Steam speed records including City of Truro and Mallard

Of course I disagree with him, he ignored air resistance and friction in his calculations.

Brunel even pointed this out at the time.

Edit: I see it’s even mentioned in the wiki page!

 

I don't think you can say that without, at least as a matter of principle, also considering the load. Which isn't stated, but if you assume this arose out of something of a dare, then potentially that could be light engine. At which point the power requirement isn't that great.

My personal view is that it is a tall tale, but I think you need to be careful with very didactic statements like "the required evaporative rate to make steam to get up to that 100mph and then maintain for a meaningful distance is surely beyond the abilities of virtually every steam locomotive built between 1829 and at least 1900" if you don't also consider the load. As a later example, the trains involved in the "Race to the North" of 1888 sustained very high speed over considerable distances - with absolutely fly-weight loads. But beyond any shadow of doubt, they did achieve the speeds required, and had the evaporation to do so.

My best guess about "Hurricane" is that it is a tall tale that has become enhanced with every re-telling: you get the sense that Nokes (Sekon) was retelling it with a sly grin as well. But you can imagine the basis for such a tale to develop if you also consider that 1839 predates the introduction of standardised time. If there is 10 minutes between London and Bristol, then you get about 3 minutes local difference between London and Twyford, and it helps you for a west-bound train. In other words, a supposed 16.8 minutes elapsed in local time determined by the station Bobbies (28 miles at 100 miles per hour) is actually about 20 minutes in reality; and add in a few more timing errors all going in the helpful direction (a passing time that was in reality xx:00 1/2 being recorded as xx:01 for example) and fairly soon you might be up around 22 minutes elapsed time. That is still pretty fast (about 75 mph) but a long way from 100. From there it then gets embellished in every retelling ...

Tom

 

Funnily enough I nearly added a joke about this, I'm glad it's not just myself who thinks it might have been a possibility...

I'm not sure where the story originates from, although I have a vague feeling it may feature in George Henry Gibbs' diary. One thing that can be said for certain, is that back then, the operation of the GWR bore more than a passing resemblance to the Wild West.

 

First off, I'm GWR through and through. It is perplexing how Brunel has quite the reputation he does though. Despite streaks of brilliance, so many of his ideas were really badly flawed.
His locomotive specifications is amongst them, But his superior design of baulk road was in reality a failure. Timber baulks set on piles sagged between the pilings creating an undulating road, so the piles had to be exposed and cut, and the track ballasted. Still a failure, the baulk road was not persevered with, and transverse sleepers and bullhead rail were in use before dual-gauging.
On broad gauge: I'm not going to get into the merits of 7' as regards speed and stability etc. In India, the GIPR was laid to 5'6" gauge, at an average capital cost of £10,095 per mile. Such a high capital cost was a considerable limiting factor in the development of a broader railway network and prevented railways being provider for poorer districts. Eventually, a secondary network of metre gauge lines was built, at a capital cost of £4,519 per mile. Bigger gauges are more expensive to build, and unless operating at capacity, there is seemingly no benefit.
The atmospheric railway was a failure as rats ate the greased leather sealing strip preventing it sealing. Black rats in towns, brown rats in the countryside, in Victorian Britain rats were an everyday scourge. How could the man who had built the floating harbour remain ignorant of the problems caused by the creatures?
He had the bright idea of hiding iron rainwater downspouts inside the masonry walls of his buildings. Had he heard of rust corrosion? His drawing office at, if I remember correctly, Chippenham, which is a listed structure of special historical significance has had fortunes expended by NWR to save it from Brunel's clever idea.
The Great Eastern? He created a ship that was too big for the trade, and having cost a mint of money to build, was laid up after 2 years of uneconomic service. Yes, its unique size made laying the Atlantic telegraph cable possible, but that was not the task Brunel had designed it for. Allegedly the only time it ever made money for its owners was as a floating billboard for Lewis's department store in Liverpool.
The man was a genius, but many seem to believe he was this prescient, infallible who was let down only by the materials of the day, but with his record tempered by some extraordinarily expensive failures, it is a legend which doesn't stand up to scrutiny

 

Tom, I often agree with your reasoning on many points, but on this I just can’t see it, with my apologies. There’s far too many factors against 100mph running ahead of 1900, and even then, I think there’s some good reasons why we didn’t get an authenticated 100mph until Flying Scotsman’s run (and if not that run, then Papyrus most certainly and that was 108mph achieved in 1935).

I’ve been looking at the dynamometer record of a locomotive built to exceed 100 mph for over eight months now, and I’ve now got in my hands that of the German equivalent, and there’s no way based on the information I have gleaned about these real world accounts that any locomotive, pre-1900, can do what was claimed in that GWR story: 100 mph sustained for 28 miles.

The boilers cannot keep up with the demand in a real world setting. That’s not said from a position of ignorance, it’s from a position of scrutiny.

Now, the races to the north are a different thing altogether and I don’t think anyone is claiming that high speeds and high average speeds, at that, were not achieved.

But the locomotives involved absolutely did not achieve 100mph, either outright or sustained for 28 miles. In fact the best performance recorded is 63.2 mph average speed to Aberdeen, which undoubtedly includes speeds in the high 80s.

But not 100 mph, and certainly not a sustained 100mph.

I think a lot of the issues with the overall discussion about 100mph and the romanticised recounting by railway enthusiasts is that we don’t take enough of a dispassionate view of the steam locomotive.

It was an imperfect machine, every single one was unique, wear and tear was uneven, they were manufactured imperfectly, fuel was imperfect, the tracks they ran on were jointed and imperfect for centuries, there were issues with braking, curves, and the size of the locomotives when you look at the claims versus the statistics, versus what was actually recorded and achieved, show a huge technological gulf between the streamlined machines and large boilered locomotives pulling short form expresses on short sections of track, achieving these one off high speeds and those we are looking at many years later and noting their designs.

So with respect Tom, I can’t agree with what you say there - 100mph was and is a difficult achievement for a steam locomotive and the number of claims that are unauthenticated versus the very, very few that are for steam all have similar things in common - the technology appearing to achieve something well ahead of the machines to come that actually did it without the advantages of those later highly developed machines.

 

I don't know I'd care to be so dogmatic. An early 19thC locomotive will do 100mph just fine off the side of a tall enough cliff. The differences between 1:90, 1;200, and level track are gulfs, as is the difference between sustained speed over a significant distance and a short boiler mortgaging burst. And that's before we even get into no train, minimal train, or reasonable test load.

Enthusiasts love to talk about this record being authenticated and that one not, but as I've said before AFAICS no steam rail speed record has ever been authenticated by an independant official body, nor would any of them meet the requirements for an authenticated record with any other form of transport.

Maybe that's why the debate gets so acrimonious - it's the sub liminal feeling that one's whole argument is founded on quicksand.

 

From that article it would seem that Simon is in august company in not agreeing with that fine gentleman on many technical and/or scientific matters...

 

I did chuckle at that first line Jim. Much to agree with you there.

But that’s why the authenticated runs need to be taken more seriously (especially 05002’s now).

They're examples of moments on the railway where the conditions were met for high speed running intentionally. A lot of effort went into preparations for each run and for some form of record to be officially made by the railway company doing the run.

The question of safety rarely comes up but needs to be.

I really must get on with writing the paper…

 

There has never been the need for a body to create rules for or authenticate rail ‘top speed’ records since there has been so little competition to achieve them. In general railways don’t build locos (or nowadays trains) in order to achieve a speed record.

Sometimes they just ‘take’ the record when they have a capable machine and a suitable opportunity (e.g. Mallard’s brake-testing, or the HST speed record coinciding with a PR event). Other claimed records emerged only due to the happy accident of somebody timing the train.

Nor does there appear to be much (if any) evidence of railway companies disputing their rivals’ records - they weren’t really bothered. The controversy, such as it is, is only amongst us railway enthusiasts.

 

I agree with a lot of that. But you have introduced lots of reasons that might prevent an early locomotive travelling at 100mph, whereas in your initial statement (which you have made several times before) you explicitly couched it in terms of the capability of the boiler to produce enough steam. My own feeling is that for Victorian locos, that is the least of your worries - the capability to use the steam is much more suspect (i.e. to get it in and out of cylinders quickly enough).

FWIW, I am sceptical too about any claim for 100mph in the 19th century (not least from 'Hurricane'!) But I don't think steam production would be the limiting factor; it would be the mechanical side of things (and of course external factors such as the track condition etc).

Really I'm just making a comment about precision of language, particularly in an academic setting: if you are going to make a categoric statement about no loco being capable of achieving 100mph, but then don't qualify it with conditions of load or gradient, then I don't think it is going to be steam production that is your limiting factor.

Tom

 

Fair comment Tom. Thank you for clarifying for me.

 

I look forward to theoretical calculations of Hardwicke down the Lickey.....

 

Probably 90+ mph - but blowing off furiously

Tom

 

That seems true for maximum speeds, but some railway companies did boast about point to point speeds.

 

Are Hardwicke's leading wheels in a pony truck or a rigid chassis? If the latter, the ride must have been interesting even on the run down from Shap.

 

That's why I said 'top speed'. I doubt increased top-speed capability was ever a significant design-goal, merely a by-product of other improvements including better point to point speeds which'd be much more relevant to a railway's commercial needs.

 

Applicable to pages of posts in this thread and many others.

 

Right. This 05 002 speed record claim. I have many issues with this.

First and foremost, some British writers have not been entirely truthful when they have been reporting on the data we have available. My understanding from the book Baureihe 05 - Schnellste Dampklok der Welt is that the original dynamometer roll does not, in fact, exist in the modern day. All writers with secondary texts have taken the speeds/interpretation of the run from Adolf Wolff’s graphs he produced to showcase the run after the event.

The 200.4kph is an instantaneous speed taken by way of a single point on the graph, and calculated by way of distance/time recordings. (Does this sound familiar….)

When you look at the graph, it is clear that the 05 002 has not achieved 200.4kph on the level. It’s actually accelerating on a very slight gradient. That in itself is not a big issue. What is an issue is that you cannot physically see how 200.4 is calculated on the graph by way of its physical scale. The graph itself shows speed and drawbar horsepower on the same graph. The graph below shows the distance travelled.

Now, my bone of contention is that if we’re talking about acceleration for Mallard and saying it’s not possible for this machine to rise up 126mph from 124mph (whilst going from a downhill section to the level and then a slightly less steep downhill section), the jump from 190kph to the claimed 200.4kph on the slightest of gradients needs some re-examining, in my view. But I don’t know how possible that is with the secondary evidence we have here.

We don’t have the time taken per distance as a separate set of stats or show graphically. We can in theory reverse engineer the timings per each part of the graph, of course, as we have done with Mallard’s dynamometer roll (where we have the distance and the time and can calculate speed). But the problem here is we want to confirm the speeds attained. Can you see the issue here?

If all previous writers have been looking at Adolf Wolff’s graphs and then making the case based on these, I have some reservations.

Not because I think he was necessarily wrong, or he made an error, or he lied (far from it) - but there’s no way, in my view, that you can check the maths behind this in a similar fashion to Mallard.

For a start, we don’t have the original roll showing how it was recorded (though the Association of Locomotive Engineers report - available from the National Archives - makes it clear it is very similar to the method used for Mallard’s run) and we therefore can’t confirm Wolff’s figures in the way we can Mallard’s. We also don’t have time, we have distance and speed.

Without knowing the actual timings recorded, we can’t confirm the speeds claimed.

So the big question is, can we take the graphs on trust without that extra level of scrutiny that is afforded by Mallard’s dynamometer roll? I have mixed feelings about accepting it verbatim - but (perhaps the thing that alarms me most) is that a closer study of the graphs we have does not show what has been claimed, which is a sustained 124mph for over a mile. In fact, it looks like the locomotive was pushed on the slight uphill gradient to 200.4kph from around 160kph and then eased back significantly once 200.4ph was attained (in the book, the author is at pains to explain the aim was 200kph and once that had been achieved in the cab, they eased off, 200.4kph was claimed later).

One thing which surprised me - the shape of 05 002’s graph looks remarkably similar to Mallard’s. I will need to do some sort of comparative overlay but it wouldn’t surprise me if the two locomotives actually behaved in very similar ways when going for that top speed record.

So - is any of the above unreasonable to think? Let me know your thoughts. For my part, I am still very impressed with the 05 002.

(But I have a funny feeling that the Henschel-Wegmann streamlined tank engine and train may in fact be the more impressive German feat).

 

One can only work with the data available; a lot of detail for Mallard, rather less for 05 002 and less still for City of Truro. In each case some of it is plausible (within allowable errors), some less so such as apparent significant changes of speed over very short intervals. For both Mallard and 05 002 the sustained performance over a considerable distance is more to be admired than the peak speed.

As for that peak speed, it does seem that the Germans were seeking the magic figure of 200 km/hr, which happens to be 124.3 mph (to 4 significant figures), just a tad below the 125 mph originally claimed for Mallard. If the definition of the metre had been 1% longer, the respective claims could have been the other way round.

 

For clarity on this particular point: my scepticism about the acceleration is not about a general acceleration over a sustained period, but rather that the speed trace upon which the 126mph claim is based appears to go from 124.5 mph to 126 mph and back to 124 mph in the space of quarter of a mile - rather less than 8 seconds at the speeds being run. It is that erratic nature that is suspect, in particular gaining 1.5 mph in about 3.5 seconds when already close to flat out.

It is that nature that is implausible and which then leads to a discussion about what is a sensible smoothing of the data, since 400 ton masses simply don't change speed in that way. The nature of any smoothing tends to have the effect that the peak speed is reduced somewhat, but the time at or around peak tends to be lengthened. If you take the graph shown in the Andrews paper for example, he removes the claim for 126 mph - but replaces it with 125 mph sustained for about a mile, arguably more impressive.

So for clarity - it is those instantaneous erratic accelerations that are being questioned, not so much the sustained acceleration over several miles to get up to a peak speed. But 124.5 --> 126 --> 124 in about 7 seconds? Implausible, IMHO. Sustained 124.5 - 125mph for 7 seconds? More likely.

Tom