The reversing rod on North Star 4-4-2

[buffalo]

I'll leave other modellers to say what interests them. One might ask why RTR manufacturers have yet to implement open frames and working inside motion. When they do, I suspect porcine aviation will have been commonplace for some time. On the other hand, many modellers do include such working items.

 

Nick

[DCB]

I think K.J.Cook refers to the difficulty in setting the valves as the reason for abandoning scissors gear in "Swindon Steam" I can't actually find the book at the moment but I am sure he refers to it having to be removed to the forge for 'Adjustment' if required. The GW 4 cylinder locos had a very precise angle in their rocker arms to get even valve events on front and back strokes of outside and inside cylinders with Walchaerts gear.    Swindon had a great deal of experience of Walchaerts gear, MSWJR no 4 was one of the first locos in the country with this gear and the GW had 3 French Atlantics and a whole fleet of Railmotors with this gear before 4001 etc were built.  But why inside gear, Maybe because it would have been ugly? as per LMS 6200 and 6201?  Or because the outside return crank is not as sound engineering solution as an inside eccentric?  

[MarkAustin]

 

MarkAustin, on 14 Apr 2014 - 19:42, said:

It has become fashionable to claim that Gibson in his writings is unreliable...

It's nothing to do with fashion or reliability, Mark, simply critical appraisal that should be applied to any source and to any views purporting to be derived from those sources.

 

Although I agree that all sources must be subject to critical analysis---the first rule of research is "only trust primary sources"; the second is "only trust a primary sources when confirmed by another", I believe that Gibson's views tend to discarded not for that reason but because he dares criticise the Great Man---Churchward.

 

 

Quote

...He worked at Swindon Works, admittedly 20 or so years after the design of the Stars. However 20 years is not sufficicient to remove all people with first hand experience, and the evidence is overwhelming that he was both knowledgable and curious about GWR locomotive practice. It beggars belief that he could have worked on the valve setting gang---as he did---without speaking to "old hands" who remembered the "old days". Indeed he quotes the chargeman as saying that it took ten days to a fortnight to set the gear of North Star, as opposed to a day or so for the others. I agree with him that this was acceptable for a single engine, requiring attention every two years or so, but not acceptable for a substantial class...

No one denies that he worked with and spoke with old hands nor that the valve gear on North Star was difficult to maintain. Perhaps the only surprise is that it was not rebuilt with conventional gear earlier. As you say, maybe acceptable on a single engine, but it never was applied to an entire class, only to a one-off experimental engine. The simple fact is that, for whatever reason, Churchward did not continue to use this gear when the Stars were built. Churchward, like Dean before him, made great use of experimental prototypes, and it is the nature of experiments that some features will be retained and others abandoned if and when production ensues.

 

I agree that Churchward was a great experimenter, but---unlike Dean---most of his experiments involved comparisons unless simply copying and modifying existing designs. If he was experimenting with valvle gear, I would have expected a second engine with Waelscharts gear, inside or outside. There is simply no evidence this was an experiment, but a solution to the problem of putting the valve gear inside, presumably for compatibility of appearance with the other designs.

 

 

Quote

...He also quotes the evidence of his father---who knew Charles Mortime, one of the GWR's Directors---who when hearing Churchward's locomotives praised, responded "Unnecessarily big and heavy, causing us to spend too much money on track and bridges", and added "and too expensive to build". This might have been a minority opinion on the Board, but I do not believe it would have been expressed if not widely held...

An unusual association of "minority view" and "widely held". Of course it was a minority view and there's no reason for surprise that there were minority views on the Board, 'twas ever thus. Fortunately, it did not prevail. Can you imagine what a mediocre railway the GWR would have become without Churchward's larger designs? Think Midland small-engine policy.

 

I used widely held, perhaps confusingly, to imply that it was not a majority opinion, but one held by a substantial minority. However, there clearly was, on the Board at least, opposition to Churchward's scheme---misguided opposition as you state and as I believe (with hindight it is obvious that by Churcward's time conventional locomotive design had reached a dead end)---however, it existed, and a visible backtracking on what was felt to be the keystone design could have increased that to the ectent of threatening the scheme, and perhaps Churchward's position. This had to be avoided.

 

 

Quote

...So, Churchward bungled with the Scissors gear--and I believe it is a reluctance to admit this that causes Gibson to be downrated---his designs had opponents at the highest level. He needed to do something that would not be an admission of failure, if only in this one small area. Hence, inside Waelscharts valve gear, with the flimsy excuse that  the two sides could be disconnected to allow an engine to work home on one side only...

 

"bungled", "flimsy excuse" and "admission of failure" are ridiculously emotive terms for a rejected experiment. Remember that the method of driving the valves by rocking shafts was retained and became the standard on subsequent GWR four cylinder designs. A critical reading of Holcroft, himself a not entirely uncritical admirer of Churchward, suggests that the decision was primarily to avoid paying fees or potential litigation from Derby, though it is debatable whether Deeley's patent would have stood up in court given the prior art dating back to the mid-1860s and specific differences between Deeley's design and Churchward's implementation. Had Churchward been particularly keen on the scissors gear, it's not unreasonable to suggest that he could have come up with an implementation that was sufficiently different from Deeley's design. As to the "flimsy excuse", Holcroft appears to present this as a minor beneficial side-effect of the modified Walschaet's implementation, not the primary reason for the change. An engine with the scissors gear would not be able to move at all under it's own steam in the event of certain failures. The Walschaert's gear was modified so that the reversing system on the failed side could be pinned in mid gear. That said, is there any record of the scissors gear failing?

 

I stand by my words. Churchward, with the scissors gear, made one of his rare mistakes---and then covered it up with the, as far as I am aware, only examples of inside Waelscharts gear.

 

Churchward had already rejected Deely's patent claim, so this cannot be used as a reason. In any case both Churchward's, Stevarts' and Joys' gear predated Deeleys patent, so it would have failed on "prior art".

 

The explanation of being able to work home on a single side is, in my view, a flimsy ex post facto justification. Can anyone produce a single instance where it was done? I doubt it

 

Gibson's views are interesting in many ways, not least because they give a wider voice to some of Churchward's detractors. They should be read and respected for what they are, not distorted in an attempt to undermine others.

 

Where am I attempting to undermine others?

 

My case is that Gibson gives a clear, logical and internally consisant explanation of why the scissors gear was anandoned, and inside Waelcharts gear adopted, and he was clse enough in time to have had the information form those who knew and I believe that his opinions are being unfairly discounted.

 

Mark A

[The Stationmaster]

Firstly I can't find the reference in Cook's book but I don't have the time to read every single page of it and alas it doesn't have an index.

 

Now to the scissors valvegear - why shouldn't Churchward experiment with it?  The GWR already had engines running with inside Walschaerts gear and with inside Stephenson gear so why not experiment with what amounted to an adaptation of Stevarts gear - which had been found perfectly reliable in its application on ships?  The gear undoubtedly offered some design advantages but it obviously quickly came to light that it had disadvantages - but, surely, is the whole point of experimentation and exploration of different methods and while valve setting in works is one thing the operational disadvantages probably carried more weight as they happened everyday.  

 

What happens in works setting up a valve gear is one thing and while the time taken has a cost that is small beer and isolated compared with the time an engine spends in traffic.  But, while it seems never to have crept into the various books written by those with an involvement at design and works level, there is firm, primary source, documentary evidence that the gear was difficult to access for oiling and was strongly disliked by Enginemen.  Ensuring correct lubrication correct is surely a fairly critical process in everyday operation and is quite likely to be treated in a less satisfactory manner if it is more difficult to perform than it need be?  And equally there is evidence that an engine with the scissors gear could not be run with one side of the gear inoperable - while there are more than a few examples in the real world of engines getting home with one side of their valve gear disconnected or partly taken down.

 

The other thing conveniently forgotten in this debate is that 'North Star' was built as a 4-4-2 - again an experiment and to compare it with the French engines.  Yet it was found that arrangement was not as successful as making the engine a 4-6-0 although admittedly it extended over more engines than the trial with the scissors gear.  and of curse Gibson also mentions Churchward's trial of the Cole superheater - quickly dropped because of accessibility problems.

 

Gibson also criticises what he regarded as a failure to use outside valve gear on the 'Stars' but it is well known, and amply proven in later years in practice, that Churchward was after the sort of bearing surface sizes which were not possible on an outside valve gear.  Everyday experience in and after WWII with the 'Castles' in particular amply proved the validity of that approach the inside components were not only able to stand up to extremely rough workaday conditions but could do so with scant regard to normal oiling procedures with oiling by bucket, instead of feeder, being commonplace at at least two large sheds.

 

As ever it is always relatively simple to look at things through main works eyes but as I have said engines spend a relatively small part of their working lives in such places.  What counts is their ability to perform as expected out on the road and to be able to undergo everyday servicing and preparation.  And in the latter respect 'North Star' would have been quickly found to present problems - somehow I can't believe that Enginemen and Swindon's Inspectors said nothing about the difficulties of oiling 'North Star' until the known evidence of the early 1920s.  

[buffalo]

Although I agree that all sources must be subject to critical analysis---the first rule of research is "only trust primary sources"; the second is "only trust a primary sources when confirmed by another", I believe that Gibson's views tend to discarded not for that reason but because he dares criticise the Great Man---Churchward...

Sorry, Mark, that's just a straw man argument. Nothing that I or others have said here supports the view that we are uncritical Churchward fans or that we question some of Gibson's statements for such a reason. As to multiple primary sources, what a luxury that would be?

 

...I agree that Churchward was a great experimenter, but---unlike Dean---most of his experiments involved comparisons unless simply copying and modifying existing designs. If he was experimenting with valvle gear, I would have expected a second engine with Waelscharts gear, inside or outside. There is simply no evidence this was an experiment, but a solution to the problem of putting the valve gear inside, presumably for compatibility of appearance with the other designs...

No 40 was an experimental prototype. You can tell that from it's number alone. As with most others, there were significant differences between prototype and subsequent production classes. As Mike has already said, the comparisons made were with the Frenchmen.

 

...a visible backtracking on what was felt to be the keystone design could have increased that to the ectent of threatening the scheme, and perhaps Churchward's position. This had to be avoided...

...I stand by my words. Churchward, with the scissors gear, made one of his rare mistakes---and then covered it up with the, as far as I am aware, only examples of inside Waelscharts gear...

"visible backtracking", "keystone design", "mistakes", "covered it up", surely you exaggerate? Frankly, I find your assertion that Churchward's position might be threatened by a single change in design is quite laughable. The language used to support it in this and your previous post owes little to critical analysis.

 

...Churchward had already rejected Deely's patent claim, so this cannot be used as a reason. In any case both Churchward's, Stevarts' and Joys' gear predated Deeleys patent, so it would have failed on "prior art"...

Where does the "already rejected" come from? Deeley's patent was not granted until June 1906, a couple of months after No 40 was completed. Otherwise, much what I said earlier, although Holcroft adds that Churchward's implementation was "...so near to that of Deeley that he could claim an infringement." This might well be the case if Deeley's patent was judged to refer only to a specific implementation of the gear. However, as I said before, it would not have been beyond Churchward to develop a sufficiently different implementation should he have felt the need.

 

..The explanation of being able to work home on a single side is, in my view, a flimsy ex post facto justification. Can anyone produce a single instance where it was done? I doubt it..

Again, much what I said. Holcroft's description supports this and does not present it as a reason for the change. Instead he mentions that the Walscaert's gear was adapted to avoid such a problem. One might ask whether they would have done this if was not seen as a potential benefit.

 

...Where am I attempting to undermine others?

I may be misunderstanding you, but talk of "bungled", "admission of failure", "mistakes" and cover-ups might reasonably be read as a quite unsupportable attempt to undermine Churchward's reputation.

 

My case is that Gibson gives a clear, logical and internally consisant explanation of why the scissors gear was anandoned, and inside Waelcharts gear adopted, and he was clse enough in time to have had the information form those who knew and I believe that his opinions are being unfairly discounted.

One can be "clear, logical and internally consisant" without being right. What was it you said earlier about primary sources? Gibson is not a primary source for what was done at Swindon in 1905-6. For that period he mostly repeats anecdotal material received from others and, quite probably, is selective in which material he includes.

 

Nick

 

ps. thanks to Mike for filling many technical details.

[DCB]

Re HowardGWR's post the Williams valve spindle movement derives from a vertical pin through the piston rod operating in a loop attached to the piston rod and thus bashing it back or forwards at the end of its throw.  It must look very good at starting on 75% cut off but I don't give much for its chances down a scale Dauntsey Bank.  I think Guy Williams book mentions a 15XX class which had Walschaerts gear which moved from forward to reverse gear.   Maybe the model trade's reluctance to represent inside valve gear is its inaccessibility for oiling....

[HowardGWR]

Re HowardGWR's post the Williams valve spindle movement derives from a vertical pin through the piston rod operating in a loop attached to the piston rod and thus bashing it back or forwards at the end of its throw.  It must look very good at starting on 75% cut off but I don't give much for its chances down a scale Dauntsey Bank.  I think Guy Williams book mentions a 15XX class which had Walschaerts gear which moved from forward to reverse gear.   Maybe the model trade's reluctance to represent inside valve gear is its inaccessibility for oiling....

Indeed, but of course it is just an invisible whirr at such speeds.   Most models I see at exhibition (including Pendon) move very sedately and in fact, on many layouts, they hardly move at all!

 

I do note some very forcible comments on this thread.  It's almost as though some have shares in the GWR still!     

[MarkAustin]

Firstly I can't find the reference in Cook's book but I don't have the time to read every single page of it and alas it doesn't have an index.

 

Now to the scissors valvegear - why shouldn't Churchward experiment with it?  The GWR already had engines running with inside Walschaerts gear and with inside Stephenson gear so why not experiment with what amounted to an adaptation of Stevarts gear - which had been found perfectly reliable in its application on ships?  The gear undoubtedly offered some design advantages but it obviously quickly came to light that it had disadvantages - but, surely, is the whole point of experimentation and exploration of different methods and while valve setting in works is one thing the operational disadvantages probably carried more weight as they happened everyday.  

 

What happens in works setting up a valve gear is one thing and while the time taken has a cost that is small beer and isolated compared with the time an engine spends in traffic.  But, while it seems never to have crept into the various books written by those with an involvement at design and works level, there is firm, primary source, documentary evidence that the gear was difficult to access for oiling and was strongly disliked by Enginemen.  Ensuring correct lubrication correct is surely a fairly critical process in everyday operation and is quite likely to be treated in a less satisfactory manner if it is more difficult to perform than it need be?  And equally there is evidence that an engine with the scissors gear could not be run with one side of the gear inoperable - while there are more than a few examples in the real world of engines getting home with one side of their valve gear disconnected or partly taken down.

 

The other thing conveniently forgotten in this debate is that 'North Star' was built as a 4-4-2 - again an experiment and to compare it with the French engines.  Yet it was found that arrangement was not as successful as making the engine a 4-6-0 although admittedly it extended over more engines than the trial with the scissors gear.  and of curse Gibson also mentions Churchward's trial of the Cole superheater - quickly dropped because of accessibility problems.

 

Gibson also criticises what he regarded as a failure to use outside valve gear on the 'Stars' but it is well known, and amply proven in later years in practice, that Churchward was after the sort of bearing surface sizes which were not possible on an outside valve gear.  Everyday experience in and after WWII with the 'Castles' in particular amply proved the validity of that approach the inside components were not only able to stand up to extremely rough workaday conditions but could do so with scant regard to normal oiling procedures with oiling by bucket, instead of feeder, being commonplace at at least two large sheds.

 

As ever it is always relatively simple to look at things through main works eyes but as I have said engines spend a relatively small part of their working lives in such places.  What counts is their ability to perform as expected out on the road and to be able to undergo everyday servicing and preparation.  And in the latter respect 'North Star' would have been quickly found to present problems - somehow I can't believe that Enginemen and Swindon's Inspectors said nothing about the difficulties of oiling 'North Star' until the known evidence of the early 1920s.  

I am not aware of any GWR design prior to the second and subsequent Stars that had inside Walschaerts valve gear. Were there any? Indeed I believe that the Stars and their derivatives were the only---certainly the only substantial---class of locomotives with inside Walschaerts valve gear. Does anyone know anything to the contrary?

 

Mark A

[The Stationmaster]

I am not aware of any GWR design prior to the second and subsequent Stars that had inside Walschaerts valve gear. Were there any? Indeed I believe that the Stars and their derivatives were the only---certainly the only substantial---class of locomotives with inside Walschaerts valve gear. Does anyone know anything to the contrary?

 

Mark A

The GWR did indeed have engines with inside Walschaerts motion in traffic prior to the construction of No. 40 - so perhaps logical, as I have already said and you quoted, that Churchward should try another type of valvegear against them; after all it was a period of experimentation and development on the GWR.

 

There were many examples around the world of locos with inside Walschaerts valvegear of course - but for a late British example the LNER built quite a few in the post-Gresley era.

[buffalo]

Off hand, I don't recall an earlier example of GWR use of inside Walschaert's gear. Historically, Gooch and Stephenson types dominated, with Armstrong being responsible for the widespread adoption of Stephenson's gear. There were also a few examples of Joy's gear. Were there any Walschaert's examples amongst the absorbed engines or those from outside builders? Ahrons only mentions a few obscure examples on minor railways in the nineteenth century, whils it had long been in favour on the continent. He notes that by 1913 the GWR had more examples than any other British railway.

However, I don't think the distinction between the one-off scissors and subsequent use of Walschaert's that is the key point here. Most important was that, with the exception of Dean's rather unsuccessful pair of tandem compounds, the Stars were the first four-cylinder engines to be built at Swindon and they brought new challenges in fitting the gear between the frames. In particular, it was not possible to have adequate big end journals and the four eccentrics that the Stephenson gear requires. Instead, the Walschaert's arrangement as used on all GWR four-cylinder engines from No 4001 onwards requires only two eccentrics. The scissors gear was itself an adaptation of the Walschaert's gear in which the drive comes from the opposite crosshead rather than from an eccentric. The difference is in fact much less than some would appear to assume.

Incidentally, Ahrons provides a lengthy description with drawings of Deeley's gear as fitted to the Midland 990 class.

Nick

edit: ps. Ahrons also mentions the inability to limp home using the cylinders on one side, "...but this is a point of minor importance." So, this was known though not considered a major deficiency only a few years after the building of No 40 and the Midland 990s.

 

pps. added the word "inside" -- of course they had outside Walschaerts on the Frenchmen and railmotors as mentioned several times already.

[The Stationmaster]

I can't find any evidence of absorbed engines with Walschaerts valvegear  (prior to the appearance of No.40)  in the RCTS work. Thus the only ones with that gear which the GWR had were almost certainly the three 'Frenchmen' which were of course delivered against Swindon Lot Numbers and were thus GWr owned from the beginning.

[tomparryharry]

Poor Stefan!

 

And he only asked about a rod.........

 

Ian