Diesel Mechanical Multiple Unit performance

[Wilts Wanderer]

Bit of a ‘shot in the dark’ question really - does anyone with intimate knowledge of driving mechanical (heritage) DMU units recall the sort of performance that could be maintained on rising gradients?

So for example, take a long 1:100 rising climb (for example, parts of the Hope Valley line). Until the late 1990s both power twin and power-trailer Class 101 units were regular performers on the all-stations service to Sheffield. I imagine there was a noticeable difference in the balancing speeds between the two, due to the respective power-to-weight ratios. As a youngster I twice experienced a power-twin on a trip to Edale and remember it being fairly (but not excessively) slow on the long climb to Cowburn Tunnel. Presumably it was much harder work with a half-powered unit.

For clarity my interest is mainly due to commercial train simulators and my perception that the DMMU models are generally overpowered. In Train Simulator the (Armstrong Powerhouse-modified) Class 101 will accelerate into the mid-60s on a 1:100 gradient and a power-trailer holds about 45-50mph in 4th gear on the same climb. To my perception this is unrealistic I tend to throttle back to notch 2-3 once 4th gear is obtained in order to maintain realism but I’m wondering if I’m doing the units a disservice?

Incidentally the other thing I remember recently reading was that a DMU on mid-life tyres would struggle to top 65mph due to the gearing. This makes me wonder if the simulators are unable to replicate the relative forces / frictions involved with an epicyclic gear box at the higher speed ranges.

 

[Magdalia]

I did hundreds of journeys on the Cambridge-Royston dmu shuttle, which was usually a MetCam 2 car motor plus trailer.

What I remember most is that a unit on only one engine would go up Meldreth Bank (1 in 120) at about 25 mph.

In those days line speed was 60 mph, with 50 mph round the Shepreth curve, so even a non-stop train didn't get a run at Meldreth bank. But I vaguely remember that a unit on two engines got slower than 50 mph not faster on the bank. That would fit in with your 45-50 mph on 1 in 100 for a 2 car motor plus trailer.

 

[Wilts Wanderer]

I did hundreds of journeys on the Cambridge-Royston dmu shuttle, which was usually a MetCam 2 car motor plus trailer.

What I remember most is that a unit on only one engine would go up Meldreth Bank (1 in 120) at about 25 mph.

In those days line speed was 60 mph, with 50 mph round the Shepreth curve, so even a non-stop train didn't get a run at Meldreth bank. But I vaguely remember that a unit on two engines got slower than 50 mph not faster on the bank. That would fit in with your 45-50 mph on 1 in 100 for a 2 car motor plus trailer.

Thanks for the response. Just out of interest when you say ‘a unit on one engine‘ do you mean a power-trailer set with one of its two 150hp engines out of action? (On the basis that one powered DMU vehicle had two engines, one on either side.) That would be exceptionally slow

 

[AndrewE]

My memory of commuting on the Severn Beach branch in Bristol is of going up the bank around Narroways Hill... The DMMUs were caught between 2 gear ratios, and slogged on at the top of one - just at the point where the exhaust drove the pipe and its silencer the resonant frequency! A great long echoing fart...
Does that get picked up by simulators?

 

[leezer3]

Whilst not a developer for RAIL Simulator, I've been maintaining OpenBVE for the past 10 years, so I've got a pretty good idea of the internals.

Any simulator is at heart a bunch of compromises heaped upon compromises. Physics simulation is generally provided by a bunch of magic numbers, and I absolutely guarantee you it won't be simulating an epicyclic gearbox, tyre wear and so-on.
The best you'll get is a scripted set of power curves, which may or may not (usually not) provide an accurate facsimile of reality.

IIRC the main developer for Armstrong Powerhouse is mid 30s, and certainly hasn't got the real-life experience some of his marketing might imply.

At the absolute best, with these sorts of things you're getting a developer who has spent some time with the 'real' thing (generally speaking on a preserved line) to get recordings, data etc.
At worst, you get something that provides a superficial cosmetic resemblance to the prototype and complete differences in behaviour. (If you read around the traction forum occasionally you'll get 'real' drivers providing some somewhat unflattering opinions when a simulator derived question comes up)

 

[randyrippley]

I've got strong memories from the 1960s of the WR Cross Country DMUs crawling up Evershot bank heading south from Yeovil Pen Mill. I doubt if they were doing more than 15-20mph in first gear, especially when towing a tail load.
Climbing north out of Weymouth toward and through Bincombe tunnel was nearly equally slow.
Class 22s were kept at Yeovil shed for banking/piloting over Evershot prior to the line being downgraded, but they never seemed to be used on the DMUs

==edit==
just found this old post at https://www.railforums.co.uk/thread...dient-in-a-station.175236/page-2#post-3787958
which rather explains the problems on that route

Digging out my old Gradient Profiles book, Chetnole halt is on Woolcombe bank on the climb up to Evershot where the gradient changes from 1 in 65 to 1 in 53. Upwey (Junction) is on Bincombe bank at 1 in 70 with the gradient changing to 1 in 50 immediately afterwards

 

[Ken H]

2 thoughts
Didnt Regional railways remove the engines from some DMU vehicles. the initial letter in the vehicle type became H for half power. I never really understood why they did that.

Also bear in mind not all DMU vehicles had the same power. The Cl 110 had more powerful engines for the Leeds - Manchester vis Hebden bridge trains. And the Cl 124 was higher powered too.

 

[Revaulx]

I've got strong memories from the 1960s of the WR Cross Country DMUs crawling up Evershot bank heading south from Yeovil Pen Mill. I doubt if they were doing more than 15-20mph in first gear, especially when towing a tail load.
Climbing north out of Weymouth toward and through Bincombe tunnel was nearly equally slow.
Class 22s were kept at Yeovil shed for banking/piloting over Evershot prior to the line being downgraded, but they never seemed to be used on the DMUs

The “premium” Swindon DMUs would have been great if they hadn’t been so underpowered.

The Trans Pennines were completely shot after little more than a decade, having been worked flat out over hilly lines. The other classes didn’t do much better, which was a shame as their interiors were pretty classy.

 

[hexagon789]

The Trans Pennines were completely shot after little more than a decade, having been worked flat out over hilly lines. The other classes didn’t do much better, which was a shame as their interiors were pretty classy.

The derating might not have helped either, in terms of powering over the gradients - all the 230bhp Leyland Albion engines were derated to 200bhp after a few years to improve reliability.

For the TPs that would've cut total gross power from 1,840bhp to 1,600bhp in original formation.

 

[Sir Felix Pole]

In the '70s Swindon XC units were used on Crewe - Stoke - Derby - Nottingham - Lincoln services, which fortuitously is pretty flat. They really struggled on the 1:80 of the Harecastle Deviation at Kidsgrove, however, often delaying following electric services.

 

[Magdalia]

when you say ‘a unit on one engine‘ do you mean a power-trailer set with one of its two 150hp engines out of action? (On the basis that one powered DMU vehicle had two engines, one on either side.)

Yes.

I've got strong memories from the 1960s of the WR Cross Country DMUs crawling up Evershot bank heading south from Yeovil Pen Mill.

In the '70s Swindon XC units were used on Crewe - Stoke - Derby - Nottingham - Lincoln services, which fortuitously is pretty flat.

The Etches Park based Swindon built Cross Country 3 car units were common on the Birmingham-East Anglia route in the 1970s. They were laboriously slow, despite having 4 motors on a 3 car train. Some of the route was flat, but going west there were Hethersett, Manton and Stockingford to climb.

especially when towing a tail load.

East Anglia used to do that too. That would have been very slow on the Cambridge-Ipswich line.

 

[D6130]

My only experience of driving diesel-mechanical units was with the Chiltern line class 115 four car high-powered Derby-built high-density suburban units, based at Marylebone....but maintained at Bletchley. These had the same Leyland-Albion 230 bhp engines as the Swindon-built TransPennine units but, as far as I am aware, they were never derated. They were designed for (comparatively) rapid acceleration between the frequent stops on the steep upgrade between Rickmansworth and Amersham....and - to a lesser extent - between Princes Risborough and Saunderton. However, by 1990, these units were also pretty knackered and you were lucky if you got a unit with all four engines working. Only once can I recall getting into fourth gear (nominally at 41 mph, but depending on the load) on the Down between Chalfont and Amersham. By this time there were also a small number of two car hybrid units - ostensibly for peak strengtheniing, but often also used on the Aylesbury-Risborough off-peak shuttle - which consisted of an ex-Tyseley gangwayed class 115 long wheelbase Driving Motor Brake Second and an ex-Carlisle Kingmoor Derby lightweight short wheelbase class 108 Driving Trailer Composite (complete with Cumbrian Coast window bars). These had a better power/weight ratio and were thus quite nippy. Finally, we had a strange three car gangwayed hybrid unit which saw very little use during the week but, on Sundays, was diagrammed for two return trips from Aylesbury to Paddington and return during the Marylebone modernisation closures. It consisted of ex-Tyseley class 115 DMBS no. 51878, ex-Reading Pressed Steel class 117 Trailer second no. 59735 and ex-Chester Derby lightweight class 108 Driving Motor Composite no. 51571. This unit was also pretty nippy and could often make up time when running late. It had two 230 bhp engines in 51878 and two 180 bhp engines in 51571.

 

[hexagon789]

These had the same Leyland-Albion 230 bhp engines as the Swindon-built TransPennine units but, as far as I am aware, they were never derated

They were, later documentation shows them also as 200bhp.

It was an engine specific not class specific modification.

 

[Taunton]

In the late 1970s a few of the WCML connections from Carstairs to Edinburgh were by dmu, and I recall a 3-car one, a typical Scottish Region odd cars lash-up of the era, might have been from three different builders, certainly an old Gloucester RCW car on the front, full and standing (the latter including me), 150 or more passengers, so another 15 tons of load. Climbing from Carstairs speed gradually fell to below 40mph, and changed down to third gear. Over the top, and it was up to 70, closed the throttle, and freewheeled all the way down to Haymarket. You can imagine the hunting. No handholds in the old Gloucester.

 

[ac6000cw]

For an example of steeper gradient performance, in the 70s I used to travel regularly on 3-car (4 x 150hp engines per unit) cl. 116 Derby 'suburban' units up the 1:50, 1.16 mile climb from Old Hill to Rowley Regis (from a standing start at Old Hill on the gradient!). I don't think the train was doing much more than 20-25mph by the summit. The 1959 WR timetable allows 5 minutes start to start, so maybe 4 minutes start to stop, or around 18mph average. (Today's DMUs are timed for 2.5 min on that climb, start to stop, so nearly 30mph average but with over twice the engine power).

On the other side of the coin, once they got onto the mainline near Smethwick Rolfe Street, it was either flat or downhill to New Street so 70mph was readily attainable free-wheeling downhill (but the ride quality on CWR at that sort of speed was 'very lively'... as Taunton alludes to in post #13 above)

The Etches Park based Swindon built Cross Country 3 car units were common on the Birmingham-East Anglia route in the 1970s. They were laboriously slow, despite having 4 motors on a 3 car train.

I agree.

 

[randyrippley]

Just to clarify - the 119 and 120 XC units had 150bhp BUT/AEC engines, later replaced by BUT/Leyland

 

[61653 HTAFC]

My memory of the class 110 Calder Valley units heading up the hill out of Bradford is that they seemed quicker (and significantly louder) than the 141s, 144s and 155s that replaced them... but that could be rose-tinted nostalgia.

 

[Wilts Wanderer]

This is all great stuff, thank you everyone for your recollections and memories.

 

[507020]

they got onto the mainline near Smethwick Rolfe Street, it was either flat or downhill to New Street

There were passenger services using the curve at Galton Junction?

It does seem that there has been a lot of mention of 1st gen DMU workings in the West Midlands, but as of yet no one has described how painfully slow the ascent of the Lickey incline was. I didn’t even realise that there were unpowered DMU trailers until encountering one on a simulator and attempting this from Bromsgrove took a while.

 

[MadMac]

Battle-hardened veteran commuter of the East Kilbride branch from 1982-99 here. For those unfamiliar, it’s more or less a steady 1 in 70 climb from Thornliebank to Hairmyres.

I usually tried to get the 17:10 “Interlink“ (generally a heavily-loaded 2x 3 car sets totalling eight engines if all was in order) which stopped at Clarkston and Hairmyres. If you got a clear run from Busby Junction, you’d start to slow while going through Giffnock. If you got stopped at Thornliebank, it was a long slow grind to Clarkston. Similarly, in the very rare event of getting a clear run through Busby on to the single line, you’d be beginning to struggle at Thorntonhall. As the Sprinters became the norm, there would be odd journeys on “power twin” 101s which coped reasonably well.

 

[ac6000cw]

There were passenger services using the curve at Galton Junction?

Yes - the Stourbridge line services were diverted to New Street in the late 1960's via the curve from Smethwick West to allow Snow Hill to close (it lingered on with a skeleton service until 1972 I think).

It does seem that there has been a lot of mention of 1st gen DMU workings in the West Midlands, but as of yet no one has described how painfully slow the ascent of the Lickey incline was. I didn’t even realise that there were unpowered DMU trailers until encountering one on a simulator and attempting this from Bromsgrove took a while.

In the 1970s most Birmingham - Worcester - Malvern - Hereford local services travelled via Stourbridge, and Bromsgrove had a very limited service - so there weren't so many DMUs over the Lickey (as far as I remember, mostly some peak hours 'fast' Worcester trains). I'm sure I did travel up it on a DMU once or twice, but it's too long ago to remember how slow/fast it was.

AFAIK, the majority of the Mk 1 DMU fleet was originally built with unpowered trailers included, either as motor+trailer or motor+trailer+motor. I don't remember there being any 'power twin' units based in the West Midlands - I didn't come across those until I moved to Bradford, where the Interchange - Leeds trains were normally 'power twins' based at Bradford Hammerton Street depot.

 

[matchmaker]

The “premium” Swindon DMUs would have been great if they hadn’t been so underpowered.

The Trans Pennines were completely shot after little more than a decade, having been worked flat out over hilly lines. The other classes didn’t do much better, which was a shame as their interiors were pretty classy.

I remember sitting behind the driver on a 126 from Stranraer to Glasgow. Between Stranraer and Girvan he was constantly shifting up and down gears. If you look at the gradient profile for this line you'll understand why!

 

[AndrewE]

AFAIK, the majority of the Mk 1 DMU fleet was originally built with unpowered trailers included, either as motor+trailer or motor+trailer+motor. I don't remember there being any 'power twin' units based in the West Midlands - I didn't come across those until I moved to Bradford, where the Interchange - Leeds trains were normally 'power twins' based at Bradford Hammerton Street depot.

Towards the end of the lives of the (reducing numbers of) units based at Chester someone had the bright idea of scrapping all the unpowered cars and taking one engine off each of the powered cars, making all the sets (half-) power twins. I suppose that as the trailer cars had a full set of controls it saved maintaining them on a car with no engine, and if a power car failed there was always the one coupled to it to get them home!

 

[ac6000cw]

Towards the end of the lives of the (reducing numbers of) units based at Chester someone had the bright idea of scrapping all the unpowered cars and taking one engine off each of the powered cars, making all the sets (half-) power twins.

I think in later years, particularly when Pacers and Sprinters started to arrive (with higher power-to-weight ratios), more power-twins were formed just to provide better performance than a motor + trailer.

I suppose that as the trailer cars had a full set of controls it saved maintaining them on a car with no engine, and if a power car failed there was always the one coupled to it to get them home!

Based on riding around on them, usually (if possible) in the seats behind the cab so you could see the 'engine running' lights, a common issue was an engine or two that wouldn't keep running. It was not uncommon to stop at a station and see the driver press the 'engine start' buttons for the non-running engines to try and get them going, especially if there was decent climb ahead... Occasionally a watering can would appear to top up the engine coolant, as I think overheating or low coolant level would cause an engine to shut down. If you got into a 'motor' car and it was too quiet, you knew you were likely to be late...

But generally they got you there, albeit slowly if things weren't going well.

 

[hexagon789]

Towards the end of the lives of the (reducing numbers of) units based at Chester someone had the bright idea of scrapping all the unpowered cars and taking one engine off each of the powered cars, making all the sets (half-) power twins. I suppose that as the trailer cars had a full set of controls it saved maintaining them on a car with no engine, and if a power car failed there was always the one coupled to it to get them home!

The half-powered power cars were modified 111s, which had higher powered Rolls-Royce engines. 180bhp vs the usual 150bhp.

 

[Taunton]

The concept of two engines on half the cars, rather than one on each, came from the original GWR railcars, designed together with the AEC bus manufacturer, located lineside at Southall. It enabled a nice weight balance of the car and drive to wheels on both bogies, all fitting under the floor, something which a single engine would not do. Their trailers were originally just normal GWR carriages, and sometimes old steam-propelled auto-trailers.

AEC had a long association with the GWR road motor department, for both buses and lorries, and their proprietary gearbox, called Preselector, which did not need a clutch like conventional road vehicles, was an obvious upside for rail application. At the end of the GWR railcar build they also devised the style of multiple unit control that, with tweaks, later became the Yellow Diamond, and later still Blue Square (a further variation on its predecessor for things like remote engine starting). Until this point the railcars had just been direct control in the car, but the last four built were two twin sets with multiple unit controls. Designed and built in 1939, it took them a further couple of years before they were actually got going in service, in about 1942.

I never quite understood how, from about their half point in life onwards, dmu sets with one engine out became rather common. The same engines in buses did not have this issue, and while there were occasional breakdowns, it was nothing like the numbers on the railway where a 6-car set would commonly have one dead. There did seem to be a way to "bump start" them, starting very slowly on one engine up to about 10mph, when suddenly the second engine would come to life, but shut down at the next station.

The power quoted for the engines, with a mechanical transmission, 150hp or whatever, is only obtained at maximum rpm, which is only at the point where you then change gear upwards. What is needed at starting is torque, to get you going. This relates to the different engine types where, although all used in equivalent buses, the engines from AEC and Leyland were used in dmus, which would spend much time at high power and rpm. The comparable power engines from Gardner went into the small shunters, classes 01 to 05, with effectively the same transmissions but where starting power was key, and much work was done at walking pace. Notably the shunters all had Gardners, the dmus all had Leyland/AEC.

 

[randyrippley]

Surely the Gardner / BUT split was a bit simpler than that?
The Gardner was a vertical engine that would fit under a shunters bonnet, while the AEC/Leyland/Albion engines were dry-sump horizontal designs that fitted under a coach floor

 

[ac6000cw]

Also, BUT (Leyland/Albion/AEC) probably had the production capacity to meet the needs of the DMU building program whereas Gardner may not (I don't know the numbers, but I assume there were far more DMU power cars built, all with two engines, than small shunters with one engine).

Didn't Gardner sell horizontal versions for bus/coach use?

 

[Grumpy]

Didn't Gardner sell horizontal versions for bus/coach use?

Gardner 6hlxb 180hp and others.
Used in the Bristol RELL, arguably the best single decker of its generation. Never understood why BR didn't use this engine as Gardner were generally regarded as superior to Leyland

 

[Ken H]

Yes - the Stourbridge line services were diverted to New Street in the late 1960's via the curve from Smethwick West to allow Snow Hill to close (it lingered on with a skeleton service until 1972 I think).


In the 1970s most Birmingham - Worcester - Malvern - Hereford local services travelled via Stourbridge, and Bromsgrove had a very limited service - so there weren't so many DMUs over the Lickey (as far as I remember, mostly some peak hours 'fast' Worcester trains). I'm sure I did travel up it on a DMU once or twice, but it's too long ago to remember how slow/fast it was.

....

There was a time when the main Litchfield - Redditch kept to that route but a bubble car was put on between Barnt Green and Bromsgrove. Might even have been after New St - Redditch electrification.
No idea how a bubble car did on Lickey, and I never caught one. But I remember them laying over in a siding at Barnt Green.