Alco diesel hydraulic

Pneudyne wrote: ↑Tue May 29, 2018 10:17 pm The DRT 1955 June issue included a summary of a paper or presentation by W.B. Gibson of Twin Disc, which in part addressed the desiderata outlined by J.S. Newton of Baldwin.

DRT 195506 p.191.jpgDRT 195506 p.192.jpg

The author did not seem to be all that keen on the idea of diesel-hydraulic/diesel-electric mixed MU operations, nor was he enthusiastic about the prospects for hydrodynamic braking that was proximate in performance to electrodynamic braking. Yet both turned out to be quite doable.

By the way, this article also contained information on the EMD DH2 prototype additional to that provided elsewhere.

In hindsight, the resistance in some quarters to mixed DH/DE MU operations is difficult to understand. By the mid-1950s, the idea that different makes and models of diesel-electric locomotives could interwork was well established in the USA to the extent of being a norm. Whilst the basic compatibility of EMD and GE power controls was largely happenstance, one could say that Baldwin took the initiative in offering the “compatible” WEMCO XM-781 master controller as an option to its standard pneumatic throttle control, I think from the production start of its road locomotives in the late 1940s. Reconciling the different EMD field loop and GE potential wire dynamic brake controls was done by Fairbanks Morse c.1954. Milwaukee developed its system by which DC electric locomotives could control trailing diesels in 1956, and UP did the same with GTEL-diesel combinations in 1958, although these were one-way systems. Against that background, by 1959-60 DH/DE interworking should surely have looked like a soluble issue simply requiring some detail development work rather than being an uncrossable barrier. Furthermore, GE, with its U25B, had shown that two-way interworking between standard 8-notch and 16-notch control systems was possible, pertinent to the DH/DE case in that by the late 1950s it had been realized that DH locomotives benefitted from multi-notch control, and so were likely to have a higher notch count than standard American DE locomotives.

Still, that history shows that some of the MU initiatives came from the railroad side rather than from the locomotive builders. So it was not so surprising that when the DH locomotive builders ostensibly said, “it can’t be done”, DRGW and SP in 1963 developed their own solutions, and interface unit in the first case and retrofitting with DE-compatible controls (derived from GE’s 16-notch work) in the second case. As a sidebar item, the DRGW interface, which one assumes provided two-way conversion between stepped electric and pneumatic protocols, probably could also have been adapted for use between standard DE and Baldwin air throttle DE locomotives.

The DE/DH MU question thus had been resolved just before Alco built the DH643 fleet for SP.


Cheers,

The earliest use of electric throttle on Baldwin road locomotives was at the end of 1949, and the earliest locomotives I can prove had it were Erie 1100-1105, model DRS-4-4-1500 built November and December 1949. Operating manual for these is Baldwin manual no. DRS-101, publication date 1-30-50, of which an example is in my collection.

Will Davis

Typewriters wrote: ↑Wed Feb 15, 2023 11:06 am POST EDITED...

A double check of official ALCO records by a friend shows that the controller for the DH643 was in fact the KC102. This later would become the "two lever" controller introduced with the GE U33 series.

Interestingly, the operator manual for the DH643 is clear that only eight engine speeds were trainlined to trailing diesels. This surely tells us that AV/BV/CV/DV were trainlined and there was no SN wire to pick up GE 16 notch control. If the DH643 was trailing diesel locomotives, the DH643 would respond only to eight notches (and radiator spray was inoperative as it would have no control from the diesel in the lead.)

Will Davis

I imagine that the 16-notch version of the KC102 master controller would have been functionally similar to the KC99, at least when configured for use by GE itself. But the Alco version might well have been different.

It would appear then that whatever Alco did (and that is unknown) to obtain the intermediate notches on its 16-notch throttle control was strictly internal, and not trainlined.

Thus, with a DH643 pair, the trailing unit would have operated in eight-notch mode. That seems odd at first glance. But a possible rationale might have been that whilst the leading unit, which was more prone to slipping, benefitted from finer control, that was not so necessary for the trailing unit. Also, for a trailing DH643, that situation was no different to what happened when it was behind a leading diesel-electric locomotive.

Also implied is that DH643 MU compatibility with the SP Krauss-Maffei ML4000 diesel-hydraulics was at the AAR eight-notch level only for the trailing locomotive(s), not at the multinotch level.


Cheers,

A friend of mine double checked and the controller is in fact the KC102. The DH643 also used FIVE solenoids on the governor, not only AV / BV / CV / DV but also the override solenoid (no load regulator so no override here) in an arrangement wherein the governor was reconfigured to use the ORS as a half-step in speed. The DH643 did the same thing as the Krauss-Maffei second series in that the first notch of the throttle was dead / not used (this would be notch 1/2 if a GE) and the second notch filled the converters. (No half fill position as on the first series of Krauss-Maffei units.) From then on up the engine speed increased by throttle position.

Thanks – that is most interesting! It leads to some further deductions and speculation.

The Woodward four solenoid engine speed and shutdown control made its debut around 1945 in the “New SI” governor, and was the used in the successor PG model which was released in the late 1940s, and then became the de facto worldwide standard locomotive governor in the pre-electronic era. The speed control mechanism was the subject of US patent 2496284 of 1950 February 07, filed 1945 May 03.


The mechanism was actually capable of providing up to 15 engine speeds using four solenoids, although the modal method of use was with eight speeds according to what became the AAR sequence.

But the 15-speed solenoid sequence was not easily compatible with the AAR sequence, which may explain why Alco chose a customised version of the PG which used a fifth speed control solenoid.

If say the fifth speed solenoid were configured to subtract a half-step in speed when active, then it could be used with the AAR sequence to obtain 15 speeds. Thus the even notches 2,4,6,8,10,12,14,16 on the DH643 would have used the standard eight-notch sequence, with the fifth solenoid inactive. Then the odd notches 3,5,7,9,11,13,15 would use the same respective four solenoid patterns as for the next-above even notches, with the fifth solenoid active to provide the half notch drop in speed. Just one extra trainwire from the master controller would be required to activate the fifth solenoid; evidently Alco chose not to trainline this in the MU bundle.

This spreadsheet shows the differences:




The standard Woodward arrangement, whereas the A, B and C solenoids acted in a summed positive sense on the hydraulic speeder valve, the D solenoid acted in a negative way by moving the speeder valve collar downwards. Speculatively, in the Alco DH643 governor, possibly the ORS solenoid also acted on speed valve collar – maybe hydraulically rather than mechanically – in a way that moved it downwards by half a step. Such motion would need to have been additive to that of the D solenoid. The maximum number of engine speeds was probably still 15, as a half-notch increment below the lowest speed shown above, i.e. with the ORS solenoid only, would probably result in shutdown. That might be why Alco sued 15, rather than 16 active motoring notches. But 15 was also the number that the German industry landed upon after its initial experience with 6 and 7 notch controls. Of course. that is just one unconfirmed possibility; probably there were other ways of utilizing the fifth solenoid.

Also, it would have been possible for Alco to have used the Woodward 15-notch sequence at the governor, but its own sequence from the master controller, with a sequence converter relay matrix in between. But that may have been seen as a more complex pathway, particularly when the ORS solenoid was there and free for use anyway.

The standard Woodward PG arrangement is shown here, with the ORS solenoid doing its primary job of shutting down normal load regulator action.


The Alco DH643 is the only example of which I am aware in which the Woodward PG governor with electro-hydraulic speed control was used with the maximum of 15 engine speeds. Some other variations that are known to have been used included six speeds, using A, B and C, D for shutdown only, by English Electric; eight speeds, but using A.B and C only, D for shutdown only, by Brush; 10 speeds, using all four solenoids, by Mitsubishi; and 14 speeds, using all four solenoids, by Hitachi and Mitsubishi.

Customizing of the PG governor to suit individual engine and locomotive builder requirements appeared to have been the norm for Woodward, so the Alco DH643 case was probably not exceptional in general, although it was probably unusual.



Cheers,

Another possibility for the Alco DH643 governor solenoid sequence was that the redeployed ORS was configured to add, rather than subtract, a half-increment of engine speed, and used thus for notches (DH643 throttle handle numbers) 3 through 15, each then being a half-speed increment above notches 2 through 14 respectively. With this arrangement, trailing locomotives would reach notch 8 only when the DH643 reached notch 16.

I have added this possibility to the spreadsheet.



Of course, I am curious as to how the ORS was reconfigured to become part of the governor speed control subsystem, but to ascertain that, it would probably be necessary to find a copy of the Woodward manual for that specific variant.


Cheers,

It is puzzling as to why Alco ostensibly did not trainline the ORS wire to allow a trailing DH643 to operate in 15-notch mode. Presumably this could have been done by using a spare within the MU bundle (clearly not the same spare that GE used for the SN function), which would not interfere with standard diesel-electric locomotive operation. (Exactly which MU trainwire GE used for the SN function remains stubbornly hidden, not mentioned in any of the GE data on hand.)

Based upon the information and assumptions to hand, I mapped out the likely master controller throttle handle outputs for both the Alco DH643 and GE 16-notch cases:



It seems at least possible that in the Alco DH643 case, there had been some overall coordination with what was done for the production K-M units. SP had ordered both these and the Alco DH643s in 1962 October. Presumably control systems and MU compatibility were worked out at the time or soon after, and the choice of the GE KC99 master controller for the K-Ms likely informed the retrofitting choice for the three K-M prototypes. Possibly Alco’s choice of the KC102 master controller came later, perhaps because it was more easily adapted to the “compact” driving position of the DH643.

If the master controller throttle handle outputs were the same for the Alco and K-M cases, then for the latter, the six trainwire outputs (GE, AV, BV, CV, DV, ORS) might have been converted to stepped air pressures for the Maybach governors. For example, the Westinghouse four-piston linear actuator driving a self-lapping air valve could have achieved this.


Cheers,