Overmod this setup is quite different from the 'articulated' rods used by Woodard et al. in the Twenties (see Bruce, pp.218-219), which are sometimes seen referred to as tandem rods.
OvermodIn these, the side rods are all kept in a line...
timz Originally, the term "tandem rod" meant the side rod from the main driver to the next rearward one was in line with the main rod-- the same distance from the locomotive centerline. Roller-bearing rods are usually sort of "tandem" in that sense: one side rod inboard of the main rod and one outboard. You don't think that's part of the definition of "tandem rod"?
Originally, the term "tandem rod" meant the side rod from the main driver to the next rearward one was in line with the main rod-- the same distance from the locomotive centerline. Roller-bearing rods are usually sort of "tandem" in that sense: one side rod inboard of the main rod and one outboard. You don't think that's part of the definition of "tandem rod"?
The classical "tandem" or "extended" main rods (that is how Alfred Bruce terms them, somewhat tellingly) use a separate bushing across a forked main-rod end. The end of the following intermediate rod is the blade in the center of the fork, and so the rods are in line, but the salient point is that the thrust of the main to the intermediate crankpin passes only through the bushing, not the main-driver crankpin. This "about halving the direct load on the conventional main crankpin" (p.215) So while I am not certain whether 'tandem' refers to the doubled pin area, or to the distribution of the thrust across multiple crankpins, that is the area of the arrangement to which 'tandem' refers.
Notably, N&W does not refer to the double-intermediate-rod setup as "tandem rod" -- they call it 'dual' in all the drawing references. Chapelon does not hesitate in using 'tandem' to refer to this setup on the J class, but he mentions clearly what the tandem arrangement does elsewhere (LLAV, p.59) and has a good illustration (fig 8, p.60). Alfred Bruce of Alco pointedly avoids using 'tandem' when he describes the dual-rod arrangement as on the J and the Niagara (fig 37, #9), but his illustration of the tandem arrangement (fig. 37, #8) shows just about what Chapelon's does. [Note that just the inverse of this, with the blade being the main-rod big end and the fork being on the forward intermediate-rod end, is what I was kinda expecting for the N&W J "single rod" configuration...]
Note that this setup is quite different from the 'articulated' rods used by Woodard et al. in the Twenties (see Bruce, pp.218-219), which are sometimes seen referred to as tandem rods. In these, the side rods are all kept in a line by putting an extension on the rear of one rod, with a knuckle pin, and a fork on the engaging end of the following rod. These knuckle pins were notorious for causing problems, and no one appears to have been sorry to see them superseded. Bruce has an interesting reference to the 'updated' version of this used on the UP 800s (his fig. 29, #3) which you will note contains the same idea of intermediate bushings (but free-floating this time, and with no knuckles; Chapelon has a good picture of this from the side (fig. 29, p.91), and of one of the bushings with all the lubrication holes drilled in it (fig. 31, p.92)
feltonhillThese rods operated in a more constricted lateral space on the Class As than they did on the 4-8-4s
A picture is worth a thousand words:
Corresponding drawing for the class J is E45409, and the drawing for the intermediate pin arrangement is E45413. Neither of those appears to have been scanned... DAVE?????? ;-}
The J eccentric crank with single-rod arrangement is D45407, and the matching eccentric rod with SKF bearings is D45404 (and if anyone wants it, the main pin is E45408)
feltonhillso there wasn’t room for tandem rods.As long as the A's main rod drove the third axle, wouldn't "tandem rods" be impossible?
feltonhillso there wasn’t room for tandem rods.
No, you could do them quite easily; there is just comparatively little point on a six-coupled engine. You'd have a spacer the width of the bearing assembly in the main-rod eye keeping the dual rods separate and parallel at the rear pin. The whole shebang would still be outboard of the main-to-front rod, and that rod would still constrain the effective power going from the main drivers to the leading and trailing sets... the whole point of the tandem rods is to distribute thrust across the two 'middle' pins when it is so high as to imperil a single main pin's structure, and when the 'second' main pin is itself driving at least one additional pin... so at least eight-coupled to get reasonable gain from all those 'extra' bearings and fittings.
timz,
First question - yes, but I need to check the available drawings regarding centerline distance to find out for sure. The Class J center-to-center distance is probably greater than the Class A. I don't live in Roanoke, so it will be October before I get there again.
Second question - I don't know about impossible, but it certainly looks improbable.
Juniatha That’s introducing a new point !
That’s introducing a new point !
Not intended as such, in the previous context ... but as I said, I'm not commenting on any further aspect of the matter right now, and I agree that we should suspend further discussion for now ... especially as I do happen to agree with most of your points in general principle.
... see if you would like to post something only to have another person go over your text scrutinizing for words that *could* be misinterpreted and promptly writing critical comments on what you didn’t mean and didn’t write . Or , as when I posted those drafts of a SE Mallet and Triplex variation as a contribution to matters *deep firebox behind drive wheel sets* you wrote a laconic ‘wipe out correction’ on an allegedly faulty position of Delta truck without offering an alternative , less so an improvement . Or as when , without substantiating your assumption , you called it a ‘misconception’ when I commented on higher flange forces with Mallet locomotive running in reversed set-up with fixed engine first – note according with the work of Heumann and others still providing basis for understanding railway vehicle guiding and tracking .
Strange -- I distinctly remember apologizing for all three of these specific issues.
Of course, I did in fact propose alternatives concerning the trailing truck design and geometry, and would have no difficulty going into my usual MEGO-inducing detail on that particular subject if necessary. But as I thought we had agreed, that's not really a major issue for an Internet board.
I am not quite sure why you disparage Johnson on balancing and then try to quote Heumann as a current authority on vehicle guiding. (For the rest of the people who care to follow the general topic, here is a convenient introduction to the relevant history; it's by no means current but will at least get interested people up to speed on some of what's involved.)
However , as I said in the beginning : I think we should better drop these controversies for good now – fighting issues on details of locomotive of over 60 years ago is not what I’m here for , it’s neither educating nor enlightening .
I thoroughly agree ... with the peripheral note that this particular discussion of rods is NOT so much 'ancient history' as a discussion of the state of the rods now on 611, and how they might best be restored (or improved) when the locomotive is...
If anyone is interested , the PRR T1 counterbalancing problems were not a secret. The problem was apparent on one of the two prototypes. T1 6110 was selected for investigation by the Test Department. Coming into the tests 6110 was cross-balanced with 19% of the reciprocating weight balanced. When it was first run on the test plant, the longitudinal vibrations were too severe to continue. No. 6110 was then statically balanced. After completing this modification, PRR estimated that 56% of the reciprocating weight was balanced. This greatly reduced the longitudinal vibration, although it was still severe approaching 80 rpm (about 19 mph), the critical period of the dynamometer spring. This did not interfere with the tests because the lowest speeds runs were 160 rpm or 38 mph. The above is paraphrased from the Altoona test report.
This is not meant to be an exhaustive reply to all the preceding details. It‘s intended to be additional perspective regarding how N&W worked and why, and also clear some of the dust that’s been raised on this thread..N&W converted at least four of the Class J 4-8-4's from tandem intermediate rods (between #2 and #3 driving axle) to single rods from about 1952 to 1956 (600, 605, 610, 611). How did this evolve and why did they do this?Initially, all modern 4-8-4's that were equipped with roller bearing rods used the so called tandem rod connection between #2 and #3 axles. In addition, because of this configuration, all of them had relatively long crankpins on #4 axle. The examples are:N&W Class J (1941-1950)ATSF 3776 and 2900 classes (1944 estimated)NYC S1a, S1b and S2 Niagaras (1945)C&O J3a (1948).Further, there were a few 4-8-2's that were equipped with roller bearing rods, either as-built or retrofitted. Two known examples are:NYC Mohawks (a few, probably around 1943-1945)UP 4-8-2 (one or two estimated, date unknown)Because all railroads used the same method (tandem rods), this was likely the industry standard for handling the forces that had to be distributed through the running gear, particularly between axles #2 and #3 of a 4-axle engine set.N&W revised the Class J rods to do away with the tandem rod setup on four locomotives, but according to this discussion, there was no obvious reason. I believe there was a missing link that has not been considered, the N&W Class A 2-6-6-4.In 1948, N&W ordered the last five Class A’s and had them equipped with roller bearing rods. The design of these rods merits a closer look. The main rod drove on axle #3 of each engine set. Therefore, the force had to be distributed to two additional axles, the same as the 4-8-2s and 4-8-4s, except the axles were forward instead of behind the main crankpin. These rods operated in a more constricted lateral space on the Class As than they did on the 4-8-4s, so there wasn’t room for tandem rods. N&W and Timken designed an single intermediate rod instead. There was no precedent for this because none of the 4-6-6-4's built were equipped with roller bearing rods. This modification is visible in photos of the last five A’s. The rod hub on #2 axle is clearly larger than the rod hub on #1 axle. The design was ultimately successful because the final Class As were no slouches in the freight-moving department. If you want to hear what the last group of Class A’s performance sounded like (shameless plug here), a new CD, Time Freight, is available from the O. Winston Link Museum gift shop and the N&W Historical Society. See their websites.As part of my drawing check last week, I noticed that there was some commonality between the Class A and Class J rods, particularly the single intermediate rods. However, the Class A drawings were generally dated 1948, and the Class J comments were added from about 1952 through 1956.It looks to me as if N&W took what it learned from the Class A rod design, and applied it to the Class Js. It would be a design improvement if the #4 crankpin could be shortened by placing the rod closer to the wheel. All railroads noted maintenance problems around the #4 crankpins. This would also improve one of the equations used to cross-balance the Js (i.e., keeping the rotating mass closer to the locomotive centerline).If there is any doubt that this application was also successful, four of the last five Class J’s retired had this rod modification. In other words, all of the modified Js lasted to the end, regardless of their construction date.
Overmod
>> You did not need to comment twice; once would have been sufficient. <<
Obviously not – I *did* write once and it seems it was *not* sufficient – however let’s drop it for good now .
>> The ACL R1 problems have been well-described, and aside from preceding the publication even of Johnson's first edition <<
If engineering wisdom contained in his book only came to R.J. at a time the book was being published , then you have a point there .
>> I see nothing strange in allowing 'extra' areas for provision of balance mass.<<
That wasn’t the point . Point was *way* provision machined *subsequently* into *existing* wheels were to be *filled with lead* ( with byways touching how well this method could provide for re-adapting cross balancing properly )
>> Please tell me more about where PRR had nonunderstood balance problems with the T1 Duplexes <<
That’s exactly the sort of pharisaical argumentation I mean : I never suggested R.J.’s scope of engineering was confined to nothing but balancing . You know perfectly well where the problems with the Baldwin T1 design were , yet you insinuate a meaning in my words I didn’t intend and which wasn’t there .
( on # 611 ) >> the degree to which the mods were actually 'cost-cutting'. <<
That’s introducing a new point ! So far , focus was on *technical quality* of the simplified rods arrangement . You shift argument from your original position questioning my decent remark the modification was a step down in quality – not just as compared with the original tandem rod system but also in relation to an *in line* coupling rod alignment with knuckles . By intention not going into details , I briefly touched implications of this simplified arrangement which for a full rebuilding really would have demanded quite thorough changes to certain components ( for example just short of everything on main drive wheel set ) as compared to a probably simplified realization in view of probably eased demands with steam traction in its final stage .
If you feel my reaction was too harsh , I’m sorry . However , if you might want to put yourself in my place see if you would like to post something only to have another person go over your text scrutinizing for words that *could* be misinterpreted and promptly writing critical comments on what you didn’t mean and didn’t write . Or , as when I posted those drafts of a SE Mallet and Triplex variation as a contribution to matters *deep firebox behind drive wheel sets* you wrote a laconic ‘wipe out correction’ on an allegedly faulty position of Delta truck without offering an alternative , less so an improvement . Or as when , without substantiating your assumption , you called it a ‘misconception’ when I commented on higher flange forces with Mallet locomotive running in reversed set-up with fixed engine first – note according with the work of Heumann and others still providing basis for understanding railway vehicle guiding and tracking .
With regards
Juniatha
You did not need to comment twice; once would have been sufficient. And relax, none of my comments are intended to be about you.
I only quoted Johnson as a reference to how balance weights were secured, not as a reference on balancing. I do not think there have been dramatic changes in the tech of balance-mass securing since that time, but I am always prepared to be mistaken and receive enlightenment. If you think we need to disagree on matters of balance-mass integrity, we can do so.
The ACL R1 problems have been well-described, and aside from preceding the publication even of Johnson's first edition, if you were questioning his knowledge, were a failure caused by reliance on then-newly-modified AAR standards (which turned out to be ghastly wrong!).
Knowing what I do about the practicalities of casting locomotive driver centers in the 1940s in the United States, I see nothing strange in allowing 'extra' areas for provision of balance mass. Certainly we can do better today.
Please tell me more about where PRR had nonunderstood balance problems with the T1 Duplexes -- all the ones I know of involve suspension and difficulties associated with the duplex configuration itself.
I am still waiting for clear-dimensioned drawings, or a picture accompanied by direct measurements, of the arrangement applied to 611. Once I have actual, measurable evidence, I can fairly decide for myself the degree to which the mods were actually 'cost-cutting'. That is by no means to say that the single-rod mods weren't made with lower expense in mind, or that anyone else is wrong to conclude so. Only that I want better proof.
Big Jim
Your photo again is interesting and provides facts - *that's* what gets this matter on .
Why do I have a bee under my bonnet ? guess everybody is talking of hybrids and bio-fuel and all that , so I thought I might try bee-power - no ?
Earnestly : sorry , usually I am a gently agreeable person and maybe I should better not have replied at all , yet after taking so much , enough is enough . I think I'm prepared to discuss next to most almost everything and then some concerning steam locomotives , however if things are just being questioned for the like of questioning that's not getting things forward .
My 'question' was not really a question but my way of trying to make think twice about the *wisdom* of doing things the way they were done at that time *long ago* (!) .
You were right about your remark concerning LC 47 and the hollow castings ready to take up filling material to suit . I suspect these may have been early efforts at some minimum of unification of castings for possible use in various locomotive types - and thus naturally adjusting counter weights to each loco's reciprocating masses . However , as you look at these massive structures it is again hard to think these were calculated for *zero overbalance* as had also been suggested earlier - together with an aberrant idea of the influence of rpm speed on balancing .
I'm certainly not suggesting any idea of how the N&W 4-8-4 *was* balanced and how it was re-balanced with the revised rods arrangement since I do not have documentation about that . All I wrote was the revised single rod arrangement was *not as good* as the original tandem rod arrangement - that's all and I keep that up on sound engineering logics and reasons evident . I really don't see what needs to be questioned about that and why .
Overmod Juniatha ... how would you suggest this be done in a cast steel wheel without casting new wheels ? Ralph Johnson discusses this and so does Chapelon. The cast pockets for counterbalance and cross-balance in the driver are oversize, usually divided into 'cells' or segments by cast-in partitions, and closed off with a welded cover plate -- my T1 drawings have some interesting notes on how the welding was to be done, indicating that tinkering with the weight was not an uncommon procedure! The material used to cast the weights is specifically chosen so it would not shrink, or become friable. Once it was set in the pockets, it is not going to move.
Juniatha ... how would you suggest this be done in a cast steel wheel without casting new wheels ?
Ralph Johnson discusses this and so does Chapelon.
The cast pockets for counterbalance and cross-balance in the driver are oversize, usually divided into 'cells' or segments by cast-in partitions, and closed off with a welded cover plate -- my T1 drawings have some interesting notes on how the welding was to be done, indicating that tinkering with the weight was not an uncommon procedure!
The material used to cast the weights is specifically chosen so it would not shrink, or become friable. Once it was set in the pockets, it is not going to move.
Juniatha,I wonder why you have a "bee in your bonnet" (x 2?)? Your question was properly answered and you get mad about it? What's up with that?
.
Hello , Prof Overmod ,
Sorry if this may not live up to your expectations – yet it appears this has to be addressed :
I have written what I wanted to comment about this matter and have nothing to add to it . I wrote it *could be done* but it was *not the same* in technological level . The replacement was essentially a low-tech application and - sorry - what you post are some quite theoretical and some pretty optimistic scholarly views and expectations about how to do this – arguably aiming at my commenting these again and so on like a ping-pong game : why some propositions are not practical or why they don't offer what you expect of them and why I have not described in detail these or in fact any other theoretically possible solutions , suggestions or substitutions – last not least : why some of them are quite ill-founded . I’m sorry , this has become burdening , especially since I feel you just want to involve me in some sort of hassle or controversy no matter what . So , I'm not going to enter this sort of discussion , especially not when about pondering if and how a low-tech solution *could* be executed , simply because with the concerning photograph it has already become blatantly plain and clear it *was* done , full stop .
It seems to me you have developed a habit to put at question anything but anything I post . If you have different views on everything please feel free to retain your view , only : please don't expect me to defend , justify , explain or prove my comments thereupon - I simply don't have that much time to spend on these things . I'm awfully sorry yet it appears I finally need to put that clear . Objections like boring out and refill by casting an earlier cast structure and expecting this to form a solid unity ignores thermal expansion / shrinking and consequential thermal stresses . Same about the idea of welding a plate to the casting to cover up a 'hole' with lead filling . Ok , so the fill-up material might not get lost , yet when having become disintegrated by beats transmitted to wheels from rail joints , pieces of it are flung around , disintegrating more of the mass , while the engine goes through the paces accelerating until reaching a speed where centrifugal forces will have pieces rest in radial position ( while that still does *not* secure their position when rattling over rail joints with loose pieces may be at 0 , 90 , 180 , 270 degrees position relative to piston or any arbitrary position in between . Misinterpretations are not helpful , like , to quote >> While there may be some lateral difference in the bending stress in the pin, it would shift inboard, not outboard << Now , *I* *never* wrote it would be “shifted outwards” , this is an imputation ! and combined with an incorrect assumption of yours that taking away the outboard part of the tandem rod would cause but >> some difference in bending stress << but some ? dramatic underrating ! ( btw what do you mean by “lateral difference” ? the stress on pin by piston force or mass inertia already *is* a vector force and not laterally but longitudinally directed ! then : how could exactly *this* rod modification result in stress to be “shifted inwards” ? if that could have been done it would have been great for it would have set free of stress just the very part of the pin in fact most submitted to those forces - especially since the simplified rods layout *did* substantially increase shearing forces ! ) More examples could be added ..
If Ralph Johnson still is your standard let me just ask : why then did Baldwin repeatedly have trouble with balancing of big engines like the Atlantic Coast Line 4-8-4 , why did they cast wheels with precaution extra counter weight blocks to be machined to suit only after manufacturing and assembling the wheel sets ( obviously to take care of not-too-small tolerances in wheel castings ?) , why did the Pennsy have trouble with the Baldwin Duplexi and so on ...?
Overmod , look ; what Johnson had written , he did more than 1/2 a century ago - technology has long since moved on and with absolutely fantastic progress . So , in earnest , what do you want to prove by instigating controversies about trivial details of a low-tech simplification of a once well composed rods system ?
No insult intended - only , could we agree to drop that .
Overmod ,
I have written what I wanted to comment about this matter and have nothing to add to it . I wrote it *could be done* but it was *not the same* in technological level . The replacement was essentially a low-tech application and - sorry - what you post are some quite theoretical and some pretty optimistic scholarly views and expectations about how to do this – arguably aiming at my commenting these again and so on like a ping-pong game : why some propositions are not practical or why they don't offer what you expect of them and why I have not described in detail these or in fact any other theoretically possible solutions , suggestions or substitutions – last not least : why some of them are quite ill-founded . This has become burdening , especially since I feel you just want to involve me in some sort of hassle or controversy no matter what . So , I'm not going to enter this sort of discussion , especially not when about pondering if and how a low-tech solution *could* be executed , simply because with the concerning photograph it has already become blatantly plain and clear it *was* done , full stop .
Something else : you seem to develop a habit to put at question anything but anything I post . If you have different views on everything please feel free to retain your view , only : please don't expect me to defend , justify , explain or prove my comments thereupon - I simply don't have that much time to spend on these things . I'm awfully sorry yet it appears I finally need to put that clear . Objections like boring out and refill by casting an earlier cast structure and expecting this to form a solid unity ignores thermal expansion / shrinking and consequential thermal stresses . Same about the idea of welding a plate to the casting to cover up a 'hole' with lead filling . Ok , so the fill-up material might not get lost , yet when having become disintegrated by beats form rail joints , pieces of it are flung around , disintegrating more of the mass , while the engine goes through the paces accelerating until reaching a speed where centrifugal forces will have pieces rest in radial position ( while that still does *not* secure their position when rattling over rail joints while loose pieces may be at 0 , 90 , 180 , 270 degrees position relative to piston or any arbitrary position in between . Misinterpretations like , to quote >> While there may be some lateral difference in the bending stress in the pin, it would shift inboard, not outboard << Now , *I* *never* wrote it would be “shifted outwards” , this is an imputation ! and combined with an incorrect assumption of yours that taking away the outboard part of the tandem rod would cause but >> some difference in bending stress << but some ? dramatic underrating ! ( btw what do you mean by “lateral difference” ? the stress on pin by piston force or mass inertia already *is* a vector force and not laterally but longitudinally directed ! then : how could exactly *this* rod modification result in stress to be “shifted inwards” ? if that could have been done it would have been great for it would set just the very part of the pin in fact most submitted to those forces free of stress - especially since the simplified rods layout *did* substantially increase shearing forces ! ) More examples could be added yet
If Ralph Johnson still is your standard let me just ask : why then did Baldwin repeatedly have trouble with balancing of big engines like the Atlantic Coast Line 4-8-4 , why did they cast wheels with precaution extra counter weight blocks to be machined to suit only after manufacturing the wheel sets , why did the Pennsy have trouble with the Baldwin Duplexi and so on ...?
Let’s agree to drop that .
Regards
The material used to cast the weights is specifically chosen so it would not shrink, or become friable. Once it was set in the pockets, it is not going to move. The alloy Chapelon cites is an alloy of lead and antimony, although he does not specify the exact percentages; it would be easy enough to determine. In a modern application you'd probably use 'crankshaft-weight' material (tungsten-based) or perhaps even, like the 747, depleted uranium (probably in this application as the oxide) embedded in layers of castable metal.
Yes , I know , you could always take *away* mass from balancing masses - yet to replace a tandem rod arrangement by this sort of a conventional rods arrangement would rather ask for *increasing* wheels counter balancing masses.
I would be tempted to expect the difference to be minimal, as determining the section and construction of a single rod possessing the same mass and mass distribution as the pair of tandem rods taken together is not a difficult exercise. While there may be some lateral difference in the bending stress in the pin, it would shift inboard, not outboard (the contribution of the 'inside' tandem rod in hammer blow being considerably less than the outer one) so I would expect removing mass would be the more common. It might be desirable to drill out and recast a portion of the 'advance' mass for cross-balance if that method is used, but that would be comparatively simple.
I would suspect that machining additional space in a cast wheel center would be relatively trivial, as would welding partitions into any opened core spaces or exposed fillets, etc. This without measurably impairing the functional strength of the wheel disc, or weakening the rim support of the tire.
I can think of ways to jig this so it could be done on assembled wheelsets, e.g. sets ready for adjustment in the quartering machine, or rigged for dynamic spin balancing. Extending or reshaping weight pockets is nowhere near as critical as main-pin alignment... even should actually placing or ensuring distribution of mass in the pockets at balance time prove to be.
Now you might reply : they could bore out and fill with lead - I tell you the British Railways had a lot of trouble with lead in counter balancing weights getting brittle , loose and leaving : how much more trouble would it have provided with higher rpm in a *much* heavier engine on jointed rails ?
See comment above about alloying. As a further note, even if the proper alloy were to be beaten loose, and even if it were to be beaten around rather than primarily held to the outside radius of the pocket by 'centrifugal force', how would you propose to get it out past the 360-degree-seal-welded cover plate?
If there should prove to be a problem with casting new alloy into recesses in old alloy, I would try preheating, perhaps with gas blanket or submerged-arc-style fluxing; getting effective homogeneity between old and new mass is just not a major metallurgical issue as far as I can see.
Ralph Johnson mentioned that American problems with 'disappearing lead' in balance masses were an issue -- in the Nineteenth Century -- and noted that issue was substantially addressed relatively early on in the Twentieth. It might be amusing to hear from our UK (ex-)brethren about why they seem to have continued to have problems -- that might wind up reminiscent of the arguments often made in some British circles, even today, about why equalization is a pointless expense for imaginary gain on locomotives (and especially between coupled and trailing axles!), or why Franklin adjustable wedges are a ridiculous waste of money when hardened liners alone do the job so well. ;-}
In the addenda: Is a slight offset in the eccentric rod permissible with Multirol needle bearings? I'd think you'd have to skew the eye alignment slightly to keep the bearings parallel-loaded -- and in any case at high speed your buckling moments might be exaggerated (the tendency toward lateral deflection was bad enough already at high speed, ne?) The catch with longer offset on the crank itself (assuming you keep most of the crank mass itself as far inboard as possible) is that the longer outboard portion will itself act as a longer 'cantilever' and deflect worse in high-inertia situations, if made for minimum impact (pun not intended) on hammer-blow momenta.
Also , mind , to keep up degree of mass balancing , this sort of revamping of rods arrangement would have asked for an according revamping of balancing masses in wheels and thus wheels counter weights should have been modified - how would you suggest this be done in a cast steel wheel without casting new wheels ?
Feltonhill
What you wrote about N&W taking care to re-design the arrangement is self-understood . This was no toy engine , the mass and steam forces evoked by this engine's drive mechanism were substantial .
What I meant was : did they care to redesign in such a way as to keep *the equivalent* of the tandem coupling rod 2nd to 3rd driver ? From that photo I clearly say : no , they didn't . Why can I say that when I wasn't there ( and you weren't neither , btw ) ? simply because the photo tells me this simplified arrangement was *not up* to the former tandem arrangement as concerns taking up mass forces and distributing them between 2nd and 3rd driver pins - beefed up 2nd driver pin notwithstanding ; a certain beefing up they *had* to do , or the pin without its support by the coupling rod *outboard* the vertical plane of action of the con rod would just have shorn off after some full application of full effort of the engine . In case deplorably you thought I would suggest they had used the *same* ( identical , unmodified ) main pin as before , sorry , then you have missed my point .
Since you can neither post drawings nor give a detailed description nor have any data on dimensions before and after the change no substantiated comparison can be made and there is little in just writing N&W took care as they usually did . Everyone will suggest nothing less , as they usually recognize N&W steam . However , it's one thing to design and build an engine with best design in mind for an intended top passenger service and later to keep it *just* running in a late hour of steam traction when there probably was no more need to keep up full speed potential . Also , mind , to keep up degree of mass balancing , this sort of revamping of rods arrangement would have asked for an according revamping of balancing masses in wheels and thus wheels counter weights should have been modified - how would you suggest this be done in a cast steel wheel without casting new wheels ? Yes , I know , you could always take *away* mass from balancing masses - yet to replace a tandem rod arrangement by this sort of a conventional rods arrangement would rather ask for *increasing* wheels counter balancing masses . Now you might reply : they could bore out and fill with lead - I tell you the British Railways had a lot of trouble with lead in counter balancing weights getting brittle , loose and leaving : how much more trouble would it have provided with higher rpm in a *much* heavier engine on jointed rails ?
addenda :
>> it appears that a lot of time was spent on the #2, #3 and #4 crankpin and driver center design revisions ..<<
Why then didn't they apply the much better ( for conventional rod system !) coupling rods arrangement with articulation joints on middle coupling rod to which the outer coupling rods reached ? This allowed for all coupling rods to be aligned in one common longtudinal center line , no offset , no full thrust taken up by intermediate pins and transmitted to next rod sideways of engaging rod . This would also have provided for main pin to be further shortened - or con rod big end to be wider .
I think they preferred to retain existing end coupling rods - to wich the former tandem middle rods could also be stored as possible replacements , since all of these were same . Carefully as it may have been carried out , this marks the change over as a budget revision - such as all too often was the case everywhere in the later days of steam with diesels ante portas .
As to the 'offset' : with a shorter main pin it was no problem to adjust seating and dimension of eccentric crank and bearings and center line of eccentric rod to keep valve gear unmodified from everything in the valve gear mounting onwards . A slight offset from strictly longitudinal plane was acceptable in a long enough eccentric rod - with the simplified coupling rods arrangement , it seems this was no longer needed . No problem at this point at all .
The eccentric crank got moved closer to the locomotive centerline. I don't recall if all of the sideways angular displacement was removed or not. The Baker frame and parts were not moved. I didn't phrase that well earlier.
Hm-- what about the eccentric crank? When they removed the rod, did the eccentric crank get shifted closer to the engine's centerline? If so, did the Baker valve gear have to shift inward too? But it can't, if the valve stays where it was?
As promised, I went to Roanoke ths past weekend and looked up about 20 or so drawings of the single-rod version of the Class J's. Unfortunately, three days goes past very quickly, and although i looked at all the drawings, I didn't get a chance to compare them in detail. I was hoping to find an equivalent to the tandem rod drawing that Overmod posted earlier. Apparently, NWHS does not have a single rod version, or maybe it never existed. However, to address Juniatha's comments, it appears that a lot of time was spent on the #2, #3 and #4 crankpin and driver center design revisions, attendant revision to the rod and bearing detail drawings, as well as the eccentric crank lateral position and size (this was perhaps to reduce the angular displacement between the eccentric crank and the first link in the Baker valve gear frame. Beyond that, I can assure all that N&W had a lot of drawings related to this change and they appeared to think through the revision with their usual thoroughness. Overmod, when can you get to Roanoke and give these drawing a good going-over?? I'll be there if at all possible.
Thank you , Big Jim , for posting this revealing picture !
Now , from what I see , this arrangement clearly is a simplification aimed at easing maintenance , no regards to keep up full rpm speed potential - probably it was realized what had been built into it was a bit above what was actually demanded by daily traffic or by traffic remaining stream hauled by then . This *should* have gone together with new , strengthened main pins of larger diameter , demanding reboring of wheel etc - however had that expense been invested by that time with steam on the way out ?
Let your journey end.
OvermodI am still looking for an adequate online drawing of the single-rod arrangement
There is a good picture of the 'old' version of tandem rod (with forked main-rod end and heavy sections) in Chapelon's La Locomotive a Vapeur (in Carpenter's translation it is on p.60). The accompanying photo appears to have been taken from a Locomotive Cyclopedia. I don't have a source for a Web-enabled version of either picture to provide as an illustration here. Still looking forward to someone providing either a picture of or a direct Web reference to the N&W drawing(s) that show the detailed single-rod arrangement applied to the Js.
I suspect the N&W version, for reasons already well-discussed, would invert the general arrangement illustrated in Chapelon, with the fork on the forward end of the 2-3 rod (essentially duplicating the layout of the double rods as far as bearings and geometry on the main pin is concerned. I would be tempted to put a single bearing on the rear of the 2-3 rod and carry the 3-4 rod inboard of it, rather than forking the rear in imitation of the double-rod arrangement, which would greatly decrease the lateral offset required for the #4 pin. That would limit the amount of 'direct pass-through' of main-rod thrust to the #4 pin, but should give better balancing at high cyclic rpm...
BTW, there is a good technical drawing of the Schwartzkopff-Eckhardt bogie on pp.110-111 of the Carpenter Chapelon translation; it is Fig. 50.
( edit some sentences on roller bearings suplemented to be more explanatory )
Hi Overmod
Just shortly - I'm in a hurry . No numbering in this thread although we could start it .
>> Just to mention: so would lateral deflection of a thin-section rod,<<
It did – yet with the proportions , shorter pin of large diameter , smaller lateral width of bearing , it simply didn’t cause a problem . On scrutinizing inspection , white metal bearing would have had to be found having worn tapered – and this again inevitably caused progressive wear since full area and uniform pressure load was no more realized in such a bearing , from then on causing inwards end to wear faster when working light , outward end when working hard , all in all working towards a convex wearing pattern that was conductive to premature squeezing out of oil ; with these bearing having had no protection against loss , actually this being part of the concept , oil consumption was increased , lubricating condition was worsened and overall wear increased . Geometrical inexactitudes by many causes was why most steam locomotives developed slack soon after shopping .
Spherical mounting in roller bearing rod end – that’s another way of taking account of these deflections from right angle .
>> I am not sure how this is supposed to be prone to jamming merely because the fork is wide enough to accommodate a full main-rod bearing. <<
Well , with roller bearings you don’t really cause jamming , you cause overload : because of pin deflection from strictly right angle – we are talking of minutes here , not degrees – the inwards end of the fork wide enough to embrace the con rod big end would run on the less deflecting , the outwards end on the more deflecting part of the pin ; this would cause a bending moment on the rod in question – again since we are talking of minutes : not such a force to bend the rod noticeably , yet it is a bending moment and as such undesirable .
>> I believe their spherical roller bearings were used both for axleboxes and rods <<
Maker notwithstanding , this is correct .
>> tapered-roller in this service <<
Ok , right , they used a lot of tapered bearings . The point is : tapered – with counter-tapering in second race – or cylindrical , no matter which , they all are non-self-adjusting against geometrical errorsas the spherical type does – that’s what counts !
>> arrangement on a Niagara <<
Principally same as in the N&W J class . The point always was to ease bending load on main pin ; with a tandem coupling rod arrangement , the above mentioned deflection of the main pin didn’t matter to load on each the bearings since the deflection was simply transmitted to the #3 pin , which again helped to contain deflection by its own rigidity . The silght remaining main pin bending - always well within elastic limits of course - was transmitted by non-spherical roller bearings to slightly bend high but thin tandem coupling rods without apparent drawbacks .
Got to leave , regards
Are we supposed to be numbering the posts in this thread? I've lost count -- can someone go back and put them in if desired?
Juniatha 1 . >> This is very true, but it may help to consider the #4 pin in context; look here (sorry if you have to squint!): << Squint ? why squint ?
1 . >> This is very true, but it may help to consider the #4 pin in context; look here (sorry if you have to squint!): <<
Squint ? why squint ?
Squint because it's an itty-bitty copy of a drawing with very fine lines, and it could be difficult to read as posted.
Well , ok : you know , I know the arrangement of tandem rods , I wrote my note because of that ; so what are you intending to convey with that drawing ?
The arrangement of the tandem rods on the N&W J class. For everyone reading this thread who is *not* familiar with the specific arrangement of tandem rods on those locomotives.
... Btw , 'using' a wide bearing to prevent bending of pin would ask for jamming of bearing - not very desirable an idea ;
Just to mention: so would lateral deflection of a thin-section rod, but I will leave that point until I have better historical information. Note that in most of the Timken roller-rod applications I'm familiar with, there is a spherical mounting between the roller-bearing assembly and the rod eye. I had always thought such provision to be highly desirable if not necessary for American thin-section roller-bearing rods.
In any case, I'n not sure I understand the point. There would be separate, opposite tapered roller bearings in each 'side' of the forked eye. Each one is thin (rather dramatically thin, for the applied load, in the N&W drawings I have seen). I am not sure how this is supposed to be prone to jamming merely because the fork is wide enough to accommodate a full main-rod bearing.
on the contrary , with roller bearings spherical bearings were used in European applications of roller rods; ...
Do you mean SKF? I believe their spherical roller bearings were used both for axleboxes and rods -- didn't Alfred Bruce discuss the differences in his book somewhere? They are certainly well-suited to the requirements of railway service...
I know earlier Timken applications used cylindrical bearings , not sure if they wouldn't have changed for - more demanding to fabricate - special bearings , too , had steam continued )
Give me a reference for this -- I don't remember any Timken bearing that was not tapered-roller in this service. (There were cylindrical rollers in part of some of the trailing-truck lateral-accommodation mechanism, including the truck on the Four Aces demonstration 4-8-4, but that doesn't really 'count' ;-}) In fact, one of the touted advantages of Timken bearings was that any wear could be accommodated merely by adjusting the outer race relative to the inner. These were certainly 'special' bearings -- very thin, very large ID, etc. -- but I doubt anything more "special" would have been required had steam continued.
How should that have worked out when that rod end in any case is acting on the pin at just one side of the con rod , not on both ? It would have been different only if you really had a widely forked end to main or coupling rod in such a way it *comprises* each the other on the pin , really providing a straight-through connection of con and coupling rods . However , nobody has so far mentioned this was the case here ;
That is more or less exactly what I thought I was mentioning (although no, I don't yet know that was the case here.). And yes, I would definitely prefer forking the 'receiving' end in the #2 - #3 rod, not the main.
... talk was of *conventional* arrangement where as known coupling rod line up was towards wheel side of pin *all through* and con rod connection to main pin was outside main coupling rod end - in any w/a . The forked end would *still* submit the concerning rod to bending moment since the pin would give more on outer bearing surface ; if at all , I would therefore highly recommend to apply the widely forked end to the shorter , stiffer coupling rod . Any of these arrangements however would mean letting go the special occasion offered in a second axle drive of an eight coupled type such as 4-8-2 and 4-8-4 ( 4-8-0 too , yes-ok , well .. ) that is captured by the tandem arrangement of main coupling #2 to #3 drivers and btw having but one type of coupling rod for all-around the engine. All good. Who has a link to a clear picture of the arrangement on a Niagara? That might be an interesting comparison to both N&W styles.
... talk was of *conventional* arrangement where as known coupling rod line up was towards wheel side of pin *all through* and con rod connection to main pin was outside main coupling rod end - in any w/a . The forked end would *still* submit the concerning rod to bending moment since the pin would give more on outer bearing surface ; if at all , I would therefore highly recommend to apply the widely forked end to the shorter , stiffer coupling rod . Any of these arrangements however would mean letting go the special occasion offered in a second axle drive of an eight coupled type such as 4-8-2 and 4-8-4 ( 4-8-0 too , yes-ok , well .. ) that is captured by the tandem arrangement of main coupling #2 to #3 drivers and btw having but one type of coupling rod for all-around the engine.
All good. Who has a link to a clear picture of the arrangement on a Niagara? That might be an interesting comparison to both N&W styles.
Squint ? why squint ? Well , ok : you know , I know the arrangement of tandem rods , I wrote my note because of that ; so what are you intending to convey with that drawing ?
2 . >> This may be a situation where the stiffer 'eye' fork in the forward end of a #2-#3 rod would make up for the equivalent in the tandems, while preserving a 'straight' thrust line from the main back to the #3 pin. Every little bit of mass reduction outboard in the rodwork helps! <<
I wrote it does not make up for a 50% relief - that's not a >> little bit << at all .
>> the stiffer 'eye' fork in the forward end of a #2-#3 rod would make up for the equivalent in the tandems <<
Actually not - no . ( Btw , 'using' a wide bearing to prevent bending of pin would ask for jamming of bearing - not very desirable an idea ; on the contrary , with roller bearings spherical bearings were used in European applications of roller rods ; I know earlier Timken applications used cylindrical bearings , not sure if they wouldn't have changed for - more demanding to fabricate - special bearings , too , had steam continued )
How should that have worked out when that rod end in any case is acting on the pin at just one side of the con rod , not on both ? It would have been different only if you really had a widely forked end to main or coupling rod in such a way it *comprises* each the other on the pin , really providing a straight-through connection of con and coupling rods . However , nobody has so far mentioned this was the case here ; talk was of *conventional* arrangement where as known coupling rod line up was towards wheel side of pin *all through* and con rod connection to main pin was outside main coupling rod end - in any w/a . The forked end would *still* submit the concerning rod to bending moment since the pin would give more on outer bearing surface ; if at all , I would therefore highly recommend to apply the widely forked end to the shorter , stiffer coupling rod . Any of these arrangements however would mean letting go the special occasion offered in a second axle drive of an eight coupled type such as 4-8-2 and 4-8-4 ( 4-8-0 too , yes-ok , well .. ) that is captured by the tandem arrangement of main coupling #2 to #3 drivers and btw having but one type of coupling rod for all-around the engine .
Juniatha Reduction of bending moment on #4 driven axle pin by having a much shorter pin with conventional rod arrangement should be the least of an argument since that pin in any case was the taking the least of steam ( piston thrust ) and mass forces in the drive;
Reduction of bending moment on #4 driven axle pin by having a much shorter pin with conventional rod arrangement should be the least of an argument since that pin in any case was the taking the least of steam ( piston thrust ) and mass forces in the drive;
This is very true, but it may help to consider the #4 pin in context; look here (sorry if you have to squint!):
I am still looking for an adequate online drawing of the single-rod arrangement; I'm sure Dave (feltonhill) can provide one, showing just how far inboard the single-rod arrangement would be.
shortening of main pin could not make up for it's relief of bending moment by 50 % with the tandem rod arrangement where con rod acts on pin *between* two coupling rods #2 to #3 drivers.
This may be a situation where the stiffer 'eye' fork in the forward end of a #2-#3 rod would make up for the equivalent in the tandems, while preserving a 'straight' thrustline from the main back to the #3 pin. Every little bit of mass reduction outboard in the rodwork helps!
I'm going to wait for the measured drawing and see what the moments work out to be...
CAZEPHYR wrote Sep 7 2013
>> If you can, see if you can find any reason for the updates to the J class so late in their service life. Did any failures or maybe cracks showing up on the rods of any J class locomotives cause this modification?? <<
Me , too - and I will add another "?" to it in regards to all that high speed high revving records of this locomotive class , allegedly having reached better than the very 531 rpm which 'metric diameter speed' requires , 178 km/h for 1778 mm wheel diameter in this case .
Reduction of bending moment on #4 driven axle pin by having a much shorter pin with conventional rod arrangement should be the least of an argument since that pin in any case was the taking the least of steam ( piston thrust ) and mass forces in the drive ; shortening of main pin could not make up for it's relief of bending moment by 50 % with the tandem rod arrangement where con rod acts on pin *between* two coupling rods #2 to #3 drivers .
Our community is FREE to join. To participate you must either login or register for an account.