Trains.com

Who Built The Highest Quality 4-8-4's?

72831 views
259 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    August 2005
  • From: At the Crossroads of the West
  • 11,013 posts
Posted by Deggesty on Sunday, September 22, 2019 9:15 PM

Flintlock76

 

 
Deggesty

Yes, the Notre Dame football team is an excellent team--but they did not sing the victory march song last night--theDawgs bit them.

 

 

 

Hmmm, is Georgia still holding that grudge dating from Uncle Billy Sherman's visit?  That would explain a lot.

 

No, I think they just like to play football and win. I doubt if anybody from Florida did any harm to Georgians--but there seems to be no love lost between Athens and Gainesville, Florida.

I don't know how much Georgians think about "Cump" Sherman nowadays. I do know that my then two year old grandfather's home, in Up Country South Carolina was not burned in early 1865 because my grandfather;s oldest sister refused to leave the house when everybody was ordered out. But, that is all gone now.

Johnny

  • Member since
    January 2019
  • From: Henrico, VA
  • 9,728 posts
Posted by Flintlock76 on Sunday, September 22, 2019 8:39 PM

Deggesty

Yes, the Notre Dame football team is an excellent team--but they did not sing the victory march song last night--theDawgs bit them.

 

Hmmm, is Georgia still holding that grudge dating from Uncle Billy Sherman's visit?  That would explain a lot.

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Sunday, September 22, 2019 8:32 PM

Jones1945

Only reason for the team to be at the IC Station when going to the West Coast would be that they had ridden the South Shore from South Bend to Chicago so they could make connections with the IC train to New Orleans and thence West.  Rather circuitous routing.

Never too old to have a happy childhood!

              

  • Member since
    August 2005
  • From: At the Crossroads of the West
  • 11,013 posts
Posted by Deggesty on Sunday, September 22, 2019 8:11 PM

Yes, the Notre Dame football team is an excellent team--but they did not sing the victory march song last night--theDawgs bit them.

Johnny

  • Member since
    January 2019
  • From: Henrico, VA
  • 9,728 posts
Posted by Flintlock76 on Sunday, September 22, 2019 7:42 PM

Oh jeez, all this Notre Dame talk is bringin' out the Irish Catholic in me.

Why fight it?

https://www.youtube.com/watch?v=Nn_-JgexeLI  

  • Member since
    September 2013
  • 6,199 posts
Posted by Miningman on Sunday, September 22, 2019 4:11 PM
Mike says:
 
"The Mets did good this year by keeping it interesting deep into September. Wish we had Angus D. McDonald. 
  • Member since
    April 2018
  • 1,618 posts
Posted by Jones1945 on Saturday, September 21, 2019 11:56 PM

Overmod

I am pretty sure your research will divulge the important points, but here are some that might help direct your attention:

Be sure to look carefully at the different systems that were applied to the N&W A class over the years, and the reasons why each was adopted and why some were to be 'preferred' or redesigned.  I think you will find the original type to be of particular interest, and its early point of failure (and correction) a fascinating example of how modern innovations were handled in the real world by intelligent people.

The technical difference between the Timken and SKF 'principle' is important to keep in mind.  A SKF bearing has barrel-shaped rollers and can tolerate a certain amount of 'twist' inherently, whereas Timken rollers are always found in 'pairs' with opposing taper and should be preloaded correctly when installed and given side-to-side structure (such as a cannon box) when used for drivers.  It was remarked on here years ago that driver axles DO flex in service to a nontrivial extent, and this is better tolerated by an SKF box than by Timkens.  Or at least so was the wisdom in the 'late days' of steam maintenance practice -- the guy I learned this from is still the only source who remembers promotional material for centrifugally-cast hollow driver axles.

When you get done looking at the driver axle bearings, look at the technology behind roller rods, which is even more fascinating.  (There is a deduction by Chapelon about the lightweight high-dynamic-steel Timken lightweight rods that you will find highly interesting if it turns out to be true...) 

This is a handy guide for me to focus on crucial things on this topic, much appreciated!

 

Miningman
 
Interesting to know Angus D. McDonald was a great team captain of Notre Dame's baseball team. He was a great man, a natural-born leader. Yes
 
  • Member since
    September 2013
  • 6,199 posts
Posted by Miningman on Saturday, September 21, 2019 1:12 AM
  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Friday, September 20, 2019 10:45 AM

Jones1945
I am doing some researching about the differences between older SKF and Timken bearings for railroading, a new topic for me to be honest.

I am pretty sure your research will divulge the important points, but here are some that might help direct your attention:

Be sure to look carefully at the different systems that were applied to the N&W A class over the years, and the reasons why each was adopted and why some were to be 'preferred' or redesigned.  I think you will find the original type to be of particular interest, and its early point of failure (and correction) a fascinating example of how modern innovations were handled in the real world by intelligent people.

The technical difference between the Timken and SKF 'principle' is important to keep in mind.  A SKF bearing has barrel-shaped rollers and can tolerate a certain amount of 'twist' inherently, whereas Timken rollers are always found in 'pairs' with opposing taper and should be preloaded correctly when installed and given side-to-side structure (such as a cannon box) when used for drivers.  It was remarked on here years ago that driver axles DO flex in service to a nontrivial extent, and this is better tolerated by an SKF box than by Timkens.  Or at least so was the wisdom in the 'late days' of steam maintenance practice -- the guy I learned this from is still the only source who remembers promotional material for centrifugally-cast hollow driver axles.

When you get done looking at the driver axle bearings, look at the technology behind roller rods, which is even more fascinating.  (There is a deduction by Chapelon about the lightweight high-dynamic-steel Timken lightweight rods that you will find highly interesting if it turns out to be true...)

  • Member since
    April 2018
  • 1,618 posts
Posted by Jones1945 on Friday, September 20, 2019 10:11 AM

Overmod

I'd think quite the opposite: you don't see more than these two purchased, or converted after the war, whereas the 'patent' spring-pad journal and hub-liner lubricators were used on many classes right to the end.  

Note, however, the difference between this design and something like a Hennessy that uses lateral motion of the drivers to actively pump lubricant -- something SP needed a separate apparatus to do.  At least one Forum member will find this article of high interest!

The fact that they got 'one of each' (Timken and SKF) is instructive ... but it also shows a fundamental ignorance of where the distinctive competence of each design was.  In my opinion at least, here are reasons SKF bearings are preferable for drivers, and Timken bearings for lead and trailing trucks, and you can see some of that in the detail pictures above -- but you see only 'bearing chauvinism' on each particular locomotive.

Thank you so much for the link, Mr. Overmod. This is the first time I see a clear pic showing how the oil cellar was "removed" from the journal box on the trailing truck of an SP steam locomotive, also a very clear description of how the spring-pad oil lubricator worked. I am doing some researching about the differences between older SKF and Timken bearings for railroading, a new topic for me to be honest. Reading some posts like this one: http://cs.trains.com/trn/f/111/t/7522.aspx and other posts from various forums as well. Idea

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Thursday, September 19, 2019 4:45 PM

Jones1945
So, I guess the spring pad lubricator invented by the SP was overshadowed by the roller bearing on the GS-5s

I'd think quite the opposite: you don't see more than these two purchased, or converted after the war, whereas the 'patent' spring-pad journal and hub-liner lubricators were used on many classes right to the end.  

Note, however, the difference between this design and something like a Hennessy that uses lateral motion of the drivers to actively pump lubricant -- something SP needed a separate apparatus to do.  At least one Forum member will find this article of high interest!

The fact that they got 'one of each' (Timken and SKF) is instructive ... but it also shows a fundamental ignorance of where the distinctive competence of each design was.  In my opinion at least, here are reasons SKF bearings are preferable for drivers, and Timken bearings for lead and trailing trucks, and you can see some of that in the detail pictures above -- but you see only 'bearing chauvinism' on each particular locomotive.

  • Member since
    April 2018
  • 1,618 posts
Posted by Jones1945 on Thursday, September 19, 2019 2:35 PM

Overmod

...You cannot predict "maximum speed" from the usual numbers quoted in power calculations, and there are some whopper examples (the C&NW E-4 and the ATSF 3460 class) where very 'high-speed' designs turned out to be incapable of actually reaching more than a few mph over 100 with a meaningful consist in tow, while on the other hand some designs of eight-drivered power could reach substantial speeds with proper valve gear, port-and-passage design, and good steam-generation capacity matched with proper 'front end' proportioning and tuning...

Thanks for catching that, Mr. Overmod. Bow What I should have said is Indicated horsepower, that was what a forumer from previous pages mentioned about the figure provided by Lima, which didn't indicate what kind of hp it represents. I just reviewed this thread for the formula Alco used to calculate and *estimate horsepower of steam engine. As you stated in this and previous posts that it is impossible to predict the top speed of a steam locomotive by power calculation, which I totally understand, the following figures and calculations are just for my curiosity

Saturated Steam HP = 0.0212 X P X A

Superheated HP = 0.0229 X P X A

HP = Horsepower

P = Boiler Pressure (in Pounds per Square Inch)

A = Area of One Cylinder (in Square Inches)

Saturated Steam HP of a GS-2 = 0.0212 X 250 X 810 = 4293

Superheated HP  of a GS-2 = 0.0229 X 250 X  810 = 4637.25

Saturated Steam HP of a GS-3 = 0.0212 X 280 X 832 = 4938.75

Superheated HP  of a GS-3 = 0.0229 X 280 X  832 = 5334.784

Saturated Steam HP of a GS-4 and GS-5 = 0.0212 X 300 X 816 = 5189.76

Superheated HP  of a GS-4 and GS-5 = 0.0229 X 300 X  816 = 5605.92

Saturated Steam HP of a GS-6 = 0.0212 X 260 X 816 = 4497.79

Superheated HP  of a GS-6 =  0.0229 X 260 X 816 = 4858.46

-------------------------------------------------------------------------

More calculation just for fun:

Saturated Steam HP of the PRR S1 = 0.0212 X 310 X 572 X 2 (Duplex) = 7518.37 (-5% = 7142.45)

Saturated Steam HP of the PRR T1 = 0.0212 X 310 X 513.5 X 2 (Duplex) = 6749.44 (-5% = 6411.97)

Saturated Steam HP of the PRR Q1 = 0.0212 X 310 X 1151 (Duplex) = 7564.37  (-5% = 7186)

Saturated Steam HP of the PRR Q2 = 0.0212 X 310 X 1227 (Duplex) = 8063.84 (-5% = 7660)

Saturated Steam HP of a NYC Niagara = 0.0212 X 275 X 816 = 4757.28

Saturated Steam HP of the ATSF Class 2900 = 0.0212 X 300 X 896 = 5698.56

Saturated Steam HP of a PRR K4s = 0.0212 X 205 X 702 = 3050.89

Saturated Steam HP of the ATSF Class 3460 = 0.0212 X 300 X 693 = 4407.48

Saturated Steam HP of the NYC Dreyfuss Hudson = 0.0212 X 265 X 653 = 3668.55

Saturated Steam HP of the MILW F7 Hudson = 0.0212 X 300 X 705 = 4483.8

 

Overmod

...I think there were relatively few opportunities for the GS-3s or -4s to run at very high speed (at least for any meaningful railroad purpose, either fast scheduling or 'making up time') and I have no idea whether their valves and passages would support even the speeds that, say, an ATSF 2900 class were reported to reach.  My opinion is that the locomotives, with the possible exception of the two that were equipped with lightweight roller-bearing rods, were nowhere near capable of reaching, far less sustaining, a speed of "120mph", although it is quite possible their balance was calculated for that speed...

So, I guess the spring pad lubricator invented by the SP was overshadowed by the roller bearing on the GS-5, the last two Daylight engines designed to be capable of reaching 110mph but probably seldom or never hit their highest design speed. Though it doesn't mean the design of the spring pad lubricator wasn't great, I guess.

I much more prefer the front end and driving cap design of GS-2, 3, 6 to the enlarged nose (to house the extra mar light) on the GS-4 and 5, that's why I didn't put too much attention on 4449. She is an example of great steam in America. Coffee

  • Member since
    February 2005
  • 2,366 posts
Posted by timz on Thursday, September 19, 2019 12:25 PM

Jones1945
what was the fastest operating speed of the Daylight engines (GS-2/3/4/5)

Timetable limit was 75. No one knows what they actually did.

Jones1945
If the GS-4/5 could really develop 5500 idhp ihp or hp, it should be possible for them to exceed their design top speed (110mph) when hauling 1000 tons or less

Maybe, if it could do 5500 hp at 110 mph, which it couldn't.

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Thursday, September 19, 2019 6:23 AM

What is idhp?  Indicated horsepower (or ihp) is a calculated figure derived from measurements at the cylinders; think of it as a measurement of the mass flow physically available to make horsepower rather than something that produces the 'drawbar pull at speed' curves that are useful in determining practical consists for locomotives in particular services to pull.  A more useful index is wheelrim torque, as measured on a locomotive test plant, which includes some of the machine and fluid-mechanical losses, and if carefully analyzed may provide interesting design information, but is still (in my opinion) not as useful as dbhp, 'drawbar horsepower', which is what a dynamometer car measures and is the "thing" against which train resistance (as for example calculated via the Davis formula) can be compared.  In my experience comparative data are not often plotted as 'horsepower'; they are presented as a Cartesian plot of drawbar pull (on the y axis) vs. speed (on the x axis) which for railroad purposes is more useful than, say, knowing where the "peak" of the nominal horsepower curve is supposed to lie.

You cannot predict "maximum speed" from the usual numbers quoted in power calculations, and there are some whopper examples (the C&NW E-4 and the ATSF 3460 class) where very 'high-speed' designs turned out to be incapable of actually reaching more than a few mph over 100 with a meaningful consist in tow, while on the other hand some designs of eight-drivered power could reach substantial speeds with proper valve gear, port-and-passage design, and good steam-generation capacity matched with proper 'front end' proportioning and tuning.

I think there were relatively few opportunities for the GS-3s or -4s to run at very high speed (at least for any meaningful railroad purpose, either fast scheduling or 'making up time') and I have no idea whether their valves and passages would support even the speeds that, say, an ATSF 2900 class were reported to reach.  My opinion is that the locomotives, with the possible exception of the two that were equipped with lightweight roller-bearing rods, were nowhere near capable of reaching, far less sustaining, a speed of "120mph", although it is quite possible their balance was calculated for that speed.

It would be interesting to hear some of the stories (some 'not for attribution', I suspect!) about the actual high speeds that might have been achieved with 4449.  My understanding is that Ross is one of the 'likelier subjects' to have tried this and he might care to 'hint' a bit about what a GS-4 with freshly refurbished but fully-broken-in running gear might have produced.  To my knowledge this locomotive has one of the most sophisticated burners applied to a conventional locomotive boiler, and that might facilitate 'forcing' to a greater extent that a regular type of burner (e.g. von Boden-Ingles or Thomas) might be able to produce effectively.

  • Member since
    April 2018
  • 1,618 posts
Posted by Jones1945 on Thursday, September 19, 2019 1:43 AM

Anonymous
QUOTE: Originally posted by jlampke

QUOTE: Originally posted by rockymidlandrr

Alco and Lima built the best. The UP super 800s and the SP Gs-4 should be plenty of evidence of who built the best. The 844 and the 4449 are the best examples of the respective classes. The 844 has never been retired, now that is craftmanship!


There's no doubt they are great 4-8-4's.The fact is I first posted the question hoping for reinforcement of my bias towards 4449. It has been said in this discussion that all 4-8-4's were great, and I like to think that is the case. Regarding 844 staying in service all these years, I think a huge amount of the credit for that is due to the efforts of Steve Lee and his fine crew. The UP should get some credit for that too.... They could've done like the SP and so many other railroads and just tossed everything to scrap. Thank God the city of Portland asked for a GS locomotive. We wouldn't have 4449 today.

I read that the boys in Southern CA did a first-class job of restoring 3751, and am looking forward to seeing 2926 on the road again. Pretty sure the Baldwin fans would have a thing or two to say about who made the best 4-8-4's!!!!

All 4-8-4's are great locomotives. I just hope they are all receiving some degree of care and preservation. Has anyone seen the others lately?

When 4449 was put on display in Oaks Park in 1958, I'll bet nobody really believed she would one day be put back into service. Who knows what the future holds for the remaining 4-8-4's on display and in storage in North America?

I'll admit though that I am partial to SP's Lima-built GS locomotives.....



A recent PBS show on the Daylights claimed some portions of the run were upgraded for 120mph speeds. The 80" drivered GS3's went into service pretty quickly after the 73" drivered GS2's and the show mentioned that SP swapped locos rather than stop for water and fuel. So I'm guessing they used the older GS2's on the tougher parts of the run or for double heading, while running the GS3's at Acela-like speeds. Was that the case or were the GS2's assigned to other trains?

Since this is a 13-year-old thread, I cannot ask the author for more details regarding the 120mph-claim by someone, maybe the Southern Pacific (probably not) in the PBS show. I really want to know what was the fastest operating speed of the Daylight engines (GS-2/3/4/5) and the sections of the route where they reached such speed.

I haven't finished the "Southern Pacific Daylight Train 98-99" by Richard K Wright yet, and my order of a copy of the "1986 1st Edition Southern Pacific Railroad Passenger Trains Volume 1 Night Trains of The Coast Route by Dennis Ryan & Joseph Shine" is on its way. I could probably get the answer from the "Southern Pacific Daylight Locomotives" by Robert J. Church, but lack of space on my bookshelf is a problem for me.

If the GS-4/5 could really develop 5500 idhp ihp or hp, it should be possible for them to exceed their design top speed (110mph) when hauling 1000 tons or less, base on the AAR road test result. Though I can't find any source to back up the 120mph claim. I have the impression that Southern Pacific never emphasis the speed but the outstanding service of the Daylight train. The schedule of Daylight train was very well arranged and maintained, the Daylight engines were designed to keep the train on schedule but not to break any speed record. 

 

Your input would be greatly appreciated! Yes

Home movie footage from about 1937-1938 including scenes of couples traveling on the "Coast Daylight" train of the Southern Pacific Railroad with footage of scenery: 

https://archive.org/details/HomeMovieOfSanFranciscoAndOfMadisonWisconsin

  • Member since
    January 2011
  • 2 posts
Posted by grand trunk western 4444 u4c on Sunday, January 23, 2011 5:38 AM

I would have to agree with cooslimited on sp's gs5  I personally love their design work. The 4449 ts a gs4 but at train festeval2009 I studyed it nonstop.

Another choice of mine is Grand Trunk western's u3b or c class northerns whigh ever GTW 6325 is.

I personally like coal fired engines But that's just my opinion.

  • Member since
    July 2004
  • 803 posts
Posted by GP40-2 on Friday, December 12, 2008 8:41 PM
Likewise, after examining the design/engineering of the N&W J class, I don't see any difference in the last locomotives produced vs. the earlier ones. Seems the N&W used the same plans with minor modifications over the years. The C&O J class did however change from the original J3 of 1935, and the J3a of 1948. Lima kept many of the dimensions the same, however, the J3a had an advanced boiler design that incorporated ideas such as a longer combustion chamber and larger diameter firetubes. Both the N&W and Lima designs were excellent, extremely powerful 4-8-4s.
  • Member since
    August 2003
  • From: Northern VA
  • 484 posts
Posted by feltonhill on Friday, December 12, 2008 6:19 PM

As near as I can determine, there  were no significant differences between the any of  the Class J's as finally turned out.  The wartime J1's were converted to Js in 1944 just prior to 610's tests on PRR in December that year.  They were highly advanced in 1941 and matched the final technology in 4-8-4's in the early and late postwar period, which is to say the NYC Niaragas and C&O J3a's through 1948. 

The only differences I know of are slightly different tender design.  Nothing on the locomotive until the siderods were changed on about 4-5 J's from tandem to single.  I'm on the road now so I don't know the exact number (s).

 The final group of J's (1950) were the last 4-8-4's for a domestic RR in the US.

  • Member since
    February 2005
  • 2,366 posts
Posted by timz on Friday, December 12, 2008 4:08 PM

superheat
A lot of people state that the N&W J class was the ultimate 4-8-4.

Could be, for all anybody knows.

superheat
...and so were able to benefit from advances in steam locomotive design and engineering?

Advances before 1941, you mean? Or advances between the first and last N&W 4-8-4s?

Anybody know how N&W's last 4-8-4s differed from their first?

  • Member since
    November 2006
  • 10 posts
4-8-4 Evolution
Posted by superheat on Friday, December 12, 2008 9:16 AM

Correct me if I'm wrong, but was the N&W the last builder of 4-8-4's? A lot of people state that the N&W J class was the ultimate 4-8-4. To what degree was that a result of the fact that their 4-8-4's were built some 10 (or thereabouts) years later than most of the other railroad's 4-8-4's, and so were able to benefit from advances in steam locomotive design and engineering? If Lima had built 4-8-4's into the 50's, utilizing advances available by then, I wonder how they would've compared? 

  • Member since
    February 2005
  • 2,366 posts
Posted by timz on Tuesday, December 9, 2008 8:54 PM

It's true, fans do overemphasize horsepower as a measure of locomotive performance-- a more useful statistic would be total work done at the drawbar per dollar of operating cost. But we don't know that.

(And the 7498-dbhp figure is particularly useless, since we don't know how long a given engine could maintain it, or how close the average engine in average condition with average coal and average water could come to matching it.)

As to whether we can "measure" horsepower, that's a semantic argument. Can we "measure" the area of a rectangle drawn on a plane? We measure the sides, and confirm the corners are 90 degrees, then we calculate the area. You're saying we haven't measured the area-- I'd say we have, but I'm no expert on the definition of "measure".

  • Member since
    December 2008
  • From: Mid-Ohio
  • 94 posts
Posted by Trace Fork on Tuesday, December 9, 2008 5:28 PM

timz

You first said larger drivers produced lower power-- "What I am inferring is that all other factors being identical, the drawbar pull exerted with 67" drivers would be less than that with 65.25" drivers.  The output of the locomotive would be less, so the dynamometer reading would reflect the reduction in measured drawbar force.  As a result the final horsepower calculation would be lower."

Now you're assuming it's smaller drivers that produce lower power. You must have gotten careless with one of your assumptions?

I see what your driving at here.  I was mistaken in stating that "the final horsepower calculation would be lower"  I should have said it would differ.  Without a physical measurement on 67" drivers, my assumption that horsepower would be reduced was erroneous.

Furthermore, I am not now assuming that the smaller drivers produce less power.  In fact the 64" driver was assumed to result in a 2.249% greater drawbar force than that of the 65.25" driver. 62350lbs vs. 61125lbs.  The end result here is that the smaller driver, due to it's inability to travel as great a distance, given equal application of energy, as the larger driver, produces lower power OVER TIME.  Force is not measured over time, it is measured at a specific point. Horsepower is a calculation of the application of measured force over time. Thus the reason for a speedometer on the dynamometer car, to determine as accurately as possible, velocity at the specific time force was measured.  I think the confusion here is from our differing definitions of power.  To clarify, when I speak of power I am referring to force and/or torque. Each a measurable value.  I don't want to get into a pi$$ing match here, so from this point forward I will concede that the formula to calculate horsepower, being the accepted one, is absolute. But I must still respectfully disagree that horsepower is the best indicator of locomotive performance.


timz

If in 1943 we wanted to know a 2-6+6-6's maximum drawbar horsepower at 46 mph with 65.5-inch drivers, would advanced physics and mathematics have told us? If so, it still will-- math and physics are just as advanced now as they were in 1943. So how about it-- think their 7498 dbhp measurement was correct?

I don't profess that math and physics would produce the same number.  And as I stated in a previous post, I don't dispute the Allegheny was capable of producing 7498 instantaneous dbhp. In fact I have the numbers in published form.  And at last, an absolute formula to verify it from the measured values. As to it's correctness?  Since it was not a measurement, but a calculation, I must assume the accepted formula was applied correctly using the measured values of force and velocity.

timz
 

Lotsa luck finding a book that tells you how to calculate a locomotive's drawbar horsepower from first principles. They'll tell you how to guess at it, but if you want to know what it actually is you'll have to measure it, and you won't be astonished if the calculated guess is wrong.

This is the point I am trying to make.....there exists no accurate method to measure horsepower, It is a calculated value only. And as you state above, math and physics would indeed only render an approximation of the drawbar horsepower (or drawbar force for that matter) that the Allegheny (or any locomotive) could exert. That is perhaps one reason the engineering staff at Lima assigned starting tractive effort values to the 2-6-6-6 at 85% of boiler capacity. A tractive effort which was determined inaccurate by subsequent dynamometer testing. Engineers had to estimate energy losses between the boiler and the rail, and those losses were based on many factors. The same calculations they used will allow the estimation of the differing effects of driver size in application of force at the railhead.  But you're right.  It is an educated guess. An accurate measurement would need to be taken at the rail and / or wheel tread to determine an actual value. Although the texts to which I elude will not specifically provide formulas to calculate drawbar horsepower of any locomotive, they will discuss the application of principles that will allow it's estimation through calculation.  And if given the availability of accurate values to calculate from, these estimates should be quite close.

I REALLY FEEL MUCH BETTER, NOW THAT I'VE GIVEN UP ALL HOPE
  • Member since
    February 2005
  • 2,366 posts
Posted by timz on Tuesday, December 9, 2008 1:23 PM

Trace Fork
Now for an example of how flawed this formula is...

Trace Fork
Even though this formula is the accepted standard for horsepower calculation, I question it's accuracy.

Everyone agrees that a 7000 dbhp locomotive won't equal exactly 7000 flesh-and-blood horses. (You remember there's also something called a metric horsepower, and 7000 of them won't be "right" either.)

But "this formula is the accepted standard" in the same sense that 12 inches = 1 foot is "the accepted standard". This unit of power has been defined as 550 ft-pounds per second, or 33000 foot-pounds per minute, or 375 mile-pounds per hour-- and it has no independent existence beyond the various forms of the definition. It's still called a "horsepower", but its definition won't change even if somebody breeds a horse that's good for 10 dbhp.

  • Member since
    February 2005
  • 2,366 posts
Posted by timz on Tuesday, December 9, 2008 12:54 PM

Trace Fork
I used 2% (actually 2.249%), and 2mph because I did not have actual measured values. These values are assumed.

You first said larger drivers produced lower power-- "What I am inferring is that all other factors being identical, the drawbar pull exerted with 67" drivers would be less than that with 65.25" drivers.  The output of the locomotive would be less, so the dynamometer reading would reflect the reduction in measured drawbar force.  As a result the final horsepower calculation would be lower."

Now you're assuming it's smaller drivers that produce lower power. You must have gotten careless with one of your assumptions?

Trace Fork
This is not the correct forum to discuss the advanced physics and mathematics involved in calculating the energy applied to a rail by a certain diameter wheel, and how that translates to the force applied to a linear object like a locomotive drawbar.

If in 1943 we wanted to know a 2-6+6-6's maximum drawbar horsepower at 46 mph with 65.5-inch drivers, would advanced physics and mathematics have told us? If so, they still will-- math and physics are just as advanced now as they were in 1943. So how about it-- think their 7498 dbhp measurement was correct?

Trace Fork
There is an abundance of text available that will better answer your questions than available band width will allow me to answer them here.  Libraries are a good first source.

 Lotsa luck finding a book that tells you how to calculate a locomotive's drawbar horsepower from first principles. They'll tell you how to guess at it, but if you want to know what it actually is you'll have to measure it, and you won't be astonished if the calculated guess is wrong.

  • Member since
    December 2008
  • From: Mid-Ohio
  • 94 posts
Posted by Trace Fork on Tuesday, December 9, 2008 11:16 AM

feltonhill

I don't see the point.  If everyone uses 33,000 ft-lbs/min, comparisons at any speed are equivalent if  purportedly "inaccurate".  So everyone using the English system of units is wrong.  What should the correct unit be if 33,000 is incorrect?  Must we throw out all the diesel and electric HP ratings?  Should we use a different size horse?

Is the metric system of units any less flawed or arbitrary? 

Sorry to ask so many questions, but there a several parts of this I don't understand.

Before this all began.  If I were to have asked you, "WHAT IS HORSEPOWER?  What would have been your response? 

Force in puonds is a direct measurement, and as such reveals the amount of energy applied to an object. Force measurements are used to calculate diesel horsepower as well.  The formula is a little different though, because the force is measured at a rotating shaft:

HP = (T X N) / 5252

Where:  T = Torque (lb/ft)      N = Speed (rpm)        5252 = (33000 / Pi) / 2

Torque = F X R

Where:   F = Force (lbs)         R = Radius (ft)

So you see the same value of 33,000 is used in this calculation.  My question is why?  What does knowing this value tell you?  And no...metric units of calculation in no way make the calculation more accurate, and / or useful.

Railroads used dynamometer readings mainly to develop tonage ratings for their locomotives (steam and diesel).  Railroad motive power, and operating department officials did not need to know how many horses could be replaced by a locomotive. They needed to know how much force is applied at the drawbar, and if that force was sufficient to overcome the forces of gravity and friction involved in moving a train from point A to point B.  By the turn of the 20th century the steam engine, and the internal combustion engine, had all but rendered the horse obsolete for industrial use.  As creative as man kind is, a century later they have yet to develop a better indicator of a machine's capability than to compare it to horses.  Or have they?  Force and torque, both direct measurements.

feltonhill

If TE is increased 2% because of a reduction in driver size, why does speed fall exactly 2 mph?  

Sorry to ask so many questions, but there a several parts of this I don't understand.

I used 2% (actually 2.249%), and 2mph because I did not have actual measured values. These values are assumed.  This is not the correct forum to discuss the advanced physics and mathematics involved in calculating the energy applied to a rail by a certain diameter wheel, and how that translates to the force applied to a linear object like a locomotive drawbar.  There are people who teach this for a living, and I am not one of them.  My training is in geodesy.  The speed will decrease (albeit at varying values) due simply to the fact that a smaller diameter wheel is also smaller in circumference, and will not travel as great a distance as a larger wheel given the same application of energy.

 Don't apologize for asking questions, that is how knowledge is gained.  There is an abundance of text available that will better answer your questions than available band width will allow me to answer them here.  Libraries are a good first source.  Seek them out.

Jim J.

 

I REALLY FEEL MUCH BETTER, NOW THAT I'VE GIVEN UP ALL HOPE
  • Member since
    August 2003
  • From: Northern VA
  • 484 posts
Posted by feltonhill on Tuesday, December 9, 2008 7:30 AM

I don't see the point.  If everyone uses 33,000 ft-lbs/min, comparisons at any speed are equivalent if  purportedly "inaccurate".  So everyone using the English system of units is wrong.  What should the correct unit be if 33,000 is incorrect?  Must we throw out all the diesel and electric HP ratings?  Should we use a different size horse?

Is the metric system of units any less flawed or arbitrary? 

If TE is increased 2% because of a reduction in driver size, why does speed fall exactly 2 mph?  

Sorry to ask so many questions, but there a several parts of this I don't understand.

  • Member since
    December 2008
  • From: Mid-Ohio
  • 94 posts
Posted by Trace Fork on Tuesday, December 9, 2008 12:32 AM

Horsepower is a calculation of force over time. The accepted formula to calculate horsepower of objects in linear motion is:

HP=    (F X V) / 33000

Where F= Force (lbs)          V=Velocity (ft/min)

Thus if an increased force is recorded at a lower velocity the calculated horsepower will be different, not necessarily constant.

So let's assume we are in the dyno car behind C&O #1608 on its record breaking trip.  A measured drawbar force of 61,125lbs has been measured at 46mph.  46mph=4048ft/min.

61125 X 4048 = 247434000       247434000 / 33000 = 7498.00dbhp

Now lets assume that we are on another trip behind #1608 and the drivers are now 64" instead of 65.25".  We will assume a 2% increase in drawbar force or 62350lbs, but this was recorded at 44mph. 44mph = 3872 ft/min.

62350 X 3872 = 241419200       241419200 / 33000 = 7315.73dbhp

Now for an example of how flawed this formula is, please read the following:

Even more interesting is how the formula came to be. It was originated by James Watt, (1736-1819) the inventor of the steam engine and the man whose name has been immortalized by the definition of Watt as a unit of power. ...

...To help sell his steam engines, Watt needed a way of rating their capabilities. The engines were replacing horses, the usual source of industrial power of the day. The typical horse, attached to a mill that ground corn or cut wood, walked a 24 foot diameter (about 75.4 feet circumference) circle. Watt calculated that the horse pulled with a force of 180 pounds, although how he came up with the figure is not known. Watt observed that a horse typically made 144 trips around the circle in an hour, or about 2.4 per minute. This meant that the horse traveled at a speed of 180.96 feet per minute. Watt rounded off the speed to 181 feet per minute and multiplied that by the 180 pounds of force the horse pulled (181 x 180) and came up with 32,580 ft.-lbs./minute. That was rounded off to 33,000 ft.-lbs./minute, the figure we use today.

Given the assumption that a horse can pull with a sustained force of 180lbs for one hour, and the arbitrary rounding to the closest whole number (or next even thousand), I'd say we need to get away from this fixation with horsepower.  Dynamometer cars are equipped with instruments that measure and record force, horsepower is calculated using the above formula.  Even though this formula is the accepted standard for horsepower calculation, I question it's accuracy.

I REALLY FEEL MUCH BETTER, NOW THAT I'VE GIVEN UP ALL HOPE
  • Member since
    February 2005
  • 2,366 posts
Posted by timz on Monday, December 8, 2008 8:50 PM

What's baffling me is... you said

Trace Fork
all other factors being identical, the drawbar pull exerted with 67" drivers would be less than that with 65.25" drivers.  The output of the locomotive would be less, so the dynamometer reading would reflect the reduction in measured drawbar force.  As a result the final horsepower calculation would be lower.

And then you said

Trace Fork
As the diameter of the drivers is reduced the tractive output [in pounds] increases proportionally.  That tractive output though, will be produced at a lower speed.

which implies the "final horsepower calculation" will remain constant. Wouldn't you say the latter makes more sense?

  • Member since
    December 2008
  • From: Mid-Ohio
  • 94 posts
Posted by Trace Fork on Monday, December 8, 2008 8:29 PM

timz

Trace Fork
As the diameter of the drivers is reduced the tractive output increases proportionally.  That tractive output though, will be produced at a lower speed.

If by "tractive output" you mean force (in pounds)-- then sure, at the same driver RPM drawbar pull will increase. If it increases (inversely) proportionally to driver diameter, then horsepower will remain constant.

If by "tractive output" you mean horsepower, then we have no way of knowing whether it goes up or down as we shrink the drivers. With 33-inch drivers it won't produce 15000 dbhp, even if we can circumvent the practical difficulties.

Please re-read some of my earlier posts.  I for one think that using a calculated value (Horsepower) is to use a flawed indicator of performance potential. Instead of using the term tractive output, which is more indicative of force on the rail, I should have said drawbar force, which is physically measured. I apologize for the confusion. Increased tractive force does relate to increased drawbar force, but I should not have mixed the terms.

The practical difficulties you mention would include the fact that the Allegheny's piston stroke was 33".  That would be very difficult with 33" drivers. And no...15,000 DBHP would not be achieved.

Jim J.

I REALLY FEEL MUCH BETTER, NOW THAT I'VE GIVEN UP ALL HOPE

SUBSCRIBER & MEMBER LOGIN

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

FREE NEWSLETTER SIGNUP

Get the Classic Trains twice-monthly newsletter