4-6-4 vs. 2-8-4 question.

I wanted to post some additional info re the N&W J and the NYC Niagara, for your consideration...

BOILER DESIGN

The Niagara boiler can be described as an open tubular design with less of a taper, larger individual flues, and a more level grate, while the J was of a more conventional design.

-The 1st ring OD of the Niagara boiler is 4" larger than the J

-The J boiler uses thicker boiler plates for the 1st and 2nd boiler courses.

-The J boiler max OD is 102" and the Niagara is 100" at the 3rd boiler course. In this area, the Niagara boiler sheet thickness was 1" and the J was (is) 7/8".

(It was not possible for NYC to use 10" larger diameter drivers and a larger diameter boiler, and stay within the roads 15-3" overall clearance diagram.)

-The J had a larger grate area than the Niagara, 107.7 vs 101 sq ft. A part of this difference was the use by N&W of a firebox water space difference of 1". The Niagara water space in this area was 6" and the J was 7".

-The rear of the Niagara boiler at the backhead and nearest the cab was 10" higher than the J, and featured an almost level grate vs the J grate which sloped. This is visible in builders photos.

All of the above is from the 1952 Loco Cyc.

The J was equipped with a "Master Mechanics" front drafting arrangement. (Pond, RME, Dec. 1946). The drafting arrangement on the Niagara was designed based on boiler drafting and development testing conducted at Selkirk, NY.

The Niagara was equipped with 4" flues. the J had 3-1/2" flues. The Niagara boiler was regarded as free steaming, such that at low throttle settings and when drifting, the use of smoke deflectors was required.

The J is equipped  with a Worthington 6SA Feedwater heater, the Niagara was equipped with a 7SA FWH, the largest offered by Worthington

EVAPORATION

Regarding evaporation, I also have  the Niagara test report and cover letter. The test report includes the following information, contained in the summary letter on page 3:

"As the evaporative capacityof the boiler is approached, the rate of firing increases much more than the rate of evaporation; for example, during the "K" series connducted with locomotive 6000 at Selkirk, the firing rate was increased from 15,000 pounds to 17,000 pounds of dry coal per hour, or 13.3%; but the combined equivalent evaporation was only increased from 147,000 pounds to 157,000 pounds per hour, or 6.8 percent."

The table of test results at 9,000 lb. of coal fired per hour show an "equivalent evaporation" of 88,900 for the S-1B and 96,300 for #6000.

The tested equivalent evaporation and the evaporative capacity of the boiler of up to 157,000 pounds per hour demonstrates the ability of this boiler to perform when "overfired".

DRAWBAR HP

The Summary Letter of the Niagara Test Report quotes 5050 Maximum Dynamometer Horsepower at 62.5 mph. (page 2.) The narrative quotes 5070 maximum drawbar HP at 62.5 mph.

The following info is from a Railway Mechanical Engineer published article by C. E. Pond and appeared in the December 1946 issue:

"Selected Data from Test Runs Developing Maximum  Horsepower"

275 psi boiler pressure

Speed 39.15   Cutoff 66% Water Evaporated per hour 104,946 Drawbar HP 4795

Speed 39.07   Cutoff 66% Water evaporated per hour 103,860 Drawbar HP 4784

300 psi boiler pressure

Speed 39.14   Cutoff 61% Water evaporated per hour 103,365 Drawbar HP 4806

Speed 41.03   Cutoff 60% Water evaporated per hour   99,942 Drawbar HP 5028

Since analog gauges were the only thing available in 1946, I am "surprised" that the N&W dynamometer engineers were able to read speeds to .03 mph.

I also calculated the pounds of drawbar pull that each engineering group read during these tests that resulted in the drawbar HP quoted:

At 62.5 mph and 5070 DBHP, the drawbar pull "read" must have been 30,420 lb.

At 62.5 mph and 5050 DBHP, the drawbar pull "read" was 30,300 lb.

For the N&W J...

at 41.03 mph and 5028 DBHP, the drawbar pull "read" was 45,954 lb.

At 39.14 mph and 4806 DBHP, the drawbar pull "read" was 46,046 lb.

Since I can't read the speed on the speedometer of my car this closely, I guess you can call me a skeptic......

Ed,

Bob Hord held two degrees from VPI, a BSME and a Chemical Engineering degree. No matter what you think of him, he must have done something right. He bequeathed over $17M to VPI. I learned this when they contacted me and ask me to review his obit. I met him just once, but heard from him almost weekly. These phone calls started before I retired from GE in 2001. He called GE with some technical questions re our locomotives and they referred him to me. When I retired, he determined my home phone number and started calling me at home.

I am sure he was a thorn in the side to NS at time. But I can assure you that he wanted what was best for the RR, as he was devoted to it completely. He was just discouraged by some things during his time working for them, which I won't go into here.

While I did not know him well, except by phone, there is no valid reason why I should "discount" some of what he told me and not other things. Based on our phone conversations, he was VERY knowledgeable on a wide range of RR subjects. I believe that he was one of the most intelligent individuals I ever talked to....

WBC,

Thank you for the "metric" update. I have not been involved with metric measurements since my time at work. We are still using "the kings nose and the size of his feet" for our measurements over here.....

Your question, as I interpret it, relates to the pulling power of two different locomotives, and the weight that they reach a "balance speed" with.

My comment re the NKP Berkshire vs. an N&W J on a grade with tonnage and the fact that there would be at most a speed difference of "a few mph" is relevant to your question.

My comment was based on a performance comparison of a GE AC4400 and a GE AC6000, and in what service each would be most effective.

The horsepower rating (net input to the traction alternator under standard conditions of temperature and altitude) is 4400 and 6000 respectively.

The continuous rating point for an AC4400 is 144,000 lb. TE at 10 mph. The rating point for a 6000 is 166,000 lb. at 12 mph.

This corresponds to a net input HP rating of 3840 for the 4400 and 5312 for the 6000.

Conclusions/claculations:

-If each locomotive was assigned to a maximum tonnage train consistent with its continuous rating point...

The AC4400 would run the grade at 10 mph and the 6000 at 12 mph.

Not a big difference despite a 1600 HP advantage (in rated HP) for the AC6000.

And not very cost effective.......for mountain operation.....

However,

IF you could apply maximum tonnage to an AC4400 and run at full power 8th notch "on flat terrain", AND

the balancing speed of that train is 50 MPH.

The AC6000 would take that SAME TRAIN to 66 mph.

The two major conclusions here are:

A big increase in horsepower is not reflected in a train speed on a grade (the reason for my Berkshire vs N&W J comment). AND

If you want to go fast, you need HP!

Before anyone posts about the rising power characteristic of steam, the difference in that vs. the curve for a diesel, etc. IT DOESN"T MATTER...

pounds of drawbar pull are the same.....

Maybe from a strip chart recorder in the dynamometer car ?  Some of those 'recording table' tapes or paper look pretty wide, nothing like the more recent portable stuff.  The pens could also have an 'offset' to just show the speed range being tested.

SZ

Hudson5432,

N&W computed average speeds using miles traveled divided by time (min-sec).  The 41.03 mph figure is the result of such a calculation, not reading a speedometer to that kind of accuracy.  Minutes and seconds doesn't strike me as much of a stretch, and calculating averages doesn't either.  I also don't see where it makes any difference if we take the speeds at 39 mph and 41 mph. Remove the decimal point altogether, the results are the same.

Look at the N&W test report, it's not a secret by any means.  Test reports exist in the N&WHS Archives.  This sort of information is also posted on the NWHS website.  You can determine the existence of a test and location of any test report they have cataloged.  The archives is open three days a month.

While we're splitting hairs, what was the evaporation and firing rate when the Niagara produced the 5,070 dhp reading? I've never found it in the test report and neither had two other members of the NYCSHS several years when I asked about it.  We were working on a possible simulation approach (based on Test 107 where complete readings and data are available) to estimate what it was when Charlie Smith passed away some years ago. It was my contention that the unit evaporation and firing rate were significantly higher than those found in Test 107, and as a result, higher than those for 604 during the BP comparison tests.  If this were so, the Niagara was being pushed much harder to make 5,070 dbhp than the J was to develop 5,028 dbhp, and the figures would not comparable.

On the other hand, we know both readings for the J at 5,028 dhp/41 mph - 99,942 lbs/hr total evap, 13,068 lbs coal/hr.   An analysis of both these two figures indicate that 604 was not being pushed to make the readings.  No reason it should have been.  The test was set up to determine the relative economies of raising the boiler pressure from 275 psi to 300 psi.  The cutoff was shortened to hold 604 down to a nominal speed of 40 mph, so the results would be comparable to the same tonnage at 275 psi.

Here's another interesting quote from the test report (pg 14) "....but the 300 pound [data] points at 40 miles per hour are under maximum capacity at this pressure...."

I don't care who wins the "horsepower race" but I do care about using inequitable comparisons.  Comparing steam locomotive performance is a nightmarish task, even with detailed information available.

feltonhill,

The Pond article for 99,942 lb. water evaporated shows 121 for the firing rate in lb./hr. If you multipy the 121 by the grate area of 107.7, the lb. of coal fired per hour is 13,032, very close to the 13.068 you refer to. The Niagara summary shows that for a coal rate of 9000 lb./hr, the equivalent evaporation of the Niagara was 88,900. If you divide the 9000 by the 101 sq. ft. grate area, the coal burned per sq. ft. of grate is 89.1 lb./sq.ft of grate per hour, which is certainly not overfiring. The standard Cole ration for rating potential HP of a boiler was based on a firing rate of 100 lb./sq.ft of grate area per hour, until the Cole ratios were obsoleted by the invention of the stoker, as you know.

To compare these locomotives directly, we would have to know the DBHP of the Niagara using a coal rate of 13,068 lb./hr, and use coal of the same heat value that N&W used.

The Niagara was probably fired at a higher rate to achieve the drawbar HP quoted, but the Niagara boiler was specifically designed for "overcapacity" operation. NYC did not require peak drawbar HP of this magnitude except for acceleration. Passenger train lengths were generally limited to 18 cars due to station platform lengths. From the test data for the Niagara, I suspect that the "design point" was for 4000 DBHP, and the overdesign approach was used because the availability and utilization of contemporary diesels was better than for steam. When the Niagara was designed, there was a school of thought in the industry that passenger train speeds would be increased to a max of 100 mph after the war ended.

I believe it is interesting to compare the boiler design for each engine. The running gear for the two designs are very similar except for cylinder and driver size. each engine was all roller bearing, used Baker valve gear, etc. There are slight differences in lap and lead, and they both used multiple bearing crossheads for low reciprocating weight, although N&W changed to alligator two bar on some J class engines. (Pond preferred the multiple bearing design, per his reply to a question in the RME article.)

There was one significant difference in the design of the running gear of the two engines. A Niagara, using two lateral motion devices, could traverse a curve having a radius of 310 feet at low speeds. the figure for a N&W J, with 70" drivers, was 477 feet. (The ATSF 3765 class was 573 feet!) I suspect that one reason N&W chose 70" for driving wheel size was for curving ability, so a higher driver may have been a non starter.

5432 -

Here, in the final analysis, is where you can't win the Niagara vs. J argument.  The Niagara couldn't come to the N&W and perform with N&W's trains the way the J did.  Too many grades and curves and too many heavy trains to start and accelerate on those grades and curves under all weather conditions.

The J certainly could have come onto the NYC and done everything the Niagara did.  Don't try to hide behind the "clearances" problem; if they had not been present there was nothing the Niagara could do the J couldn't, and without overfiring.  NYC didn't demand sustained 100 MPH running, and the J's operated at that speed daily on N&W.  If you read the N&W Society's magazine writeups on the tests of J 610 on the PRR (the J was too big for lots of places on Pennsy, the so-called standard railroad of the world), you would have your eyes opened up.  Since it's obvious that you haven't, then you need further education.  If you're obsessed with boiler plate thicknesses, etc., you're really reaching for arguments that still won't hold up. Your curvature argument holds no water, either, since the J coped with N&W's curvature, much worse than NYC's, very nicely, and without excessive tire wear.

As far as Mr. Hord is concerned, if you canvas all the retired N&W folks still around who knew him, you'll find most, if not all, shared my views.  What he did that was right was to choose a wealthy parent, from whom sprang his VPI legacy.  I'm glad you knew and liked him.  Everybody needs at least one fan, and you're his.

EdKing

N&W tests 1948

Norfolk Division [Poe to Suffolk]

For the test runs with both N&W J # 604 and PRR T # 5511, the train was made up of 20 passenger coaches and the dynamometer car, and weighed approximately 1506 tons. N&W assigned this tonnage because it wouldenable both locomotives to develop maximum outputaccelerating to the operating speeds in each series of tests. In the highest speed group, both locomotives would be operating at near maximum capacity with a balancing speed of about 93 mph for the J and 97 mph for the T1

The test runs in this district were divided into three speed groups 65, 75, 85 mph. For the first group of the two round trips the train was accelerated from start as rapidly as possible to a nominal maximum speed of 65 mph. For the second group the same procedure was followed, accelerating to a nominal maximum of 75 mph and attempting to maintain this speed. For the third group the nominal maximum speed was 85 mph.  An attempt was made to maintain this speed after it was reach, but this was not possible over a considerable portion of the district. A speed restriction at Waverly (about 45 mph) reduced the average speed and required a second acceration run for each trip.

The J showed performance superiority in the 65 mph and most of the 75 mph tests. The T1 showed superiority in some of the 75 mph and in all of the 85 mph tests. This was due to the T1's higher drawbar horsepower above 65 mph.

Author: David Stepheson

Keystone Vol 42, No2

Thanks for the comment. The operative words is the "same train". A higher horsepower locomotive will pull the same train faster than a lower horsepower locomotive. Makes perfect sense.

We have some definite numbers posted on the thread.  For instance, 6000 HP  for the Big Boy 4-8-8-4 and 6000 HP for the AC6000CW. Using the Watt, both of these locomotives can theoretically hold a constant velocity one meter per second against a force of 4476000 Newtons.

Looking at GE spec sheet, the starting tractive effort of the AC6000 is 181,000 lbs which I can convert to 805 kN, which is the force the locomotive can exert to pull a train. 4476000 divided by 805000 is 5.56. Thus, the AC6000CW can hold a train at a constant velocity of 5.56 meters per second against a force of 805000 Newtons (181000 lbs). Converting m/s to mi/hr gives 12.4 miles per hour, calculated from the Watt. Pretty close to Hudson5432's numbers.

Looking at the Big Boy stats, it gives a starting tractive effort of 135,500 lbs or 602 kN, which is the force of the locomotive can deliver.  4476000 divided by 602000 is 7.43. Thus, the Big Boy can hold a train at a constant velocity of 7.43 meters per second against a force of 602000 Newtons (135500 lbs).  Converting m/s to mi/hr gives 16.6 miles per hour. I could not find a velocity for Big Boy pulling against its maximum force. Is this close?

Which is more powerful. The answer is simple; neither. They are the same power. Which is doing more work? Neither, they are doing the same work.

If you want to pull a lighter load more quickly, then Big Boy is your machine. If you want to pull a heavier load more slowly, then the AC6000CW is your machine.  Ultimately they are delivering the same power and doing the same work.

To go to the original question of 4-6-4 vs. 2-8-4. The NKP 174 has tractive effort of roughly 40,000 lbs, the NKP 765 has tractive effort of roughly 60,000 lbs. You can figure it out from there using the horsepower and thus Wattage of the two. 

Perhaps not the right place, but I wonder if the experts might comment on a proposed Class N Northern that Norfolk and Western considered at the late date of 1955.  Recently saw a reference, but could not find any real information.

5432

I apologize to you for coming down so hard on Bob Hord, but it's obvious that you and I had differing experiences with him.

EdKing

Ok, "Honest Broker" time: earlier in this discussion I rather dogmatically stated that the DBHP of the N&W Class A was 6300. Upon doing a bit of digging, 2 sources corroborate Ed King's statement that the class A DBHP is 5300.  P.29 of Gene Huddleston's "The World's Greatest Mallets, the C&O H8 vs. N&W Class A" sites drawbar horsepower max of 5300 (estimated) for the 2-6-6-4. Additionally, the chart contained in Bob LeMassena's "N&W's Secret Weapons", Nov. 1991 "Trains", p.66 states the A's DBHP/MPH as 5300/35-45. All well and good.

      But Mr. Huddleston's piece lists the data from the 1952 Class A #1239 vs. EMD's "juiced up" 1700 horsepower F7's test. The 1239 handled 175 cars, 16,028 tons, 112 miles, in 3 hours/31 minutes and produced 124,000 pounds drawbar pull. The hotrod F7's pulled 176 cars, 15,763 tons and displayed 247,000 pounds of drawbar pull (a rated figure). But wait a minute! How does a locomotive (#1239) with half the drawbar pull produce equal performance results? And with 1700 less horsepower? The average running speed for the diesel was 31.4 mph; the 1239 was 31.6. Could the electric motor torque loss (as noted previously by Ted Hikel) have been so severe so as to render the diesel only competitive with a machine with half it's drawbar pull? Or could the 1239 be producing DBHP more in line with the diesel locomotive (more like 6300 DBHP)? From my reading, it appears that EMD's horsepower ratings are pretty accurate and EMD certainly went to these tests with every intention of knocking off the last steam stronghold in the country.

       So what gives? Can anyone illuminate this seeming mystery? Bob LeMassena theorized that the N&W raised the boiler pressure of the 1239 to 315 psi and added lead to the frame. That thesis has been discredited, as Ed King noted.

       Additionally, the LeMassena article addresses the Y6 and the "interceptor valve" and states that "although this raised steam pressure slightly for the low pressure cylinders, it greatly increased the temperature of the steam. Hotter steam produced a larger tractive effort from the low pressure cylinders.....". And perhaps greater horsepower, leading to Ed's 5500 DBHP figure for the Y6.

        Finally, if the Class A 5300 DBHP figure is correct, why is it correct? It has a larger boiler with seemingly greater steam production capacity then either the J 4-8-4 or the Y6 2-8-8-2. Could it be that at higher speed the A produced the 6300 DBHP? Could the 5300 DBHP figure have been developed in early testing when the Class A's had a 275 psi? I confess, I am mystified. 

Gilbert - The Class N 4-8-4 was not considered in 1955.  It exists as an undated diagram from about 1926 as one of the proposals that led to the class K-3 4-8-2.  The diagram is in the N&WHS Archives and was reproduced in my book on the Class A.

The big difference between the "N" and the K-3 as actually built was in the firebox, which was to have 100.1 Sq. Ft. of grate area, thus necessitating the use of a four-wheel trailer.  The other mechanical details - cylinder size, 63" drivers, etc., were to have been the same as the K-3.' The firebox was cut down to 84 Sq. Ft. on the K-3 which allowed the use of the two-wheeled trailer.  It had no bad effect on the K-3 steaming capacity; it was known that the locomotive was a superb steamer, probably the only good thing that could be said about it.

If built in 1926 the N would have been the world's first 4-8-4 (Norfolk type instead of Northern?), and with those small drivers, certainly the strangest.  But it would have the same counterbalance problems that dogged the K-3 throughout its lifetime.

EdKing 

There's just NO WAY I'm gonna jump on this N&W A= 5200DBHP business.

Especially the 1238-42 series...with thirteen years of tuning and tinkering

with the design.  Too much good machinery, and too much talent at Roanoke

for this figure to be realistic.  Maybe if you had a load of bad coal (highly

unlikely on N&W) or bad water, or some moron for an engineer...but not

otherwise.  These machines had to be at least as potent as the big AT&SF

5011 2-10-4s, and these beasties were good for 5660 DBHP.  This whole

5200HP thing sounds like a Pennsy plot to try to embarass someone  who

might not be into kissing the ring or sucking the toe...Philly style!

  AS for the 4-8-4s,  IF you pushed all of them to their absolute upper

limits, on the same scale...there can be little doubt which machine would

be the boss...red stripe and all !   The only question remains, who is second,

third and forth in the great 4-8-4 shoot-out??   Anyway, all of this just further

 proves  that ...Nobody ever boiled water like N&W...Nobody!

BTW, I got no dog in this fight, being an absolute, unrepentant Dieselover !

Very much appreciate the reply.

jaygee

The one thing that's particularly frustrating about N&W's rating of its locomotives is that they simply didn't care about traditional maximum figures.  They used very conservative evaporation and firing rates in their calculations (the calculation books exist so this isn't a guess) and they didn't flog their locomotives to get high figures during a test.  They wanted to know what they would do any day, every day, with any crew.  As a result, you really can't compare N&W's ratings with anybody else's and get anywhere.  I've been at this game for over 50 years, and I can guarantee you that all locomotive comparisons are just about impossible.  Even knowing all the details of the tests, it's still hard to do an a reasonable basis.

For example, the A's 6,300 test rating, however clouded by dynamometer car problems, may have been possible, even at 275 psi.  I've given simulation a try, and the figure is feasible using 116,000 lbs total evaporation (water from tender plus condensate from the FWH).  It would have been pushed significantly above its economical operation zone to do it though.  Any other railroad would have probably rated the A at something like 5,800 DBHP at 45 mph, still not using anywhere close to a high evaporation rate (more than 100 lbs of water per SF of direct heating surface). 

However, N&W was more interested in the relatively flat DBHP curve of the A.  It may have peaked in the 40-45 mph range, but the curve was essentially flat up to 60-65 mph.  This is exactly where the A was used - coal service at the lower end and time freights at the upper end.

If you've never heard or seen an A on a merchandise freight at speeds hovering right around 60 mph mile after mile, you just haven't lived!!  OK, it was just a bit before my time, but we still have Winston Link's recordings to give us an idea what it was like.

Now good old Uncle Ed was actually there first hand back in 1957-58. Talk about lucky!  It's no wonder the A's seduced an entire railroad (his words, not mine).

Lets go back to that Le Messina article in TRAINS back in Dec. '91.  While some

of the conclusions may be in error, one thing is not... and that would be his

analysis of the ride Link took on N&W A # 1238 (IIRC).  While the hill climbing

portion of the ride was exciting, it was the level portion that raises the eyebrows.

Was his analysis accurate?  I'd imagine it was pretty close, if not exact.  There's

another interesting point here.  While N&W may not have gone for the big numbers

in their steam evaluaton tests, and instead concentrated on the bottom line....

the locomotives, and especially the A, J and Y6 classes were some of the most

outrageous machinery ever built; designed to operate at the screaming edge

of steam power operability.

jaygee,

I don't know exactly what "level" portion you're referring to in Link's recordings (Mainline to Panther?), but LeMassena overstated the grade by a factor of 10, 0.5% instead of the actual 0.05%.  I'm referring to the average grade westbound, and just west of Bedford, if that's the part of the article you're referring to.  This overstatement of the grade led to a similar overstatement of the A's horsepower.  I'm a very definite critic of the entire LeMassena article in Trains, so you may label me as biased if you want.  His research for the article must have been minimal, because it has that many errors in it. Here again, this is no secret.  There have been two rebuttals published in the N&W Historical  Society's magazine, The Arrow, rebutting it in considerable detail (May/June 1994 and Jan/Feb 1998). 

Link took several rides behind a Class A in Dec 1958 and Jan 1959.  The 1958 recordings survive, although the others are lost (%@#$*) These recordings are not secret either, and can be heard at a kiosk at the O. Winston Link Museum in Roanoke.

But Mr. Huddleston's piece lists the data from the 1952 Class A #1239 vs. EMD's "juiced up" 1700 horsepower F7's test. The 1239 handled 175 cars, 16,028 tons, 112 miles, in 3 hours/31 minutes and produced 124,000 pounds drawbar pull. The hotrod F7's pulled 176 cars, 15,763 tons and displayed 247,000 pounds of drawbar pull (a rated figure). But wait a minute! How does a locomotive (#1239) with half the drawbar pull produce equal performance results? And with 1700 less horsepower? The average running speed for the diesel was 31.4 mph; the 1239 was 31.6. Could the electric motor torque loss (as noted previously by Ted Hikel) have been so severe so as to render the diesel only competitive with a machine with half it's drawbar pull? Or could the 1239 be producing DBHP more in line with the diesel locomotive (more like 6300 DBHP)? From my reading, it appears that EMD's horsepower ratings are pretty accurate and EMD certainly went to these tests with every intention of knocking off the last steam stronghold in the country.

Mark

I can only add a couple of points for consideration to the expert discussion we have all been enjoying here.  Ed King has already pointed to some of them and I'll try and add to that.

First, the discussion has focused on the measurement of drawbar horse power of steam locomotives.  To compare the actual ability to pull a train this is the measurement that matters.  Other steam locomotive horsepower measurements, boiler horsepower or indicated horse power may be important for designers but from an operations perspective the draw bar is where the train meats the locomotive and that is what matters.

Second, there is more than one place where the horsepower of a diesel locomotive can be measured.  In Europe it is common to refer to the output of the prime mover.  American Diesel locomotive builders have long used the more conservative measure of "horse power available for traction".  This is the output of the generator/alternator.

But traction motors are not 100% efficient so the traction motor output will always be less than the generator output.  If energy wasn't lost to heat in the motor diesel-electrics wouldn't need traction motor blowers.  And the locomotive must move itself.  In the case of an A-B-B-A set of F7s that is almost a million pounds of locomotive to move and that takes some horse power too.  So a 6000 horse power set of diesel-electrics (even with the injectors set to put out about 6800 HP) may in fact produce +/- 5300 horse power measured at the drawbar.

A diesel-electric can produce a tremendous amount of starting tractive effort.  But early diesel-electrics couldn't do it very long without causing damage to the traction motors with the very high amperage load.  I imagine the reason EMD was anxious to test the F7 on the N&W was due to its much more robust traction motors when compared to the early F3s.  That made the F7 a much better locomotive for hauling heavy trains at low to moderate speeds.

The F7 set may have a starting tractive effort of over 240,000 pounds but it can't sustain that figure very long without overheating the traction motors.  I'm sure Ed King or Hot Water can confirm this but the maximum tractive effort at minimum continuous speed for a four unit F7 set should be about 160,000 pounds at 10 mph.  That is more than an articulated steam locomotive but not vastly more like the 240,000 pound figure would indicate.  Both the steam locomotive and the diesel will show a fall off in tractive effort as speed increases above 10 MPH but the fall off by the steam locomotive will be at a little slower rate.

I posted a link with a horsepower and tractive effort curve for the J back on the first page of this discussion.  If anyone has charts for the A, Y6 and F7s it would be very helpful if you could post them.  A graphic depiction of comparative tractive effort and DBHP might be worth more than three pages of words.

To amplify feltonhill's evaluation of N&W's testing procedures . . .

After the diesel-influenced Kenova District tests of 1952, Bob Pilcher (N&W's Test Engineer) felt that an A could achieve 500,000 GTM/TH on coal trains there.  He made three runs after the initial diesel tests with the 1239 and did, in fact, achieve that figure.  The As were then given comparable tonnages for every trip and they all turned in equivalent performance; my educated guess was the some trips would exceed the half-million, some would be a little short, but probably the 500,000 figure was a good average.  And this is for every engine that happened to work in that pool, from the 1200 to the 1242 - all of them.  Y's were occasionally used on those trains in a pinch, and the crews didn't like them nearly as well.  Seeing themselves marked up for a 2100 on a coal train at Williamson, they'd say "we won't get home so fast, today; we got one o' them ol' "thousand leggers".  We know that the Y wouldn't burn as much coal and use as much water, but the availability and consequent utilization of the As in that service made it worthwhile.  Over a month's time, the quickness of the trip and rapid turnaround meant more trips per engine.

Similarly, Dave Stephenson has worked up the data and determined that a Y-6 would bring a 10,000-ton train up New River at 30-32 MPH, which required a reading behind the auxiliary tank of an average of 5,400+ DBHP.  If you can find any living crew member to ask today, you'll be told that that was the speed that was normal.  Again, that was achieved by every engine that happened to work in that pool, from Y-5 2101 to Y-6b 2200.

This policy is in direct contrast to other railroads, who'd really tune an engine up before subjecting it to a test - especially the PRR.  Correspondence unearthed by Stephenson indicates that PRR's officialdom actually expected Test Plant performance to be duplicated, or nearly so, in the field.  It's true that the 1239 and 2195 were freshly shopped before the diesel tests, but the fact that subsequent performance by all the engines in the pool matched them negates the importance of the shopping.

EdKing

quote:

but they did not have the power to run at speed with the same load that the J could.

Rich,

Tell that to the Bristol and Shenandoah Div. engineers when you get to heaven. As the late C.I. Robbins related to me, "63mph is as about as fast as you wanted to run a Y6. After that, it got a little shakey!"

Granted, a Y6 would never be able to run as fast as a J, but, with only 1,700 tons (the same load that a J could), I don't think a Y6 would have a problem keeping up...to a certain point that is.

Too bad you haven't heard Bud Swearer's N&W recordings. I think they would change your mind. I wish that I had the money at the time to purchase a set of all of his different RR recordings. What I do have, is quite enlightening.

I also have a recording of a Niagara from "Living With Steam" as she accelerates from Tower 47 in the Buffalo area. A great recording as the author, John Prophat III, paces train #78 and her 4-8-4 up to a "breathtaking clip". Only thing is, it took forever to do get there. The real bummer about this recording is that Mr. Prophat is eventually stopped by a grade crossing getting in the way of his going any further. As the Niagara approaches the crossing, the poor Niagara blows its horn!  

Compare this with O.W. Link's recording on "The Fading Giant - Sounds of Steam railroading Vol. 2" (version II or Remix) of a J "blowing in the flag" and then accelerating train #15, "The Cavalier", away from Ripplemead .

If the trains could be placed side by side, the folks on the rear of #15 would be waving Bye-bye to #78.

Ted,

Here's a copy of Chart 1 from the 1945 tests comparing the fuel and water usages for 275 and 300 psi settings.  Please note, that these curves are theoretical, calculated using a method developed by Baldwin.  This procedure was widely  used in the industry.  It is also extremely conservative for modern steam locos.  The only actual performance data on the chart is designated.

You will also note that the 300 psi curves conflict with those in the ASME booklet.  The 300 psi curve in the ASME book was calculated using an undisclosed method and is incorrect.  Based on my research the curve closely matches the Baldwin method estimate of Indicated HP for a 275 psi J.  Further, the DBP aand DBHP cuves for 300 psi are inconsistent above 90 mph. These two quantities are linked by a formula mentioned earlier in this thread.  They do not vary independently of each other.  That's another problem with the ASME curves.

How did this happen?? I  have no idea, but the attached chart is what the N&W said, and the method they used is clearly indicated. 

Ain't numbers fun??

Great data and analysis! Let me ask my last question one last time, then I will slink away into the night and never bother anyone again! Where/when was the Class A's 5300 DBHP measure obtained? Could it be a 275 psi measure? I note in Feltohill's horsepower curve graph that the 275 psi J peaked out at about 4800 DBHP (forgive the crude read of the chart - was employing a ruler on my computer screen!) and the 300 psi DBHP read appears to be about 5300, which is about a 10.4% increment. How is it that a locomotive - the A - with a much bigger firebox and evaporative surface + the same boiler pressure, did not produce more DBHP then two locomotives with smaller boilers? All the boilers were designed by the same bright fellows in Roanoke.

Mark,

The only way I've ever been able to estimate how N&W came up with such a low figure is use a unit evaporation rate of about 71 lbs water/SF of direct heating surface for a starting point, a +10% FWH allowance and a -10% deduction for auxiliary steam usage.  This would limit the boiler to about 100,000 lbs of steam per hour, and produce about 5,300 DBHP.  This is a super-conservative way of estimating locomotive performance.  You would not get in any trouble using these unit evap and percentage figures.

It can get a bit more interesting.  For example, +12% and -7% were more typical allowances for FWH and auxiliaries.  Making only this change, the A would develop 102,000 lbs of steam/hr and 5,500 DBHP.

The DBHP figure I mentioned earlier in this thread used a commonly accepted average of 80 lbs/SF DHS/hr, +12% and -7% allowances,  for 108,000 lbs steam/hr and 5,800 DBHP.

There are limits as to how far you can go with the unit evap rate, but 100 lbs/sf DHS/hr is not out of the question based on test data and other railroad industry information.  As you can see, the figure I used was way below that.

There are my estimates, not N&W's.  I used generally accepted methods of the industry for modern steam power.  The resistance equations were adjusted to reflect cast frames and roller bearing applications.

The key to a locomotive's output is the lbs of steam delivered to the cylinders per hour.  From there, it's up to the machinery to extract the work from the steam.  I believe this is the summary you were looking for.

It was N&W's railroad and they could rate their power anyway they wanted.  I have no other answer than that.

Waay too bad to find out that the Trains article from Dec. '91 is such a FARB!

Won't be the first time we've been hoodwinked, but I really had no idea

about the grade fiasco.   OTOH, the boiler output work-ups would seem to

indicate a vastly under rated machine, and makes you wonder what might

have resulted if the "A" had really been flogged, NYCS or PRR style! With

regard to the 1952 tests in Ohio, wouldn't EMD have had to rework the

electricals a touch to be able to fully use that 1700HP per unit as opposed

to the factory 1500HP set-up?  FWIW, this has to be one of the best threads

I've seen.....EVER !   ...but I'm still looking for my beloved 65:12 SD7...

somewhere !

No, not at all. The only issue might have been the cooling systems ability to reject that additional heat from the 16-567 at the increased horsepower. But for short duration, even THAT would not have been a problem. 

No taker in this quarter! The N&W, hands down, was the champion of steam locomotive utilization and productivity. The only other road that made extensive/almost universal useage of auxiliary water cars, was the Illinois Central. The IC used steam locomotives as late as the N&W (1960), had an enviable red ball freight record, and I believe a strong record of profitability during the steam era. Other roads used auxiliary water cars, but primarily in limited water availability or bad water districts (CNR, CPR, L&N, SP, GN, CB&Q, D&RGW, NYC, NP, PRR, DM&IR, MP, to site some examples).

Feltonhill

Thank you for posting the chart.  Although the data points are in a limited range and only from a 275 PSI test it looks like the calculations for the rest of the curves are fairly conservative.  That seems to be in keeping with what everyone here has said about the N&Ws use of test data, to determine what could be expected every day.

Ed,

I accept your apology.

WBC,

Your are on the right track, but your comparison of the UP Big Boy vs an AC6000 requires a further explanation and some clarifications...

-The Max drawbar HP of a Big Boy from the Kratville book was 6290 at 41.1 mph. The drawbar HP of a Big Boy was not a constant as it is for diesels (see below). The Kratville book contains too few data (i.e. speed) points to make a reasonable Speed Tractive effort (STE) curve.

-The STE curve of any diesel on its "constant HP portion" resembles the lower left quadrant of a parabola. The tractive effort for a diesel is plotted "at wheel rims". (Note you need a chassis dynamometer (ie rollers) to confirm these numbers. They are relatively uncommon (PRR had one at Altoona) and DOT at Pueblo, CO has one.) So the tractive effort on a STE curve INCLUDES system efficiency at each speed point.

-The builders generate this curve and info under "standard AAR conditions", which is 1000 ft above sea level and usually 60 degr. F. To make a true simulation, you may also have to make corrections for site conditions.

-The system efficiency of older 4 axle diesels was usually about 85%. (EMD used a 320 divisor vs the correct 375 divisor to calculate "rail HP", which indicates an efficiency of 85.3%). The system efficiency of older 6 axle diesels was about 82% due principally to the use of two more DC traction motors compared with a 4 axle of the same HP. Current locomotives are higher. 

-The actual shape of the curve is not a true parabola which would exist if the product of multiplying the speed in MPH by the tractive effort in lb. were a constant. The full HP shape of that curve is skewed because at low speeds, a diesel locomotive is providing tractive effort at high amperage, and at high speed it is supplying TE at high voltage. (Remember volts x amps = watts (or Kilowatts at these power levels.) Power delivered at high amperage is less efficient than power delivered at high voltage.

-The area UNDER the STE curve represents the amount of HP that the locomotive can develop over its entire speed range.

-You have to deduct FROM the lb. of tractive effort available from the locomotive at each speed point the amount of energy to move the locomotive (or locomotives) to obtain drawbar pull available at the rear coupler of the locomotive. The energy required to move the locomotive will vary with speed, frontal area, flange resistance, etc. The AAR has developed formulas for this loco resistance, and also for car resistance. Grade resistance is 20#/ton/1% grade, which is the major reason why it requires a LOT more HP to run even a slight grade vs. running on level track.

-For steam locomotives with STE curves not developed from test car data, you have to subtract wind resistance and also tender resistance from the "roller tractive effort" to obtain drawbar pull (and drawbar HP) at the coupler at the rear of the tender. (All PRR locos tested on the Altoona Test Plant require this correction.)

-A STE for a steam locomotive was considerably less precise than one for diesels for several reasons in addition to instrumentation. As a result, the various measurement points were averaged with a "least squares" mathematical method to yoield a smooth curve. In addition, curves were "corrected" for locomotive resistance, tangent track, etc. Results were not always reproducible due to changes in throttle position, cutoff, minor speed changes, fuel quality, etc. (This was one reason why NYC used a "brake engine" between the test engine and the train so that speed points in particular could be held constant so multiple readings could be taken.)

I  have worked with locomotive performance for about 50 years with a career of 37 years, and am one of six people  who performed this work for GE since 1964. (Another of the six is my older son.) I am very familiar with and have participated in testing of all kinds, including generation of curves and other performance data, and am a student of steam locomotive performance and technology. For example, there are at least five different methods that were historically used to calculate steam loco HP. I grew up at the end of the steam age, have ridden steam, and have photographed NKP, N&W and the Canadian roads. My library, in addition to information not generally available to others, includes performance data for ATSF, PRR, NYC, N&W, and UP, and of course testing of recent locomotives. I also have isolated files for B&O and C&O.

So I do have some knowlege of and also experience with this subject.

Whew! After all of that, no wonder GE locomotives sound like a box of rocks...and run like that too!

GE's last great engine was UP #30 !  Been all downhill ever since.  OTOH, if

GE had obtained the Jawn Henry contract from N&W, the results would have

almost certainly been different, and far more positive!  

Whew! After all of that, no wonder GE locomotives sound like a box of rocks...and run like that too!

 

GE's last great engine was UP #30 !

I hate to throw Cold Water on you guys but have you noticed that GE has been number one in locomotive sales?

For nearly 30 years!

Maybe it is because they are better at math than the competition.  Railroad accounting departments seem to think so.

Maybe we should have titled this thread, which we hijacked by the way, "Back in the Day".

quote:

Railroad accounting departments seem to think so.

Right, it's all about how much money the bean counters have to shell out for each unit.

Actually, from what little I have been able to run them, the new AC units seem to be the best performing diesel loco GE has yet made. It's taken them long enough.

Hudson5432

When you referred to the STE-speed curve for BB, were you also referring to its relative, the IHP-speed curve?  I always thought that DBPull/DBHP curves were more useful in determining steam locomotive performance.  In Kratville's book there's a pretty decent set of DBHP/DBPull curves (p20), which you probably already have, and wouldn't these go a least a little way toward determining IHP and TE, with some of the adjustments you mentioned?

I gave it a try a few years ago using as much of the data in Kratville's book as possible and played the DBPull values backward through some slightly modified Kiesel equations to subtract all the resistance components.  I've had good results doing this with other locomotives where I had more complete test data to confirm the results.  The  IHP curve derived using this method looks relatively normal even though I had no actual test data to confirm it.

Hudson5432, thanks for the reply.  I take it that the Big Boy's measurement of max HP was not against a force of 135,000 lbs as it would not be possible on track perpendicular to Earth's core (flat track) or a vector against Earth's core (up hill) for the locomotive to move against a force of 135,000 lbs at a velocity of 41 mph. With the aid of the Earth's core it may be possible.

This is pretty easily explained. The drawbar pull is a force measurement not a power measurement. The F7's exert a greater force on the train than the Class A. 

An example we can all relate to is our home.  The core of Earth exerts a force on all of us. However, you can stand still regardless of that force and move independently against that force, though at greater expenditure of energy. However, there is a great deal of power released by the core. Thermal power from the core of Earth drives volcanoes and earthquakes which can certainly move objects and people at a velocity. 

The same train is being used in your comparison between the Class A and the F7. They were both pretty much the same number of cars with the same tonnage. In this comparison the steam locomotive compares well (ignoring fuel consumption and pollution). Of course, these conditions are advantageous to the steam locomotive and as such it compares well. It is doing what the machine was designed to do. 

However, the diesel is being under utilized. Those F7's are not being utilized to their fullest potential. The F7's can exert twice the force upon a train with the same horsepower. Thus, double the train. Increase the number of cars to 350 and tonnage to 32,000. The F7s can still exert the force and apply the power to move the train at velocity (10-15 mph).  The Class A cannot (assuming that it took all 124,000 lbf to move the 175 car/16,000 ton train) as it would slip. Now the diesels would succeed where the steam locomotive would fail. The situation is now advantageous to the diesel. This may not be physically possible to constraints such as coupler strength.

I have a book called Portrait of the Rails. In it there is a passage that talks about the steam locomotives being dray horses and the diesels being race horses (hot rod). However, the example shows quite well that the reverse is true.  It is the diesel that is the dray horse and the steamer the race horse. Two different machines designed differently to do tasks differently.  However, the same work is being accomplished per horsepower regardless.