It is currently Sat Apr 04, 2020 8:19 am

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 30 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Steam locomotive wheel balancing
PostPosted: Wed Feb 12, 2020 2:00 pm 

Joined: Fri Jan 01, 2016 5:47 pm
Posts: 16
At the former Southern Railway Spencer, N.C., shops, now home to the North Carolina Transportation Museum, there is a wheel balancing shed with scales allowing each wheel of a steam locomotive to be weighed separately — up to four axles at a time.

Attachment:
IMG-2173_1000px.jpg
IMG-2173_1000px.jpg [ 219.64 KiB | Viewed 1799 times ]


Location on Google Maps:
https://goo.gl/maps/Fd8jR6ycBbUYbfmo8

I haven't been able to find definitive information on this building or how it was used back in the day. Does anyone know if scales like these were commonly used in steam locomotive maintenance facilities? I've spoken with a former shop worker who says it was installed by management in an attempt to reduce post-repair shakedown runs on the main line (an important step which, as you might imagine, it failed to replace). But my knowledge of steam locomotive maintenance is minimal.

How is wheel balancing done today? Do all wheels of a locomotive carry equal weight or do some intentionally carry more?

Thanks in advance.


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Wed Feb 12, 2020 2:48 pm 

Joined: Thu Aug 26, 2004 2:50 pm
Posts: 2522
Location: Northern Illinois
I would think the most important function was to provide an accurate weight-on-drivers figure for each class of locomotive, since the BLE agreement used to base the engineer's pay rate on that. I would imagine that while the pay was paid according to the class of locomotive, the brotherhood reps would want confirmation that the road wasn't basing the figure on one specially lightened example, but rather the factor used was representative of the class.

_________________
Dennis Storzek


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Wed Feb 12, 2020 5:06 pm 

Joined: Sun Jul 22, 2018 2:03 pm
Posts: 19
My gut feel is that axle scales would not provide any significant advantage in reducing or eliminating shake-down runs after major shoppings.

Instead, I would expect that such a scale would be useful when a new steam locomotive class was first produced, to allow fine-tuning the spring rigging for even weight distribution. Also it would be used to determine axle loading, which is necessary to determine compatibility of a design with various bridge (and turntable!) designs.

If I recall correctly, Southern was one of the railroads which designed and built its own steamers, so the above supposition "make sense" in that context.

Such scales would also be useful for the process of adjusting the locomotive's spring equalization after a significant change in weight distribution, such as after addition of various appurtenances, such as superheaters and feedwater heaters. This supposition can apply to any major shop which routinely performs these significant modifications to steam locos. If this were the primary purpose, then I suspect that such scales would be used at a large variety of major steam backshops, across all regions and railroads.

So far as I am aware, the ideas behind "balancing" steam locomotive drivers depend more on weight and weight distribution of rotating parts, as well as weight and weight distribution of reciprocating parts. While a static measure of total weight on a driver may be part of the calculations when balancing a driver, I would not expect it to be a particularly significant part of the calculations.

Bob Milhaupt


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Wed Feb 12, 2020 8:04 pm 

Joined: Thu Oct 08, 2015 11:54 am
Posts: 720
Location: New Franklin, OH
Bob Milhaupt steamy wrote:
Instead, I would expect that such a scale would be useful when a new steam locomotive class was first produced, to allow fine-tuning the spring rigging for even weight distribution. Also it would be used to determine axle loading, which is necessary to determine compatibility of a design with various bridge (and turntable!) designs.

Speculation here but I think you’re on to something. Note the set of scales for each short section of rail. But I’m a diesel guy - what do I know,,,,

_________________
Eric Schlentner
Orrville Railroad Heritage Society
Car Knocker, Gandy Dancer & Hog Jockey
https://orrvillerailroad.com


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Wed Feb 12, 2020 9:38 pm 

Joined: Wed Jan 15, 2014 9:14 am
Posts: 277
Definitely for axle weight rather than ballancing. If you would like to see how to properly balance steam locomotives, see if you can find a copy of The Steam Locomotive by Ralph P. Johnson. Chapter 15 is all about ballancing, and it is fascinating how many variables come into play on a reciprocating steam locomotive.


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Wed Feb 12, 2020 10:55 pm 

Joined: Tue Aug 24, 2004 10:34 pm
Posts: 783
I am curious about balancing steam locomotive wheels ever since I saw some molten lead leak out of a main driver once upon a time. Had never given it a thought about balancing locomotive wheels? Does anybody do that now days when a set of tires are installed and wheels reworked? On a slow moving locomotive maybe it is not that important but a larger road engine at speed it probably does? I am asking as I never hear anybody mention it when wheel work is done. Thought originally that the counter weight would of been cast accordingly. But after seeing lead, maybe a pounds worth leak out I figured it had to have been to fine tune the wheel balancing? Am I wrong about this? Don't recall ever hearing about it. Saw the title on this thread and remembered the question. Thanks in advance to all the Mechanics who know about this kind of stuff past and present.

In the paper industry the rolls all had to be balanced at run speeds. Knew guys who ran the balancing machine. It is kind of interesting, to me anyway. Can't imagine the variables of a locomotive going down the track mt or pulling a loaded train. All the forces moving around and the more wore out or out of adjustment they are the more variables I would think? Would it matter on a logging locomotive that was totally worn out? But when I look at say the #765 or UP Big Boy at speed I would think it does?

Regards, John.


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 9:58 am 

Joined: Sun Jul 22, 2018 2:03 pm
Posts: 19
I write as an occasional student of steam locomotive technology, and I write what I remember. It has been perhaps 10 years since I've done any reading on this subject. Consult appropriate reference books for details and accurate representations of the concepts behind what I write.

John Risley wrote:
I am curious about balancing steam locomotive wheels ever since I saw some molten lead leak out of a main driver once upon a time... On a slow moving locomotive maybe it is not that important but a larger road engine at speed it probably does?"

Proper balancing can be important on just about any size locomotive.

A poorly-balanced steam locomotive can behave poorly at certain speeds, and those speeds are not necessarily at the highest which the locomotive can operate. Ask a track guy about steam locomotives, and they will say that (reciprocating) steam locos pound the track and do massive track damage. This "pounding" of the track structures is, in part, a function of the speed of the loco and how well balanced the loco is at that speed.

In the early days of steam locos in America, it was not uncommon, as trackwork improved, boiler sizes increased, and locomotive top speeds increased, that locomotives were found to ride roughly at certain speeds. Shops often bolted on additional counterweight materials to existing counterweights. Such efforts often changed the speed at which the locomotive "ran rough".

John Risley wrote:
I am asking as I never hear anybody mention it when wheel work is done.

Typically, a wheel's design and/or post-production "tweaks" to the couterweight size, location, weight, and weight-distribution will give a locomotive which behaves as well as it can or as well as it can within the amount of effort that the shops are willing/able to give. Assuming that the counterweight materials stay in-place, secured, and at intended weight, there ought to be no need to adjust and/or check the counterweight if simply performing wheel work. If, however, any "added lead" happens to leak out, or if a chunk of counterweight falls off or breaks off, then perhaps re-balancing is justified.

And, if the rods were modified in some significant way, like a change to a side rod from a heavier locomotive, or perhaps swap from plain bearings to roller bearings on the rods, or a change in rod material, change of rod dimensions, change in crosshead type, change in piston/cylinder diameter, etc., was made, then it might be desirable to re-visit wheel balancing.

If a driver set overheated account sticking brakes, any added lead counterweight material could become liquid, and leak out at a "plug" in the cast counterweight. Apparantly that happened in the case you mentioned.

John Risley wrote:
Thought originally that the counter weight would of been cast accordingly. But after seeing lead, maybe a pounds worth leak out I figured it had to have been to fine tune the wheel balancing?

Yes - on more-recent locomotive designs, the driver centers and their counterweights were typically pre-cast based on best estimate of the forces to be "balanced". The counterweights could be modified at a later date to suit actual conditions, or to try to make the loco run more smoothly at certain speeds.

Some driver center designs were specifically designed to make balancing easier and to allow space for adding lead to tweak the dynamic balance.

John Risley wrote:
In the paper industry the rolls all had to be balanced at run speeds. Knew guys who ran the balancing machine. It is kind of interesting, to me anyway. Can't imagine the variables of a locomotive going down the track mt or pulling a loaded train.

I would expect that the rolls in a printing press would primarily experience "rotating" balance issues. But a rodded steam locomotive also experiences "reciprocating" forces.

The primary wheelset balancing concerns I remember relate to "rotating mass" and "reciprocating mass". So far as I remember, weight of train, weight of tender, etc., did not have an impact on the calculations.

Rotating masses include crank pins and counterweights, and distance from the axle's rotation point. Those masses can be balanced using the same sorts of techniques used for a "simple" rotating mass like a roll in a press.

But a printing press roll would, I suspect, typically be a very regular cylindrical shape, where a drive wheel set is decidedly odd. Consider the off-axis crank pin which extends from the face of the driver. An obvious counterweight would be to place an equivalent (balancing) mass to the inside of the face of the wheel on the opposite side of the axle. But that the frame prevents that.

Reciprocating masses which need balance include side rods, main rods, crossheads, crosshead pins, cylinder rods, and pistons. The reciprocating masses have a much more complicated relationship to the forces on a given wheelset. The masses on one side of the locomotive have a non-trivial impact on the forces seen on the other side of the locomotive. And the force of a main rod on the main pin is in non-trivially different directions depending on the position in the stroke and direction of movement of the piston.

Balancing these reciprocating forces generally requires adding weight to the counterweights. But balancing these forces is best done by applying huge weights at very small locations on the drive center - something that generally is not possible due to various mechanical constraints. Thus the counterweight becomes thicker and covers a larger angle on the wheel center, and even might have some portion of the required counter-weight mass moved from one axle's driver center to another axle's. But adding mass to the counterweight thows the rotating masses out of balance, so the work previously done to balance the rotating weights is no longer accurate, so that has to be re-done...

If I understand correctly, "balancing" a steam locomotive's rotating and reciprocating masses was pretty much an "art" because the practical limitations of steam locomotive mechanical systems prevents complete application of scientific principles.

Regards,
Bob Milhaupt


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 11:37 am 

Joined: Thu May 24, 2012 1:37 pm
Posts: 1267
We need to carefully distinguish 'balancing' for dynamic augment reduction from the use of the static balances to determine and set driving-axle load as mentioned in the original post.

There is probably as much or more actual 'science' and engineering skill involved in late balancing practice as anywhere else in steam-locomotive design. Do not mistake 'compromise' between two competing sources of augment for fundamental ignorance either of their magnitudes or their significance for fast vs. smooth running...

Some Europeans were far more concerned with proper dynamic balancing (perhaps because their organization priorities were driven more by civil than mechanical engineers). You can find instances on the Web, for example, of balancing rigs from the Great Western that suspend a driver pair in an arrangement of springs and levers and spin it up to high speed to assess very accurately how the rotating balance needs to be adjusted; a comparison of lever deflections on 'both sides' will give you some of the hammer-blow moments -- these of course being relatively minimal on the four-cylinder power the GWR adopted for high-speed locomotives. (The appropriate correction by using bobweights on the crankpins with lateral center of mass 'matching' the corresponding rod arrangement will have occurred to the astute reader already; these are only slightly more refined versions of the bobweights that roads like NYC used to move power 'rods-off'.

I particularly refer you all to the theory of angle balancing, which ATSF used so well, as it is a form of 'cross-balancing' that simplifies some of the ways that balance weight is put in the wheel, particularly when that pernicious and evil thing 'overbalance' comes into play.

You will see much of the discussion in Johnson involve the calculation of effective overbalance, but you should not forget that significant work was done in this area after 1944. You will see it given as a truism that you can't rotationally balance a quartered 2-cylinder reciprocating steam locomotive 'perfectly', because the momentum effects of the 'linear' component of the linkage (both in the rods and valve gear) -- for example, the center-of-percussion-determined forward portion of the main rod, the crosshead, and piston rod/piston) -- are increasingly substantial with increasing speed. (If you wonder why the piston rods on Niagaras are bored hollow and the piston is such a thin section, now you have a better idea...)

To achieve proper high speed on a reciprocating steam locomotive you need as little overbalance in the main as possible (Glaze only put 80# or so on the J class 4-8-4s, and it's there nominally to compensate the vertical component of piston thrust on the suspension) and the Australians, for example, experimented in zero overbalance, even on some remarkably unlikely classes of engine. We might note that on a large modern North American engine the 'nosing' effect or 'hunting' tendency of inertia is relatively small, since the wheelbase can be so long, the connection to tender via radial buffer so tight, and lateral control via the trucks so peculiarly effective. We also observe the invention of the Langer balancer (patented 1947) which selectively relieves the particular effect of surge from needing to be 'overbalanced' (it does little or nothing for nosing couples) and the particular balancing conventions adopted on the British 'Evening Star' 2-10-0s that allowed them to reach surprisingly high speeds without destructive effect on the track ... and the lessons on what to balance that this offers us.

Incidentally, there was a very interesting expedient used to provide more effective counterbalance mass in relatively small-drivered heavy freight power in the period just before WWI. One example is on CB&Q. These use (quartered) webs like those on a cranked axle inside the frames (between the bearings) with 'supplemental' rotating balance weight in them. This allows a greater proportion of the weight 'outboard' in the wheel counterweights to be used for overbalance, as well as reduce the size of the physical pockets in the driver-center castings. It does come at the cost that much of the 'counterweighting' force effectively does have to go through the axle fits, which can make rotational slip and therefore loss of quarter more of a concern, and the additional cost for the special axle, potential loss of lateral rigidity in the axle if fully 'cranked', etc. are factors too. (What we see happening instead is the move to better rod alloy steels, recognition via such things as the two different 'tandem' rod arrangements that keeping inertial mass well inboard is a Good Thing, and ultimately the kind of refined 'retrofit kit' that turned the track-wrecking T&P 600 class into something of a silk purse in the 1930s...

_________________
R.M.Ellsworth


Last edited by Overmod on Thu Feb 13, 2020 12:00 pm, edited 1 time in total.

Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 11:51 am 

Joined: Fri Nov 16, 2007 10:21 pm
Posts: 116
Tyler,

You might be confusing balancing (which would be most important when running at speed) with weight distribution.

We use a portable scale on a drop table to weigh each driver individually to determine load on the bearings for even weight distribution, which would then be adjusted by shimming the spring rigging.

Eric


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 12:30 pm 

Joined: Tue Aug 24, 2004 10:34 pm
Posts: 783
I know most of this mechanical engineering mentioned is voodoo way over my head. But in understanding stresses and how it demands to be set free both in boilers and running gears helps understand why things start to wear out and tear up things. Some locomotives more chronic than others. Also explains finding the "sweet spot" while in operation. Again it is above my conceptual ability to think of balancing a steam locomotive to the best of abilities with all the rotational forces added to the equation. So I thank all who have contributed to the questions and this thread in general. Fascinating science.

Have often pondered stress from worn out running gears causing problems elsewhere in the "system". But never really thought about the "balancing" act of mass starting from the beginning of movement.

Regards, John.


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 2:14 pm 

Joined: Thu May 24, 2012 1:37 pm
Posts: 1267
Some additional considerations:

Note that dynamic augment force can act longitudinally as well -- this is a major reason for the importance of Franklin self-adjusting wedges, and the effectiveness of 360-degree support for driving axles in bearings (a vaunted characteristic of roller-bearing boxes, and a frequently-misunderstood consideration for plain-bearing lubrication schemes...) The idea is to permit relatively free vertical motion while holding longitudinal motion to as absolutely close limits as possible -- recognizing that much of the "motion" is related to compression of the lubricant film between the hardened liner surface and the engaging surface of the box.

Likewise, a substantial amount of augment 'balance' can be accommodated vertically via the suspension without severe effect on the track, as it is nominally 'sinusoidal' motion and both smoothly applied and smoothly released. There is substantial 'preload' imposed on the equalization by the imposed weight acting through the spring range, and the rotating augment forces of course have to accelerate not only the mass of the wheelset and boxes but also the proportion of the spring rigging that moves, so you can get surprisingly close to the 'balance point' where traction adhesion matches the nominal weight 'seen' by the driver contact patch. (Beyond which you get more and more severe propensity to slip (or slide on braking) before you start to see things like rail kinking...)

But all bets are off as soon as you go below "zero adhesion" on a driver, which is one of the issues supposedly recorded on film during the AAR tests on the C&NW E-4s in the late Thirties. If the inertial force pulls the wheel up off the rail, the combination of spring and gravity force accelerates it down with, now, true hammer-blow kinetics, and even a small amount of lift creates not only dramatically larger force, but a much shorter time for the rail to absorb and dissipate it. This will be where the distinctive kinking or breaking of rails becomes most observed... see the ACL R1 testing, where the amount of overbalance was remarkably overcalculated as a matter of industry 'best practice'!

Now, surge is the effect when inertial forces can't produce either 'hammer' against the pedestals or play of any kind ... but start accelerating the whole chassis fore and aft with wheel revolutions. Naturally the higher the engine mass, the less this peak acceleration will be (and the shorter the stroke, the less time there will be for it to build up vehicle acceleration per revolution, which will of course nearly reverse on the 'opposite stroke'). Equally naturally, you can see that tying the mass of the tender in with the locomotive mass in the longitudinal plane ... which is one thing the Franklin radial buffer does very effectively at any angle ... and tuning the buffer springs appropriately to snub any otherwise 'critical frequency' effects will reduce the nominal effect of surge on the cars of a following train. (It might be noted that several people, including H.T.Harley of PRR, have distinctly felt the effects of surge at very high rotational speed, in Harley's case from at least two classes of duplex locomotives as they came into synchronization.)

The effect of cross-balancing also applies to quartered double-acting engines in a less obvious way. You will note that the inertial effects on one side will always be 'leading' those on the other side by about 90 degrees, so one side will be lifting/falling in the pedestals while the other side is producing more or less longitudinal surge effect. The acceleration of mass produced by these two sides is very different, and the effect is to lift the 'vertical' side relative to the other. (This was the functional reason for the relatively utter failure of Bethlehem auxiliary locomotives, a design that could be precisely 'statically balanced' as it was gear-driven, in high-speed service). A primitive way to address this is to restrict lateral in the pedestals tightly ... but this is a disaster in a great many modern respects, not least among them the practical use of lateral-motion devices.

In my opinion, the relative importance of this is too often understated in discussions of balancing: the great unsung take-home message of the technique applied to the late British 2-10-0s was not that they were balanced more 'perfectly' (of course they weren't!) or that they had advanced methods of putting weight in the wheel centers (they certainly did not!) ... and a moment's reflection will show they had pathetically little lateral restoring force at either end against yaw/hunting force couples, or particularly effective tender-mass utilization. So there was tremendous dynamic augment present in the design even at comparatively middling speeds ... why, then, could the locomotives reach speeds of well over 90mph? Think about it.

_________________
R.M.Ellsworth


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 5:20 pm 

Joined: Fri Jan 01, 2016 5:47 pm
Posts: 16
Thank you, all, for the very interesting information and insights.

hadder wrote:
Tyler,

You might be confusing balancing (which would be most important when running at speed) with weight distribution.

We use a portable scale on a drop table to weigh each driver individually to determine load on the bearings for even weight distribution, which would then be adjusted by shimming the spring rigging.

Eric


Yes, you're correct. All my reference materials call the building the "Wheel Balancing Shed," thus my phrasing, but I suspect that it was used for static weight distribution. Thanks for clarifying, and for explaining how you do it. What triggers a need to check the weight distribution? Routine maintenance and inspection, or only major overhauls?

I don't doubt that these scales were also used for weighing engines for engine crew wages, bridge and turntable ratings, etc., but there must have been a reason for the independent scales and the name — assuming that is indeed what the railroad called it.


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 5:33 pm 

Joined: Thu May 24, 2012 1:37 pm
Posts: 1267
If it is the 'Wheel Balancing Shed' then it's highly likely it would have had -- and perhaps still does have -- a dynamic balancing rig. As well, perhaps, as facilities for determining center of percussion of mains, weighing each end of rods, making up and casting counterbalance-mass alloy, etc.

.

_________________
R.M.Ellsworth


Last edited by Overmod on Thu Feb 13, 2020 7:01 pm, edited 1 time in total.

Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 5:42 pm 

Joined: Fri Jan 01, 2016 5:47 pm
Posts: 16
Overmod wrote:
If it is the 'Wheel Balancing Shed' then it's highly likely it would have had -- and perhaps still does have -- a dynamic balancing rig. As well, perhaps, as facilities for determining center of percussion of mains, weighing each end of rods, making up and casting counterbalance-mass alloy, etc.


What might such a device look like? There's currently nothing else in the building (besides a scale test car and the two shower enclosures stored in there — both pictured in my original post).

For context, the building is located on the other side of the roundhouse from the Back Shop and the various workshop buildings and doesn't have any attached workshop or office space. It's just four walls, a roof, and end doors to keep the scales out of the weather.


Offline
 Profile  
 
 Post subject: Re: Steam locomotive wheel balancing
PostPosted: Thu Feb 13, 2020 5:45 pm 

Joined: Sun Aug 22, 2004 7:19 am
Posts: 5746
Location: southeastern USA
Way back in my days at Spencer it was simply know as the "scale house." No idea where the new designation came from

_________________
Santayana: "He who does not remember the past is condemned to repeat it."
Corollary: "He who does is doomed to watch those who don't repeat it anyway."


Online
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 30 posts ]  Go to page 1, 2  Next

All times are UTC - 5 hours [ DST ]


 Who is online

Users browsing this forum: Dave, Google [Bot] and 58 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: