A facts and corections, and a repeat of an idea that I presented some time ago. First, do any of the GG-1’s that have been preserved have frames without any sign of cracking. If so, a restored and practical GG-1 is possible. Much of the catenary, nearly all south of Harold Tower and Sunnyside in New York City, is still 25 Hz, and the raised voltage from 11,000 to 12,500 volts should not be a problem if old cotton insulation is replaced with modern plastic, which should be done in anycase. But eventually, the entire NEC will be 60Hz, and a completely restored GG-1 just won’t work. So, what to do? The transformers have to be replaced anyway, because the old ones used a toxic cooling flued that won’t pass modern environmental rules. The saving point, though, is that AC-commutator motors work fine on dc, without any modification. (Not induction or any type of synchronous motor, just commutator motors.) My solution was to save all the elecdtricals of three DC AEM-7s, transformers, rectifiers, electronics, etc, and adapting all this to the twelve motors that power a GG-1(two-per-axle). Whether the multi-notch GG-1 controller could be retained might be a problem, and possibly a new power control stand would have to be designed and built. But the result should be a powerful and useful locomotive that can run on any existing or futre NEC power for occasional excursion service and not give any exterior visual evidence of internal changes. Sort of what could have been done if Amtrak or NJT really wanted to keep GG-1s in service.
From what I have researched, all preserved examples have had both the transformers and the mercury rectifiers removed and replaced with concrete ballast of the same weight. These engines will never run again under their own power.
At least one GG1 sent to the National Train Day event in Washington DC in recent years had gondolas loaded with stone fore and aft of the GG1 acting as brake cars because the brakes were not functional on it. I don’t know if they ran auxiliary brake lines to connect the two gondolas or if they used the ancient lines in the GG1 but this thing hadn’t gone anywhere by itself in decades.
EDIT: PRR 4935 would be a notable exception since it was given special attention by Amtrak and rail fans. Does anyone know if 4935 was visiting National Train Day with those other GG1s a few years ago?
Never say never after 50+years a big boy has taken to the main line again and is in Cheyenne WY at the steam shops of the UPRR getting ready for 2019 celebration of the golden spike. She will run on her own power when finished. She is being completely gutted and re-machined according to official UP drawings. A GG1 could have no problem with the electric’s it’s the frame that holds back a complete restoration. Solve that and the rest is feasible.
I have to wonder whether the technical knowledge of some of these posters concerning what has been done to preserved GG1s, or what supposedly makes it likely a GG1 will ‘never run again,’ is trustworthy. If you provide me with the blueprints of the mercury rectifiers used on the GG1, I will personally make the replacement components for you in a very short time, at small cost to whoever will be doing the restoration.
One point being made is a good one, but counter to what the poster seems to want to establish: to my knowledge, all the PCB-containing transformers have indeed been removed from preserved GG1s (with associated structural damage and crude re-ballasting attempts). However, it should be noted that this involves less, rather than more, actual restoration work (if we assume, as most do, that there would be no way to remediate the PCB problem during the restoration, or ‘grandfather’ the use of PCBs through a proper safeguarding program if the transformers remained intact (but drained or passivated).
As Dave Klepper has repeatedly noted, there is no particular reason why an ‘external restoration’ of an operable GG1 needs to replicate the precise multitap main transformer architecture that PRR used, let alone the 1930s construction of the main transformers themselves. However, when I looked into this a few years ago, I found a number of academic and commercial sources that could design (and if necessary provide) an electrical replacement with the appropriate windings and taps to make the locomotive operable, either at full or reduced power. Since that time, both the cost and expertise needed to produce technology for transversion of high-voltage power-frequency AC t
That said, after all this time the transformers would have to be replaced regardless of the cooling medium used. It is usually transformer breakdown or failure that sidelines an ageing electric.
The big obsticle to returning a G to life is the truck frames. They can be removed, stripped to the casting, flipped over, heated back into form, gaps filed, cracks welded and, finally, heated for stress relief. It is not a technological problem; it is a money problem. How many fantrips will you need to run to recoup the cost? One could never break even on a venture like this. It would take a fan or group of fans with deep pockets to make it happen. Should it happen? Well, duh.
One procedure that was being discussed in the mid-'70s involved stripping, cleaning and etching the underframes, then constructing an appropriate jig and enclosure to heat them for an extended period to redissolve the “crystallization” cracking, then slowly taking them down through normalization as appropriate for original casting. Then you’d do NDT and some residual metalstitching (keyhole welding, more likely, today) on cracks that did not correctly close.
The point is the same, however: it’s a money problem, and a significant one. And yes, you’d likely never break even. And yes, it should happen if there’s any way to accomplish it*.
*Including perhaps the sort of money-throwing efforts done with 1361, if the chance ever comes again – but with better planning, scheduling, and knowledge about project management…
I cannot speak for every steam locomotive but none of the ones I have ever worked with showed manifistations of the cracking seen on the Gs. Then again, not too many 40+ year old steamers were called upon to run at 100+ MPH on a daily basis.
There’s a bit of a difference: cast beds were generally designed with larger sections, less requirement for weight reduction, and of course were sited “between the drivers” (where the most longitudinal or oscillating stress loading occurred) in almost all cases. Where these kind of design principles could not be or were not applied (for example as in some Australian Garratts) alarming or expensive cast-bed cracking could be and was observed. There is no inherent unbreakable magic in a cast-steel underframe, whether or not GSC propaganda was effective…
The G cracking was in part a result of either the alloy or the treatment used in the detail casting, the “crystallization” leading to preferential stress raising. Had the art of finite-element analysis (or the use of computers in design) been available to GE or Baldwin at the time the GG1 underframes were designed, some of the stresses could have been better adjusted for, but the fundamental alloy change leading to the cracking was (at the time, anyway) the stated reason why rework of the existing underframes, or any extensive rebuilding of the GG1s into more modern power, was not undertaken even as a demonstration project.
Somewhere I have the drawings of the ‘welded-frame’ alternative, including chevron springs for the driving axles (in part to get rid of some of the pedestal loading issues) that was at one point in the mid-Seventies being considered for rebuilt high-speed GG1s. The tire problem killed that reason, it was thought that truck-borne power made better sense if true high speed weren’t going to be required, and (in the event) toasters made much better sense for Amtrak service than a large quill-drive 4-6-6-4.
I’m pretty sure that a recast GG1 frame, put back into heavy ser
Much more likely that individual complex sections would be cast using lost-foam, then jigged and assembled using laser keyhole welding. That method eliminates both the difficulties involved in keeping all the molds and cores aligned and in making such a large pour and then controlling the various sections’ cooldown correctly.
if you are going to recast the frame, using an existing frame as the mold for the mold, keep the mold around, so after 20, 30, 40 years, cast a new frame.
A new welded frame might be less expensive and more practical?
GG1 did not have a liquid reostat. No reostat for traction power. A multi-tap transformer instead. Remember an ac COMMUTATOR motor was used, the motor design being to make ac at low enough frequency behave like DC in the motor, well, close enough to dc to get torque and traction. So you just reduce power by reducing voltage to the motor, as is done with resistors for dc motors but more efficiently with multi-tap transformers with ac. And, yes, the New Haven power that ran into Grand Central Terminal did have essentially two separae control systems, grid resistors contolled by relays (and series connection of two motors to parallel, and possibly field shunting [weakening]for dc, and multitap transformers for ac. The rebuiled would use AEM-7-dc technology, using electronics to control dc voltage to the motors.
To paraphrase Arthur C. Clarke, ‘you can’t do that, Dave.’
Matching shrinkages to get a proper one-shot casting from molds made directly from an existing frame is an extremely difficult exercise, on top of the already-difficult matter of assuring proper gating, etc. for a quality pour of a casting that complex.
Perhaps the best approach today would be to make pointcloud measurements of the existing frame, translate these into CAD, and make the patterns, including the lost-foam patterns for the cores, directly with the correct shrinkage allowance. And then make sure someone keeps up with changes in the computer industry, CAD/CAM practice, etc. so that in 40 to 50 years the data will still be in usable shape to make more (or drive whatever cheaper and better processes are available by then).
The point of decrystallizing is that the existing frame represents a near-net-shape pattern already containing all the metal needed for a good casting, so a self-mold not taken fully to liquidus (but only to where the crystallization structures redissolve) and then properly and very slowly cooled will give you a proper underframe (needing little more than metalspraying to reach proper service dimensions) with good strength. Attempting to re-cast the underframes ‘from scratch’ is likely to be far more expensive, even if you have enthusiast-level donation of professional goods and services and you get good castings on the first pour.
Might be interesting, just for historical purposes, to make a set of fabricated underframes as proposed for the high-speed project. Nothing in those that a good shipbuilder couldn’t watercut, hydroform, and laser-weld, and if you wanted a true high-speed GG1 that could operate at modern Corridor speeds you’d
I seldom stray from the Classic Trains and General Discussion Forum’s or else I would be behind a 'puter all the day long and never take the garbage to the curb or anything else done, but in this case I will make an exception.
Very informative discussion here. You could conclude the Pennsy was quite primitive in this frame design without the aid of CAD/CAM, watercuts, hydrofoam or laser anything and yet they built an outstanding locomotive, let alone the frame itself.
!st Semester students in my class are taught real Drafting with T squares, triangles and pencils and inking pens, on drafting and light tables. I have always received a lot of flak and resistance for this as the powers that be continually cannot see this as part of the curriculum, instead wanting the students to go directly to AutoCAD and Vulcan or GemCom. However, its still my course dammit!
I find the students have a far better understanding of what they are doing at the computerized levels with a semester of getting it done by hand on paper and actually thinking. They appreciate the beauty of their work, as did the designers of the GG1 and its components.
There is nothing whatsoever either primitive or defective about the GG1 underframes; they were at least as complex a casting as what GSC was doing with cast engine beds at the time, and the fact that they represented the only locomotives that could achieve practical Metroliner speeds at one time should tell you all you need to know. It is worth considering how the design was made, and then the patterns and cores made, and how GSC cast them – the difficulty today being that doing the job that way involves more metal than required, the suspension is still steam-era (excellent, mind you, but the only damping provided other than friction and gravity was ‘snubbers’ - which were nothing but coil springs wound to a different period from the main suspension springs, similar to what’s done on modern three-piece freight trucks. Those were taken off the GG1 comparatively early in their lives, which I didn’t understand when I thought they were still energy-absorbing devices…
The big problem with the cast frames for high speed was that they were very heavy for the job they actually had to do. In part this was rationalizable because all the buff and draft force went through the underframe, but there got to ba a major problem (as with steam locomotives) when Metroliner service provided only relatively short Amfleet consists (with those inside-frame disc-braked trucks) which were (obviously) not intended to help stop hundreds of tons of locomotive too. The fun came in when you look at how you apply braking effort to a GG1 traveling at high sustained speed – the engine trucks don’t do much good, and the quill axles … have shrunk tires on them, like 57"-drivered steam locomotives. These had to provide a more-than-expected share of the brake force, which led to what we really should have seen coming…
For any practical rebuilding of GG1s for 120mph service, you’d need reliable braking from that speed using consists capable of running
Well heck. Thanks for the in depth informative answer. A fellow colleague I work with daily is a PhD Metallurgist. I will bring up this crystallization factor with him and get his take on the new process required.
Share with me what you find, in detail. Particularly any improvements in alloy metallurgy (or casting technique) that would produce better castings in this particular service.
You might also see what he thinks about eliminating the pedestals and stress associated with them through the use of the canted chevron springs to the axleboxes…
RME- Will do. His PhD is in Physics and he has specialized in Metallurgy for many years now, mostly with aluminum and tungsten studies with aircraft materials and also works on the Synchrotron in Saskatoon. He is also somewhat of a Railfan as his dad was an engineer (driver) in India. You never know what we may find out.