made a note that N67 was for 1031 to 1045, apparently taking this from the drawing register.

RETURN TO INDEX