FB-3 variant queries

[Pneudyne]

On the other hand, GE's operator manual for the U50C is available at George Elwood's invaluable "Fallen Flags" rail image site, at http://www.rr-fallenflags.org/manual/U50-OM.pdf" onclick="window.open(this.href);return false;
The arrangement of equipment diagram at the end of the manual shows the GTEL trucks, so it, at least, was prepared with knowledge of the final configuration. And the weight quoted in the data table near the beginning is 417,000 pounds.?

That’s an interesting operator’s manual. The data tabulation shows a truck wheelbase of 13’7”, which I suspect would be associated with the FB3 design, as the GTEL8500 trucks were of 14’6” wheelbase. The way the manual is written, it creates the impression that dynamic brakes were standard, not optional equipment. Anyway, it looks as if the manual was written with mixed information, from both the core design and the actual production variant for UP. Perhaps not quite the best editing job by GE....

Cheers,

[Allen Hazen]

Pneudyne--
Thanks for the essay (two posts back) on the history of GE's C-truck designs! I think that's going to be one of the posts to this forum with permanent value.
As for the Pennsylvania E-44… This design was basically, was it not, just an uprated version of the E-33 built for the Virginian. So I think that the instinct of GE's engineering staff would have been to use the same truck design.
I believe that GE was initially unwilling to go for six-axle designs for 2500 hp diesels, feeling that the U25B's wheelslip control, if properly maintained, would allow competent locomotive drivers to use 2500 hp with four axles. (One of the very nice essays, based on original GE documentation, on the Davis brothers' "Railroad Locomotives" website, b.t.w., is about the history of GE's wheelslip control schemes for high-horsepower domestic U-series locomotives:
http://railroadlocomotives.blogspot.ca/ ... 5-u36.html" onclick="window.open(this.href);return false; )
They gave in to customer demand and offered the U25C after the first E-44 were built. So the choice of truck for the U25C was just to use an off-the-shelf design already in use on another model.

As for GE never being really an enthusiast for the trimount style of truck… Well, it was one thing they dropped at the first major redesign of the domestic U-series in 1966!

[Pneudyne]

I believe that GE was initially unwilling to go for six-axle designs for 2500 hp diesels, feeling that the U25B's wheelslip control, if properly maintained, would allow competent locomotive drivers to use 2500 hp with four axles. (One of the very nice essays, based on original GE documentation, on the Davis brothers' "Railroad Locomotives" website, b.t.w., is about the history of GE's wheelslip control schemes for high-horsepower domestic U-series locomotives:
http://railroadlocomotives.blogspot.ca/ ... 5-u36.html" onclick="window.open(this.href);return false; )

Yes, it would appear that GE thought that 625 hp/axle should have been manageable, in part perhaps because the GTEL8500 was at 708 hp/axle. The Davis brothers essay is excellent. The 16-notch throttle was also part of the package, intended to provide finer graduation of starting and low-speed tractive effort. That followed the precedent of the GTELs, which had 20-notch throttles in deference to their high power-per-axle as compared with diesel practice. Presumably the choice of 16 notches (and the manner in which they were obtained) was dictated by the need for not-too-complicated backward compatibility with existing diesel 8-notch throttles. Although that said, the GTELs were later fitted to control trailing 8-notch diesels, although the exact manner in which that was done seems not to have come to light, at least in the railfan domain.

They gave in to customer demand and offered the U25C after the first E-44 were built. So the choice of truck for the U25C was just to use an off-the-shelf design already in use on another model.

Possibly there was an element of: “we don’t think that you need six axles, but if you must have them, then take these trucks....”

As for GE never being really an enthusiast for the trimount style of truck… Well, it was one thing they dropped at the first major redesign of the domestic U-series in 1966!

By then GE had likely moved from denial through resistance and acceptance to commitment, and had decided to do a proper job of the domestic C-truck, fully aligned with its own precepts.

Re the GTEL8500, apparently as originally envisioned, and as shown in an early model, it had outside-equalized rather than inside-equalized trucks. A photograph of this model is shown in Cockle, page 60, and also in an article in Diesel Railway Traction, 1957 May, copy attached. The cutaway drawing shown by Lee (page 26) was also of this original with outside-equalized trucks. Outside equalization was consistent with GE’s established practice; possibly the change to inside equalization was done to save some weight when it was seen that there was some weight growth from the original estimate.

from DRT 195705 p.188.jpg

As for the Pennsylvania E-44… This design was basically, was it not, just an uprated version of the E-33 built for the Virginian. So I think that the instinct of GE's engineering staff would have been to use the same truck design.

I wonder though whether the Pennsy might have wanted something better, but was dissuaded by GE because of the cost and/or timing for the required re-engineering.

Pneudyne--
Thanks for the essay (two posts back) on the history of GE's C-truck designs! I think that's going to be one of the posts to this forum with permanent value.

Thanks, Allen. Please bear in mind though that it was a “back-of-the-envelope” exercise based upon readily available data and not any deep research.

Cheers,

[Allen Hazen]

Re:
"I wonder though whether the Pennsy might have wanted something better, but was dissuaded by GE because of the cost and/or timing for the required re-engineering."

Given the financial status of the PRR by 1960, I doubt they would have asked for anything but the cheapest option!

[Engineer Spike]

I am wondering if the developments of trucks on various GE electric locomotives, and the turbines has nothing to do with diesel development. We may be overthinking this whole thing. I think Alco has more to do with this than anything else.

GE had been in partnership with Alco until GE became concerned with Alco’s quality control, and their after sale service and parts. In the 1940s and 1950s, some railroads bought some of each brand of diesel. Some of this was the thought that any diesel was better than steam. Other lines were willing to pay to see who actually made the best mousetrap.

By the late 1950s, EMD and Alco has proven themselves to have the best products. Both had significant market share, and had driven out also rans Baldwin, Lima, and FM. GE couldn’t come in with something totally foreign. Except for the CB diesel, the same parts that were stocked for Alcos worked on GE. Truck/traction motor sets might be no exception. A railroad could have spare trucks that could be swapped around its 6 axle Alco and GE fleet, much like New Haven, for example could do so for its U25B/C425 fleet. It was more to the late 1960s that both GE and Alco had developed new truck designs. This very well may have been due to the deficiencies of the tri mount trucks, and the much higher outputs of 3600 hp locomotives.

Two suppliers may have just been an anti trust deal. It just as well could be so any supplier issues such as cost could be controlled. I’m surprised this FB design lasted until ratings exceeded 4000 hp. GM came out with a truck with all motors facing the same direction way back with-2. MLW did likewise shortly earlier.

[Allen Hazen]

Engineer Spike--
Somewhere, a long time ago, I read a claim that when the first MLW M-630 or M636 were traded in to GE, GE's engineers were impressed by the Dofasco trucks of the MLW units… with the suggestion that this had something to do with the adoption of a new truck design on the Dash-9 and AC4400 units. I don't know whether there is any truth to this: it would be surprising if so, since GE had used truck similar to MLW's on many export CC units before this.

[trainiac]

Somewhere, a long time ago, I read a claim that when the first MLW M-630 or M636 were traded in to GE, GE's engineers were impressed by the Dofasco trucks of the MLW units… with the suggestion that this had something to do with the adoption of a new truck design on the Dash-9 and AC4400 units.

It would not surprise me that the MLW Hi-Ad was instrumental in the development of the GE Hi-Ad truck. After all, the AC4400CW prototype and all Dash 8-40CM's rode on MLW trucks. But here's the most telling detail: the rubber pad secondary suspension on the GE Hi-Ad is almost a carbon copy of the MLW design, right down to the shape, size and positions of the rubber pads - which are angled around a traction pin outboard of the middle axle (4" farther outboard on the GE design).

High-adhesion trucks from all builders tend to use a stiff secondary suspension (to reduce weight transfer), a soft primary suspension (to equalize axle loads) and motors all facing the same direction (to balance motor forces). The EMD HT-C and MLW Hi-Ad incorporated all three; both used tall primary springs (resulting in a tall truck frame). The HT-C had wear plates that kept the bolster horizontal while the MLW version went bolsterless (as did the later EMD HTCR and GE Hi-Ad).

By adopting the "roller blade" axle housing, GE was able to use tall primary springs on their Hi-Ad while making the truck frame lower than either the MLW Hi-Ad or EMD HT-C. EMD did the same by bringing the springs down to axle level on the HTCR.

I’m surprised this FB design lasted until ratings exceeded 4000 hp. GM came out with a truck with all motors facing the same direction way back with-2. MLW did likewise shortly earlier.

I am too, actually. The FB-3 was possibly a cut above the EMD Flexicoil in terms of adhesion in that it had rubber pads instead of springs for the secondary suspension. (It also might have benefited from GE's bigger traction motors). However, it's true that it didn't have the characteristics of later high-adhesion designs - the motors faced different directions, the primary springs weren't particularly tall, and movement of the bolster was not restricted.

[bogieman]

Somewhere, a long time ago, I read a claim that when the first MLW M-630 or M636 were traded in to GE, GE's engineers were impressed by the Dofasco trucks of the MLW units… with the suggestion that this had something to do with the adoption of a new truck design on the Dash-9 and AC4400 units.

It would not surprise me that the MLW Hi-Ad was instrumental in the development of the GE Hi-Ad truck. After all, the AC4400CW prototype and all Dash 8-40CM's rode on MLW trucks. But here's the most telling detail: the rubber pad secondary suspension on the GE Hi-Ad is almost a carbon copy of the MLW design, right down to the shape, size and positions of the rubber pads - which are angled around a traction pin outboard of the middle axle (4" farther outboard on the GE design).

At the time the Hi-Ad was developed Dofasco had a staff of truck design engineers who I believe actually did the truck design as they were the first supplier of those truck frame castings.

High-adhesion trucks from all builders tend to use a stiff secondary suspension (to reduce weight transfer), a soft primary suspension (to equalize axle loads) and motors all facing the same direction (to balance motor forces). The EMD HT-C and MLW Hi-Ad incorporated all three; both used tall primary springs (resulting in a tall truck frame). The HT-C had wear plates that kept the bolster horizontal while the MLW version went bolsterless (as did the later EMD HTCR and GE Hi-Ad).

The plates between the underframe and bolster on the HT-C truck are officially side bearing wear plates and are not intended to contact in operation. Per FRA rule, the clearance is to be between 1/4" and 1/2" stationary and are there to limit roll between the bolster and underframe. They do, however, come in contact at an adhesion level of about 25% as the traction force acting on the bolster from the truck frame exceeds the center bearing ability to prevent the bolster from tipping (the net moment acting on the bolster falls outside the center bearing diameter).

By adopting the "roller blade" axle housing, GE was able to use tall primary springs on their Hi-Ad while making the truck frame lower than either the MLW Hi-Ad or EMD HT-C. EMD did the same by bringing the springs down to axle level on the HTCR.

GE had severe problems with primary spring failures on the FB-3 truck and I was told that the "roller blade" bearing adapter was to provide for a longer spring which lowers the stress for the same static deflection. I also believe that the spring failures were a result of their source not doing as thorough an inspection of the springs during manufacture. When I designed the HTCR truck, I made the springs taller to eliminate the inner coils to allow for a single coil on each side of the bearing adapter. I didn't want the inner and outer coils to wear on each other as the axle ends displaced longitudinally during axle yaw. To insure the springs survived, they are made from "peeled" bar which adds some cost but removes the surfaces defects which are the main source of fatigue cracks in service.

The FB-3 was possibly a cut above the EMD Flexicoil in terms of adhesion in that it had rubber pads instead of springs for the secondary suspension. (It also might have benefited from GE's bigger traction motors). However, it's true that it didn't have the characteristics of later high-adhesion designs - the motors faced different directions, the primary springs weren't particularly tall, and movement of the bolster was not restricted.

The weight shift performance of the FB-3 is not significantly different than the Flexicoil. Using rubber secondary springs actually inhibits truck frame pitch which helps weight shift when the motors are not facing the same direction. The opposite is true with the motors all facing the same direction - a stiff secondary is required to limit pitch of the truck frame, if there is no pitch of the frame or bolster, there is no weight shift within the truck since all axles see equal spring and motor nose forces (a bit of an oversimplification). For reference, weight shift between axles in a locomotive is evaluated using the term "adhesion efficiency" which is the percent of nominal static load on the lightest axle at a locomotive adhesion level of 25%. For HT-C and HTCR trucks, the lightest axle is about 93% of static at 25% adhesion, for the Flexicoil, it is about 78% and similar for the FB-3. The GP (Blomberg to railfans) truck regardless of secondary suspension type (rubber or elliptic) is about 84%. These numbers are for the locomotive and include weight shift between trucks which is about 4% depending on truck center spacing.

Dave

[trainiac]

Thanks Dave! It's a delight to have clarifications from someone who actually knows things first-hand (as opposed to someone like me who doesn't).

[Pneudyne]

The starting point for GE domestic-market C-trucks – at least in the post-WWII era - seems to have been the New Haven EP-5 rectifier electric locomotive. This had an inside-equalized, single-swing bolster trimount type truck, evidently designed for good riding and tracking at high speeds. As well as the two (per side) customary between-the-axles primary coil springsets, it also had outer, stirrup-mounted springsets. I am not sure when such were first used, but they seem to be strongly associated with the Fairbanks-Morse Train Master, which had rigid-bolster trimount trucks. Another feature of the EP-5 trucks was that the spring pair between the centre and outer axles were separated to allow room for the bolster elliptic springs between the coil springs.

Next was the Virginian E33 electric locomotive. One imagines that in deference to its lower speed freight haulage role, it had simpler outside-equalized rigid bolster trimount trucks, with the usual two primary springsets per side. VGN itself may have seen swing-bolster trucks as an unneeded luxury for its operations. For example, its preceding EL-2B motor-generator electric locomotives had rigid-bolster B-trucks, whereas for example GE’s more-or-less contemporary GTEL prototype had swing-bolster B-trucks, also with span-bolsters.

Some more background:

In the case of the VGN EL-C, the railroad wanted trucks and traction motors that were interchangeable with those of its GE-equipped F-M Trainmaster, so that effectively dictated the rigid-bolster, trimount type; it was not a GE choice (1). The EL-C and T-M trucks were somewhat different, for example the former was outside-equalized and the latter inside-equalized with outboard extensions of the equalizing bars carrying stirrup-mounted springs. But they conformed to the same major dimensions, including wheelbases and pivot placement. So they may well have been interchangeable. The EL-C was built to a 396 000 lb maximum weight, implying a 66 000 lb axle loading, so in that regard it was also similar to the VGN T-M. (Sidebar: It is interesting that the VGN was by then working to a much lower axle loading that had been the case for the steam era. The EL-2B had a 62 500 lb axle loading, but that appears to have been a circumstantial number rather than a limiting value.)

With the NH EP-5, its 58 000 lb axle loading, as compared with the 45 000 lb of the earlier EP-4, was allowed in part because of bridge improvements meanwhile, and in part because it had widely spaced trucks, as compared with the EP-4, whose wheelbase was relatively close-coupled throughout. I also suspect, although it is unconfirmed, that the use of relatively long wheelbase trucks was part of the equation. The trimount choice was made to allow adequate room for three motors, and presumably, the swing bolster. (2)

The floating bolster truck itself was patented by GE, as US2907282 of 1959 October 06, application date 1957 September 04. The diagrams in the patent document show a C-truck that looks broadly like the export type.

(Another sidebar: the swing motion/lateral motion C-truck appears to have been a post-WWII development, with an interesting cross-Atlantic early history.)


(1) from: “General Electric Review” 1958 May p.36.

(2) from: AIEE Transactions 1955 July p.183ff.


Cheers,

[Allen Hazen]

Thanks for following up! (Some of these old strings continue to attract interesting new information.)
Re: "So they may well have been interchangeable. " They were. There is a photograph (I don't remember whether it is at the "Fallen Flags" site or in a book about the VGN) showing a VGN C-C locomotive (I don't remember whether it is a TM or an E-33) with mismatched trucks, one of each type. The dimensions of the FM and GE trucks are almost exactly the same, so (for example) traction motor air ducts for ventilating the motors on one type of truck would work for the other. In contrast, Alco's trucks differed in relevant dimensions by a few inches: when N&W wanted to use trucks from retired TM on an order of Alco C-630, they had to be built with non-standard underframes.
(((One of my fantasies -- it could be used to justify an "imagineered" model -- is of a railroad that decided, in the mid-1960s, to get rid of minority-make power, and ordered U-25C or (early) U-28C with the trucks of traded in TM.)))
About New Haven axle weights: I'm not enough of an engineer to have much clue about what matters to a bridge, but maybe it is also relevant that the total locomotive weight of a 2-C+C-2 EP4 is greater than that of a C-C EP-5 despite its lighter axle loading.
Thanks again!

[Pneudyne]

The attached excerpt from “Railway Locomotives and Cars” 1965 September has an illustration of what looks to have been the original FB3 truck, although the supplier variant was not identified.

RLC 196509 p.49.jpg (329.83 KiB) Viewed 959 times

I suppose that the fact that the GE U28C was offered as being suitable for higher speed passenger service indicated that it should be equipped with a lateral motion truck with secondary springing, rather than a rigid-bolster type with primary springing only.


Cheers,

[Allen Hazen]

Pneudyne--
Thank you for continuing to search out relevant information!
If the images are from a September 1965 issue of the magazine (and so copy must have been submitted at least somewhat before that), the artwork was done before there were any U28C locomotives, and, I would guess, before there were any FB-3 truck frames in existence: the first "new carbody" U28C, and so the first locomotive to be equipped with this truck, was built in May of 1967 (for the Louisville and Nashville). So GE's artist may have been working off the specification that GE had sent out to the two truck-frame suppliers, without seeing how either of them would realize the design.
Certainly the truck in the picture doesn't have obvious spotting features for either GSC or Adirondack (though when I stared at it the first time I thought the squared-off profile of the "hammerheads" over each of the outer axles might be a bit more Adirondack-ish -- but I don't trust my judgment on that sort of thing). Early Adirondack versions had a prominent "shelf" over the middle axle, which certainly isn't shown here. (The origin of this "shelf" is a mystery. Obvious guess would be that Adirondack added it to their version of the design as some sort of structural reinforcement, which GSC didn't see the need for ... and which Adirondack itself thought better of, since it was eliminated from later versions of the Adirondack casting.)
There are two circular holes in the side of the frame casting for an FB-3 truck: about half-way between the centre and outside axles in the GSC version, and a bit closer to the outside axles in the Adirondack.(*) They aren't shown at all in this image: further evidence, perhaps, that it is an artist's impression based on the specification and not drawn "from the life" in front of a physical casting.
(*) Maybe the holes themselves were a later addition to the design, not yet included when this picture was prepared? Well, not MUCH later: both the Adirondack trucks on L&N's May 1966 U28C and the GSC trucks on PRR's U28C built late in 1966 have the holes.
---
Re:
"I suppose that the fact that the GE U28C was offered as being suitable for higher speed passenger service indicated that it should be equipped with a lateral motion truck with secondary springing, rather than a rigid-bolster type with primary springing only."
---It also, perhaps, indicated more confidence in the design than was warranted. The early "passenger" units with this truck -- Santa Fe's (1966) U28CG and (1967) U30CG -- were reassigned to freight duties after at least one derailment. The truck was used successfully on some later passenger locomotives -- NJ Transit's U34CH and Amtrak's P30CH -- but the design might have been tweaked a bit by then.
---
Tangent: the traction generator in the drawing is huge. I assume it is a DC generator, since no location is shown for rectifiers, and since I believe that L&N's units were built with DC generators.

[Pneudyne]

For high-speed passenger work, where it is less necessary to chase every last increment of adhesion, the GSC double-swing bolster C-truck would probably have been a good choice. (GE did use this truck in at least one application, the NSWGR 43 class.)

The generator was the GT598, which GE stated was suitable for power inputs above 3000 hp. In the article it was mentioned that in 1966 there would be limited production of units with alternator/rectifier combinations.


Cheers,

[Allen Hazen]

Thanks again!
re: generator. GT-598 was the big generator introduced on the U25B. I had forgotten whether U28C (DC) had GT-598 or GT-586.
GE seems to have been a bit slow in adopting AC-DC transmission on its own locomotives: they started U28 production with DC generators, even though they had already designed and built the GTA-9 alternator for Alco to use on the C-630!
(Perhaps GETS's engineers were a bit in love with their great DC creation, the GT-598!). For once the intensely conservative railroad industry jumped at a new technology: it was reported some years back on this forum (or one of its ancestors) that GE experienced a severe drought of orders in 1966. And then pivoted to AC-DC applications.
The suddenness of the pivot can be seen in the history of GE's test-demo set 300-303 (or is it 301-304). It is clear from the builder's numbers that GE originally planned for two units of this set to have DC generators and two alternators, and that then -- realizing that the alternator version was the only one for which there would be demand, switched to a set of four AC-DC units. (The two DC versions were probably already under construction when plans were changed: the New York Central, clearly in need of all the power it could get, bought the them. They became NYC 2822 and 2823, the two U28B belonging to the NYC proper as opposed to the P&LE subsidiary.)
I think GE's total production of domestic locomotives in 1966 was lower than in either 1965 or 1967, and a contemporary article in "Trains" opined that GE had stumbled in cataloguing the U28B and U28C when its competitors had moved directly to the GP40/SD40 and the C-630. (Though Alco continued to offer the C-628 as an alternative to the C-630, and sold a reasonable number: many, i.i.r.c., to Mexico: perhaps NdeM's mechanical department was particularly cautious! But note that, though Alco offered a C-428, they didn't sell any.)