Sir Nigel Gresley - The L.N.E.R.’s First C.M.E.

The posts below are on the "Articulated Locos of North America" thread, but I think the discussion belongs here on the Gresley thread.

I think the total number of engines with Gresley conjugated gear was between 1100 and 1200 - of which about two-thirds were on the LNER and the remaining one-third in a number of overseas countries.

Thanks for that post - it fills some of the gaps in what I knew.

3-cylinder propulsion seems to have enjoyed brief popularity for fast freight engines in the USA in the 1920s. ES Cox ("World Steam in 20th Century" - Page 37) estimated that about 250 were built. Most appear to have had the Gresley gear, which likely made the 3-cylinder drive more acceptable in a country where inside cylinders were disliked. I have not come across a full list of the relevant railroads and engine classes, but the most numerous appear to have been Union Pacific 4-12-2s (88 engines) and Southern Pacific 4-10-2s (49 engines).

Other overseas uses of the Gresley conjugated gear were the Japanese Class C53 4-6-2 (97 engines), NSW Class D57 4-8-2 (25 engines) and a few small classes elsewhere. Totting up the numbers, we are likely close to 400 overseas engines. It would be interesting to know whether Gresley received any royalties for overseas use, but I doubt whether LNER records will shed any light on that.

As @Jon Pegler has noted, overseas interest dried up in the early 1930s, not just in respect of the Gresley gear but of 3-cylinder propulsion in general. I think it is important to note that both Vincent Raven and Nigel Gresley really did stand-out from other engineers in their strong commitment to the 3-cylinder design option. 3-cylinder engines were of course also common on the SR and LMS and in a limited number of other European countries, but that was about it.

 

Perhaps you need to do a little more research? Chapelon locomotives were used almost to the end of steam in France. The 141Rs were not a case of KISS; the damage to the railway system in France during the war meant that the means to repair let alone build new locomotives were just not there. At the time the USA was the only country with the productive means to do so.

 

I can well understand why three cylinders is a good idea in principle. You get a relatively even torque and balancing is better. It also allows you to get more tractive force within a tight loading gauge as three smaller cylinders are easier to fit in than two larger ones. However, I can well understand why it fell into disfavour. Three cylinder locos are at a significant disadvantage at the moment of starting and, if you want to haul a heavy train you have problems. If you are the operating department, you don't want a three cylinder loco. Some of Gresleys locos were even worse as he deliberately limited maximum cut-off to 65%. I don't know how widely that idea was copied.

 

Why would a 3 cyl be any more challenging to start than a 2 cyl loco? Or is that not what you mean? Why would hauling a heavy trailer create problems?

To me at least, the 3 cylinder design looks good against the UK's very tight loading gauge. In the states it makes a lot more sense to have massive outside cylinders as loading gauges are wider.

 

Hi Steve - I'm interested in why you would say this. The evidence we have in terms of the LNER suggests the issues of the 3-cylinder locomotive was not an ability to start or haul a train, it was more whether you chose three sets of valve gear or conjugated and the issues arising from the pros/cons of both types.

Yes, this was intended to prevent over-travel of the middle valve on the conjugated gear: but with long travel introduced to replace the short version was, I think, raised from 65% on the classes affected, though I will need to consult my notes on this.

 

I'm getting confused now. My impression that Chapelon's influence on the actual operation of the railways was minimal was based on what I've been reading on this very thread.
What were the respective periods of introduction of the 141Rs and the 141Ps?

 

On any steam loco, there is a cyclic variation in force on each piston throughout one revolution.

On a multi-cylinder design, because the pistons are arranged out of phase, the peaks in force on one piston to some extent align with the "troughs" on another, so the combined force applied at the wheel is smoothed out. That smoothing is better on a three cylinder design than on two cylinders (and even better on a 4 cylinder Lord Nelson, but not on other four cylinder designs). If you draw the diagrams, the difference between max and minimum force applied at the wheel is within narrower bands on a three cylinder design, hence the general comment about smoother torque delivery.

But ... that only applies when running at constant speed. On starting, there are certain positions in which steam pressure can only be applied to one cylinder, the positions of the valves meaning that the other cylinders receive no steam. The effect of that is that only one cylinder is actually providing any force. If you then take two locos of nominally equivalent TE but different numbers of cylinders (say a Maunsell N and N1 mogul), the three cylinder design has smaller cylinders and therefore can only exert a lower force at starting. So starting a heavy train is worse with a three cylinder design; you only get the benefit of smoother torque delivery once the loco is moving.

Tom

 

Tom has essentially beaten me to it but I'll try and add a bit more info. With a 2 cylinder loco the cranks are at 90° to each other, thus when one cylinder is at the end of its stroke, the other cylinder is at mid stroke. Because it is at the end of its stroke the first cylinder is producing no force at the wheel. Neglecting angularity, the second cylinder is at the point at which the piston is producing its maximum force at the wheel so you are getting the full force from one cylinder but none from the other. With a 3 cylinder locomotive the cranks are at 120° to each other. If we therefore put one piston at the end of its stroke, one of the other pistons is 120° in front of it and the other is 120° behind it. Now, neglecting angularity, 120° is equivalent to 75% of piston stroke. Most locomotives have a maximum cut off of about 75% so, at that point it is getting no steam because the valve is closed. The second cylinder is, as we said, at the end of its stroke so is not producing any force at the wheel. This leaves just the third cylinder producing any force at the wheel and the piston is not yet at the point where it applies the maximum force as that will not occur until it reaches mid stroke. If we look at a 2 cylinder loco with 18" cylinders, everything else being equal (wheel dia & boiler pressure), a 3 cylinder loco would only require cylinders of approx 14.7" diameter so would have a much reduced starting tractive effort from that one cylinder compered with an equivalent 2 cylinder locomotive. Thus, a 3 cylinder loco is at a significant disadvantage if it is stopped at the point where one cylinder is at or near the end of stroke. Not only that, but there are six positions in a wheel revolution where this point of lowest torque occurs compared with only four on a two cylinder loco.

The above ignores the effect of angularity which skews things slightly. It also ignores the effect of lead steam, which can also affect things negatively but the principle remains the same. Once the loco is moving, it get the benefit of the other cylinders but it is at the point of starting that the problem lies.

Gresley, so I understand, thought that there was no need to have a maximum cut-off of 75% and he limited his pacifics to 65% to avoid them being thrashed as much. (I've read that somewhere but can't think where). With 3 cylinders, this created an even worse situation at starting because it increased the distance over which only one cylinder was available for starting. They were later altered to 75%, I think in BR days, but only for forward gear and reverse gear remained at 65%. Anyone who has ever tried to start a train in reverse on a gradient with Gresley will know how much of a problem this presents. 3 cylinder locos really need a cut off greater than 75% to help overcome the problem but that doesn't solve it completely as longer cut offs introduce different problems.[/QUOTE]

 

That's where you might be going wrong (for some posts)!

They were introduced at about the same time. The 141Rs outlasted the 141Ps by about 4 to 5 years.

 

This bemuses me. If the difference between max and minimum is greater on a two cylinder type, how come the minimum power on a 3 is smaller than that of a two cyl? I ought to look up the power curves some time.

 

[/QUOTE]

Would this be the reason that I was once told by a driver that Gresley 3-cylinder locomotives should be started in full (forward or reverse) gear then quickly turned to a lower gearing hence the occasional slipping from an inexperienced driver ?

 

Firstly, it’s a question of tractive effort, not power. And secondly, the issue is specifically about starting, when there are situations in which only one cylinder might actually be providing any force. On a three cylinder loco, that cylinder is smaller than on a two cylinder loco identical in all other ways.

Once you are moving, you get the benefit of the smoother torque. But for that first wheel revolution, the three cylinder design does worse.

Tom

 

It's fairly normal to start all locomotives in full gear as that ensures that at least one cylinder is open to steam. However, quite often you can get away without dropping it into full gear and it will start, especially if it is a small train or light engine. With a three cylinder loco of any sort, not just Gresley, the chances of doing so are less. Whether you pull it up immediately I'd suggest is personal preference. I usually wind it up to about 45% fairly quickly but, if the rail conditions are bad I might leave it in full gear and carefully open the regulator, keeping an eye on the steam chest pressure gauge (if fitted) to keep the maximum pressure down and hope it will start without too much pressure. It depends very much on the loco and how controllable the regulator is. If it is stiff, it is more of a challenge. With a 9F, for example, I'll open the reg and quickly shut it whilst I wait for the steam to work its way through the superheater and steam chest and eventually lean on the piston a few seconds later. If I left it open until something actually happened I'd probably end up with a good (?) slip. With the Q6, I'll just open the reg and let it do what it wants 'cos I know it is unlikely to slip.

 

Was the maximum cut-off ever increased from 65% on the V2s?

 

V2s got long travel valves so one would suspect it was. Open to correction though.

 

Can someone point me at a chart/graph that shows the torque at the wheel over the full cycle. Still trying to get my head round this.

 

If you go and crunch the numbers on no. of Chapelon locomotives built/rebuilt from others, in service, and for how long in service, you will come to similar conclusions as me, I believe.

I don't believe so.

 

Thank you, that is what I thought (as that is the dimension given in the BR P&E report on 60845 and I could not see anything in Locos of the LNER).

 

@Steve does I believe.

Tom

 

Varying Torque by Hermod posted Jan 4, 2023 at 5:02 PM

The german G12 and baureihe 44 and 45 were three-cylinder freight 2-10-0 and totalled close to 4000 very longlived and was surefooted everywhere.
The bad starting manners of the Gresley conjugators was due to the 65% cutoff that was mandated by the conjugation when trying to go fast.
The Prussian G12 from 1917 had a conjugation system that was a maintenance disaster and was rebuilt in 1950 in East germany with relief.
No German threecylinder got conjugation after 1917 and as heavy freigth were not going fast.