Here is a description of the C44ACi locomotives:
Wayback Machine (archive.org)
This shows the a diagram of the bogies which do indeed seem to have standard gauge traction motors. The diagram of the intercooler also answers questions asked in this forum. The "variable hosepower" feature was dropped, and most locomotives were described as C44ACi although the "AC Class" locomotives were listed as "C43 ACi" on the builder's plates.
Peter
BDA No , it didn't limit to 11500 . Don't know why . And no , the reason I'm asking about 8 motor performance is because I think this would be the only way to get decent locomotive performance in Australia . I doubt the Feds , States and ARTC are going to pay for US domestic standards for the rail infrastructure so we could run 5020 type units (heavier 180 tonne C44ACi) on 6 axles . Logically the only other way to achieve this is more powered axles at the supposedly allowable 22.3 tonnes . 6 by equals 134 tonnes , 8 by equals roughly 176.5 tonnes . Would it be too simplistic to assume that the tractive effort increases by the same 30 odd percent . If you scale the C/ES/44ACi up by 30% thats more like 1700 to 2200 tonnes on 1:40 grades .
No , it didn't limit to 11500 . Don't know why .
And no , the reason I'm asking about 8 motor performance is because I think this would be the only way to get decent locomotive performance in Australia . I doubt the Feds , States and ARTC are going to pay for US domestic standards for the rail infrastructure so we could run 5020 type units (heavier 180 tonne C44ACi) on 6 axles .
Logically the only other way to achieve this is more powered axles at the supposedly allowable 22.3 tonnes . 6 by equals 134 tonnes , 8 by equals roughly 176.5 tonnes .
Would it be too simplistic to assume that the tractive effort increases by the same 30 odd percent . If you scale the C/ES/44ACi up by 30% thats more like 1700 to 2200 tonnes on 1:40 grades .
I'm not sure that there is the demand for the 180 tonne locomotives. Only one operator, Aurizon, uses these units. They started their Hunter Valley coal operation with 12 4000 HP units, 5000-5012, and later purchased another 25 4400HP units 5021 - 5045. These are used on ECP fitted coal trains, usually around ninety 120 tonne hoppers with one loco on each end in wired distributed power.
There are around 190 134 tonne units of the same power which are used widely over the whole country, but the majority are used by other operators in the Hunter Valley coal traffic, usually with three units at the front of a train with fewer, sometimes 84 hoppers. These can be fuelled up to 139 tonnes on the heavy track in the coal fields. In earlier times Aurizon used one 180 tonne and one 139 tonne unit leading on the same trains as the pairs of 180 tonne units, then both leading. This seemed to work since the lighter unit was trailing and was less likely to slip. The locomotives have the same power, but the heavy units have GEB 13 motors and the lighter units have GEB30 motors.
Pacific National had been using 165 tonne DC units, GT46CWM type, but as discussed in my earlier post above, standardised on 134 tonne AC traction units which could match the heavier DC locomotives. These were usually run in threes at the head of the train, but after ECP braking became more common, one unit was moved back to the 2/3 point in the train.
The third major operator, last known as One Rail, has all ECP trains but uses three 134 tonne locomotives at the front of trains.
With only 37 units against 190, I don't think there would be the demand for 180 tonne units for use outside the Hunter. Aurizon have expanded their coal haulage fleet with more 134 tonne units, some moved from iron ore traffic in Western Australia, although this traffic has built up again more recently.
The eight axle locomotives would be more costly than either 134 tonne or 180 tonne six axle units, and not even Aurizon have purchased any 180 tonne locomotives since 2015, although they have continued to buy 134 tonne units.
The Progress Rail 1TB2622 is a narrow gauge motor, I believe for 42" gauge. The SD70MAC motor is the 1TB2630, which fully uses the space between standard gauge wheels.
Indeed, the GT46C ACe traction motor is the 1TB2630, not the 1TB2622 which is fitted to the earlier GT42CU AC units. I understand that the 1TB2622 was considered initially but the larger motor was actually fitted. My apologies for the error.
The GEB 30 was found in early trials in Australia to not provide the same performance as locomotives fitted with the 1TB2630 in wheelslip conditions with wet rails on a 1 in 40 (2.5%) grade. Later tests with revised wheelslip control software showed that the UGL built locomotives could match the Progress Rail locomotives.
However, Pacific National, one of the larger operators, ended up with 49 GT46C ACe units and 39 C44ACi units. Most of these are used in coal or crushed rock traffic.
M636C The Progress locomotives use the 1TB2622 motor which is a standard gauge motor. It was used on the SD70MAC in the USA. Does this help? Does anyone know if the GEB30 is sutable for narrow gauge? Peter
The Progress locomotives use the 1TB2622 motor which is a standard gauge motor. It was used on the SD70MAC in the USA.
Does this help? Does anyone know if the GEB30 is sutable for narrow gauge?
The Progress Rail 1TB2622 is a narrow gauge motor, I believe for 42" gauge. The SD70MAC motor is the 1TB2630, which fully uses the space between standard gauge wheels. The first two digits after the 1TB indicate the motor diameter, not in any particular units; the second two digits indicate the core length, again not in particular units but bigger number is longer core. The ACe motors, A3432, likewise, have numbering related to dimensions. The 1TBxxxx motors are the Siemens designation, the similar size current PR motors are A29xx.
I am quite certain based on the bogie design work I did for GE & MPI on the MBTA HSP46 locos that the GE GEB30 is a standard gauge motor similar to the GEB15 but using a smaller axle diameter.
Dave
In the case at interest though, the implied question is would the conceptual Australian B-B-B-B unit employ standard gauge or CM gauge motors?
The present locomotives in this category are the Wabtec/UGL C44 ACi and the Progress Rail GT46C-ACe
The Wabtec locomotives use the GEB 30 motor. I have no idea of its dimensions, so it may be a narrow gauge motor.
The problem is solely that the permissible axle load is 22 long tons. For some reason, when the country changed to the Metric system in January 1973, anlthough lengths changed to metres and mass to tonnes, apparenly kiloNewtons became too hard, so it is still 22 long tons.
This limits a locomotive with six axles to 132 tons if it to be allowed to run at 115km/h. To increase confusion, the mass is expressed as 134 tonnes.
This is only a problem because the 7FDL16 engine is about two tonnes heavier than the 16-710G3B.
So while a GT46C ACe can carry 10000 litres of fuel within the 132 ton limit, the C44ACi can only manage 7400 litres. The desired capacity is about 12500 litres, which the Cv40-9i could carry until its motors were upgraded to the 796A1 type, which reduced the capacity to 11500 litres.
This would allow the locomotive to run from Melbourne to Brisbane without refuelling.
PneudyneOne of the latter, not previously mentioned, was by Deutz for a 4000 hp, twin-engined diesel-hydraulic unit with axle loadings in the 23 to 30 tonne range.
That should have been Henschel, not Deutz, and 2 x 4000 hp, not 4000 hp.
Upthread, I said:
‘Four-truck interurban locomotives were built until at least 1941-42, e.g. Piedmont & Northern #5600 by GE,’
That was not correct. The GE-built unit was #5611, described in Railway Age (RA) 1942 April 04. It weighed 236 000 lb. It was said that its B-B+B-B wheel arrangement was dictated by the desire to duplicate existing traction motor equipment, to maintain individual axle loading at less than 35 000 lb, and to handle 1800 ton trailing loads with a single locomotive.
I also said:
‘Illinois Terminal used B-B-B-B, with independent span bolsters. In some of these cases then span bolsters may have been associated with lateral motion trucks.’
Illinois Terminal home-built 20 of its C class from 1924 onwards. The photographic evidence indicates that these had swing bolster trucks with secondary suspension. Five of the C class were rebuilt into the D class in 1940-42.
Also, Oregon Electric home-built five four truck locomotives in the 1940-44 period. The photographic evidence suggests that these were of the articulated span bolster B-B+B-B type. They were later sold to the Chicago, North Shore & Milwaukee.
Thus there were at least 29 span bolster electric locomotives used by interurban systems, 20 non-articulated B-B-B-B and nine articulated B-B+B-B. And the build period was at least 1924 through 1944.
Evidently some of the interurban operators were of the viewpoint that for the heavier haulage jobs, eight-axle locomotives would be more economical than a pair of four-axle locomotives in MU. The span-bolster running gear enabled this concept without any loss of curving capability whilst using the same trucks and motors as for the B-B types, and keeping axle loadings within acceptable boundaries.
Regarding the solitary EMD T model, in this case the eight axle form appears to have been chosen for axle loading reasons. It was actually lighter, at 342 000 lb, than the contemporary pair of C-C transfer locomotives of similar power that the IC acquired. They weighed in at 342 000 lb (GE/IR) and 346 000 lb (Busch-Sulzer/GE). Kirkland noted that the EMC T had an axle loading of 40 500lb, as compared to 57 000 lb and 57 700 lb for the two C-C units, enabling it to negotiate more lightly laid track. A reasonable inference that the IC had some lighter track in the envisaged operating area.
Thus the reasons for the choice of span-bolster running gear were quite diverse. The UP Streamliner, VGN EL2B, GTEL4500, GE U50 and Alco C855 cases have already been covered.
The CEM 4B case derived from a request by Oferom, then the authority looking after motive power, etc., for the French overseas (Outre Mer) railway systems. Oferom wanted a single-engined locomotive of above 3000 hp, but within a 16 tonne (roundly 35 000 lb) axle loading that would be more effective and efficient than the relatively low-powered B-B units that then made up the bulk of the Outre Mer fleet. The immediately preceding Alsthom CC2400 model was a stepping stone, but its short wheelbase monomoteur C truck had turned out to have poor tracking capability. (Why that was so I have never seen – perhaps it had something to do with its relatively high centre of mass.) But that may well have created an aversion to C trucks on Oferom’s part. The axle loading constraint anyway pointed to eight axles, and the curving requirements favoured B trucks, so that the span bolster arrangement was logical, although in this case executed somewhat differently. Then concomitantly the 3B arrangement provided a six-axle locomotive without using C trucks. It may be noted though that Alsthom had used the single-frame tribo, B-B-B running gear since 1939. Within the Outre Mer and associated orbit, such had been supplied to both Madagascar and Algeria, but not elsewhere in Africa. Perhaps Oferom did not favour this type of locomotive. Note that the CEM monomoteur trucks did not have the two-speed gearing often used in French domestic practice. (But the Alsthom CC2400 C trucks did have two-speed gearing.)
That list I think closed out the “historical” use of span bolsters, which could be summarized as follows:
US interurban: At least 20 B-B-B-B and nine B-B+B-B, from 1924 (if not earlier) through 1944.
EMC T: One only B-B+B-B, 1936.
UP Streamliner: Five articulated body power cars, one of 2100 hp and four of 2400 hp in 1936. (These had just one span bolster per unit.)
VGN EL2B: Eight B-B+B-B units forming four two-unit locomotives in 1948.
GTEL4500: 26 (including prototype) B-B-B-B 1949-1954.
N&W STEL: One only C-C-C-C, 1954.
GE U50: 26 B-B-B-B 1963-64 (23 for UP, using recycled running gear, and three for SP).
Alco C855: three B-B-B-B in 1964 for UP (using recycled running gear).
CEM 4B: 27 (CFM Madagascar one, 1969; CFCO 10 1969-77; 16 RFC Cameroun 1975)
CEM 3B: 22 for RAN, Abidjan-Niger, 1970-75. (These had just one span bolster per unit, and unlike the rest of the group, six rather than eight axles.)
Forward to 1991, and the GE BB40-8M model for EFVM, Brasil. Here it would appear that the primary reason for employing eight axles was to provide a sufficient number of narrow gauge traction motors to handle the 4000 hp power output; six motors were not enough. Reduced axle loading may have been a secondary factor. Of course, the same reason had applied previously in the case of the EMD DDM40 supplied to EFVM from 1970, although in this case D trucks were used. In the GE case, given its history, it was not surprising that it chose span-bolster running gear for the BB40-8M (and subsequent BB models.) The EFVM experience with D trucks might also have been a factor.
Then, and for the same traction motor count reason, came the EMD GBB truck, which might be described as a fully integrated span bolster unit, rather than a span bolster placed over more-or-less standard B trucks. The earlier CEM unit was somewhat integrated, but in that case the B trucks were configured also for independent use.
And turning to the “fellow-travellers” with eight axles, all powered:
As previously noted, the NYC T class electrics had four truck running gear, but not of the span bolster type. Apparently NYC’s starting position was that it wanted an improved version of its S3 class 2-D-2 design, including powered pilot trucks, and with those pilot trucks spaced further away from the rigid wheelbase in order to enhance tracking and riding at higher speeds. As a better way of addressing the needs, GE split the wheelbase into two halves, articulated together at the centre, and had the superstructure ride on each half. That would have allowed more freedom in pilot truck placement, as well as significantly shortening the rigid wheelbase. Whether anyone thought of the span bolster alternative at the time is unknown. But that would have been something of a sidestep, whereas the chosen running gear was more-or-less a lineal descendant of the 2-D-2 type, via a notional B-D-B, with the D part effectively split in two. Four sub-classes were built for a total of 36 in the 1913-1926 period.
In respect of its solitary GTEL prototype of 1950, Westinghouse gave its reason for its choice of non-span bolster four truck running gear, as follows:
‘This unconventional arrangement of running-gear has advantages from the standpoint of tracking, simplicity, ease of maintenance, and light weight. The tracking stability of a truck type locomotive at high speed is dependent largely on the center-pin spacing. In the case of this locomotive, the trucks with lateral restraint are at the ends, so that the effective center-pin spacing is large. If the more conventional span bolster arrangement were used to connect the trucks together in pairs, the center-pin spacing would be much smaller.’
When Westinghouse had previously proposed this kind of running gear, in both three- and four-truck forms, for use in a standard range of electric locomotives, the PRR had expressed concern about its ability to negotiate vertical curvature without significant weight transfer. Perhaps that is why when the PRR did order its electric prototypes (originally in AC form, later changed to the AC-DC rectifier type), the E2c C-C type was included as well as the E3b B-B-B type, the latter probably being Westinghouse’ preference. (As an aside, the E2c might have been the first US domestic locomotive to be fitted with lateral motion C trucks, in this case of the single swing-bolster trimount type, probably by GSC.)
As previously mentioned, the D-D wheel arrangement was an EMD specialty, although proposed by other builders. One of the latter, not previously mentioned, was by Deutz for a 4000 hp, twin-engined diesel-hydraulic unit with axle loadings in the 23 to 30 tonne range.
Cheers,
BDA It would be interesting to see the performance difference between 8 motor diesels in the US and export versions . By this I don't mean the double engine types like DD40X etc . More a comparison of a type with the same power assembly etc .
It would be interesting to see the performance difference between 8 motor diesels in the US and export versions .
By this I don't mean the double engine types like DD40X etc . More a comparison of a type with the same power assembly etc .
OvermodI am still amazed that a short little wheelbase like this improves high-speed stability (they describe easily running at 100mph). Presumably the very low polar moment made possible by centrally locating the motor facilitates it. If there is any discussion of how performance changes as the wheels wear, I didn't see it.
Having observed the simple and rapid gear change on the model bogie, I was a little surprised at the apparent complexity of the manual operation. However, after a little thought, I realised that the change had to occur on both trucks simultaneously, so the mechanism had to stretch the length of the locomotive, changing both pinions at the same time....
Amusingly the subtitles refer to "B-B" locomotives consistently as 'bebes'. I don't know if this is the same happy thing as 'deesse' Citroens... but I'll take it.
Amusingly -- I first came across this truck in the Ransome-Wallis Encyclopedia of World Railway Locomotives... where I learned that the gear ratio was adjustable by the engineman. I later 'learned' that the conversion could only be done in the shop, and believed that for many years. The rocker approach to a common central motor pinion -- combined with clear film describing the quick-change -- establishes pretty conclusively that my original understanding was right.
I am still amazed that a short little wheelbase like this improves high-speed stability (they describe easily running at 100mph). Presumably the very low polar moment made possible by centrally locating the motor facilitates it.
If there is any discussion of how performance changes as the wheels wear, I didn't see it.
Earlier in this thread Pneudyne introduced the French built B'B'B'B' locomotives with monomoteur bogies. Today I found a video giving very good detail of the BB16500 class and its monomoteur bogies. In particular, it shows the procss for manually changing the gear ratio between passenger and goods traffic.
La passion des trains - La grande famille des BB (n°36) - YouTube
The section of the video starts at 8min 17 sec and is entitled "Light Cavalry". Sadly the commentary is in French but there are closed captions available, also in French, but these should help anyone with a basic understanding of French.
However the use of a large scale model with a perspex gearcase and detailed video of the assembly of a bogie show clearly the principles and construction of the bogie. The assembly of the prototype three axle monomoteur bogie is shown at the end of this section of the video.
The rest of the video is also interesting but covers other aspects of B'B' locomotives, including an experimental locomotive with synchronous AC motors, something the French took more seriously than others.
M636C Thus we may deduce that GE opted to provide on eight axles a GTEL “equivalent” to a typical three-unit diesel-electric freight locomotive, which had twelve axles. As an aside, the capacity of the then relatively new GE761 traction motor facilitated this. I assume you mean the GE 752 motor, which indeed remained in production for new GE locomotives up until the ES 44DC and is probably still available. The GE 761 is a metre gauge motor, which while extremely successful and still in wide use would not be useful for USA Domestic situations. Peter
Thus we may deduce that GE opted to provide on eight axles a GTEL “equivalent” to a typical three-unit diesel-electric freight locomotive, which had twelve axles. As an aside, the capacity of the then relatively new GE761 traction motor facilitated this.
I assume you mean the GE 752 motor, which indeed remained in production for new GE locomotives up until the ES 44DC and is probably still available. The GE 761 is a metre gauge motor, which while extremely successful and still in wide use would not be useful for USA Domestic situations.
As previously mentioned, the CEM 4B design of 1969 had its couplers mounted on the frame ends. It was 62’8” long over couplers, 59’1” over end frames, with span bolster centres at 35’1”. Evidently the coupler overthrow was not excessive, even though the locomotive was intended to negotiate 50 metre (164 ft) radius curves. It was not though the longest CMT (Cape/metre/three foot) gauge single-frame locomotive extant. One candidate for that at the time would have been the 1959 Ghana Railways 1401 class C-C, Henschel/EMD model TT12, 64’4” over couplers with 35’6” truck centres.
The TT12 was a double ended verandah cab unit, and the rounded EMD noses would have reduced the clearance problem compared to the French locomotives with straight boxcab bodies.
The longest 1067mm gauge locomotive in Australia is the GT46CU-ACe which has been posted about in this forum. It is 23.7 metres over couplers (around 77 feet 9 inches). It probably is restricted to recently constructed heavy haul lines, or lines upgraded to that standard. It is of course more recent than the other units discussed here.
Returning to the span bolster case, we may look for GE’s reason for using that type of running gear for the GTEL4500.
Three paragraphs from AIEE paper 50-77 (1) are pertinent:
‘The Alco-GE gas-turbine electric locomotive is designed for freight service. It is an 8-axle 8-motor B-B-B-B locomotive, weighing 253 tons (average) and rated at 4,500 horsepower for traction (at 80 degrees Fahrenheit and 1,500-feet elevation). The locomotive is 83-feet 7½-inches long, and 14-feet 3-inches high over the roof sheets. It will negotiate curves of 288-feet radius.’
‘The general design, especially the power rating and weight of the locomotive, was developed from analyses of freight and passenger locomotives in use in this country in 1941 and 1946. These analyses indicated that about 75 per cent of the freight locomotives were in the 4,000-to-5,000-horsepower range and had approximately 500,000 pounds on drivers. The studies also showed that approximately 90 per cent of the total locomotives in use in road service handled freight. Consequently, it was decided to concentrate on designing a freight locomotive in this range of power and weight. To be attractive and profitable to the railroads, this locomotive must show a low initial investment, low fuel cost, and low maintenance cost.’
‘The trucks are of the conventional 2-axle swing-bolster type. Each pair of trucks is connected by a span bolster which, in addition to spanning the two trucks and supporting the locomotive cab, also acts as a traction-motor air duct. This arrangement gives a high degree of flexibility on curves. It also has sufficient vertical flexibility to give satisfactory riding qualities even on comparatively rough track. Resonant vibration between the truck equalizers and the swing bolsters was encountered at low speeds. Damping means were provided in the spring system and no further trouble has been encountered.’
Thus we may deduce that GE opted to provide on eight axles a GTEL “equivalent” to a typical three-unit diesel-electric freight locomotive, which had twelve axles. As an aside, the capacity of the then relatively new GE761 traction motor facilitated this. And with eight axles chosen, the four truck span-bolster arrangement presented itself as the optimum choice. Inter alia, it did allow the use of standard two-axle trucks from diesel-electric practice.
Fairly recently, GE had used the span bolster form, albeit articulated, for the VGN EL2B motor generator electric locomotive. This was designed at the same time as the GN W1 class. In both cases the starting point was the GE746 traction motor, from which the respective axle counts, and then the wheel arrangements were derived. From ASME paper 49-SA-7 (2):
‘A study of the service requirements and the performance of the existing motive power indicated that a locomotive carrying approximately a million pounds on drivers would be required. Since the top speed involved (50 mph) was moderate, it was decided that all weight would be on drivers and no special guiding axle would be required.’
‘The selection of an eight-axle running gear for each of the two cabs narrowed down to two designs, a B-D-B arrangement employing a main frame, integral with the cab structure, and two two-axle swivel truck, or a B-B+B-B arrangement, employing four identical two-axle swivel trucks.
‘The second arrangement was selected in the interests of easy maintenance, interchangeability. of trucks, and over-all superior flexibility of operation. The resulting locomotive nomenclature of this arrangement of running gear is 2 (B-B+B-B), and each of the two units composing the locomotive., for all practical purposes, a duplicate of the other.’
In this case the moderate top speed allowed the use of both articulation between the span bolsters (without any stability aids) and rigid bolster trucks. GE’s work on the EL2B case evidently informed its choice for the GTEL4500. In the latter case the locomotive length would anyway have made articulation of the span bolsters more difficult, but GE may have also decided that it was not appropriate.
Not covered in any of the references that I have seen is the choice to mount the couplers on the span bolster outer ends rather than the main frame ends. This might have been simply a dimensional issue in respect of coupler overthrow on curves. The span bolster centres were at 41’6” against an overall length of 83’7½”, so the former was not a large fraction of the latter. Coupler overthrow in a given curve is a function of both truck centres and overhang, increasing with both but also increasing with the overhang-to-truck centres ratio. (There is probably a standard formula for this.) So perhaps keeping the overthrow within bounds was the reason for GE’s choice. It may be noted that the later EMD DDA40X case showed that with a very long locomotive, the couplers could still be mounted on the frame ends provided that the overhang was not excessive.
But there might also have been a structural element here as well. The GTEL4500 main frame was effectively a more-or-less full-width and very deep box section that formed the fuel tank, upon which the superstructure and equipment was mounted. Possibly it was thought that it was preferable to have this structure carry the buff and drag forces only between the span bolster pivots, and not at the outer ends.
As a possible precedent, one may also look at the Illinois Central Busch-Sulzer prototype #9201 of 1936, whose mechanical parts were designed by GE, who said (3):
‘By taking advantage of the maximum bridge loading permitted by the railroad, it was possible to limit the number of axles to six and thereby design a simple running gear consisting of two three-axle, non-articulated swivel trucks upon which the cab is mounted.
‘The problem of holding the total weight within limits, permitting the use of six axles, was a serious one, without resorting to extensive use of special material in the cab. However, by putting the draft gear on the trucks, no part of the platform carries more than one-half of the drawbar pull; and by using a heavy centerplate, a suitable design was obtained.’
Perhaps the same thinking applied to the GTEL4500.
The running gear decisions made in respect of the GTEL4500 then informed the GE U50 and the Alco C855. Whether GE might have done differently in a “clean sheet” situation is unknown. But when it came to the BB40, it opted to mount the couplers on the frame ends, and not on the span bolsters. In part this may have been because the frame (derived from the C40?) was designed this way, and in part because the specific dimensions allowed it.
(1) AIEE 50-77 ‘The Alco-GE 4,500-Horsepower Gas-Turbine Electric Locomotive’ by A.H. Morey (GE), 1950 January
(2) ASME 49-SA-7 ‘’Motor-Generator Locomotives, Their Design and Operating Characteristics’ by J.C. Fox (VGN), J.F.N. Gaynor (GN) & F.D. Gowans (GE), 1949 June.
(3) Railway Mechanical Engineer 1963 September, p.383ff, ‘Busch-Sulzer 2,000-Hp. Switcher.’
As I understand it, when Victorian Railways (VR) ordered the Clyde G8, it was at the 43’0” standard length. (VR Newsletter 1954 October). Presumably it then agreed to the 44’6” length requested by QR for the G12.
The drawing I mentioned of a scaled down GP appears on page 9 of Diesel Railway Traction for January 1956, with an article reporting the delivery of the T class to Victorian Railways. The only dimensions are length, 43 feet over headstocks and bogie centres, 25 feet.
While those dimensions match the EMD standard G8, the loco illustrated is entirely different. Small platforms extend past the headstocks at each end with railings. The vee shaped ends allowed more room on the end walkways particularly just above the steps. The plating on the frame sides has seven louvred access panels, like those seen on SD7s. The fuel tanks and the battery boxes are arranged differently, and the roof grille over the radiators extends down the body sides requiring the hand rail to be lowered beside the radiators. The drawing shows the outline of full VR lining.
A photo of T324 appears in the article, and nobody appears to have noticed that it looks nothing like the drawing.
It is possible that the longer frame was offered by Clyde as providing at least one cross walkway, which the revised EMD design eliminated.
Interestingly, the shorter G8B locomotives reverted to the Vee shaped hood ends, but without walkways at either end .
As I understand it, when Victorian Railways (VR) ordered the Clyde G8, it was at the 43’0” standard length. (VR Newsletter 1954 October). Presumably it then agreed to the 44’6” length requested by QR for the G12. But then subsequent deliveries (of the G8B) were shortened to 40’8”. Clyde was first to build the GR12 model, for QR (1450 class). Its version had 12’6” wheelbase trucks and was 49’2” over end frames. The EMD version had 12’2” wheelbase trucks and was 47’4” long. If one goes back to the first QR 90-ton locomotives, the 1150 class from GE and the 1200 class from English Electric (EE), both had a 45’0” total wheelbase as well as 12’6” wheelbase trucks. (At the time, GE’s customary export C truck was of 10’7” wheelbase.) In both cases the total wheelbase was a larger fraction of overall length than was usual, and both locomotives were evidently longer than they needed to be to accommodate the requisite equipment. So perhaps QR had specified not only the 12’6” truck wheelbase, but also the 45’0” total wheelbase for bridge loading reasons, but soon relaxed the latter requirement? The information I have found for the Australian version of the Alco DL531 is that it had a longer frame, at 44’3” over headstocks as compared with 42’0” standard. The total wheelbase was the same, at 34’4”, as were the truck centres, at 28’0”. It had a longer truck wheelbase, though, at 12’0” rather than 11’6”. This was obtained by increasing the distance between the centre and inner truck axles, thus reducing the inter-truck space. Might that have been done to meet a bridge loading requirement from either NSWGR or SAR?
When the VR order for the G8 was announced, it was accompanied by drawings and artist's impressions of a locomotive that looked nothing like the EMD export unit actually built. It had small platforms at each end, shaped hood ends and a curved cab roof looking like a scaled down GP7. I think these images could be found in copies of Diesel Railway Traction, sometimes even after the locomotive was actually built. I will have to check the dimensions of that design some time....
An interesting feature of the G8 and G12 was that the frame was completely standard, even to the provision for traction motor blower ducts in the appropriate places for both types of bogie. So 1400 ran trials in NSW on standard gauge VR T class bogies, and with the QR buffers relocated to the standard gauge location. 1400 was built at Clyde's own expense as a demonstrator and was initially hired by QR until acceptance after extensive trials.
But I never regarded the units from T347 onward as G8s. They were more closely related to the later GL8, although 347 to 356 had the G8 radiators, horizontal in the hood roof. 357 onward had the vertical GL8 radiator indicated externally by the full height narrow radiator inlet. This shorter frame reduced the weight of the locomotive compared to 320 to 346 and eventually, allowed them on even lighter lines than originally permitted. The actual axleload was of course much higher than NSW allowed, since their light line units were all six axle.
While the 1450 class were initially referred to as Model GR12, Clyde later called them G12C. I think the frame between the bogie pivots was the same as the G12, but was lengthened beyond the bogie centres to allow for the longer bogies. This resulted in the 1450 having a shorter nose but longer engine hood than the EMD GR12. Someone, presumably at Clyde Engineering, worried about this enough to physically cut a hole in the builders plates to remove the engraved "GR12" leaving a hole showing the blue paint of the cabside.
I wasn't aware of any other similar units being built at the time the 1150 class appeared in 1952. In the QR Laboratories at the Ipswich Workshops they had quite a collection of failed CB FVBL-12 components and made unkind comments about GE providing them with completely untested prototypes to be developed at the expense of the Queensland Government. I felt this was a bit "over the top" but that was their opinion. But I'd believe that a 12'6" bogie wheelbase might have been in the specification.
I don't know who wanted the longer truck on the DL 531, but the SAR ran them on narrow gauge lines which certainly looked less secure than NSW branchlines. The Australian DL531s had small platforms at each end, which might have been the reason for the length increase. Fuel capacity was a problem for the NSWGR, and they moved the air reservoirs to a location in the short hood and had to put the batteries on the frame in front of the fireman's position. The SAR didn't need the extra fuel, which increased the weight, of course.
To return to the E626, the conventional subscript zero still applied to the two rigid axles since they are individually driven by motors and not coupled. The trucks appear to be pivoted at the inner end inside the inner axle, so they might be described as Bo+Bo+Bo, while the PRR P5B which always comes to mind would have been Bo-Co-Bo, since those were trucks with centre pins. But Bo+Bo+Bo still counts as TriBo to me....
While on the subject of TriBos, I had a cab ride on the Glenbrook Railway in New Zealand in a DJ class locomotive. It seemed to be somewhat uncertain, at least to me, when entering curves...
It wasn't just the G12s that were 18 inches longer... All the G8s and all the DL-531s were also exactly 18 inches longer.
Anyway they replaced the inline 6CSRKT with a 12 CSVT Mk II, which wasn't much longer, but of course the radiators needed to be bigger and the fuel tank had to be larger. The intention was to use the same trucks on the new 1300 class as the 60 ton unit but calculations showed that the Cooper bridge restrictions would be exceeded. The quick answer was to lengthen the "short" end of the bogie by 12 inches. Since the lighter units were still in production, a removable wooden plug was placed in the mould used for casting the frame. On the finished casting, two faint lines were visible at each end of the plug.
Thanks for that. I have long wondered how the QR 1300 class managed to escape the established QR 12’6” truck wheelbase requirement for 90-ton locomotives. Presumably at 11’10½” it just made the Cooper requirement.
To go back to an earlier post regarding Italian Tri-Bo locomotives, the pre WWII units of class E626 were rigid body units with the central two axles in a rigid frame and the outer axles in trucks. The articulated bodies started with the post WWII E636 and continued through four further classes until superseded by rigid body B-B-B units with monomotor trucks.
It might be something of a stretch to describe the E626 as a tribo, given that the two centre axles were in the mainframe, and not in a truck that was allowed rotational and/or lateral motion. More-or-less it could be viewed as a 2-B-2 type with powered pilot trucks. (Maybe analogous to the solitary PRR P5 rebuild as a B-C-B type?) At the time, FS was thinking in terms of rigid-frame locomotives. As well as the E626, there was a 2-C-2 type, and a 2-D-2 was planned, but abandoned when the 2-C-2 was found to be not too brilliant in the riding and tracking department. In place of the planned 2-D-2 FS chose a 2-B+B-2 type.
As to centralisers, The NSW 46 class were built by Metropolitan Vickers -Beyer Peacock in a factory built for diesel and electric locomotives (in Stockton on Tees, if I recall correctly). Metropolitan Vickers had articulation links on locomotives supplied to South Africa and to Japan. In fact M-V built the first of the 2-C+C-2 locomotives in Japan.
Certainly M-V had supplied B+B and C+C locomotives to SAR. These had conventional single point articulations through which passed all buff and drag forces, and as I understand it, had the usual riding and tracking problems at higher speeds associated with this form of construction. These problems could be addressed by the use of pilot trucks and stability devices, but the use of independent, lateral motion trucks (with or without centralizers as appropriate to situation) proved to be a better option, and became the norm in the post-WWII era. The NSWGR 46 was an unusual case. It was very late for a new design with articulated trucks, but in using a bar-type articulation (which avoided a “hard” connection between the two trucks) in conjunction with the newer centralizer, M-V might have been looking to offset some of the disadvantages of articulation without discarding it altogether. Nonetheless, I have heard anecdotally that the 46 was somewhat hard on the track. Re the Japan 2-C+C-2 locomotives, I think that the first lot was supplied by English Electric (with mechanical parts by NBL), rather than M-V. The rest were all built by various makers in Japan.
One of the high points of my 1977 visit to the USA was to see an SP U50 leading a DD35 on a transfer run in the Los Angeles area.
That was an interesting sighting– both types of eight-axle running gear in the same consist! Although it pales in comparison, I have seen the EMD DDA40X that was on static display at Dallas. If nothing else, that provided the opportunity for a close look at the running gear. I recall that the bolsters seemed to be quite small in relationship to the trucks themselves. I think that the UP also did some mixing and matching of its eight-axle units in MU consists. Whether that ever resulted in a U50, C855 and DD35 appearing together is unknown.
bogie_engineer M636C Regarding South Africa, the truck interconnection was not exclusive to GE. The diagrams of the EMD GT26C units in South Africa also show some form of interconnection under the fuel tank. If my HO scale model of an SAR GT26C is accurate, these have diagonal links between the opposite inner sides of the trucks. But the shallower fuel tanks must be lower in capacity. The GT26MC's had what we called at EMD "inter-bogie control". Long triangular frames that had vertical pivots on the bogie end transom met under the fuel tank and were connected laterally with a spring assembly that required a difficult adjustment so it didn't adversely affect tangent track running. I believe, but am not certain because it was before my time in the bogie design group, that this was added at the insistence of Dr. Scheffel at SAR. EMD was not a proponent of it but the customer is always right. It adversely affects curve entry and exit lateral wheel forces while improving them in the body of the curve. It also hinders bogie rotation thru crossovers. Dave
M636C Regarding South Africa, the truck interconnection was not exclusive to GE. The diagrams of the EMD GT26C units in South Africa also show some form of interconnection under the fuel tank. If my HO scale model of an SAR GT26C is accurate, these have diagonal links between the opposite inner sides of the trucks. But the shallower fuel tanks must be lower in capacity.
Regarding South Africa, the truck interconnection was not exclusive to GE. The diagrams of the EMD GT26C units in South Africa also show some form of interconnection under the fuel tank. If my HO scale model of an SAR GT26C is accurate, these have diagonal links between the opposite inner sides of the trucks. But the shallower fuel tanks must be lower in capacity.
The GT26MC's had what we called at EMD "inter-bogie control". Long triangular frames that had vertical pivots on the bogie end transom met under the fuel tank and were connected laterally with a spring assembly that required a difficult adjustment so it didn't adversely affect tangent track running. I believe, but am not certain because it was before my time in the bogie design group, that this was added at the insistence of Dr. Scheffel at SAR. EMD was not a proponent of it but the customer is always right. It adversely affects curve entry and exit lateral wheel forces while improving them in the body of the curve. It also hinders bogie rotation thru crossovers.
Dave,
Thanks for the explanation. The system you describe is similar in principle to that fitted to the Commonwealth Railways NT class locomotives, more correctly to the first three units. These were built by Tulloch Limited in Sydney to a Metro Cammell design and used one of the earliest bogie designs using four rubber-metal pads to take the load directly to the frame. The axle loads were arranged so that the centre axle had the heaviest axle load (about 12 tons) while the outer axles carried less than 10.5 tons in order to reduce lateral forces in curves. The roughly triangular links were attached at bogie frame level on the inner ends and connected to some form of rubber link at the centre. This resulted in the fuel tank looking like an inverted letter "U". The bogies also had Alsthom link axle location.
On the really poor track on the Central Australian and North Australian Railways, these units rode badly, and the remainder of the class came without the links and with conventional axle guides. They also had wider carbodies, the first units being built to a design intended for export to multiple countries, while the earlier Sulzer units from BRCW were around a foot wider. The other oddity was that the original design was double ended, but the Commonwealth units only had the cab at the No 2 end next to the radiators. I guess that gave a marginal weight reduction. By the time I saw these units, in 1975, the interconnection had been removed and the early locomotives had bogies with conventional axle guides, although I found a pair of the original bogies at the back of the Darwin workshop. The only locomotive of this type preserved is the last one built NT75, which never had the interconnection. The interconnection arrangement was illustrated in Dunlop Metalastik advertising, often on the back page of British railway engineering magazines. These appeared alternately with illustrations of the contemporary MLW truck design for the C630M, which used similar secondary suspension.
I'm not surprised that Dr Herbert Scheffel was involved. He was responsible for freight trucks that had diagonal links between the opposite axleboxes of the two axles. He was a German resident in South Africa. I met him at the Heavy Haul Railway Conference in Perth Western Australia in November 1978.
My presentation at that conference included some recordings of one of his freight bogies under an ore car on the Mt Newman line. This had special wheel profiles which were supposed to improve curving based on his South African experience. As is often the case, his profiles designed for extremely sharp curves on a narrow gauge line didn't match the profiles on the relatively straight Mt Newman line with much more heavily laden wagons. The result was a form of hunting on tangent track, due to the mismatch of wheel and rail profiles. I don't know if Mt Newman ever got the Scheffel bogie to work, but they never used any in normal traffic.
But you can understand why I thought the GT26MC links could be diagonal links, rather than the Metro Cammell system...
In Switzerland, the centralizer was much used, and was also developed for tribo locomotives, starting with the Rhaetian Railway Ge6/6 II class. English Electric adopted the centralizer in 1951 as one approach to combatting excessive flange wear on then-new C-C electric locomotives in Brasil.
Some references on centralizers are:
ILE paper #484 of 1949 January, ‘The latest development of the Electric Locomotive in Switzerland - Its Mechanism and some Problems, Dr. Gaston Borgeaud (SLM).
ILE paper #603 of 1959 December, ‘Methods of Reducing Flangewear on Diesel and Electric bogie Locomotives, by W.L. Topham (Vulcan Foundry/English Electric).
Mathematical Gazette 1964 October, ‘The Theory of the Centralizer or Transverse Coupling in Electric Locomotives’, by R. B. M. Jenkins, pp.296-304.
UK patent 611237 (SLM)
UK patent 742129 (English Electric)
US patent 3054361 (GE)
US patent 2994284 (GSC, mentioned in connection with the 1-C truck))
Wow, there's quite a range of points there....
I'll try to address them, if not in the order they appear.
I'm not sure that the main line locomotives can be lengthened. For the last thirty years, all Australian main line locomotives have been 22 metres long. This may be a coincidence but I don't think so. There are standards out there but I'm not sure that I have access to them. The only units longer than 22m are the new Progress units for use in Queensland on 1067mm gauge. Also height restrictions are more severe than even the former New York Central clearances. Australian locomotive trucks are generally smaller and lighter than domestic USA trucks. The few 180 tonne units have trucks that look like USA domestic trucks but I don't know if the height is the same.
It wasn't just the G12s that were 18 inches longer... All the G8s and all the DL-531s were also exactly 18 inches longer. But as I point out above, we may have hit the limit on length.
Queensland were very serious in meeting the Cooper loadings on their bridges. Over our summer of 1970-71 I was able to talk my way into assembling locomotives at the English Electric plant in Rocklea, a suburb of Brisbane. I was given a week in the design office which was really great fun. EE wanted to build a new "90 ton" locomotive (the metric system was adopted in January 1973) for Queensland, and decided to modify an existing design of "60 ton" locomotive (QR 1620 class). The 60 tons had been relaxed to 61.5 tons, 10.25 tons per axle, although the EMD units actually met the nominal weight. Anyway they replaced the inline 6CSRKT with a 12 CSVT Mk II, which wasn't much longer, but of course the radiators needed to be bigger and the fuel tank had to be larger. The intention was to use the same trucks on the new 1300 class as the 60 ton unit but calculations showed that the Cooper bridge restrictions would be exceeded. The quick answer was to lengthen the "short" end of the bogie by 12 inches. Since the lighter units were still in production, a removable wooden plug was placed in the mould used for casting the frame. On the finished casting, two faint lines were visible at each end of the plug.
I was aware of the SP U50s and I didn't expect that they had second hand UP trucks. One of the high points of my 1977 visit to the USA was to see an SP U50 leading a DD35 on a transfer run in the Los Angeles area.
‘The + sign is used only when the buffing and drawgear are fitted on the bogies and when the buffing and drag stresses are transmitted directly bogie to bogie and not up to the cab under-frames at the pivots. Those subsidiary articulation systems between bogies which are for guiding purposes only are not considered.’
The subsidiary articulation systems comment almost certainly referred to devices such as the centralizer, originally developed by SLM, but very quickly taken up by English Electric. GE, which had a history of developing and patenting various stability devices, most notably in connection with 2-C+C-2 running gear, also developed and patented its own version of the centralizer, initially for the South African Railways (SAR) 32 class (U18C1 model) with 1-C-C-1 running gear (which was equipped with GSC trucks designed to be used with a centralizer), but later for C-C locomotives (U20C et seq) for SAR and other railways.
An interesting case of wheel arrangement mis-designation was that of the New Zealand Railways (NZR) Df class diesel-electric of 1954, which initially and until the early 1960s was listed as 2-Co+Co-2. But an early 1960s article in the New Zealand Railway Observer magazine explained that it was in fact 2-Co-Co-2. There was no articulation joint between the bogies, the buff and drag forces passing through the main frame between the main truck pivots. But there was an English Electric centralizer (lateral coupling) device between the inner ends of the main trucks; this did not transmit any longitudinal forces. Thus by happenstance I learned about the distinction quite early on. I suppose that given that the couplers were mounted on the outer ends of the main trucks, and there was a visible interconnecting device, assuming that it was a full articulation joint was an easy mistake to make. Later I learned that 2-Co-Co-2 was an extremely rare wheel arrangement, with a worldwide total of 12, namely the 10 in New Zealand and the pair of GE steam turbine electric protypes tested on the UP. The latter had their stability devices set up for cab-leading operation, hence the nose-to-back orientation when used in pairs. On the other hand, 2-Co+Co-2 was more common, with an estimated worldwide total of 495, all being electric, of which Japanese National Railways had the first (1925), the most (237), and the last (1958), although probably not the most famous.
An unusual case was that of the New South Wales 46 class electric locomotives. These definitely had Co+Co (articulated) running gear, but with a bar-type rather than a single-point truck interconnection. They also had what looks like (from the photographic evidence) a centralizer.
Re the ‘competition’ between running gear and fuel tanks in respect of longitudinal space, that evidently came up with the SAR 32 class (GE U18C1), which had longer 1-C trucks in place of the usual C trucks of the U18C. In that case it was addressed by placing an auxiliary tank in the bottom of the body space normally reserved for the optional steam heating boiler, not required by SAR. (The upper part of that space was occupied by the inertial filtration equipment, an extra that SAR required.)
With the extra weight that eight axles could carry for a given axle loading, simply extending the locomotive length as required to fit the desired size of fuel tank might be another possibility. There is some history of that kind of frame extension in Australian practice. For example, I understand that Clyde stretched the G12 model from 43’0” to 44’6” over end frames specifically for Queensland Railways (QR) to allow the fitting of a suitably large fuel tank. Much later, something similar seems to have happened with the QR 2600 class, GE U22C model. Notwithstanding its designation, this was derived from the light-frame version of the U26C, not the baseline U22C (which was simply an uprated U20C). QR required 12’6” wheel base trucks (I think for bridge loading reasons) instead of the standard 10’5½”. Accommodating these without any loss (in fact with a small increase) in intertruck (i.e. fuel tank) space required that frame be lengthened from 55’6” to 60’0” (and with no increase in total weight in this case).
I suppose though the question would be whether the putative haulage capacity gains of the eight axle span bolster running gear would offset the extra first and ongoing costs. And even if the numbers stacked up, there is also the perceptual issue that tends to work against anything that looks to be more complicated, complex and difficult, even if it works well in practice.
I guess UP finally realised that with the LA & SF 4-5-6 sets, they were dealing with full size trains, not dedicated articulated lightweight trains. All the E3s and E6s had steam heating and could be used on any passenger train. While the fuel consumption increased, it was still far less than would be used by oil burning steam power, and they could stop maintaining the power vans and convert them to baggage cars. The overall simplification might have saved money in 1941 terms.
In Australia, the broad gauge states went with axle driven air conditioning, there was no steam heat used in Australia in the early 1950s, but the standard gauge operations decided on power vans as the source of power. This situation changed in 1962 when trains began to run from Sydney to Melbourne without change at the border, and the cars converted from broad gauge were fitted for HEP. By 1970, the Melbourne Adelaide broad gauge "Overland" was converted to HEP. In the early 1950s it gained the nickname of "Overdue" since two 1500HP EMDs lost time dragging the heavy train with axle driven generators up the steep grades at each end of the line. When 1800HP units arrived around 1960, things improved but by 1970 they decided they had to make the change.
In Australia the various governments are encouraging homeowners to change from gas heating to electric, although there are relatively few locations where real heating is needed even in Winter. Oil heating died out quickly due to the oil price increases in the 1970s.
M636C I should have credited the Kratville "Steamliners" book as my source for the span bolsters on the later M10000 series locomotives, although I think the feature was mentioned in most references. However, the photos in the Kratville book made things quite clear. That book is an excellent source for most things related to the UP Streamliners, although the change from head end power in dedicated power cars to steam heating and axle generators, a retrograde step in nearly every way, is only indicated by a careful reading of the actual vehicle diagrams. This occurred between the LA 1-2-3 and SF 1-2-3 sets and the following 4-5-6 sets (so before WWII).
I should have credited the Kratville "Steamliners" book as my source for the span bolsters on the later M10000 series locomotives, although I think the feature was mentioned in most references. However, the photos in the Kratville book made things quite clear. That book is an excellent source for most things related to the UP Streamliners, although the change from head end power in dedicated power cars to steam heating and axle generators, a retrograde step in nearly every way, is only indicated by a careful reading of the actual vehicle diagrams. This occurred between the LA 1-2-3 and SF 1-2-3 sets and the following 4-5-6 sets (so before WWII).
I've also found Kratville's "Streamliners" book to be very informative, picked up my copy in spring of 1977 and didn't fully realize how detailed the book was until getting White's book(s) on American RR pasenger cars.
From my reading, the motive to go back to steam heat and axle generators was to allow interchange with UP's existing passenger car fleet (and the fleets of the C&NW and Espee). What amazed me about the head end ower was that using electric resistance heat from the diesel generators used less fuel than steam heat despite the inefficient conversion of diesel fuel to electric power. I'd also expect that the head end power used less fuel than what was needed to overcome the drag of axle driven generators.
Our community is FREE to join. To participate you must either login or register for an account.