The research alone involved in this project is no small task. We’ve read most of the published literature on the T1 to get a feel for the real problems that existed with the design, and develop plans to address them. T1 Trust volunteers have gathered over 1,000 original engineering drawings and test reports for the T1 from the PRR and Baldwin Locomotive Works collections. The archival blueprints have been used to generate 3D CAD models for many of the parts in Solidworks, and we’ve contacted suppliers for tooling and component manufacturing. The massive and complex wooden patterns for the T1’s driving wheels were constructed at Liberty Pattern in Youngstown, Ohio and on February 26, 2016 the first driving wheel was cast at Beaver Valley Alloy in Monaca, Pennsylvania.
The T1 Trust’s engineers have developed plans for an all welded boiler, and as the CAD modeling effort continues, more attention will be focused on the design of a revised locomotive frame. Ultimately the design will be subjected to multi physics testing to ensure optimal performance. In addition, The T1 Trust plans to produce a few more “showpiece” parts for public display, to raise awareness and to encourage donations.
Our current estimated completion date is 2030. This was based on our own internal estimates of the number of man hours required to complete certain tasks, as well as the duration of the A1 “Tornado” project in the UK. In reality, the program timing will likely be dependent on manpower and funding. If we don’t get the volunteers needed to complete the engineering and construction, or the funds to produce the parts, it could take much longer. Conversely – if we received a donation of $20 Million tomorrow, we could hire a full time professional engineering and fabrication staff, and the project could be completed in as little as 5-6 years.
A more relevant question would be “where could it run?” The answer to that is – just about anywhere. Unlike the S1, the T1 was designed to operate anywhere on the PRR mainline circa 1942. With the original lateral motion configuration, the T1 could negotiate 16 degree curves, and according to the timetable, could operate in areas where even the M1 was restricted. A specific problem with 130 lb no.8 switches prevented them operating through Pittsburgh – but an increase in lateral motion in 1946, and track realignments in the modern era (required to handle longer freight cars than the 1940’s) mean that this particular issue has been resolved. Based on the revised lateral motion, and the overall dimensions, we’re confident that the T1 can operate anywhere on the current mainline network that the N&W J class can. As part of our project, we will investigate a further increase in lateral motion to allow negotiation of 20 degree curves, which would let the T1 operate on any track currently accessible to a NKP Berkshire.
At this point in time, the Trust does not have an agreement in place to operate the locomotive on any class 1 railroad. We will attempt to secure support from a Class 1 carrier once the project is further along.
The T1 Trust has received letters of intent from three well established tourist railroads; the Steamtown National Historic Site in Scranton, PA, the Steam Railroading Institute in Owosso, MI and the Cuyahoga Valley Scenic Railroad in Independence, OH. All three of these fine organizations are experienced in handling mainline steam locomotives and can easily support the operation of the T1 locomotive when complete.
Presently there are only two possibilities, neither of which is likely for revenue service – The USDOT test loop in Pueblo, and portions of the Northeast corridor. The DOT facility is where we would intend to perform high speed testing to confirm the locomotive’s tracking qualities and top speed potential, with an instrumented test train, and only in compliance with all applicable DOT regulations. High speed running is not necessarily part of the routine service plan. Our intent is to maintain schedule on whatever railroad is willing to host the locomotive for excursion service. We anticipate this will be limited to 79mph top speeds on one or more of the Class 1 railroads. If, however, Amtrak can be persuaded to allow excursion trains on their system, we would plan on operating at speeds of 85-110 mph plus to match their timetable.
We haven’t researched the full service career of the T1 fleet, so we can’t say for certain how often it happened, but the answer is definitely yes. 25 of them were built at the Baldwin Locomotive works in Eddystone, PA, just south of Philadelphia, and there are references to one being displayed at railroad equipment shows in Reading PA, and Asbury Park, New Jersey. T1 #5542 was assigned to pulling train No. 17 between Broad Street Philadelphia and Pittsburgh station at least once. However, there was little need for the T1 to run east of Harrisburg, as that region was electrified, and blue ribbon trains were serviced by the GG1.
From what we’ve seen, the PRR solved this problem by 1947, by changing the valves from mild steel to a higher strength alloy that was better able to cope with the fatigue issues at service speeds. We will run durability and fatigue simulations for speeds in excess of the T1’s rumored top speed, and select alloys and manufacturing processes to maximize reliability.
The wheel slip issue had two root causes. The first was ineffective spring equalization. As originally designed (engines 6110 and 6111), the engine truck was not equalized with the drivers, and all four pairs of drivers were equalized together. When entering curves or moving over track that was less than perfectly level, weight was transferred off the front engine, causing the front pairs of drivers to slip. This condition was observed at all speeds, and we believe is the basis for the “uncontrollable” reputation the T1 has. The PRR addressed this in the production fleet by splitting the spring rigging in two – the front engine was equalized with the engine truck, and the rear engine was equalized with the trailing truck. The other root cause was improper handling. Engineers assigned to T1s were given no formal training on how to operate them, and their performance was very different than the K4’s most of them were accustomed to. The front end throttle, high boiler pressure, very large diameter steam delivery pipes, and poppet valves combined to make the T1’s very responsive to throttle application compared to a K4. Too much power applied too quickly resulted in wheel slip, especially at speeds around 15-25 mph. We will be performing kinematic and compliance simulations of the spring rigging and equalization to determine whether further improvements in adhesion are possible. We will be applying a wheel slip alarm, so the engineer would be made aware of a wheel slip more quickly should it occur, and reduce power manually. We will also investigate fitting an electro-mechanical anti-slip device similar in concept to that fitted to the Q2, but with more reliable valves and modern electronics, so no involvement from the engineer would be required.
The short answer is 10 million dollars.
How did we get there? That’s the fun part, but only if you like math. There are several ways to approach the question.
The most obvious way to estimate cost might be to consider inflation. The average cost of a T1 in 1945 was about $320,000. Using data from the Federal Reserve, and its Consumer Price Index (CPI), the cost of a new T1 in 2013 is an estimated $4,175,324.68. Unfortunately, that number does not take into account lost skills, knowledge, and tooling that will have to be relearned, rebuilt, or replaced with modern alternatives as the T1 project progresses. In the worst case scenario, the cost could be seven times as high. Consider for a moment the following example. An original A1 built in Darlington cost £16,000 in 1948. The inflation in Britain over the time period 1948 to 2008 was 2,623%. At that rate, one would expect the final cost of Tornado to be £419,680. It was in fact more, seven times more. The final price tag for Tornado was in excess of £3 million. Why is that? In many instances batch production tends to spread cost, whereas the production of a single unit tends to add cost. There is however a silver lining. In the case of Tornado cost savings of up to 33% of the original cost were achieved during some stages of construction. For example, fabricating a disposable mold used for one part is less expensive than manufacturing a mold which will be used repeatedly to produce 50 parts. In order to reduce expense, the 5550’s construction will employ modern techniques such as CNC, and rapid prototyping when, and where-ever possible. Smaller castings with specialized joints for welding may help to further reduce costs, especially in the case of the T1’s large frame.
Another method of calculating cost, is to do so by weight. Tornado weighs 167 tons and cost 5 million dollars. That’s a cost of $30,000 per ton for Tornado, and we’ll use that to calculate the T1’s cost based on its weight. Depending on who you read, production model T1 weight is reported from 318 to 346 tons. The average is 332 tons, almost exactly twice the weight of Tornado. So that should be just about twice the cost, or $9,960,000. Let’s call it 10 million. Next, we consider total heating surface, and firegrate area. Total heating surface for Tornado is 2,461 sqft, and at a total cost of 5 million dollars, that’s $2,031 per sqft. Total heating surface for the T1 is 5,639 sqft, at $2031 per sqft, that’s $11,452,809. Turning to firegrate area, Tornado has a grate area of 50 sqft, and that’s pricey real estate at $100,000 per square foot. Grate area for a T1 is 92 sqft, so 9.2 million dollars.
Finally, we look at length. Tornado measures 73′ buffer to buffer. That’s $68,500 per foot. The T1’s wheelbase is 107′ which gives us $7,329,500. That helps take the edge off the earlier 11.45 million dollar figure. In the end, it’s going to come in really close to 10 million dollars.
The Type A gear, while effective, presented a challenge with regard to maintenance, especially for the rear engine. To illustrate this, it’s necessary to explain the main features of the Type A system.
Power for the gear was taken from a lever attached to the crossheads of both engines. This lever actuated a bell crank, which actuated an adjustable length rod, which was attached to another bell crank, which actuated the input shaft of the gearbox. The gearbox itself was a sealed, cast steel box which was located above (front engine) or between (rear engine) the locomotive frames. Inside each gearbox were two complete sets of miniature Walshaerts’ valve motion, which were immersed in about 30 gallons of SAE 30 oil. One set provided drive for the intake valves, the other for the exhaust valves. Each set of motion actuated a separate output shaft from the gearbox. Each of these output shafts had a bell crank, which actuated another adjustable length rod, which then actuated another bell crank attached to the inboard end of one of the camshafts. The camshafts were mounted in a second sealed box, mounted between the steam chests, and filled with 2 gallons of cylinder oil. The Camboxes then opened and closed the valves, which were mounted in the steam chests.
When the gearboxes needed to be serviced, they were very difficult to access. To reach the front gearbox, located on the frame just ahead of the front cylinders, all of the streamlining on the smokebox and pilot had to be removed. The rear gearbox was almost completely inaccessible, and had to be removed via a drop pit for major servicing. The gear boxes weighted about 3700 pounds each, so removing one from beneath a locomotive was no easy task. Once serviced, all of those adjustable connecting rods between the crosshead and gearbox, and the gearbox and camboxes, had to be re-adjusted to keep the valve events square. That’s a total of six rods, whose lengths were specified to the thousandth of an inch, for each engine.
The Type B system avoids all of this. Power is taken from a gearbox driven by a return crank on the main driver crankpin. This gearbox is attached via a jointed driveshaft to a matching gearbox on the outboard end of the cambox. There are very few moving parts, no adjustments required, and everything is accessible from the outside of the locomotive without having to dismantle anything. It’s simpler, lighter, easier and cheaper to fabricate, and much easier to maintain.
The Type B is also not without precedent on a T1. In 1948, locomotive #5500 was involved in a sideswipe accident with a K4 on the St. Louis division, resulting in heavy damage. Instead of repairing the locomotive with the Type A Oscillating Cam gear, it was rebuilt with the B2 Rotary Cam gear. Afterward, it gained a reputation as being the best of the fleet, by both engineers and maintenance men alike. There is evidence in PRR correspondence that consideration was given to fitting the type B gear to as many as five T1’s, but this idea was not acted upon. It’s reasonable to speculate that, had the T1 not been replaced so quickly by Diesels, that additional units would have been similarly modified.
The locomotive will be used as a test bed for alternative environmentally friendly fuels to allow operation of America’s steam locomotives into the foreseeable future. The locomotive would be a national touring education center when complete while testing coal alternative fuel sources such as torrefied biomass, natural gas, vegetable oil, recycled oils and propane. As well as fuel sources being tested, combustion and drafting would also be tested for increased combustion performance and reduction in carbon output. Results would be shared with operators around the country as well as plans for coal to other fuel conversions.
