[N&W] Re: What if?...

[nw-mailing-list at nwhs.org]

A dissertation such as this deserves a better future than being simply
deleted in the due course of mail management.  This is interesting stuff that
should be saved for those future members who are familiar with steam only
through books, videos and static displays of locomotives and cars.

It won't be too many years before the people who really know this stuff will
no longer be around.  A wonderful resource will be gone.  I hope that we are
making every possible effort to tap into that knowledge and then retain the
information in a permanent form for the future.

C. H. Chamberlayne
____________________________________________________
A good description of Franklin and the poppet valves can be found in Vernon 
K. Smith's autobiography "Locomotives in My Life" ( I hope I got that 
right!) now out of print.  Smith worked for Franklin during the 
introduction of the product.  Note that the C&O put the poppet valve gear 
on a some of their 4-6-4's in the much simplified worm drive version.

The original Franklin design as used by PRR used a complex gearbox and cam 
system mounted to the front of the engine.  The system wasn't very 
accessible for maintenance and required special maintenance training to set 
the system up correctly.  This design had a lot of parts and was costly to 
manufacture.  The later worm drive design required ( by eye-ball estimate) 
about 75% less parts and was accessible as it was mounted off of the 
drivers as previous valve gears were.  But, it was "too little- too late" 
to make any difference in the impending obsolescence of steam.

Gary Rolih
____________________________________________________
Nathan,

      Let me add my two cents to what Ed King wrote 11/13/00.  My comments
will be of a more general nature, indicating what poppet valves could do if
the downside issues could have been successfully addressed.  As he points
out, it may have been difficult to realize any benefits, given the
characteristics of individual engines and their service applications.

      One of the best sources of general information on poppet valve
locomotives and what improvements could be expected can be found in Vernon
Smith's book One Man's Locomotives, Trans Anglo Books, 1987, pgs 61-85 and
pgs 100-104.  It contains his personal experiences and opinions regarding
the application of poppet valves to many different locomotives.

      Keep in mind that the most spectacular gains in performance and
drawbar horsepower (sometimes exceeding 25%) were generally on older
locomotives that were rebuilt to varying degrees and equipped with poppet
valves.  The PRR K4 Ed mentioned (#5399 in particular) is a good example of
this.  The huge increases could not be reasonably expected with more modern
power.  The only "apples to apples" comparison that I know of which
documents the effects of poppet valves on a modern locomotive is found in
the NYC Niagara tests.  The major difference between no. 6023, a
conventional Baker valve gear equipped 4-8-4 and no. 5500, a poppet valve
version on the same locomotive, was a 15% savings in steam usage at similar
power output.  There were extenuating circumstances to the tests, however.
There is the possibility that the valves were not properly sized, and could
not handle the full boiler capacity of the Niagaras. Consequently, the only
conclusive part of the test was the savings in steam usage at the same DHP
output, because 5500 was unable to reach significantly higher DHP than the
conventional 6023.  The 15% figure is also similar to that mentioned by Mr.
Smith in his book, where refers to the improved performance of ATSF 4-8-4
no. 3752 after rebuilding with poppet valves in 1948.

      Regarding the N&W A, any attempt to obtain additional DHP output would
probably have been misplaced, due to the reasons Ed gave.  Further, any
increase in DHP output would mean increased stress on the machinery,
causing increased maintenance costs.  From everything I've read, the A was
a high-stress design to begin with.  N&W did not "throw iron at the
problem" in order to beef up the engine.  If you want an example of a lean,
mean machine, the A is it.  And its full capacity was used, frequently for
hours on end. Because of this high output operation, conditions were
favorable for savings theoretically possible through the use of poppet
valves, stress on the word theoretical.  Maybe N&W could have used them to
reduce the fuel and water requirements at the same DHP, thereby increasing
boiler efficiency through reduced firing and evaporation rates.  Maybe they
could have used slightly smaller cylinders to eliminate some of the
potential for adhesion problems.  But there would have been the first cost
of the modifications to contend with, additional mechanical complexity in
the valves, potential drafting problems.....  Who can say.  Would it have
been worthwhile economically?  Probably not.

      Don't get me wrong.  Poppet valves have definite advantages if
properly applied.  Elsewhere in the world, poppets were used successfully,
but always on engines smaller than ours and in service conditions far
different from ours.  In this country, poppets were probably too much
gadgetry, too late in the game.  There was no time left to work out the
bugs.  PRR succeeded in getting many of them out of its T1s through late
modifications.  But IMHO, N&W took the right approach by sticking with
mechanical simplicity and sophisticated operation to move its traffic.

      I also want to give some perspective to the T1 tests on the N&W,
particularly on Christiansburg Mtn.  The two engines involved were N&W 604
and PRR 5511.  Several tests were run, with train weights of 1,003, 1,220,
1,440 and 1,758 tons.  For ease of comparison, I'll use starting tractive
effort (TE) for each locomotive as a basis of comparison, because at low
speeds (0 to 15 mph or so), TE doesn't fall off too much.  During the
tests, 604 handled a maximum load of 1,758 tons.  Since a J had a TE of
80,000 lbs, it was moving about 21.9 trailing tons per 1,000 lbs. of
tractive effort.  In the case of the T1, the tests were run after #5511 was
modified with 18¾" cylinders, so its TE would have been about 58,000 lbs.,
considerably less than 604.  With a trailing weight of 1,220 tons, the 5511
pulled 21.0 tons per 1,000 lbs. TE, just about the same as 604 handled
successfully.  So, the two locomotives performed on an equivalent basis, if
you consider trailing tons per lb. TE. At 1,440 tons, a load that stalled
the Tl, 5511 was trying to manage 24.8 tons per 1,000 lbs. TE, a
significantly higher proportional load than 604 handled during the tests.
If 604 were similarly loaded, it would have had to pull 1,984 tons up
C-burg Mtn.  Possible?

      With a little imagination (and a selection of videotapes), we can
experience part of these tests today, at least as far as 604 is concerned.
During the 1987 NRHS convention in Roanoke, #611 pulled 23 cars up
Christiansburg grade at about 15 to 20 mph on dry rail with sand, and
appeared to be working near full capacity at about the limit of adhesion.
If the average weight per car were the same as the 1948 tests, #611 was
dragging 1,771 tons up the grade, approximately duplicating sister 604's
performance with 1,758 tons.  BTW, these are my estimates.  Is there anyone
out there who confirm weight/speed/capacity information from 1987?

Dave Stephenson
____________________________________________________
Thanks to Ed King for a thorough dissertation on poppet valves, A's, J's, etc.

Here's some added history re Pennsy's troublesome T-1.  Thought some of the
younger members might be interested.

PRR steam locomotive development went dormant after the period 1910 - 1925,
when the family designs ( classes H, K, L, M, etc.) were designed and
produced.
 From about 1920 to 1935, PRR focused on electrification of the New York -
Washington - Harrisburg triangle.  In doing so, they failed to keep up with
the significant improvement in steam design known as "Super Power".  When PRR
resumed interest in steam, they depended too heavily on their own antiquated
technology and the unproven duplex drive concept advanced by Baldwin.

The presumed advantages of duplex drive were reduced piston thrust, lighter
rods, lessened need for counterbalancing, lessened wear on bearings, etc.
But even in the early T-1 design stage, at least one consultant identified
the disadvantages:
Rigid wheelbase longer ( 3 ft. ) than a conventional 4-8-4; increased valve
maintenance; and potential for drivers slipping at speed.  One suggestion,
rejected by PRR, was for driver sets having different diameters, so that the
two sets would not
rotate in sync.

The published figure for F.A. for the T-1 is 4.3; normally an acceptable
figure for a Northern.  But with divided drive and poppet valves; and no time
to study performance
in service, the T-1 never lived up to the promises.

Jerry Crosson
_______________________________________________________
[Ed King wrote:]

>Nathan -
>
>Poppet valves were, IMHO, developed too late in the day to benefit the steam
>locomotive; the poppet's main feature was a reduction in exhaust back
>pressure by separation of intake and exhaust valve events, unlike the piston
>valve which reduced exhaust port openings while the reverse lever was being
>hooked up, to the same extent that it was reducing intake port openings.
>Poppet exhaust valves operated independently of the intake valves, and could
>be open longer to accomplish the reduction in back pressure.  Franklin
>equipped a Pennsy K4 Pacific with a set and the performance improvement was
>amazing, but the K4 was a 'teens design with much room for amazing
>improvement from a number of sources.  One of the K4's features was a factor
>of adhesion of more than 4.0, meaning that its rated tractive effort was
>less than 25% of its weight on drivers.  This was a great help, as it helped
>keep the engine from being slippery due to the reduction in exhaust back
>pressure.  Heartened by this, Pennsy equipped its T1 4-4-4-4 with poppet
>valves, and wound up with a locomotive that was known to be extremely
>slippery - it had a lower factor of adhesion than the K4.  The T1 had many
>novel features, none of which had been tested in a fleet environment - the
>duplex drive, the poppet valves, etc.  It is almost miraculous that the T1s
>did as well as they did.
>
>The J had a factor of adhesion of much less than 4.0, and relied on a number
>of factors to make it successful - well designed and maintained running
>gear, well designed and maintained valve gear, well designed and maintained
>track structure (proper contact between wheel tire and rail is vitally
>important), efficient sanding machinery, etc.  Even so, there were occasions
>where Js had trouble holding the rail when the sand quit running for some
>reason.  Again, IMHO, if the J had experienced the reduction in back
>pressure afforded by the poppet valves, it might have been totally
>unmanageable under bad rail conditions.  And, understand, the J was as fast
>as it needed to be for the N&W.  The low drivers were there for a reason,
>and poppet valves might have negated their benefit.  The Js horsepower curve
>peaked just where it needed to, for the demands of N&W's service.  In normal
>service, a J could take a heavy Pocahontas up Alleghany Mountain and
>Bluefield Mountain at track speed under any rail conditions, and run the
>same train fast enough west of Ironton to make up time into Portsmouth.  One
>also has to consider the expense - would the cost of application of the
>poppet valves (they'd have had to buy them from Franklin) have paid for
>itself?
>
>As far as the A is concerned - the same reasoning applies.  The A would run
>as fast as the demands of N&W's service required.  A poppet-valve A might
>have run a 16,000 ton train faster between Williamson and Portsmouth, but
>the lack of back pressure might have resulted in the engine not being to
>accelerate the train to that speed because of slipping.  As also had a
>factor of adhesion of less than 4.0.
>
>Again, the poppet valve probably didn't get a fair shake, because it came
>along too late.  Apart from the T1, nobody made a fleet application of them.
>But studies needed to be made of factors of adhesion and weight
>distribution; if such studies were made on the two original T1s they weren't
>extensive; the Pennsy was in a hurry for a locomotive to replace the aging
>K4s and that was what they decided they wanted.
>
>Pennsy sent a T1 down to the N&W, where it was tested between Roanoke and
>Norfolk; it was also tested on Alleghany Mountain, where it slipped still
>with tonnage routinely handled by a J.
>
>Pennsy also had a Duplex-drive freight engine, the 4-4-6-4 Q2.  This
>locomotive also suffered from adhesion problems, even though it was equipped
>with piston valves actuated by Walschaerts valve gear.  Engine 6180 of this
>class was actually sent to the N&W and tested on the Kenova District, where
>it did not approach the performance of the 1210, which had been tested with
>the dynamometer car some years before.  The Q2 was heavier than the A but
>had less weight on drivers.  And there was a Pennsy Road Foreman with the
>engine, who should have been able to show the N&W folks what tricks, if
>there were any, to the 6180's operation.
>
>  I am not
> >including the Y6's because I don't believe they would have benefitted from
> >the poppits due to the fact that they were not high speed engines.
> >Could any increased performance and efficiency  have staved off the
> >advance of diesels?
>
>Not on your life.
>
>of diesels for a longer period?
> >
>What killed the N&W steam engines was not their performance or lack
>thereof - it was the performance per dollar.  When the diesels became
>capable of supplying the performance N&W needed for less money, steam was
>done.
>
>EdKing