User:Marquis DeCarabas/The Signalling of the Duffws Line

This article is a work in progress and as such I would really rather prefer a) no (further) non-MdeC editing yet, please and b) that no links pointing directly at this page are created. There is a long road yet to travel before I'm happy to move this into a more public arena. ta very much

The signalling history of the Duffws Branch (later the Main) is very complex, and to tell the story fully requires a slight broadening of the term 'Signal Box' to include locations where block instruments were situated. In order to cover the story completely, there will also need to be an analysis of the transition between the single track branch line and the latter twin single lines.

I propose to analyse the areas with reference to the later signal boxes. This may mean that during the contstruction phase of the article that the chronological order of each section is slightly muddled.

The three conventional signal boxes in Blaenau Ffestiniog were built between late 1878 and 1881 and all had similar features - built of timber, clad with horizontal shiplap with sliding windows, slated roof and access by porch from a stairway. They had lockers and stoves. In geographical order west to east the locations where the trains were block signalled on the branch were {Old Dinas Junction points}, No 3 signal box (the first), {1899 Dinas Junction}, {1881 Dinas Loop Line Junction points}, Glanypwll Level Crossing, No 3 signal box (the second), {Festiniog and Blaenau Junction points}, No 2 signalbox (all iterations), No 1 signalbox. There were many more connections into the branch line for sidings and goods yards.

Blaenau No 3. box (the first) Old Dinas Junction:
This was situated in the Eastern angle of the triangle junction of the line to Duffws and the original route to Dinas. It had 12 levers, one spare. This (old) No. 3 box was replaced when the 1899 deviation of the Dinas branch opened. As evidenced in the Spooner Album, when originally built this was a hybrid Disc/Semaphore signalling installation, with at the very least the Down Dinas Branch and Up Main Line using Disc signals as Distants. At this remove it is hard to be completely sure, but the first controlled signal from Duffws was a semaphore home signal after November 1897. McKenzie and Holland commissioned lever No 1 in November 1897 to act as the Up Distant. The Spooner Album shews a former double-arm block semaphore at the Duffws end of LNW Exchange that may have been controlled by lever No 1 prior to 1897.

Glanypwll Level Crossing:
Block instruments were provided in the crossing keeper's house (now part of Glanypwll Hostel) until the second No 3 box opened.

1885 Block Working Instructions Blaenau No 3 - GyP XD97/50006

Blaenau No 3 box (the second) Dinas Junction:
Built during 1899 by a box on the clock side of the line between Glan y Pwll level crossing and the footbridge to the school. This location had views towards Station Fein and back down the line across the level crossing. This signal box had 19 levers with one spare.

1885 Block Working Instructions between Blaenau No 2 and Blaenau No 3 XD97/50005

There is a cryptic set of correspondence in XD97/19495, ostensibly covering the working of the Tanygrisiau - Blaenau METS system when Cwmorthin incline is in use. Robert Evans circularised the staff on 22/8/25 commenting "On and from Monday next, the 24th instant, the Electric Staff instrument and also the telephone switch will be at Blaenau Festiniog station. Same has been removed from No 3 Signal Box." In the same bundle is a typewritten note dated 29/8/25 from the Accountant's Office to Piercey at Duffws reading: " As soon as the Tanygrisiau staff is put in, the Glanypwll box is to be closed. Please note and arrange accordingly. Also, directly the box is closed, the new arrangement as regards the five-lever frame is to come into force Piercey replied to Evans on the 1st September, datemarked 19th September acknowledging the note, but commenting that as No 3 box was still open, it was impossible to bring into force the 5 lever frame as instructed!

The next set of correspondence chronologically, from September/October 1925 suggests that Col. Stephens wanted Train Staff and Ticket working instituted from "Festiniog L. M. & S. to Duffws" (Memo from Tonbridge, dated 7/9/25). Evans acknowledges Stephens' request on 17/9/25 and instructs GL Griffiths on the day after to arrange to have the staff ticket boxes put in at both Blaenau Festiniog and Duffws:

Letter from Evans to HFS, giving the origin of the staff boxes for the new, extended section and suggesting a closure date (or period) for No 2 Signalbox), extract from XD97/194646

As No 2 box seems to have been removed from regulating the traffic, this letter suggests some evidence for the closure of No 2 box, perhaps by September 1925. There seems to have been little progress in the matter, until 15/10/25 when Griffiths replies to another request for the matter to be progressed from Evans - Griffiths writes on 15/10/25: 'I find it difficult to get a carpenter to do this work. I have to-day shewn the Signal Lineman what is required and doubtless these will be put up to-day or tomorrow". Evans then officially informs the Station Masters at Duffws and Bl. Festiniog (presumably LMS, from context) on 17/10/25 that the staff boxes are now installed and that Train Staff and Ticket working can be instituted, Evans sent the station masters appropriate Staff Tickets on the same day.

Blaenau No 2 box (the first):
Called Duffws Junction in 1879 signalling arrangement diagrams ; controlled what Boyd himself named 'Dolgarreg-ddu Junction' (the junction between the Festiniog proper and the Festiniog and Blaenau Railway). Dolgarreg-ddu was the name given to the general area of Newborough lands that became the site of Duffws station proper. This box has some of the earliest known correspondence about the signalling along the Blaenau branch: understandable, given that it was the junction of two railway companies and the signalbox was to be jointly provided by both the FRCo and the Festiniog and Blaenau Railway Company. The first known plan for the signalling here is dated 20th December 1877, and is the first 'conventional' (i.e. semaphore) plan for a location along the branch. The turnpike road has been bridged, and the Down Distant is mounted on the double arm semaphore in Duffws Station.

Duffws Station throat, as of December 1877. Note Double Arm block semaphore at Duffws, depicted with two arms. Extract from McKenzie and Holland signalling arrangement plan

Note the recess on the clock side in the lee of the Turnpike Bridge, this was to become the site of the later Blaenau No 1 box. During March 1878, there was some discussion between McK&H and the FRCo about the junction at Duffws, concerning the apportionment of levers and outside equipment between the FRCo and the F&BRCo.

6th March 1878 apportionment list to the FR from McK&H

It would be foolhardy to try and reconstruct a signalling diagram from the partial information in the apportionment list, although it is worth noting that the 'erecting existing signal' most probably refers to the double arm block signal in Duffws Station. This apportionment list was followed up by a letter from the Vulcan Iron Works to CES on the 14th March 1878:

We are in receipt of your favor of yesterday's date with copy of Mr. Kelley's letters. He seems surprised at the difference between the two estimates given. Our first estimate was given at the request of Mr Kelley in a letter of Jany 3rd, and was of course based up on the idea that each Company would bear its proportion of the whole, not knowing of any special agreement between the Co's; we estimated then as follows viz:
Festiniog Rly Co £325-0-0
Festiniog and Blaenau Co £115-0-0
Our next estimate was based on the information contained in your letter of the 6th inst. and the proportion allotted to each Company is there of an arbitrary character and we took out the quantities for each in accordance therewith as near as we possibly could and gave you the figures in ours of the 7th inst.
We now enclose you the particulars which we will enable you to settle the matter, in case however you find it necessary to alter the proportion, if you give us the details of what you finally consider belongs to each Company we will give you an estimate as near as we possibly can.

Pending further research the actual nature of the junction in pre-semaphore days between the Festiniog and F&BR is open to a high degree of conjecture. It is a case of the truth is out there, somewhere, buried in fragment. Boyd quotes 4 conditions, purporting to stem from the May 1868 inspection of the F&BR by Captain Tyler. These are:

1. The removal of certain facing points near to the Blaenau terminus of the line, and also opposite the main junction signal box at that point.
2. The removal of the main junction semaphore and to reposition it at or nearly opposite the junction between the Festiniog and F&B Railways.
3. The said junction points to be fitted with locking apparatus and until this is done, to be locked between the passage of trains and subjected to a speed of 3mph.
4. The viaduct at 3 miles to be strengthened as suggested.

Of these, clearly Item 4 is of relevance only to the F&B and can be disregarded. Interpreting items 1 and 2 presents quite a challenge. To the best of my knowledge, there was no 'main junction signal box' in the modern understanding of the term. However, there are some very vestigial clues contained within the Archives.

Land Swap plan for the Festiniog and Blaenau, undated but pre 1877; XD97/418951

This plan gives us a tantalising glimpse of the situation in Blaenau, and at first sight apparently has no signalling interest whatsoever. However, if you zoom into the drawing some interesting information can be gleaned from the wall boundaries. The sites for the original Duffws approach signal (more to come in the section on Duffws station) and, just possibly, the site for the junction signal mentioned in condition 2 and the putative site for the 'main junction signal box'.

Southern Detail from XD97/418951

The area that especially interests us in this analysis is just to the left of the 'A' and underneath the upside-down figures "110" and "1.50" on the clock side of the line. Just to the left of the 'A' is a semi-circular recess away from the track, this is most logically the altered site of the junction signal. Without sight of the original inspection report I am highly dubious that this signal was of semaphore form: there is scant evidence for the ordering of more semaphores in the Archives: this does not mean that more were ordered, just that the evidence does not survive. It is tenable that the Disc signal moved from the upper side of the gateposts before the level crossing to before the new junction. There is ancillary evidence from the 1880s that whistle signals were in use, and I suggest that this Disc signal was normally at Danger and only cleared when the crews whistled up. Movements to the Festiniog and Blaenau would most likely be controlled by a white flag or lamp from the junction signal box.

What 'junction signal box'? Examination of the drawing underneath the upside-down figures "110" and "1.50" on the clock side of the line shews a pencilled black square. Is it a leap too far to consider this to be the same size as one of the square signal cabins (pointsmans bothies) used at Minffordd Yard and Old Dinas Junction?

Suggested reconstruction of the Festiniog/F&B junction in pre-semaphore days

Of course, this image is provisional, pending further research. It may well be possible that Tyler deliberately specified a 'semaphore' signal, and at this stage there may well have been one spare from Rhiw Goch (although that is a moot point). I suspect Spooner pragmatism prevailed and the Disc signal was simply moved down the line. If it were a Disc, then as at Minffordd the pointsman would merely change the points when they were free of the wirelocking of the twin wires to the Disc. Without any further evidence I suggest that this is the most likely situation, as the use of Discs to lock points was established since their introduction: there was no mechanism for locking points using semaphores. Well... There's an exception to every rule, and the block semaphore at Tanygrisiau apparently locked the top end loop points in the 1920s at least.

It seems that the F&B used double arm semaphores; we know that McKenzie, Clunes and Holland were the contractors for the line and there is photographic evidence of there being a double arm semaphore at the Llan Festiniog terminus as well as Tanymanod .

Further digging in the Archives has produced this Board resolution, as a result of an accident to Little Wonder that Boyd reports in NGRISC:

Facsimile Board Resolution for the working of "Dolgarreg-ddu Junction", wording courtesy Christopher Jones.

Although interpretation of these instructions is axiomatic on assuming that they post-date the doubling of the Duffws Line they still give us a useful indication of what existed on the Duffws Line before semaphore signalling was installed. Admittedly, the instructions do not give us any indications of the signalling on the F&B side, but the phrase 'the man in charge of the Block below Duffws Bridge' indicates that the pointsman's bothy at the level crossing (see discussion of No 1 box) was indeed in use as a Block Post. In someways, this resolution suggests that there was no specific 'junction signal box' until one was provided by McKenzie and Holland. In and of itself, this then raises the question of what the black square in the suggested reconstruction above actually symbolises? I would venture to suggest that the square is purely an idea of another double arm semaphore. Clearly, there is no semaphore present at the junction on Festiniog Railway Company land (otherwise the working of the signal would be reflected in the wording of the resolution) - in that case is there any evidence of a signal on the Portmadoc side of the level crossing (and later) the 1875 Turnpike Bridge? The answer is Yes!

Extract from XD97/419009, shewing possible site for a semaphore, below Duffws LC

If we assume that the squares drawn on earlier drawings symbolise the posts of a semaphore signal, then I posit that the square in this drawing above also symbolises a semaphore. The full drawing (not XD97/419009 - check ref, possibly XD97/419084) also indicates that there may be a semaphore in the 'traditional' location at the end of the pre-1870 station building. I feel that it may be unlikely that the semaphores were contemporaneous - a more plausible "signalling explanation" is that the semaphore was moved below the level crossing to centralise control and just perhaps to protect the Turnpike Bridge during construction. The extract of XD97/419009 above does have a red emendation of "Site of Proposed Bridge" running diagonally across the level crossing.

Combined XD97/419009 and XD97/418951, shewing all possible pre-interlocked signalling

This combined image gives an indication of the signalling in the period between Duffws Yard being remodelled and the doubling of the branch. I have drawn a semaphore signal on the approaches to "Dolgarreg-Ddu" Junction, as suggested by the black square in XD97/418951, this now seems unlikely after a close reading of both the 1872 and 1876 rule books where Rule 40 clearly refers to Old Dinas Junction (and is backed up by photographic evidence). I would tentatively suggest that in comparision to the provisional image c. June 1868 above that there was *not* in fact a semaphore signal fitted at the junction - unless of course it was a F&B owned and maintained signal. I am now sure that the square is a 'doodle' in response to Tyler's report and the four conditions. Due to the ordering records being so patchy, it is impossible to be absolutely certain that more semaphores were acquired and in 1868 the complete position is still under research, but I feel that it is unlikely that a semaphore would have been erected as there were no known 'spares' nearby . If there were a semaphore actually at "Dolgarreg-Ddu" Junction I think the bothy adjacent to the level crossing would have been moved much nearer to the site of the junction and there would be a lot more evidence apparent in the Archives of the bothy's movement. As it stands, the wording of the 1878 instruction would cover the single line working between 1868 and c.1875 (or whenever the line was doubled); the wirelocking enacted by the Disc would lock the junction points. It is natural to assume that Spooner pragmatism prevailed and the junction points were locked by the Disc below the Market Hall as would happen at any other interlocked FR stage station: that then raises the question about why the (un)locking of the points and Disc are covered as two separate operations in the 1878 instruction. I suggest that the separate operations are an accident of timing - they are written in the transition stage between the doubling of the line and full signalling being installed from McK&H.

Even the doubling of the line is not especially clear in the Archives. The next drawing clearly post-dates the construction of the Turnpike Bridge, but has Dorvil Road bridge marked as 'new':

XD97/418919, drawing of new sidings in the area of the later No2 signal box.

As can be seen, the drawing itself presents quite a puzzle - the land boundaries are not quite the same (especially around the Turnpike Bridge) and there are interesting pencil emendations that deserve further analysis.

Blaenau No 2 box (the second) FR/GWR joint box:
Rebuilt from the original box on approximately the same site with the change of gauge in 1883 and then two frames were installed - one for the FR and one for the GWR. This was at the end of the island platform at the GWR interchange station. It controlled narrow gauge points and signals and also the standard gauge. It was manned by an FR employee and the GWR contributed towards wages and upkeep of the box. The FR-side signals were dismantled during 1925. The GWR removed their frame in January 1929 and the FR theirs in April 1929. The box was dismantled by late April 1929. The GWR moved the control of their standard gauge signalling to a new central signal box which came into use in September 1926. As discussed in the section about the second No 3 box, it appears from one strand of evidence that the second No 2 box closed on the FR side during September/October 1925. There is in the Archives (XD97/19494) a memo dated 28/7/25 from Evans to Griffiths reading: "The driver of the 11.0 a.m. train on Saturday reports that points by No 2 box were half open due to not being oiled, and shall be glad if you will give instructions for this to be attended to at once". This strand of evidence suggest that No. 2 box was still in some form of existence in the summer of 1925.

1897 Block Working Instructions between Blaenau No 2 FR Frame and Blaenau No 2 GW Frame XD97/19521

Blaenau No. 1 box Duffws:

Fortunately, in the case of Duffws there is much more primary evidence to base a suggested development timeline of the signalling.

Plan dated 23rd December 1869, XD97/419014

XD97/419014, although by no means the earliest plan of Duffws to mention signalling gives us the most complete picture of the layout before the semaphore installations of the late 1870s/early 1880s. There is a Disc signal on the approaches to the level crossing and a Semaphore Block signal in the station grounds. In this respect Duffws is no different from any other 'traditionally Festiniog-signalled' location and is a mirror image of Portmadoc pre-1923. It is tempting to suggest that the installation at Dinas was very similar, but there is a real paucity of hard evidence about the signalling of the Dinas line.

XD97/419014 annotated

This file is patently a copy of an earlier drawing that must have been drawn up to cover the arrival of the Festiniog and Blaenau Railway, however it does indicate what the signalling must have been for the 5 years before the F&B arrived in Blaenau. Actually, strictly it only does indicate where signals were sited, there are two possibilities. XD97/410977 mentions a 'signal post'.

Duffwys, dated 1864 in the Archives Catalogue. N.B. Image and title block rotated ninety degrees for display and comparison purposes.

This layout is entitled 'Proposed Alterations at Duffwys Station'; unfortunately it is not exactly clear what the proposed alterations entail. Note, from signalling interest the square post denoted at the top left hand corner of the new station building; and (just perhaps) the suggestion of a semaphore point indicator right at the station throat where the passenger line diverges from the main line (the first hard evidence for those that I have seen comes in 1872 with the opening of Tanybwlch station). It is within the bounds of possibility that the alterations are just for the crossovers at the upper end of the passenger landing - creating a separate passenger line from what may have been the line to Diphwys Casson Quarry incline.

Track plan from XD97/419077 superimposed (approximately) on XD97/419014

This is what many would see as the 'traditional' Festiniog-style terminus signalling. It is a carbon-copy of the original Portmadoc layout and with the double arm block semaphore also matches the layout known at Llan Festiniog on the Festiniog and Blaenau Railway, though the F&B did not (as far as is known) use Disc signals.

Track plan from XD97/419077 superimposed (approximately) on XD97/419014; Disc signal only variant

Given that XD97/419014 has the same symbol of a quartered circle either side of the level crossing, it is worth considering the possibility that the Church St. level crossing was signalled in the same manner as Job William's (Lottie's) crossing at Minffordd. Unfortunately this can only be speculative at this stage of the research, but there may be a clue present in the following photograph:

Church Street level crossing, pre 1875, post 1868.

Note what the arrow points to - this is awfully reminscent of one of the FR's pointsman's bothies. If this were where the level crossing keeper 'lived' then it would be a carbon copy of Lottie's crossing. If the Disc capstans were sited either side of the road then there would be no need to pipe wires under the road. From the perspective of this photograph it is difficult to divine the precise position of the bothy. I suggest that it is on the clock side of the line; fortunately we have a clue from a slightly later drawing.

Church Street level crossing, extract from XD97/429018 (turned through 180 degrees and annotated)

This drawing suggests that there was indeed a pointsman's bothy to control the level crossing: chronologically, the photograph of Church Street above suggests that the bothy was not moved down to the junction with the Festiniog and Blaenau Railway, but stayed guarding the level crossing. There are apparent suggestions that once Dinas (FR) had closed as a passenger station that there was a similar bothy up near the Welsh Slate Company viaduct - this may have come from Church Street Level Crossing after 1875 but much further research would be needed to validate that point.

It is also worth commenting that the siding running into the walled compound behind seems to make use of the former location of the Approach Disc to Duffws - the bottom end of the semi-circular enclosure can just be delineated on the Engine side - when the siding was extended into the walled compound behind Davies's office obviously part of the formation was built on the disc enclosure.

Just to try and round off the pre-signalbox status of Duffws, I've drawn two further layouts on top of a trimmed XD97/419014: on just Disc Signals and one Disc and Semaphore. Although the drawing has the same symbol for both signals, I feel that the most likely solution would have been an Up Disc and an Up and Down block Semaphore.

Church Street level crossing, extract from XD97/419014, Disc Signals only.

Church Street level crossing, extract from XD97/419014, this seems the most likely arrangement.

So why is the Semaphore and Disc the most likely arrangement? Well, both XD97/419077 and XD97/429018 shew a square signal post. If it were a Disc signal then I'm sure a round post would be drawn. I cannot, at this remove, explain why XD97/419014 uses the quartered circle symbol both sides of the road: it may just have been the convention of the draughtsman.

Just before we move onto the signal box 'proper' there are a few more points to be addressed............

XD97/429018 - pre-1870 track layout accentuated in black

XD97/419088 post 1870 station throat

If you look carefully into XD97/419088, you will just notice that the Pointsman's Bothy is still by the road; there have been some other alterations with the sidings on the Engine side of the line and what appears to be a new semi-circular alcove in the same fashion as the earlier one for a Disc signal, but this time it is opposite the remaining gatepost. The reworked siding connections, with a single lead into the Davies' and Jones' Coal yard appears to be the same gap in the wall that Bleasdale photographed in 1887, and the 'gap' at least appears in the December 1877 signal arrangement plan mentioned further up this page. Why would there be further excavations into the tump? Perhaps it is a drafting error, perhaps it is an accurate reflection of the simplification of the connections in the area immediately prior to the building of Church Street bridge.

This was in a recess in a retaining wall opposite the GWR Exchange platform and close to the Queens Road Bridge. It had 14 levers. Levers disconnected from the signalling apparatus w.e.f. 10am, Wednesday 29th August 1923 and new catch point opposite the former signalbox put in. Final recovery of equipment decommissioned since 1923 took place during the first half of 1929.