Tangmere

i think both the RAIB and all the posts here have failed to regard the significance of the other key namely that which goes through the tapered 'inside' end of the gudgeon pin and into the crosshead preventing same from turning. there is quite complicated method in fullsize of securing this key but totally missed by the RAIB. this key goes into the crosshead and then into the tapered end of the gudgeon pin preventing the gudgeon pin from either falling out if the nut and cotter on the 'front' comes loose and falls off, and prevents the gudgeon pin from turning.

or have i missed something?

cheers,
julian

 

Is it wrong to think that repairs to Tangmere shall take a bit longer? Seeing as to the locomotives inconsistency have really been going on since last summer.

 

Steve

I mentioned the cotter should be vertical in post 324 .

 

Julian,
I have checked the drawings in the report and as expected the key you refer to only stops the gudgeon pin rotating. I think the fixing you are referring to is a screw to hold the key in the gudgeon pin. To fit the pin in the crosshead the keyway in the crosshead would need to be open at the rear and hence not stop the gudgeon pin falling out. If you enlarge fig 4 p15 you can clearly see the open keyway in the lower view of the crosshead

Dave

 

There is no screw needed to hold the key in the gudgeon pin. Once the key is fitted, it cannot go anywhere.

 

See fig 7 p21 and there is provision for a screw to hold the key in place using a 1/2 in whitworth screw. Needed or not a screw is shown on the drawing

Dave

Oops this shows the value of more than one view. Re looking at the drawind there is a drilling through the gudgeon pim, an oilway?, which would need to be plugged to hold the oil in. It just looks like its holding the key

 

I'd suggest that the screw blanks off the end of the oil passage drilled through the centre of the gudgeon pin, with communicatiom drillings down to it from the small tapered end and in the middle up to the journal surface.

 

At a totally uneducated and ill informed guess I wonder if there might be a lot of copies of drawing 7/SL-DN-K 569 pinned to walls above workbenches and a lot of replacement cotter pins made over the next few months...

 

I guess I don't understand how that latter happens? The change means that the cotter is parallel to the motion of the crosshead rather than at right angles, but how is that going to change significantly the forces exerted on it by the nut (in trying to rotate)?

On thinking quickly, the magnitude of such forces might vary over the motion's cycle, but since the forces are constrained by the thread to operate in a circular way (i.e. the only significant resultant motion on the part of the nut can be to try and rotate), and since the interaction of the nut and cotter is constrained by their mutual geometry (which should constrain the directions of the forces on the latter produced by their interaction), so on that front I don't see how the orientation of the cotter will change anything?

Although (thinking out loud) should the gudgeon pin become loose, it might change the forces transmitted through the nut to the cotter pin?

Or are you thinking that the inertial forces on the cotter (of the mass of the cotter pin itself as it moves through its cycle) are now different because it's oriented so that those are parallel to its length, rather than crosswise?


I'm not sure (without detailed numerical modelling that is beyond my capabilities) that any of those effects would be large enough to be a problem, but my intuition says not.

And as the report points out (para. 50) there are lots of locomotives running around out there with horizontal cotter pins. And FWIW, the RAIB called out the cotter pin design as something to check - but not the orientation of the cotter pin.

Noel

 

As someone on The Works put it "45 pages to say it's just one of those things"

Still having trouble with the embedment thing. I can understand the idea of surface roughness being deformed during loading to increase bearing area, but none of the surfaces mentioned are plain compression such as washer on plate. The taper is basically sliding linear and the other two are sliding rotary - the thread and the nut on the crosshead. Any roughness would be wiped off as the metal slides. If the metal continues to distort after tightening is finished, that is creep. There's no mention of friction either, during tightening or release. Ah well, I guess we'll never know unless someone finds the nut, or bits of it.

 

If they believe that then they need to read the report again *very* carefully. The RAIB reports are ridiculously verbose and contain far too much duplication, but that report most definitely does not say "one of those things". I'm no expert in this stuff at all, so take what I'm saying with a very large pinch of salt, but it wouldn't surprise me if there was more to come from other bodies involved in rail regulation.

 

I went looking through photos of various Bulleids (keeping to them for comparative purposes):

34007 has plain nuts and vertical sawn cotters
34016 has castellated nuts with square slots and vertical sawn cotters
34046 has plain nuts and vertical cotters that I think are sawn
35005 has castellated nuts with vertical cotters (I think, not the clearest shot)
35028 has plain nuts, and cotters folded from strip, with an eye at the top and possibly welded up below the eye (as per the RAIB report)
34067 we know had horizontal sawn cotters and castellated nuts.
Then I went to the books and found a picture of a rebuilt crosshead dated 1963, with the same arrangement as 35028. That didn't surprise me much.

34067 has probably done more mainline miles than any of the others with the castellated nuts, but I doubt the nut is the issue, even with square slots (although I don't like them). I tend to think that the main issue with the horizontal cotter is that once the legs break there isn't even gravity holding it in, and every time the piston reaches the relevant end of the cylinder it's going to try to leave. A vertical cotter will take a bit more time to leave.

Anyway, thought experiment here. We take the cotter out on A.N. Other Bulleid at Waterloo. How far does she get down the SW main line before the pin comes out? I'm not advocating trying this, I'm just wondering whether the cotter is vital or insurance. If it's vital she'll get as far as Tangmere did. If it's insurance, then I think it's more likely that the nut and pin got loose enough that they broke the poorly made cotter, leaving nothing to hold it all together.

Cards on the table: I've never worked on full size steam (or anything bigger than 5" gauge). I am a MechE, but working with very different stuff. Part of the reason for my comments on En3B is that the old spec is so loose that it's practically shorthand for 'that crappy bit of mild steel we found out back'. Yield for EN3B is about half tensile strength, and EN8 (or 080M40) has a higher yield than EN3B's tensile strength. EN16 and EN24 are very good stuff, but in percentage terms they're not as big a step forward over EN8 as EN3B is a step back.

For bolts it's fine, as long as you accept the low strength, and it'll give a long way before actually failing. This isn't a bolt though, I think the most important thing for the gudgeon pin is that it grinds or laps to a good finish, and then stays the same shape. Given the low yield and cylic loading on the gudgeon pin, I wonder if EN3B would start to work and deform at any stress concentrations, reducing the ability of the taper to hold and the nut to preload it, especially if you put the cotter back in the same slot with no changes after taking the pin out to replace the connecting rod bushes. I'm speculating on incomplete information and little experience, there are plenty of people on this forum who probably have a much better idea of things than I do.

 

Also from the Works it is said the modified arrangement was done during the latter restoration period on the basis of the owners representatives spec and drawings.
This being the case is there no obligation to verify that it met the required standard? Assuming the client brief was to meet mt276 or subsequent group standards it strikes me as odd that no further professional enquiries were seemingly made.

 

Who is'seriously questioning' the RAIB report? Several people (me included) are discussing it, which is no bad thing, and most of us that are doing so have some relevant experience. My humble qualifications are that of a Chartered Mechanical Engineer and, although never professionally involved, with over 50 years experience of steam locomotive operation and maintenance. Not an expert but, I like to think, with a pretty good all-round knowledge of practical engineering.

 

The forces that the cotter is not designed to take are the inertia forces along its axis. This is the equivalent of pushing it back and forth in its slot and, if these forces exceed the frictional resistance of the cotter against the nut& slot it will eventually start to move back & forth. That would not happen with a vertical cotter. If it starts to move back & forth it is effectively hammering the legs and the head of the cotter and, eventually, something is going to give. To me, the prime factor in this incident is the fact that the cotter was horizontal. The design and execution of the cotter itself may be poor but I suspect that it would have continued to do its job satisfactorily if it had been vertical. The RAIB report gives a causal factor as being the design changes without full consideration of the effects and in its recommendations it says that the design of cotters fitted to steam locomotives should be reviewed but it stops short of saying that gudgeon pin cotters should be set vertically, which I consider to be the most important point to come out of the investigation.

 

I agree with Steve that there are a number of people discussing the report. The only "question" of the reports conclusions is the relative importance of the gudgeon pin material and its manufacture. Whilst my professional training and experience is in Electrical Engineering most equipment is mechanical in operation so even Electrical Engineers need a solid grounding in Mechanical Engineering including surface finish and metalurgy. Also over 40 years maintaining my own cars gives a good insight to fit and function.

The "experts" as you label them, have simply queried the conclusion that the gudgeon pin material was not causual. Dismissing the making of a critical component out of an inferior material is the only questionable part of the report.

 

I don't think we're saying that anything they said is wrong (e.g. about fabricated cotter design), just wondering if they considered all the potential factors. I personally think it's a pretty good report. But if it missed a thing or two, in the long run it will do nobody any good if we have to learn them the hard way, later on.

Noel

 

I am not a qualified engineer or locomotive mechanic but - as someone who worked in the steel industry (including administering the library of BS specifications which were maintained for the local steelworks and tube manufacturing plant) for some time I wonder if someone has got confused with specifications ?

EN3A / 07OM20 is the old BS970 specification for mild steel used for general purposes. Suitable for lightly stressed fasteners ,shafts etc. Can be easily machined and welded. Available hot rolled, normalised, cold drawn or turned. Supplied in Black round or square , bright round square, flat and hexagon. Presumably this replaces the 1949 (LMA recommended) BS Report 24 Part 4 No 3 which is the recommended steel for springs (Helical, Volute Laminated) on various sections of the locomotive but NOT for any parts involved with the Coupled Wheels; Cylinders and Motion of steam locomotives

EN8 / 08OM40 is the base BS970 specification for a medium strength steel. Suitable for stressed pins, shafts studs, keys etc. Available as rolled or normalised. Supplied as square bar, or round bar or flat which has the 4 classes EN8A (08OA35); EN8B (08OA37); EN8C (08OA40) and EN8D (08OA42) that I understand equate to the 1949 BS Report 24 Part 4 Number 8 Classes A, B, C, D respectively. According to LMA this specification should be used for Gudgeon Pins with either Class 8A or 8D material used and should be the material used for 34067's motion.

Worryingly to my mind Paragraph 77 (and note at bottom of page) of the RAIB report notes the use of lower quality steel but its statement that "The use of a lower grade of steel than originally specified, although it may wear more rapidly, is unlikely to have been significant in practice, as the gudgeon pin and nut are not highly stressed" (my italics) does bring into question the competence of the "expert(s)" who allowed that statement to be made and printed.

This is not meant as a criticism of WCRC as the maintainer of the locomotive as that company can only work on the basis of paperwork supplied when the locomotive was purchased and therefore attention needs to be drawn to the importance of that paperwork trail in identifying WHAT was done by WHOM and WHEN; in fairness the RAIB report has identified this importance but offers little information on what action to takes in cases of deficiency thus leaving the owners / maintainers / overhaulers in some limbo asto whether to accept the old adage of "If it ain't broke - don't fix it" or take the expensive step of "If in doubt - replace!"

As many have noted the RAIB report is like the curate's egg - good in parts but suspect in others; to my "untrained" mind there is still some doubt as to whether all the factors involved have either been identified or investigated"