Imaginary Locomotives

[Northmoor]

7 hours ago, The Johnster said:

And that's the problem with them on hills; what goes up must come down. Carnforth, Carlisle, and Glasgow are not far above sea level but the bits in between are lumpy, so there is no so much opportunity for high power over extended periods; it has to be shut off when you are coming down the hills.  The Deltic prime mover is firstly a marine engine for fast patrol boats, configured for running at a continuous speed under a specified load; the ECML provides more of this than the WCML.

While most marine diesels do operate like that, a fast patrol boat's operating cycle is something like, chugg-chugg-chugg-chugg, Zooooommmmmm, chugg-chugg-chugg-chugg, Zooooommmm, chugg-chugg-chugg-chugg, get off and go for tea.  The Deltic engine has more in common with aero engines than conventional marine diesels (in fact it is derived from an aero-engine design).  Also remember that they were last used in Hunt-Class minesweepers, which really did just go chugg-chugg-chugg-chugg-chugg-chugg-chugg-chugg, all day.

[LNWR18901910]

Anyone familiar with Budget Model Railways on YouTube and Instagram? They have done some interesting diesel conversions:

My favourites so far are the Mini Class 17 and shortened Class 35.

[tythatguy1312]

Well I do believe this thread's name isn't specifically "Imaginary British Locomotives" so I do have an intriguing idea. In the US, the Southern Pacific Railroad often struggled with locomotives being underpowered, leading to the acquisition of a series of Kraus Maffei Diesel Hydraulics rated at 4500hp. Could a locomotive using 2-3 Deltic prime movers have been used in its place?

[The Johnster]

Why not.  There’s no advantage to two, 3,300hp, which the Kraus-Maffei beats all round, but three Deltic prime movers would probably fit on a big American loco frame and there is no real problem with the loading gauge or axle weight, at least not in comparison with UK practice.  A Deltic prime mover as used in the Class 55 is rated at 1,650hp (there were two on that loco for 3,300hp), so three would deliver 4,950hp.  Whether the loco would be suitable for the work is moot, and they needed a specialised maintenance regime in the UK, but the Americans are good at that sort of thing and would not baulk at providing the necessary facilities
 

 

[BernardTPM]

A Deltic engine was used in a US fire appliance, so it was not an entirely unknown quantity.

[tythatguy1312]

2 hours ago, The Johnster said:

Why not.  There’s no advantage to two, 3,300hp, which the Kraus-Maffei beats all round, but three Deltic prime movers would probably fit on a big American loco frame and there is no real problem with the loading gauge or axle weight, at least not in comparison with UK practice.  A Deltic prime mover as used in the Class 55 is rated at 1,650hp (there were two on that loco for 3,300hp), so three would deliver 4,950hp.  Whether the loco would be suitable for the work is moot, and they needed a specialised maintenance regime in the UK, but the Americans are good at that sort of thing and would not baulk at providing the necessary facilities
 

 

If its of any help, I have heard that the Deltic engines used in the Norwegian Tjeld class patrol boats were rated at 3,100hp each. Whether the engine from a Scandinavian PT boat would even fit in the US's generous loading gauge is up for debate though.

[Flying Pig]

Tjeld type boats were operated by the a number of countries, including the US Navy as Nasty Class patrol boats .  I believe the engines were all pretty much of an external size, just rated much higher for military use.  Compactness was one of the original design goals of the Deltic engine.

 

Meanwhile I'm wondering why we've stopped at three engines.  Surely we can conceive a four-Deltic Co-Co-Co?  No-one in their right mind would build one of course, but that's not the point.

[The Johnster]

Deltic engines came in two sizes (so far as British locomotives are concerned), the 1,650hp 18-cylinder unit used in pairs in DP1 and the production 55s, and the 9-cylinder 1,100hp unit used in the Class 23 ‘Baby Deltic’ Class 23.  Pick whatever combination you want for the imaginary Southern Pacific loco!   Deltic engines by definition have multiples of 3 cylinders. 

 

Edited October 29, 2022 by The Johnster

[DenysW]

Almost all of the Southern Pacific cab-forwards steam locos were rated at 6,000 hp, plus whatever you get short-term from mortgaging the boiler. Maybe 9,000 hp peak, 6,000 hp continuous at 25 mph? Did they just had the same confusion on deiselisation that we had - that it wasn't just the continuous output that determined a Really Useful Engine, but how well it performs on real-world tasks.

 

Summary: I don't think you can create even an imaginary locomotive without a specification.

Edited October 30, 2022 by DenysW

[tythatguy1312]

1 hour ago, DenysW said:

Almost all of the Southern Pacific cab-forwards steam locos were rated at 6,000 hp, plus whatever you get short-term from mortgaging the boiler. Maybe 9,000 hp peak, 6,000 hp continuous at 25 mph? Did they just had the same confusion on deiselisation that we had - that it wasn't just the continuous output that determined a Really Useful Engine, but how well it performs on real-world tasks.

 

Summary: I don't think you can create even an imaginary locomotive without a specification.

It really came down to the issue of available power. The most common freight diesels in the US only had 1500-1600hp, being MU'd into groups of 2-6 to accommodate. Unfortunately this massively drove up maintenance costs (on account of having to maintain 4 diesels for the equivalent power of 1 steam locomotive), which is what prompted Southern Pacific to buy the 3500hp Kraus Maffei units. Similar issues on Union Pacific lead to perhaps the single most successful turbine locos ever built.

 

 

 

Turns out I misremembered and the Kraus Maffei ML-4000 units were only rated for 3500hp, though operation in groups of 2-3 wasn't uncommon for them

Edited October 30, 2022 by tythatguy1312

[rockershovel]

Since I've never driven a steam locomotive and the key words don't bring up anything relevant in context, what is meant by "mortgaging the boiler"? It clearly refers to over-extending the boilers reserves in some way to achieve a short-term power boost, but how? 

[Michael Edge]

Simply letting the boiler pressure drop faster than it can be maintained for short periods, then when the loco isn't working as hard - e.g. down the next hill - shovelling a lot more to get the pressure back up. A steam loco boiler contains a big reserve of power.

[rockershovel]

1 hour ago, Michael Edge said:

Simply letting the boiler pressure drop faster than it can be maintained for short periods, then when the loco isn't working as hard - e.g. down the next hill - shovelling a lot more to get the pressure back up. A steam loco boiler contains a big reserve of power.

From what No 2 Son tells me, and he should know, hybrid petrol/electric vehicles operate in a similar fashion in order to produce short-term power outputs considerably beyond the abilities of their charging engines.

 

Driving one involves striking a balance between initial charge in tbe battery, charging available while driving and consumption while driving. 

[The Johnster]

Gross simplification.

 

A steam locomotive accellerating hard or climbing a bank with a heavy load can be regarded as a balance between boiler pressure and how quickly the steam is used by the cylinders then chucked out the chimney.  The amount of steam supplied to the cylinders can be adjusted by the driver with the 'cut off', a function of the forward/reverse control which operates the valve gear.  When the loco starts off under load, the valve gear allows the maximum amount of steam into the cylinders, but as soon as the loco is under way and has a bit of momentum, the driver lessens, 'cuts off' the amount of steam to the cylinders; he has to, because if he didn't he'd empty the boiler of steam altogether within half a mile or so, and the fireman would not be happy.

 

Eventually, the loco will be purring along nicely on level track with 400tons of train at 90mph or more, and hardly using any steam at all, just a smidge to keep it going, at a high rate of cutoff and the regulator barely open; this is steam at it's most efficient; a diesel would be working pretty hard in this situation.  But even so the cylinders are pumping out steam from the exhaust pretty quickly and if the fireman doesn't keep an eye on things and keep stoking frequently pressure will fall, and then speed will as well and he's in  trouble!  Don't forget he must keep an eye on the water level in the boiler as well and there's no point in wasting effort by putting too much coal on and having the safety valves blowing off and wasting steam/all your hard work.  The ideal situation is to have the safety valves 'feathering', on the point of blowing off but not doing so fully.

 

Righty ho, we're running along through Carnforth on the down with the safety valves feathering and a good fire, but now we've got to pull that 400ton over Shap, which is a bit of a haul once you're through Tebay, so you build the fire ready for the big effort, and the driver has already opened up the regulator and decreased the cutoff; the loco is already using steam at a much faster rate.  The fireman can probably keep up with the demand through Tebay, but the proper climbing has started now, and he is firing continually.  The steam is being used very quickly, speed is falling, and if we're not careful time will be lost. 

 

So you mortgage the boiler.  We're on full regulator and maybe 30% cutoff now, and the engine is making a bit of noise and trying to blow a hole in the sky, but Scout Green is only a mile or so away, so you can let the pressure fall, and the driver can push her as hard as he wants to, so long as you get over the summit.  Don't worry if pressure or speed falls and the fire gets in a bit of a state, just keep her moving until her nose dips towards Penrith.  She wouldn't have kept that up for much longer but the train is more than half over the summit at 20mph, you and the driver exchange a look of satisfaction, and he shuts the regulator and eases the cutoff back to mid gear, 100%; Newton's First Law is now pulling the train.

 

The mortgage has to be paid, of course, but this is not too difficult, with the loco coasting down the bank, and the driver putting a bit of brake in now and then to keep her steady.  You have time, since those 4 big cylinders are not using any steam now, and you can sort out your fire and rebuild boiler pressure at what passes for your leisure as a fireman on a top link express job.  Might take a little while, but the safety valves will be feathering and the fire will be in good steamraising condition by the time the driver needs it again.

 

You can use this example to envisage the same sequence of footplate events on any and all steam jobs that involve a bit of heavy pulling that consumes steam quickly and has a following period of respite.  It is not feasible if there is a longer but less steep bank with less respite at the top, and while the steam is used up less quickly, you need to keep up your firing effort.  Imagine the fireman on a Britannia-hauled 'Red Dragon' from Cardiff-Paddington stopping Newport only, 14 bogies unassisted from the bottom of the Severn Tunnel to Badminton, about 20 miles, and the loco is bound to be winded at the top.  So are you; no boiler mortgaging here!  There's no respite, you can't repay the mortgage because you've inevitably lost time, keep shovelling, in order to bring her in to Paddinton on time (and it's a point of honour to try for a couple of minutes early) you've got to keep it up until she's eased for the 50mph at Wootton Bassett, which is only a minute or so respite until you're at it again until the 40mph through Reading, then back to the shovel for the final 80mph dash into Paddington, all of which is done with the regulator on the collar until you get to Southall.  You've shifted 5 and half tons of coal in just over 3 hours, but luckily the Brit's boiler can steam on a candle and you've been using best Welsh; the fireman on the Royal Scot, was relieved at Carlisle and is less exhausted than you...

 

You can't do this on a diesel or an electric (no sh*t, Sherlock).  Essentially what you are doing is allowing the driver to use steam at a rate beyond that which the locomotive can sustain for any but a short period; with non-steam traction he has no choice on level track but to thrash the loco as hard as he can to keep time, and keep it up on the banks in the hope that he can clear the top within the timings; there is no possibility of increasing the loco's power output without putting the ammeter into the red and tripping the circuit breakers, not what you need ascending a bank!  I've many times sat in the secondman's seat and, with speed dropping alarmingly, the driver has commented 'at least with steam we could mortgage the boiler'!

[DenysW]

4 hours ago, rockershovel said:

over-extending the boilers reserves

In parallel to @The Johnster'd detailed explanation: steam locomotives have several tons of boiling water stood in them - this is a store of energy that does not have a parallel in diesel or electric engines.

 

I think this explains the disappointment generally seen with the early diesels. If you compare a steam engine running at its nominal design point, it's still got some grunt left to do acceleration, and to cope with gradients.  A diesel doesn't. The LMS twins did the duty of Black 5s, but their design point spec looks more like a 7P than a 5P. 

[The Johnster]

Pretty much, DenysW.  Steam has that reserve on tap and (in terms of the size of the cylinders), effectively limitless expansive capacity; a diesel or electric flat out is flat out, and has no more to give.  To paraphrase Popeye the Sailor Man, 'it's had all it can stands an' it cain't stands no more, Olive, ke ke ke ke ke'.

 

Another factor in the dissapointment with early details was a fundamental flaw in the data obtained from the Rugby Testing Station, which was built with the purpose of objectively determining the sustainable power output of steam locomotives with the intention of using the data to replace them with adequately powered diesels.  This resulted in the 1955 Modernisation Plan's locomotive policy specifiying Type 4 power band locos to replace 8P and 9F steam, and Type 2 to replace 5MT.  In the event, and for a reason I am not party to, the Rugby data was clearly off the mark, and by an appreciable degree, I'd suggest between 2 and 33%.  The Type 2s were not really capable of 5MT work, and the Type 4s, especially the early 2,000hp examples, barely capable of mainting 7P loads and timings without being thrashed to the point at which availability was compromised until they were cascaded to lighter work.  When Sir Brian Roberston, the then BRB Chairman and a businessman rather than an experienced railwayman, rode in the cab of D200 on it's inaugural Norwich-Liverpool run, not one noted for it's banks, he commented that it wasn't as good as a 7MT Britannia, and it wasn't, but there was a fair bit of 'King's New Clothes' about in those days.  The late 50s and early 60s were characterised by attmpts to increase the power output of Type 4s, but they still couldn't quite manage 8P work unless the loads were reduced, and when airconditioning was brought in the loads were reduced further, even for the Deltics, the only diesels that could replicate 8P steam performance.  The 25kv AL series electrics for the Euston/BNS/Manchester/Liverpool scheme could also manage 8P work, but needed to be 1,300hp north of a D200 and 50tons lighter!

[DenysW]

15 hours ago, The Johnster said:

the Rugby data was clearly off the mark

I think it would be very difficult indeed to build a rolling-road test bed that could accurately (and safely) reproduce the loads delivered to a locomotive pulling unfitted freight and an (operated) brake van up and down the roller-coaster that is most of the UK's network. It would be even more difficult to get the funding to go through several, increasingly costly, iterations until you got it close to right - taking perhaps 7-15 years of studying a technology regarded as obsolete, and irritating your political paymasters as soon as you'd exceeded their attention span (2 months after the initial attempts at data?).

[john new]

21 hours ago, The Johnster said:

Pretty much, DenysW.  Steam has that reserve on tap and (in terms of the size of the cylinders), effectively limitless expansive capacity; a diesel or electric flat out is flat out, and has no more to give.  To paraphrase Popeye the Sailor Man, 'it's had all it can stands an' it cain't stands no more, Olive, ke ke ke ke ke'.

 

Another factor in the dissapointment with early details was a fundamental flaw in the data obtained from the Rugby Testing Station, which was built with the purpose of objectively determining the sustainable power output of steam locomotives with the intention of using the data to replace them with adequately powered diesels.  This resulted in the 1955 Modernisation Plan's locomotive policy specifiying Type 4 power band locos to replace 8P and 9F steam, and Type 2 to replace 5MT.  In the event, and for a reason I am not party to, the Rugby data was clearly off the mark, and by an appreciable degree, I'd suggest between 2 and 33%.  The Type 2s were not really capable of 5MT work, and the Type 4s, especially the early 2,000hp examples, barely capable of mainting 7P loads and timings without being thrashed to the point at which availability was compromised until they were cascaded to lighter work.  When Sir Brian Roberston, the then BRB Chairman and a businessman rather than an experienced railwayman, rode in the cab of D200 on it's inaugural Norwich-Liverpool run, not one noted for it's banks, he commented that it wasn't as good as a 7MT Britannia, and it wasn't, but there was a fair bit of 'King's New Clothes' about in those days.  The late 50s and early 60s were characterised by attmpts to increase the power output of Type 4s, but they still couldn't quite manage 8P work unless the loads were reduced, and when airconditioning was brought in the loads were reduced further, even for the Deltics, the only diesels that could replicate 8P steam performance.  The 25kv AL series electrics for the Euston/BNS/Manchester/Liverpool scheme could also manage 8P work, but needed to be 1,300hp north of a D200 and 50tons lighter!

Back when I was editing the SLS Journal I ran an article on the L116 report based on analysis of steam locomotives. The correspondence got vitryolic very quickly and we had to cut the thread. I don’t want to open that debate here but am puzzled as to why it is perceived by some that Rugby got it wrong when the French equivalent at Vitry influenced the design teams of both the LMS and LNER sufficiently for them to commission Rugby? What did the French do differently (clearly with success) that the British engineers then later ignored?

 

Edited October 31, 2022 by john new
Spelling error corrected.

[The Johnster]

Rugby did produce some usable information (I agree with your point about loose-coupled freight trains); it was British facility that attempted to provide objectively quantified data about the power output of steam locomotives, and at least enabled comparisons to be made between them that were more objectively based than the various exchange trials, which of course were subject to the considerable number of variables that affect a steam locomotive 'out on the road', not to mention a considerable degree of partisan opinion. 

 

Driving a steam locomotive in service to time successfully and with economic use of coal and water depends on route knowledge as well as driving and firing skill, variable coal, and so on.  An example of this is the unsuccessful attempt to replace the S&DJR 2-8-0s with WD Riddles 2-8-0s in the early 50s.  The two designs are of a comparable power and capacity, in fact the 8F WD is 'better' on paper than the 7F Derby loco, but the WDs were an abject failure when they were trialled on the 7F, and the crews complained that they were not suitable for the work.

 

Locomen are notoriously small c conservative and resistant to new equipment; the old engines have done the job for years and they don't see why new ones are needed.  They know by experience how to do the jobs with the old engines, where to fire, where to open or close the regulator, where to feed the boiler, where to alter the cutoff, where and how to apply brakes, so when a new loco is provided all that is upset and a new routine must be learned to get the best out of it.  It is therefore almost standard practice for locomen to complain about new locos and want the old ones back, but they eventually work the problems out and get used to the new locos.  Of course, they still grumble, they wouldn't be enginemen if they didn't, but then a couple of decades later when another new loco comes along, they complain and want the old one back.

 

Had the WDs been given more time the Green Park men would have managed fine with them, they are not bad engines for North Somerset coal trains, but of course they'd have continued to grumble about them.  The locos used for the trials were borrowed and probably a bit rough anyway; a WD on a good day is a cloud of steam with a clanking noise in the middle.  It is a classic example of how difficult it is to get reliable quantitative information about the performance of steam locomotives, and many engines that were fundamentally unsuitable for their work were managed with as the men learned how to time the jobs with them.

 

But BR needed quantative information about the performance of their steam fleet before the leap in the dark (which it was in the 50s, as it was difficult to design a diesel loco that ticked all the boxes within the British loading gauge and route availability restrictions, never mind that the traffic patterns were about to change radically) of dieselisation to replace said steam fleet.  Rugby attempted this, but the data was deeply flawed and steam's capability very considerably underestimated; there are probably good engineering reasons for this but don't ask me, I'm not an engineer.  Recieved wisdom at the time of the 1955 Modernisation Plan was that 8P steam could be replaced by 2,000hp Type 4 diesels, and recieved wisdom was wrong,  This was a shame, and it took BR 20 years to close the gap and improve on steam timings, albeit with lessened loads. 

 

Your points about the time it took to gather usable data and political paymasters who need results before the next election are entirely valid.  Railways are investment-heavy and you have to play the long game to get any financial benefit from them, by which time you have become economically dependent on them while they haemorrage cash, and the British investing classes got burned badly twice with them in the 19th century, with the Hudson Bubble and the Overend Gurney collapse.  Their genetic memory is long and unforgiving, and we are still suffering from the British desire for a Pullman railway at Parliamentary rates, and while other countries electrify and build high-speed networks without demur and have done for the last 70 years, Swansea, Plymouth, Filton Bank, the MML, and Trans-Pennine must wait for their sparklers.  We are British, dammit Carruthers, and good at half-baked compromise and the ingenuity needed to manage with less than perfect situations because it was good enough for our parents and their parents and their parents, hence multi-mode electro-diesel trains, a typically Brit 'it'll do for now' response that will no doubt become a permanent solution rather than the temporary one it was originally intended to be. 

 

[The Johnster]

2 minutes ago, john new said:

Back when I was editing the SLS Journal I ran an article on the L116 report based on analysis of steam locomotives. The correspondence got vitriolic very quickly and we had to cut the thread. I don’t want to open that debate here but am puzzled as to why it is perceived by some that Rugby got it wrong when the French equivalent at Vitry influenced the design teams of both the LMS and LNER sufficiently for them to commission Rugby? What did the French do differently (clearly with success) that the British engineers then later ignored?

Did the debate become Vitryolic, John?

 

Sorry...

[DenysW]

Possibly under-informed guesswork follows...

 

My suspicion is that Rugby was designed, in effect, to re-solve the LNER and LMS problems of Grouping - which was how to integrate different classes of steam locomotives of all ages from many different designers into a functioning fleet operating over a wide variety of route availabilities. So it was revisiting the previous struggle better, not addressing the poorly-defined and understood new need.  It was comparing apples with apples (the old problem), not apples against pork pies. 

29 minutes ago, The Johnster said:

Received wisdom at the time of the 1955 Modernisation Plan was that 8P steam could be replaced by 2,000hp Type 4 diesels,

I agree. However, this is the thinking that compares continuous output as the strategy for parity, which Rugby and Vitry could measure. But if you are comparing against steam that can give you 30-50% more for long enough to get up most inclines, it isn't just continuous that matters, it's reasonably-sustained peak as well.

[Compound2632]

31 minutes ago, DenysW said:

Possibly under-informed guesswork follows...

 

The decision to build the Rugby test station was taken in 1937, as a joint LMS / LNER facility, but the proposal goes back to Gresley's Presidential Address to the Institution of Locomotive Engineers in 1927. [NRM document here.]

 

As far as I can see from that, its intended purpose was not the comparison of different classes of locomotive but evaluation of the effect of design modifications to a particular class, or example thereof.

Edited October 31, 2022 by Compound2632

[tythatguy1312]

I suspect at least part of the reason for "conventional wisdom" may have been America's success with 2,000hp diesels replacing even larger locomotives, seemingly in ignorance of the fact that American diesels conventionally worked in groups, as well as the fact that American passenger trains of the day were rarely heavy enough to warrant anything more powerful than 2000hp. This would've undeniably been compounded by the fact that steam locomotives could easily put on more power than their continuous maximum, albeit only for a short time. Making the issue worse for Diesel-Electric designs was the fact that some of their engine power was siphoned off to power electronic systems in the locomotive or the train itself, with 1 notable example, the Class 31/4, losing 1/3rd of its power to the electric components, such as air-con. It all seems to have been a perfect storm against Rugby's calculations in my opinion.

[JimC]

1 hour ago, DenysW said:

However, this is the thinking that compares continuous output as the strategy for parity, which Rugby and Vitry could measure. But if you are comparing against steam that can give you 30-50% more for long enough to get up most inclines, it isn't just continuous that matters, it's reasonably-sustained peak as well.

That's true of course. But dynamometer testing should have been telling them exactly what power the steam engines were actually delivering to get up the hills. And given even a half **** prototype testing it should have been obvious from the LMS twins and the SR prototypes what could actually be delivered. Of course its never been a problem for management to fail to hear the unpalatable.

 

Edited October 31, 2022 by JimC

[rogerzilla]

Well, a loco boiler contains very little steam reserve at all, but it does contain a lot of very hot water that really, really wants to be steam, except that the pressure keeps it from doing so.  As you let steam out through the regulator, more water instantly becomes steam to replace it but this takes a lot of latent heat, so the temperature and pressure both start to fall unless the fireman and the tubes can keep up. 

 

Loco boilers are very efficient - maybe 80% of the heat from the coal goes into the water if the coal doesn't contain too much small stuff and the fireman is skilled.  The woeful overall efficiency comes from the cylinders, where temperatures and pressures are far too low for a good result on the Rankine cycle, even given perfect valve events and zero mechanical, heat or fluid losses.  When you add up all these inefficiencies, theoretical and practical, you end up with 5-10% for the loco overall.