You're definitely getting it.
gregcFrom what I read, dynamic augment forces are common in reciprocating engines and are perpendicular to the desired force.
They are, but the particular issues with augment (particularly that from 'inertial' forces) on locomotives are not found in things like IC engines. If you look at a typical gasoline automobile engine you will see precisely the kind of very-low-clearance 360-degree bearing support that the Timken bearing in a proper cannon box offers. However, the reason why the severe augment forces manifest as they do on reciprocating locomotives is that they require suspension 'clearance' -- both above and below the centerline of driver axles (with the locomotive ready for service and not 'dry' -- in case you were wondering why the piston centerline in diagrams was always about 2" above the axle centerlines) with the only restoring force being springs loaded by gravity above the box. There is little way to damp the considerable inertial forces frictionally (or hydraulically) with something that will not grossly impede suspension motion.
Meanwhile, in a locomotive with any 'play' in its driver pedestals you can get dramatically increasing wear and shock from alternating augment forces that try to 'walk' the axle in the horizontal plane as well as the vertical. This is a primary reason for adoption of the Franklin wedge, which nips the problem in the bud, as it were.
The simple bearings if properly 'scrape'-fit and kept lubricated will have much the same resistance as a roller bearing; in fact, it can be slightly less with grease lubrication because of the necessary hydraulic displacement of the viscous grease as the rollers and cage move around. The 'catch' of course is that with typical 'total-loss' bearing lubrication the time plain bearings remain well-lubricated ... or even well enough lubricated ... may be comparatively short. One advantageous method is that actually involved in Alemite lubrication: when a rod bearing is 'greased' with the Alemite pressure gun, the grease is being driven into a spring-loaded cellar, which then works a bit like a grease gun to dispense it out in shear against the journal. To an extent this is self-regulating: if the bearing begins to heat under load, the solid grease becomes more fluid at the point of contact and more will 'flow' to create hydrostatic effect.
I am looking for numerical sources for the UP high-speed rod design, which will give us a reasonable comparison between it and the components of the Timken lightweight system. Some of the initial T1 tech discussions revealed the point that the rod roller bearings were heavier than a plain-babbitted brass alternative but that much of the rotational effect of this was negligible because it could be counterweight-balanced and be much thinner (and hence closer to the main-pin seat and thrust liner)
I didn't specifically mention this, but you seem to have recognized it: the only actual contribution of piston thrust to dynamic augment directly is the vertical component of thrust in the plane of suspension action. This of course is not directly proportional to MEP but to instantaneous thrust at any particular position. Voyce Glaze actually measured this for the N&W J, which has very substantial peak piston thrust, and the actual value appears to be dramatically smaller than I'd have expected (the value he used for balancing the J to 110mph being the 80lb he retains in the main driver).
All the rest of the pernicious effect of piston thrust is the relative vector against the area of the bearing 'counter' to the thrust vector (which is a little hard to visualize in the third-degree lever setup that is reciprocating locomotive drive; a model may help you show where the forces appear in the bearings proper in order to produce the effect at the drawbar). The point here is that over much of the driver revolution there is force where a conventional crown brass isn't expecting to 'see' it, and where it is difficult to provide conformal structure close to the axle to maximize the effect of lube film and minimize distortion.
Against this it should be remembered that an engine with Timken bearings will almost always have dramatically higher unsprung mass because the whole cannon box and bath structure has to move in compliance with the suspension. This is bad for track following, but nice in reducing the peak augment accelerations that can cause much of the damage if they occur after complete unloading of the wheelrim has occurred in 'bounce'.
gmpullmanI hope this supports what Overmod has been explaining.
appreciate your help. didn't explain the relavence of quaternions and quantum chromodynamics
OvermodFor instance 'quaternions' were invented to "simplify" some aspects of steam-engine performance calculation.
OvermodIt's similarly hard to explain why entropy is used rather than 'heat content' to describe thermodynamics in steam-engine design ... or to get from quantum chromodynamics to understanding The Dancing Wu Li Masters.
greg - Philadelphia & Reading / Reading
gregcTimken bearings reduce weight, which does appear relavent, and by reducing friction, they also reduce the required cylinder force which also has a direct impact of dynamic augment force.
gregcThanks to Overmod for making me aware of Augment force and that it is not directly related to locomotive resistance.
Another advantage to the roller-bearing main and side rod design is the reduction of the "overhanging weight" in other words, how far away from the center-line of the driver counterbalance is to the center of the mass of the main rod, crank pin, eccentric crank and side rods.
This reduces the required weight of the "cross-counterbalancing" needed to offset this effect.
Counterbalance_crop by Edmund, on Flickr
Again, better steel alloys = lighter reciprocating weight = less counterbalancing. I added four more pages to the "Rods for 100" album if you care to read further on counterbalancing.
I hope this supports what Overmod has been explaining.
Regards, Ed
gmpullmanhttps://www.flickr.com/photos/gmpullman/albums/72157673726478087
From what I read, dynamic augment forces are common in reciprocating engines and are perpendicular to the desired force.
In the same way that the cylinder force is sinsoidal and reversing in the direction the locomotive is moving, the vertical motion of the main rod is equally sinsoidal and reversing.
these forces stress locomotive and track components. At least locomotive components have limits before rapid unscheduled disassembly occurs, or at least deformation. These forces are literally trying to pull the locomotive apart.
since these forces are related to speed, they limit speed. in other words, these forces need to be reduced to increase the speed of passenger trains.
one way to reduce them is to reduce the mass of the reciprocating components. The article, Locomotive Rods for 10 ... describes reduction of main rod and piston mass.
i'm still not completely sure about the relationship between the above and Timken bearings. The article Timken Tapered roller Bearings mentions that Timken bearings reduce weight, which does appear relavent, and by reducing friction, they also reduce the required cylinder force which also has a direct impact of dynamic augment force.
I also wonder if simple bearings and lubrication have speed and temperature limits that the Timken bearings exceed.
for my interest, i'm still curious about the (quantitative) impact of Timken bearings on locomotive resistance. I'll guess is reduces the factor of 25 (lb/ton) in the equation (Alco data) I posted above not insignificantly. Seems like there's a lot factors between that value and the Armstrong value of 2 lb/ton for a loaded freight car.
Thanks to Overmod for making me aware of Augment force and that it is not directly related to locomotive resistance.
my study of steam locomotives is making me aware of many subtle design and operating aspects of steam locomotives.
Overmod gregc Overmod gregc I'll just have to stick with numbers Good luck with finding numbers that don't involve mathematics who said anything about not involving math You did, just a few posts up, regarding How a Steam Locomotive Really Works. If that math is trouble, you're likely going to have trouble developing correct 'numbers' from data or fully incorporating many of the actual sources of potential resistance.
gregc Overmod gregc I'll just have to stick with numbers Good luck with finding numbers that don't involve mathematics who said anything about not involving math
Overmod gregc I'll just have to stick with numbers Good luck with finding numbers that don't involve mathematics
gregc I'll just have to stick with numbers
Good luck with finding numbers that don't involve mathematics
who said anything about not involving math
gregc SeeYou190 "How A Steam Locomotive Really Works" 90% of the book is mathematics, so be ready for that. read it. Not enough math
SeeYou190 "How A Steam Locomotive Really Works" 90% of the book is mathematics, so be ready for that.
90% of the book is mathematics, so be ready for that.
read it. Not enough math
SeeYou190 said the book was 90% math. It's not (< ~5%)
I was suggesting there was not enough math.
i was happy to see confirmation of my understanding of the combined effect of piston force from two cylinders. There was an example for calculating cylinder efficiency without any explaination of where the values he used came from. Not much more math.
the author was fond of discussing interesting features of specific locomotives.
OvermodKnowles is a frequent poster on steam_tech and if anything has 'happened' to him we can find out fairly directly. Greg is welcome to join that Yahoo Group and ask the question directly in group, where he will likely get better answers than I can provide.
could you provide a link?
a
Overmod Knowles is a frequent poster on steam_tech and if anything has 'happened' to him we can find out fairly directly. Greg is welcome to join that Yahoo Group and ask the question directly in group, where he will likely get better answers than I can provide.
Knowles is a frequent poster on steam_tech and if anything has 'happened' to him we can find out fairly directly. Greg is welcome to join that Yahoo Group and ask the question directly in group, where he will likely get better answers than I can provide.
If someone could find out if Knowles is active, I would be very pleased if I could correspond with him about his machinery resistance calculations. This may be "inside baseball", but that too has its enthusiasts.
If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?
Thanks, Ed.
OvermodHere's what's affected by Timken bearings (ask Ed for the detail cross-sections and graph back if he's removed them from the postings)
I will provide this link to my Flickr album which contains the "Timken Rods for 100 MPH" pages plus the added roller bearing pages:
https://www.flickr.com/photos/gmpullman/albums/72157673726478087
Anyone interested may look there and, by using the download option, save, copy or print the pages as needed.
Additional pages from the Alco handbook can be found here: (but you'll have to sort through additional items)
https://www.flickr.com/photos/gmpullman/albums/72157689157947401
This will save valuable space here but make the information available for anyone interested.
Almost missed out on what is to me a fascinating thread -- I usually browse Steam and Preservation over at the Trains forum.
Anyone here know of a man named John Knowles, who I believe was from Australia? He had posted his article/report/treatise on steam locomotive machinery friction on the Web a while back. His expressed reason for delving into this topic is that he wanted to understand horsepower claims of different historical steam engines, which are sometime quoted as indicated (or "cylinder") horsepower (presumably calculated from indicator diagrams) and other times as drawbar horsepower, from dynamometer car readings out on the road and in rarer instances as "wheel rim" horsepower measured from roller dynamometers from the few places that had "test plants."
I printed his report but it is stuck under piles of other printed matter in my study at home, and I cannot find it anymore on the Web. Do you think Mr. Knowles could have passed on, and the persons managing his estate didn't see fit to pay the charges to keep his writings up on the Web?
Mr. Knowles had a calculation, an estimate, perhaps a crude prediction of machinery friction based on a theoretical model and claimed to get good agreement with data from the Rugby Test Plan in England? Along the many things he mentioned is the seemingly high machinery friction of The Red Devil locomotive described in Wardale's book. That high friction was a burr-in-the-saddle for Wardale -- he blamed it on the South African Railway having a dynamometer car that was out of calibration, but he couldn't be certain, and it threw off many of his claims of improved thermal efficiency for his master work
gregcI'll just have to stick with numbers
Good luck with finding numbers that don't involve mathematics, and that are largely determined by empirical constants related to factors not intercomparable between locomotive designs. I can only tell you where to find the water, how to recognize it, and how to avoid the more usual kinds of poison there; I can't actually get you to do the work of drinking.
?? i'll just have to stick with numbers
gregcI think the aerodynamic force is not significant.
For most of what you're concerned with, that is true. I believe there is a very good discussion of actual aerodynamic resistance on locomotives somewhere in Kantola's work (ISTR there is an actual value calculated for the 'bathtub' Commodore Vanderbilt streamlining on the 'horsepower' increase at something like 100mph, which I remember as being in the 300 to 400hp range, which is particularly significant as most of the losses in practical steam-locomotive power are beginning to be severe even at this speed range.) By the time you're considering 120mph power or better, good streamlining begins to look attractive. (even more attractive for conventional reciprocating steam, which begins to have severe efficiency problems above 9rps, than for constant-horsepower motor trains that are down to an effective drawbar pull of under 9000lb even at high nominal prime-mover hp as you get into the range -- there are tables from ATSF that show it by speed for several of the early diesel designs, and one of Brashear's books contains one)
The argument can, and has, been made that the relative contribution to train resistance attributable to Cd reduction only at the front of what is a long and relatively thin mass is little. A good example of this is the test Metroliner train composed of T-1 to T-4, which could run in the 150mph range with almost a blunt front end; were economy or absolute top speed (or absolute top speed within the capability of the catenary to feed power) a prime consideration, you might have seen more effective frontal streamlining, but for suspension and other ride testing that's a secondary set of considerations.
The TL;DR is that even as fast as 80-85mph the effect of full streamlining over the Kamm-type effects around a blunt or complex outline is relatively small, and in a first-order approximation could be reasonably neglected or deprecated.
gregcnot sure how this is affected by Timken bearings
Let's try this some more.
As frequently for wiki sources, knothead Britishers ignore the semantic difference between hammer blow and what Americans usually mean by dynamic augment. Perhaps part of the difference is due to suspension and axlebox-design considerations that are relatively or wholly absent on most British steam power.
Hammer-blow is a lateral effect, seen on the 'outside' wheel (where the lifting or dropping force is expressing) as lever action with the fulcrum on the inside wheel tread. It can't be expressed unless the driver pair in question has enough effective free lateral to be able to pivot in the pedestals (what the British call 'horns') freely enough, and the "friction" between axlebox guides and pedestals is very low (this is part of why the British spent such care and time on regular adjustment of fixed pedestal surfaces rather than pay for Franklin wedges ... but I digress)
Dynamic augment occurs when the whole axle rises and drops more or less 'together', constrained by good cannon-box construction (as needed for Timken driver bearings to 'live') and Franklin wedges. The bouncing-driver films made by AAR (if I remember correctly, on the hapless C&NW E-4 class) and some of the measurements in Kiefer's greased-rail J3 Hudson testing involve this kind of action, where I think there is comparatively little 'cock' and the stiffness of the axle and its support from the cannon box results in nearly-identical force on the outer driver as the one being driven (via its pin) by rod forces.
Here's what's affected by Timken bearings (ask Ed for the detail cross-sections and graph back if he's removed them from the postings)
The bearings themselves are very special. The rollers in them are very short and relatively fat, in order to have as little length as possible and still work correctly as self-locating taper bearings (in pairs, the effect being like thrust on a herringbone gear). SKF bearings won't do as well as a putative replacement because 'barrels' of this short length don't have enough self-steering action not to 'walk' in their cages and cause the wrong kinds of stress and wear.
Both inner journal or race surface and outer race surface are likewise made just as thin as contemporary metallurgy and fabrication practice can make them.
A plain hydrodynamic bearing in this location, no matter what you use to lubricate it, that has only these bearings' lateral dimension will almost certainly not develop enough wedge or lube integrity to take peak rod thrust or inertial forces against the pins that compromise the lube film. Remember this phenomenon.
Now, looking at 'hammer blow' as a critical effect, the closer the driving force outside the driver sectional center of mass is, the less the developed (lever) force working up and down will be. It was well-recognized as early as the 1920s that keeping the rods as far 'inboard' as possible made for reduced augment, and this remains a reasonably 'key' design principle (although the early J lightweight rod configuration provides an interesting alternate argument I'll take up in a little while)
Now, back at the Timken bearings, we note that these have a decidedly nontrivial diameter increase over where the pin-to-brass (or pin-to-intermediate-bronze-sleeve in the UP designs) will be, even if we neglect the necessary bronze 'liner' between bearing and rod eye. As a result the forged eyes in the rods have to be much greater in inside diameter, and of course this means that strength in the eyes has to be greater still. Meanwhile, the rod has to be made as thin as possible (both at the eyes, to match the thin bearings, and in its section, to keep its mass as far 'inboard' as possible.
This explains the various shapes in the Timken rod. If you study the available drawings you will see how very intricate (and carefully calculated!) the shape of the eyes and ends are. The actual drawings for the 'train' of forging dies to make these survives in the NWHS library, and this was one of the most valuable resources the T1 Trust has documented. Adequate tensile strength in a very thin rod requires increased section -- which is achieved by making the web very deep. (The problem here, already noted, is that it makes the rods almost exquisitely sensitive to bending deflections, which progress with almost laughable speed to bending beyond the yield limit and disastrous 'rapid unanticipated disassembly' of components....)
As you may have understood from the mess of verbiage, the main rod is the greatest source of potential 'contribution to augment', and at various times engineers have made efforts to keep the main-rod big ends as far inboard as possible to reduce this. You will note by looking at many 'successful' Timken-rod installations, particularly on Niagaras, that the big ends are in fact outside the line of action of the side rods. That will give you an idea of where the relative contribution to augment of lightweight roller-bearing rods vs. mains lies!
Now we come to another irritating detail: there are two kinds of 'tandem' rods, and they are very different in construction although doing the same general thing. In order to get the side rods as far inboard as possible, we have to be concerned with how they pivot on the pins (otherwise the suspension would bind!). One obvious solution ... and we see this often in the 1920s ... is to make the rod junction fork-and-blade, and sometimes this was done not directly over a pin, but on a short extension (these are sometimes called 'articulated' rods). As you can understand from the above description, doing this with plain bearings will increase the required lateral section of rod eyes, and require a better kind of EP lubricant ... hence the rise of Mr. Alemite Man ... and this of necessity increases augment resultant.
The 'tandem' concept applied to Timken rods involves using a duplicated pair of rods between #2 and #3 pairs on a 4-8-4, so the net thrust goes through all the rods in a straight line. As I recall, the original idea was to keep the main all the way inboard (and its roller bearing has to be the stoutest and hence widest of any -- look at it on a Niagara, for instance) and the problem there, on a 4-8-4, is that the bearing on #4 winds up being way outboard of its driver face. On the original Js that resulted in a very long, tapered pin .. which had multiple problems with fracturing, seat distortion, etc. This resulted in the 'revised standard version' which is on 611 today; official pravda from N&W sources says this is qualified up to the same speed as the original arrangement (which is 10% over peak anticipated service speed, or 110mph)
gregcthe forces produced by the cylinders is obviouly involved. I'm told there are no good books quantitatively discussing the physics of a steam engine.
The problem is that most of the ones that do "quantitatively" discuss the physics involved will do so in mathematics, frequently abstruse mathematics. For instance 'quaternions' were invented to "simplify" some aspects of steam-engine performance calculation.
It's similarly hard to explain why entropy is used rather than 'heat content' to describe thermodynamics in steam-engine design ... or to get from quantum chromodynamics to understanding The Dancing Wu Li Masters.
Now, having said that, I have something of a grim lack of regard for "How A Steam Locomotive Really Works" because much of the 'mathematics' appears to be there for show-off value rather than actual explication of phenomena for the intended audience. A much better reference for understanding how to do it yourself would be to obtain a copy of Wardale's full FDCs ("fundamental design calculations") from the 5AT people -- this is his legacy to the steam community, and fantastically valuable for the wisdom it encapsulates. It is not quite as easy to plug in 'sample' numbers to calculate effectiveness of a particular legacy design, but it can be done. Where you see empirical numbers slipped in, he explains how they were derived (which is considerably better than Fry did in his supposedly magisterial work in the early '20s!)
The references I've seen don't include the augment forces in first-order tractive-effort calculations; they show up much more in understanding of 'slip propensity' which winds up being a first-order consequence (you can't just yank the throttle open and then wind up the reverser, cough, cough, T1s, and passengers may slosh forward and backward at starting speeds) but not something you see in average drawbar pull measurement.
As you yourself have already mentioned, a reasonably good idea of wheelrim torque at speed can be derived from PLAN or indicator diagrams with the appropriate correction for nominal boiler pressure. Note that at least one source says the ".85" correction factor is structural, related to limited cutoff and not heat losses from boiler to dead space cylinder ports, so a little bit more calculation may be in order to see if a double loss calculation along the lines of that for diesel-electric horsepower is in order.
Overmod gregc perhaps you can explain the difference ... in layman's terms. I can try. I'll at least get far enough that someone else can translate defective jargon into English for you.
gregc perhaps you can explain the difference ... in layman's terms.
I can try. I'll at least get far enough that someone else can translate defective jargon into English for you.
from wiki
In rail terminology, the hammer blow is a vertical force which alternately adds to and subtracts from the locomotive's weight on a wheel.
It is transferred to the track by the driving wheels[1] of many steam locomotives. It is an out-of-balance force on the wheel (known as overbalance[2]). It is the result of a compromise when a locomotive's wheels are unbalanced to off-set horizontal reciprocating masses, such as connecting rods and pistons, to improve the ride. The hammer blow may cause damage to the locomotive and track if the wheel/rail force is high enough. 'Dynamic augment' is the US term for the same force.
not sure how this is affected by Timken bearings
SeeYou190"How A Steam Locomotive Really Works" 90% of the book is mathematics, so be ready for that.
gregc I'm told there are no good books quantitatively discussing the physics of a steam engine.
.
"How A Steam Locomotive Really Works"
-Kevin
Living the dream.
gmpullmanSometimes it is challenging to figure out exactly what Greg is looking for.
i'm studying the motion of a steam locomotive. that requires an understanding of the forces affecting motion. They include the forces produced by the cylinders and the opposing forces.
the forces produced by the cylinders is obviouly involved. I'm told there are no good books quantitatively discussing the physics of a steam engine.
opposing forces include the resistance (friction) of the cars which Armstrong discusses in his book. The resistance of a steam locomotive is more complicated but Ed provided some data from Alco. The last significant component is grades which can both oppose or contribute to the force produced by the cylinders. (i think the aerodynamic force is no significant).
i'm just looking for first order effects. Based on some discussions with 5at.com.uk, i'm more than satisfied with the results.
i appreciate the help from the forum
??
gregcperhaps you can explain the difference ... in layman's terms.
Most of resistance is a resultant of any forces on the locomotive that act in the plane of the drawbar, opposite the motion of the locomotive. Wind, bearing friction, and gravity are all examples of this. There are also some forces that act laterally to induce added friction or to move the locomotive laterally or in other planes, usually cyclically with some complex resonant harmonics, and since these are generated with 'power' that could have been used for forward motion, they have the effect of adding to practical 'resistance' even though they can be difficult to measure directly as such.
Augment, on the other hand, is a force that acts at right angles to the plane of running resistance, and it is related to several aspects of outside reciprocating rod drive. There are two 'actions' that I think need to be kept distinct when thinking of this in layman's terms; three if you include a discussion of 'overbalance'.
First is the action of quartered rods. As you know, a 2-cylinder steam locomotive with its cranks 180 degrees apart would fail to start at or even near a dead center (here usually called fore and aft dead center, not TDC or BDC as in automotive engines). So to counteract this the rods are 'quartered' so one side is near the center of its maximum developed torque while the other is near one of its dead centers. The problem with this is that there is a vertical reaction due to momentum of the mass of the side rods on one side that is not balanced on the other (because the rod mass is moving fore-and-aft on that other side) and this is difficult to balance in-plane -- usually there are weights in the wheel to balance the mass of the side rod, but the center-of-mass action of these weights is inboard of the center of mass action of the rods, perhaps by several inches, so for "perfect" rotating balance there is still a torque about a fulcrum calculated at the opposite wheel's wheeltread, trying to rotate the driver up against the suspension. At high cyclic rpm this force can become substantial, and a good locomotive suspension and equalization system is designed to permit 'cross-level' accommodation to maintain reasonable anti-slip, so the small rotation is tolerated even if Franklin wedges keep the axleboxes perfectly aligned longitudinally.
Meanwhile, there is some component of the main rod that is not 'balanceable' by rotating weights at all; it's the part of the rod close to the crosshead, which is constrained to slide nearly completely fore and aft. Ralph Johnson calculated that the percentage of rotating weight that can be counterbalanced effectively with counterweights is related to the geometric 'center of percussion' (and gives formulae to find this, and the corresponding percentage of main rod mass that is taken as reciprocating). In small locomotives, or those with restricted wheelbase, the inertial effects of the reciprocating portion of the rods can become so severe at high speed as to cause the locomotive to nose (and by coupling effects through the suspension, to 'hunt') severely; there can also be the effect of 'surge' amplified beyond just what the instantaneous piston-thrust reaction produces as short-term acceleration. To 'fix' this a little additional weight, called 'overbalance', is put in the drivers, and some other empirically-derived adjustments to angular location of some of the balance weight is done ... the effect of which is to put the rotating balance of a wheel so fabricated 'out of round' to a degree which can dramatically change the effective weight on drivers for that pair, even to the point that it bounces free of the rail on uplift.
Now, we can stop for a moment and look at what we can do to relieve these. One obvious step is to reduce the absolute mass of the rods in question, particularly the reciprocating 'part' of the mains. Another is to keep the rods as far inboard as possible, to line them up as much as possible with their counterweighting.
An astute reader will immediately recognize why an inside-cylinder engine has an advantage here; in fact, the Belgian "Hiawatha-style" 4-4-2s, of which one has survived to be restored to operable condition (!), have inside cylinders precisely to reduce the need for overbalance correction (and hence to tolerate much higher speeds with low European axle loading).
Now, the energy to lift and drop the rods does 'count' as an opportunity cost that could have been used as more drawbar pull. The problem is that there's usually no way to 'recover' the energy so it pulls against the drawbar, and hence no need to try calculating it as there's no point; it's "lost" no matter what you do, so we simply take empirical steps to minimize 'whatever it is' as absolutely as is economically practical.
One of Voyce Glaze's techniques in balancing the N&W class J to permit high cyclic rpm with relatively low drivers is to take advantage of the rod connections to transmit some of the overbalance correction -- he specifies only enough overbalance in the main driver to compensate for the peak vertical component of piston thrust. All the other overbalance is distributed in the other driver pairs (I forget the exact number, but it was covered in an '80s RMC article on J balancing, and I recall it being around 2200#).
The heinous effects of low overbalance are also 'corrected' somewhat on the J by having very stiff lateral compliance, stiff to the point the effective rigid wheelbase is extended somewhat and there is increased risk of derailment on some kinds of poor (or 'diesel-optimized') track. You will find online discussions of this as well.
The surge issue was addressed, and rather effectively solved in principle, by Langer in 1947. He uses a geared rotating weight at the centerline of the chassis to counteract the effect of longitudinal inertial force, and I think there is no fault in his approach; you see some people calling this an 'overbalance' solution like the salmon rods in Cossart valve gear, but being on the centerline it explicitly has no contribution to reduction of nosing tendency, therefore we can conclude it is an anti-surge device.
gmpullmanNever one to cause a commotion. Sorry. Posts edited for integrity.
Sorry, Ed; didn't want to leave the impression I was yelling at you or complaining. That was valuable information about augment. The concern was that gregc would see it and assume those curves meant something like rolling resistance.
I don't know if anyone has actually modeled the relationship between increasing augment force and the increase in 'resistance' it poses as a drain on some combination of thrust and momentum. Something has to be providing the energy to jack the wheels up and down against the suspension-spring rigging 'preload'. I'd lay money on Chapelon having this in the volume 2 of LLAV that was never published...
OK... disclaimer... I know nearly nothing about steam locomotives.
However, I know tons about heavy trucks.
There is so much that goes into the calculation of rolling resistance, that it is impossible to make a blanket statement of "What the rolling resistance of a 40 ton mine material truck is".
Surely, tire pressure would not be a factor for a steam locomotive, nor would road material.
However, wheel bearing grease weight, internal axle bearing material, grease clearance fill ratio, wheel diameter, ambient temperature, etc, would all be factors.
So, while an average or typical answer might be possible, there is no way to know if it would be true for any specific locomotive.
OvermodI call foul, because this has nothing to do with what gregc is concerned about
Sometimes it is challenging to figure out exactly what Greg is looking for.
Never one to cause a commotion. Sorry. Posts edited for integrity.
Ed
My engineer told me that when the Y6 came along with its roller bearings, they had to learn to run the train all over again! And, that they did. No slide rule, no graphs, no high math needed. Just "the seat of their pants"!
Overmod(resistance, not augment)
perhaps you can explain the difference ... in layman's terms.
gmpullmangregcis there any info on the impact of roller bearing on locomotive resistance? Roller-bearing_fig7 by Edmund, on Flickr
gregcis there any info on the impact of roller bearing on locomotive resistance?
Roller-bearing_fig7 by Edmund, on Flickr
I call foul, because this has nothing to do with what gregc is concerned about (resistance, not augment) and comparatively little to do with roller bearings (the lion's share of the actual improvement being in the lightweight rods and inboard location only incidentally facilitated by the thin eyes and reduced 'end liner' requirement allowed by the rod roller bearings. (The bearings themselves were, and are, recognized as increasing weight and therefore both inertial and offset-couple (albeit comparatively slight) augment contribution.)
Although you can bet Timken, which I suspect designed the whole High Dynamic Steel rod project around selling more bearings, isn't going to brag up that this is so.
Take a look (it's covered in Kratville's Mighty 800 book) at the UP developments in the mid-Thirties for low-augment high-speed rod construction that does not include roller bearings. It would be interesting to see the FEF (which is conspicuous by its absence in this supposedly objective graph) plotted in comparison with at least the two converted 'legacy' engines (the 5344 and the indicated K4s) to see where the actual advantage of the rollers themselves in reducing augment might lie.
Meanwhile, in case this is not familiar, the two Union Pacific locomotives mentioned are the ones that were converted (and a bit later streamline shrouded) to be protect power for the Streamliners. This is the 'whole package' of augment reduction including angle balancing, new disc mains, and I believe some suspension changes and snubbing; to attribute the dynamic reductions solely to Timken (as appears to be the clear intent of semantics here) is more than a little specious.