• BR Class 28 query

  • General discussion about locomotives, rolling stock, and equipment
General discussion about locomotives, rolling stock, and equipment

Moderator: John_Perkowski

  by Allen Hazen
My wife just bought a collection of "Thomas the Tank Engine" stories for some children we know, leading to a dinner-table conversation about BoCo.
One of the bits of history that gets repeated about the Class 28 (five axle: Co-Bo) locomotives is that they were used in pairs of a particular fast freight run. Does anybody know: when so used, were they
--coupled B end to B end
--coupled C end to C end
--operated elephant style
--randomly coupled?

(I have found one photo of a pair on a passenger train: coupled at their B ends. But I don't know if this was standard practice. Given the bits of scuttlebutt about various 4-axle and 6-axle locomotive types, I would ***guess*** that the engineman's ride would have been more comfortable with the C end leading, but I don't ***know*** this for certain, and I don't know how much attention BR paid to locomotive drivers' comfort!)


Their engines -- two-stroke Crossley V-8s -- were problematic. (The locomotive builder, Metro-Vick, used the same engines on the Class A diesels, 01-061, for the Irish railways: these locomotives were re-engined with EMD diesels. Apparently some thought was given to re-engining BR's Class 28 ... leading to interesting alternative history speculations!) The engine type had been suggested by Bulleid: I assume he, as motive power chief of CIE, made the suggestion to Metro-Vick. Why, I don't know, but a speculative hypothesis would be that, as an innovative railroad mechanical engineer, he was aware of what was going on overseas, and had noticed the successful use of two-stroke diesels in locomotives in the U.S. EMD might not, at that stage, have been keen on the idea of providing engines (or engine blueprints) to other builders to use with non-EMD electrical equipment, so an alternative two-stroke was sought. The Crossley engine ran (full speed) at 625 rpm, significantly more slowly than the EMD 567, so I assume it had correspondingly bigger cylinders. (When the Irish A class were re-engined, their 8-cylinder Crossleys were replaced with 12-cylinder EMDs.). My guess is that they weren't very similar to the EMD engine in detailed design.
  by JayBee
Look on the Wikipedia page under British Rail Class 28. The fast freight train was nicknamed the "Condor" and operated between London and Glasgow. There is a little bit more
information in the book "Freightliner: Life and Times", but not a lot. I believe the locomotives normally operated with the end supported by the Bo truck coupled together from the few photographs I have seen.
  by Allen Hazen
Thanks for reply! The photo on the Wikipedia page (two units on a passenger train) is the one I had seen, and, alas, I don't have (or no how to get access to) the Freightliner book. But: we know (from the photo) that they were at least sometimes coupled at the Bo ends, confirmed by other photos you have seen, and I ***think*** this would give a better ride for the engineman. So, unless someone tells me otherwise, I'm happy assuming that this was the usual configuration.
British Rail was big enough that it didn't mind scrapping a a few minority-type units. CIE was a lot smaller, perhaps even less well supported financially than BR, and their A class were the majority of their larger diesel locomotive fleet. So it's maybe not surprising that CIE did and BR did not choose to re-engine their Crossley-engined locomotives. But alternate history speculation is fun: some BR modeller please post a photo of an HO/OO Class 28 ... with roof modified to show a pair of EMD stacks! (Grin!)
  by Pneudyne
There is a book on the BR class 28 that might answer your question, although I haven’t seen it.

Amazon.com: The Metropolitan-Vickers Type 2 Co-Bo Diesel-Electric Locomotives: From Design to Destruction (Locomotive Portfolios) (9781526742810): Sayer, Anthony P: Books

I don’t recall seeing anything in what I have read elsewhere that spoke to the relative ride qualities at the “Bo” and “Co” ends.

The bogies (trucks) were of the Commonwealth cast steel swing bolster type, but without any secondary springing. That was an arrangement that MetroVick had used on the CIE A and C classes, and which Birmingham Railway Carriage & Wagon (BRCW) had used on the CR (Australia) NSU and CIE B classes, evidently without problems. But Brush’s use of that form on the Rhodesian DE4 class (with a straddle-type bolster à la English Electric, but who did use secondary springs) evidently was one cause of bogie frame cracking. I’d guess though that at high speeds, the BR 28 might not have ridden as well as the Brush class 30/31, which also had the Commonwealth swing bolster bogies with secondary springs.

Apparently BR did consider re-engining the 28 with the EE 12SVT engine, as it did for the class 30. But the fewness of the 28 fleet made it not worthwhile. In Australia, WAGR persevered with its X class, and eventually got the Crossley engine to work tolerably well.

Apparently BR was concerned about the likely performance of the Crossley engine even before it placed the order for the class 28. The rationale for that order was BR though that it should try one locomotive type fitted with a medium-speed two-stroke engine. The “Type 2” (originally Type B) BR power range into which it fitted was in some ways BR’s “sandbox”, including oddities as well as locomotives expected to become part of the mainstream fleet. Thus the class 23 “Baby Deltic” – really a silly idea in that power class - was ordered because BR thought it should try out the Deltic powerplant, although at the time it didn’t want to buy a replica of the EE Deltic prototype. Thus EE then had to tractionize the 9-cylinder Deltic engine, probably reluctantly so. (The production fleet of the big Deltics, class 55 came about from a business case submitted by the Eastern Region, who, denied electrification, wanted another way of significantly accelerating its passenger services, and showed that as costly as the Deltic would be to operate, a small, intensely maintained fleet doing around 200 000 miles a year each on one route on premium passenger services would be profitable. And there was nothing else even remotely close in the diesel field at the time in terms of power-to-weight ratio that would also meet the 16.5 long ton axle loading limit. The Deltic prototype was the result of “technology push”, but the production fleet was the outcome of “market pull”.) Another “oddity” was the class 21, which was ordered simply because BR thought it might be a good idea, for comparison purposes with the similar class 22 diesel-hydraulic, already ordered.

  by Pneudyne
Allen Hazen wrote: Tue Dec 22, 2020 12:25 am My guess is that they weren't very similar to the EMD engine in detailed design.
Very different. To start with, the Crossley engine was loop-scavenged, whereas the EMD was uniflow....

from DRT 195510 p.295.jpg

You do not have the required permissions to view the files attached to this post.
  by Pneudyne
Also see this site:


If you go to the “Service History” section, there is a photograph of a pair of the Co-Bo locomotives hauling a pre-service Condor test train. In this case they are coupled No. 1 (Co) end to No. 1 end, thus with a No.2 end (Bo) leading.

Going to the “Technical Description” section and then to the “Locomotive Description” page will find a picture of the bogies. Noticeable is that the Bo bogie has a lower pivot than the Co bogie. The latter has the customary double bolster that sits above the centre traction motor, and so of necessity has a highish pivot. The Bo bogie has the usual drop-centre bolster; evidently no effort was made to match the pivot height of the Co bogie. Qualitatively, one might then expect having the Co end leading might be slightly better from a heavy pulling viewpoint, with lesser unloading of the leading bogie than the other way around, attributable to the lower pivot of the Bo bogie, and that it was also closer to the locomotive end. But that might not have been material, given that these locomotives had an abundance of adhesive weight relative to their modest power output, and that they had (of necessity) permanent all-parallel (5P) motor connection, such that wheelslip was unlikely to have been much of a problem.

An unusual feature of the bogies was that they had outside drop-type equalizer beams, which were quite scarce in British practice, the inside type being customary where the drop-type were used. That was not an established MetroVick feature, either, as the CIE A and C classes had the inside type. MetroVick’s own idea had been the non-bolster, twisting swing link type of bogie, as used on BR GTEL prototype #18100, and some export electric locomotives. Apparently BR HQ had (quite justified in the event) misgivings about it, and had advised the Western Region (WR) accordingly, suggesting instead that the LMS design, as used on diesel prototypes 10000/1, be used. Of course, the WR was not going to listen to BR HQ, and even less adopt an LMS design, however good it was. (As well as being a good bogie, as unlikely as it may seem, as best I can work out it was also the very first lateral motion Co bogie on a worldwide basis.) On the other hand, LNER, GWR, and SR all contributed some bottom-decile bogie designs to the BR effort. MetroVick did abandon its twisting swing link design, and quite sensibly I think, opted for the established Commonwealth type for its later productions, such as the CIE A and C, BR 28, and SAR 5E1.

  by Pneudyne
Re the association between MetroVick and Crossley, this may have predated the CIE interest and perhaps the Bulleid influence. The WAGR X class order for Crossley-engined locomotives was announced in DRT 1950 October, so MetroVick was working with Crossley before then. The CIE A and C class order was placed in 1954 May (according to Crossley Chronicles #198).

As I understand it, MetroVick, having in 1949 formed a JV with Beyer Peacock for the building of electric and diesel locomotives, then saw forming an alliance with a diesel engine builder as a logical complementary step. Apparently it had chosen National as a likely supplier, but National was absorbed into the Brush group, who was a direct competitor. Crossley, who wanted to get into the traction business, was about the only uncommitted supplier left. Paxman was by choice uncommitted, and anyway, did not yet have any engines in the 1000+ hp range, whereas MetroVick was looking for a committed supplier.

CIE’s choice of MetroVick as a complete diesel locomotive supplier may well have had something to do with the fact that MetroVick had supplied the electrical equipment for its two home-built Sulzer-engined prototypes, the first of which was completed in 1950. After this, I think, Sulzer had formed a tripartite alliance with Crompton Parkinson and Birmingham Carriage & Wagon in the UK, but was still prepared to work with other builders. But MetroVick, having aligned with Crossley, may have been fairly insistent on using that engine.

Although the unusual Co-Bo wheel arrangement has been much commented upon, nothing I have seen addresses MetroVick’s reasoning that led to it - perhaps that is lost in the mists of time. Thus we are left to imagine what it might have been. BR’s de facto target for Type 2 locomotives appear to be a Bo-Bo unit of around 72 long tons. This was just about doable with the Sulzer 6LDA28 engine (roundly 19 000 lb dry for 1160 hp) in a double-cab unit equipped with a train heating boiler. But MetroVick’s powerplant of choice in this range, the Crossley HSTV8, was a lot heavier, at around 30 000 lb dry for 1200 hp. In fact weightwise, it was comparable to the Mirrlees JVS12T, then 1250 hp, and only marginally lighter than the English Electric 12SVT, then 1500 hp. With a 625 rev/min rated speed, the Crossley engine might also have required a larger and heavier main generator than the others. So with little doubt MetroVick needed to use more than four axles. Moving to an A1A-A1A wheel arrangement would have been the modal choice in this situation, and BR’s haulage requirements probably did not require more than 72 long tons of adhesive weight. A Co-Co locomotive, with six motors, would probably have been extravagant for the application. But somewhere along the line, someone in MetroVick may have been thinking that whilst more than four axles were needed, six may have been more than strictly necessary, so why not consider five. There was a Fairbanks Morse precedent for the Bo-A1A wheel arrangement, which probably could have carried the required machinery within an 18 long ton axle loading. But there was also a Co-Bo precedent (actually referred to as Bo-Co by its builder and user) in the form of the Ganz-Mavag built, Hungarian State Railways V55 class single-phase, Kando-system electric locomotive of 1950. Perhaps this was MetroVick’s inspiration. And perhaps it thought that its Type 2 candidate, with 5 driving axles as compared with the 4 of all of the other contenders, would show some in-service advantage to BR that justified the extra traction motor and led to repeat orders. That the use of five traction motors necessitated the 5P connection would not have been an issue, given that the CIE A had six of the same motor in 6P, fed from the same engine-generator system, albeit with a slightly different control system.

In the event, the BR class 28 seems to have come out somewhat on the heavy side, at 97 long tons, whereas perhaps 90, but at least no more than 95, might have been a reasonable target, bearing in mind that the CIE A, with one more axle and motor, but without a train heating boiler, was 87 long tons. I think that axle loading of the 28 at the Bo end was around 20 long tons, which might have restricted its running rights. MetroVick was not alone, though. The Brush class 30 was heavy at 104 long tons, as compared with the 87 long tons of the Ceylon M1 class from which it was derived, although it stayed within an 18 long ton axle loading. And the English Electric class 23 came out overweight. One suspects that the EE engineers, required to use the very light Deltic engine in a locomotive that did not really need it, might have thrown caution to the wind when designing the mechanical parts, thinking that they had room to spare and then some, quite a contrast to the care and attention needed for and given to the class 55.

  by Pneudyne
Here is a diagram of the BR class 28:

from DRT 195812 p.472.gif

And for comparison, the preceding CIE A class, which used the same engine and main generator, and the same type of traction motor, albeit 6 of them rather than 5:

Crossley Chronicles 198 p.06,07.jpg

Then the Co-Bo precedent, the Hungarian V55 class AC electric:

from RG 19490325 p.324.gif

Although having an unusual wheel arrangement, this not-so-well-known locomotive was nonetheless better known as the final exponent of the Kando phase-converter system, previously used in rigid-frame locomotives (2-Do-2, 1-D-1, and 0-E-0) but here transposed to the bogie type with individual axle drive. Apparently its designers were inspired by the all-adhesion example of the BLS (Swiss) Ae4/4 class, but found that four axles would not carry the weight of the rotating conversion machinery within the allowed axle loading. Hence five axles were used. Unusually for its time, it had tandem-mounted motors on both the Co and the Bo bogies. That later became commonplace on Co bogies, but remained rare on Bo bogies. (The only other example that readily comes to mind is that of the Argentinean Fiats of the 1960s.)

If nothing else the V55 was said to be the inspiration for the Oerlikon built SNCF CC14000 class, which took the phase converter concept even further, endeavouring to do electromechanically what is now done with power electronics.

Be that as it may, I think it is reasonable to assume that MetroVick was aware of the V55 during its deliberations that led to its choice of the Co-Bo wheel arrangement.

You do not have the required permissions to view the files attached to this post.
  by Allen Hazen
Thank you for the several very informative replies: I now know a lot more about the Class 28 than I did before!
Not sure I have much to say in response.
----The Crossley engines seem to have had several unusual features. There was a "semi-compound" or "pseudo-compound" aspect to the exhaust: pressure from one cylinder was supposed to help scavenge another. This strikes me (usual warning: I'm not an engineer, and my layman's guesses on technical matters are just that) as a system that would be hard to get right: it must depend on very precise design of the gas manifolds. (Weird stuff like this, it would seem to me, might be easiest to get right if you assumed the engine would run at one constant speed, so maybe practical for stationary applications... but in a locomotive?). But if the WAGR people could make it work in their locomotives...
----(Mind you, WAGR may not have had much choice: my impression when I lived in Oz, in the 1980s and 1990s, was that Australian state railways were all desperately underfunded, and had to keep using their "first generation" diesel locomotives well past what a North American Class 1 railroad would have deemed their retirement age!)
----Axle loads. I'm afraid I have no further information: my basic source for everything I knew about Class 28 before reading your posts is B.K. Cooper's "BR Motive Power Since 1948" (Ian Allan, 1985): some of the diagrams in this book have weights on individual axles noted, but not the one for Class 28.
----On a purely aesthetic note, Class 23 has long seemed "cute" to me, for about the same reasons that the (North American) GE U18B has seemed cute, and the (original) Boeing 737 seemed cute it you were familiar with the 707 and 727. I think we see things as "cute" if they seem baby-like, and Class 23's carboy design (with its noses in front of the cabs) makes it clearly a member of the same English Electric "family" as the Deltic and the Class 37, and if you perceive it as a member of that family it is obviously the baby.
  by Pneudyne
You are right that the Crossley pulse-scavenging system was not well-suited to locomotive applications, and in fact that was noted by engineer-authors such as Tufnell (*). But in practice it was probably a minor issue, as the engine was beset by a multitude of more basic mechanical faults and failures. Evidently, right towards the end of the BR class 28 service period, Crossley supplied a redesigned crankcase that was installed on one of the fleet, and this was viewed as being satisfactory. But by then it was too late.

I don’t know for sure, but I imagine that this redesigned crankcase was taken up by WAGR. Apparently WAGR did consider re-engining in the early days, but found this to be too costly when compared with buying new locomotives. Replacement by new light-axle loading (10.5 long tons) locomotives was contemplated in the later 1960s, and apparently English Electric (EE) Australia proposed a Co-Co unit that would have been a more powerful derivative of its NZR Di class/QR 1620 class design. But that did not proceed; how serious was the WAGR is unknown. At that stage, the expected lifetime of new locomotives was probably noticeably longer than the projected need for light axle loading types.

Re the BR class 23, well it was known as the “Baby Deltic”, so if baby equates with cute, then I guess it was cute. As to the Boeing 737, “cute” does not seem to align all that well with its “Fat Albert” nickname. I thought that the 737-200, which operated here for many years, had rather a squat look about it.

(*) R.M. Tufnell; The Diesel Impact on British Rail; Mechanical Engineering Publications (MEP), 1979; ISBN 0 85298 438 3. Of those that I have or have seen, this is about the best source of information on the engineering side of BR’s first generation diesels. I imagine that it was published by MEP because the likes of Ian Allan might well have found it to be “too technical” for their markets.

  by Allen Hazen
Thanks again! The Crossley engine doesn't seem to be covered in glory: the reason why its unsuitability for locomotives wasn't an issue was that other problems surfaced first! (Maybe, if anyone ever has a "worst imaginable locomotive design" competition, we can submit something with the Crossley engine and some chunky 1-C C-1 running gear!)
As for cuteness... De gustibus, as they say. At a guess the "Fat" in the "Fat Albert" nickname refers to the same bit of proportioning that I though of as contributing to cuteness: the fuselage of a 737 has the same cross section as the other Boeing first generation jetliners, but it is shorter, and so gives more of a visual impression of plumpness.