Flying Scotsman Speed

Thanks for that!

Tom

 

I trust the accuracy of the LNER dyno car - I just think that if the dyno roll still survived it would show a speed of 98 mph. I wonder whether the LNER announced the 100 mph as soon as they arrived at Kings Cross (perhaps based on the loco speedo or a speedo in the dyno car) - before they had examined the dyno roll.

The blip in the curve would take an impossible amount of power to achieve. Just to repeat again - the curve was not produced in the dyno car - it is human interpretation of the dyno time and distance data. And repeating my very first post - the LNER produced two curves, one (without a blip) clearly shows 98 mph and the other (with a blip) seems to show 99 mph. (Perhaps the draughtsman was happy to let the difference between 99 and 100 get lost in the thickness of the lines)

 

Not disputing the speed issue (don't care either way to be honest), but one remark: it is often said that such and such a maximum speed would require an "impossible" or "unfeasible" amount of power. However this is always supported by calculations based on *sustained* power requirements or steady-state. Remember, for a very short period, the cylinders can produce more power than the boiler can keep up with, and there forces do not need to balance - if it is a true transient condition. Obviously only very briefly.
In transient conditions extremely large forces/powers can occur which would indeed be impossible in steady or quasi-steady/gradually varying conditions.
I don't know if this would have relevance to the speed maxima and required power.

 

That may be so, however at very high speed it is almost impossible to get instant extra power. The exhaust velocities
may go transsonic in which case the classical orifice does not allow these higher exhaust velocities. One would need
a de Laval orifice to free the extra power.
Kind regards
Jos Koopmans

 

I agree with Courier and Jos that with a loco that is working flat out then a sudden 'blip' of speed is almost impossible.

If you look at the descriptions of what happened on the footplate on the LNER footplate runs, the cut off was lengthened not shortened, which would make any sudden 'blips' of speed impossible.

The City of Truro run continues to stir debate. The Rous-Marten log does not add up. This is apparent from the original log published in The Engineer published a week or so after the run in 1904 which is much earlier than that quoted by Courier. There is considerable disagreement as to when the braking occured after the descent of Whiteball summit, which markedly affects Rous-Marten's log. You also have to read into this the embargo imposed by the GWR, plus the publication the following day of the run of the timings by the post van worker in the local paper which escaped the embargo!

Cheers,
Julian

 

It produces a very reasonable chart though.

.

This attempts to show the way the limited resolution of the stopwatch timing affects what you see.

The broad green band in the centre is the limit of capability of CR-Ms stop watches. If you believe CR-M operated his stopwatches absolutely precisely then COTs speed at each point was within the dark green band.
The pale green band each side attempts to represent the likely range of human error operating a stop watch, generally reckoned to be about 1/10 of a second.
The darker blue band represents an error of one stopwatch click - a 1/5 of a second.
The pale blue band represents an additional tenth of a second outside that.

The actual speed of the locomotive would have been a line somewhere across the band. The two lines shown give examples of how much these can vary.

A better statistician than I could work out probabilities, but basically the actual speed of COT would most often have fallen within the dark green band. It would not be unlikely for the actual speed to have sometimes fallen within the light green band, and the dark blue band - representing an error on the stopwatch - is quite feasible. The pale blue band represents two simultaneous errors in the same direction, and this is a bit more unlikely.

The first surprise was that I could draw a straight line - the red one - pretty much solely through the green band with quite a slope on it. This does tend to suggest that the locomotive/bank combination (and we mustn't forget that the bank - ie gravity - was doing an awful lot of the work) was still accelerating the train substantially. That gives CR-Ms data some credibility. The other thing is that if so inclined one can draw a fairly reasonable line - the yellow one - through the chart that doesn't get to 100mph. Another data point would have settled things beyond doubt hence CR-Ms oft publicised disappointment/anger at the braking.

But the main thing is that, looking at the chart, you have to really try hard to draw a reasonably probable line that doesn't pass through the 100mph mark.
The more time a line spends in the dark green the better, and the more its centred on the dark green the better. Lines that pass into the light green a bit are very credible.
Lines that go into the dark blue regularly are getting more unlikely, and the pale blue more unlikely still.
Any line that does not average into the dark green is very difficult to believe.

My own feeling: the last quarter was probably done in around 8.9, on the cusp between 8.8 and 9.0 second quarter miles. Clements braked at over 100mph, but not by very much. But we'll never know unless someone invents a time machine. I once wrote a short story about that as doubtless many others have. Wasn't very good though!

Of course if you want to believe that Rous-Marten made up his figures and the engine never went near that speed that's your privilege. But I think if I'd been faking numbers I'd have got that last 8.8 on the list: the very uncertainty adds to the credibility to me.

(30/6. I've amended the chart to show secs per 1/4 mile and horizontal gridlines on the secs/mile boundaries. Think it makes the granularity of the stopwatch data clearer.)

 

It's a pity that 98 mph wasn't the 'magic'speed that people become obsessed about, that would make everything a hell of a lot simpler!

 

I'd say that CoT was almost certainly the first to break the magic 160km/h barrier

Tom

 

I wonder if it would be a good guess that a broad gauge GWR locos was the first to reach 100km/h?

 

Yes, good point.

 

If logs of two high speed runs with different locomotives both have blips at exactly the same milepost, there must be some reason for that. Is there an irregularity in the gradient? Could the old NE dynamometer car be brought out of the museum and taken for a run down Stoke bank behind an electric loco to see what happens?

 

On the 17th March 1840 the newly delivered Firefly achieved an average of 50 mph, already quite close.
Kind regards
Jos Koopmans

 

Its a nice thought, but with 50 years of reballasting and track replacements in between I don't know how much it would tell you. Especially true if it were a dynamometer car blip rather than a track one. I wonder if the expertise would still exist to put the internal machinery of the car in full working order.

However... I imagine there must be folk who take GPS speed traces of steam tours, and for all the limitations of GPS its still pretty damn good. If one took a good number of GPS tracks across that spot, and there was a tendency for a bit of a blip that might be rather interesting. Would service train tracks show anything, or is (as I would expect) modern traction sufficiently regulated that it would tend to smooth such things out and stick to line speed?

 

Thank you very much JimC for your excellent post with graph.

Give me a couple of days and I will hopefully post something that might be of interest.

Cheers,
Julian

 

I recently had the opportunity to see the graph that the NRM have of Flying Scotsman's speed. This is different to both the graphs shown in Semmens' book (which come from the Jan 1935 edition of the Railway Magazine) - the NRM graph is on a much larger scale and clearly shows 100mph - albeit with the same odd peak in the speed trace. Just to be 100% clear - this is not the original dyno roll. However I did study the original dyno roll for Mallard's run (which even for a GWR fan is quite an emotional experience when you unroll it).

 

Looks like the old LNER spin doctors we're at work perhaps.
But its all water under the Bridge. the first recorded 100mph has been allocated to FS in the Public and Journalistic sub consciousness, and its so much easier to pin all the tales to one donkey. As a result the famous engine saved and restored several times over the tickets train sets and T-shirts continue to sell and another generation knows at least one 'Steam Train' world without end amen.

 

I'm only delighted* to see doubt thrown at FSs record as it brings CoT back into centre stage.

102.3 is widely discredited but a speed of around 100mph seems to be widely accepted by the pundits of the day.

That it took several decades and a loco much larger to equal it only raises the opinion of Truro in my eyes.

*delighted not quite the right word, but ,sure, you know what I mean

 

I think it highlights how unlikely it is that CoT got anywhere near 100…

;-)

 

obviously you know a lot more than the Allen's and Nock's of this world.

 

Were they there? Did they think Truro did 100mph?

Simon