Alternative High Speed Steam?

This incident is referred to in E. S. Cox's book 'Locomotive Panorama', Volume 1, on pages 74 to 76.

 

I don't think that you need to. A stopwatch or chronograph is an accurate and precise, calibratable piece of equipment. As I think we're all agreed on, a man looking out the side of a moving train at quarter mile posts isn't either of those, unless you really increase the value of what you accept as a suitable tolerance. The general phrase I would argue would be; 'recording equipment good, human influence bad'. We accept that with a man with a stopwatch we're going to get a substantial tolerance on the recorded speed. It's easier to argue that COT was doing 90mph than it was doing 110, even if one outcome doesn't make the other impossible.

I wonder how valuable these speed records were to the individual railway companies? We know the GWR had a dynamometer car available at the time, did they chose not to authenticate an attempt because the value of holding that particular title didn't justify the effort to them, or because they doubted the validity of the claim at the time? What I find most odd is that I am not aware of any similar 100mph claims until Flying Scotsman's authenticated record 30 years later? Excusing the period around the 1st world war, it just strikes me as strange that if that record (and the overall record) was so easily in reach, why did it take 30 years for someone to get around to properly recording it?

On a similar vein, how valuable was it to Rous-Martin to be known as the first person to 'authenticate' a speed of 100mph?

 

Let's come back to the point of agreement we two reached - that the records are unauthenticated, and at this remove unprovable as true or false.

 

You can’t say dynamometer car records are accurate at all. You have the trace of the event, which you can study but you can’t prove the accuracy of that trace unless you have the before and after calibration traces that are accurate and don't require adjustment, not simply details of the techniques used to calibrate. The distance ( x axis) can be accurate if it is a pure mechanical drive and correctly geared but the speed trace has got to be related to both distance and time and that requires an integrator of some sort, both of which are subject to possible error. It has already been argued that time can be subject to error. (In that context I have a modern digital clock that loses two minutes a week which does intrigue me.) I've enough experience of measuring speed using distance and time on relatively modern instrumentation to know that you need to be very wary of the results you get. Even on steam and diesel locos that have speedos regularly calibrated you can get a gut feeling that they are not right at times.
Coming back to COT and Rous-Martens quarter mile times, with readings of 10.6, 10.2, 10.0, 9.8, 9.4, 9.2 and 8.8 seconds, one thing that immediately stands out is the inconsistency in the rate of reduction of consecutive times. We have 0.4, 0.2, 0.2, 0.4, 0.2 and 0.4 seconds. That indicates a constant acceleration of 0.3 mph/s and no reduction as speed increased, improbable but not impossible. If true, a potential speed well in excess of 102.3 is being suggested. The 102.3 mph is based purely on that last reading. We really need a reading or two after that 8.8 seconds but, unfortunately, the driver put the brake in.

 

While a lot of this discussion is fascinating, surely the question must be what did the passenger want, and surely ot was Sir Herbert Walker on the Southern who cracked it with a frequent clockface service of fast - but not that silly fast electric trains

 

I don't believe I have ever said that they are wholly accurate or indeed perfectly accurate, but they are definitely a more accurate representation of something that happened than any single Timekeeper's log. We can also study the output, something you can't do in the same way with the timekeeper's log.

Actually we can prove to a good degree the relative accuracy of the machine as we know all of the variables where the L.N.E.R. dynamometer car is concerned - others have already studied this and concluded it works very accurately. Even adjusting for a potential 1/4mph out with Mallard's run would not reduce it below 125mph in any event.

On the L.N.E.R. dynamometer car, it works by marking one second time intervals across one foot of paper. The paper was propelled by the turning of the 12ft diameter measuring wheel outside of the coach, applied to the running rail. This gave an accuracy to within a 1/4 mph of the speed.

The time was recorded by an electromagnetic pen that was controlled by a chronometer, speed being calculated by way of a graduated scale at 5 second intervals.

For Mallard's run, we have a greater degree of accuracy because the Dynamometer car had the ability to trace across different lengths of paper roll and this was done by changing the measuring apparatus to record the distance/speed readouts across two feet of paper, instead of the usual one-foot width. This meant that the recording doubled the sensitivity of the data recorded across the paper roll.

That is why we calibrate the instruments regularly in the first place. I don't think that's a good enough reason to declare a dynamometer car as inadequate for the task yet the human with the stop watch is somehow adequate.

That is suggesting CoT is capable of maintaining a rate of acceleration that even Mallard did not achieve in 1938. Highly sceptical.

 

Really 12ft?

 

Is there any information about the make up of CoTs train, or can someone please extrapolate what it was from the description "hauling a light load of eight-wheeled postal vans with around 1300 large bags of mail on board, total weight 148 tons"


At a very rough guess, based on figures elsewhere, to maintain a speed of 100mph would need around 1500hp to overcome air resistance, although i suspect the real figure could be lower.

 

[QUOTE="
I wonder how valuable these speed records were to the individual railway companies? We know the GWR had a dynamometer car available at the time, did they chose not to authenticate an attempt because the value of holding that particular title didn't justify the effort to them"[/QUOTE]

There was no dynamometer involved as this was not in any way an official speed attempt. True, it wasn't a normal service train but not something run with the aim of recording a particularly high speed either. I believe the train makeup was 6 70ft vans.

In the context of the time (May 1904) and soon after, this is something that the famous GWR publicity department would not have been making much noise about. In quick succession there where 3 major rail accidents (Salisbury July 1906 , Grantham Sept 1906, Shrewsbury October 1907) all blamed on excessive speed. The travelling public had become nervous of such things and even at this early date, the GWR were PR aware enough, that the event would not have been much publicised outside of the timing fraternity.

 

No, I have mistyped most likely!

Edit: I really should have more faith in myself…

 

RCTS 2A.

Edit:

I’m going mad, I’m looking at photographs of the wheel. It cannot be 12ft. All the main LNER secondary sources say 12ft but it doesn’t look like 12ft.

I think I need a lie down. Who’d write a book?!

 

Could it be 12 ft circumference?

Edit. On looking at the above from 2A it does say circumference.

 

12ft circumference, which is 3.8ft in diameter.

 

Given their 7ft 3in drivers, one might expect the LYR Atlantics to be very fast. In his book "British Atlantic Locomotives", Cecil J Allen mentions that a number of legends circulated, including an improbable 117 mph. Allen gives a number of timing logs for the class, with a highest speed of 82 mph at Kirkby on a 200-ton Manchester to Liverpool train.

Allen's logs also quote speeds in the 80s mph for some of the other Atlantic classes, including 86 mph on a Birmingham to Paddington express by one of the GWR's French 4-cylinder compounds, much faster than the 75 mph (120km/hr) legal limit on these engines back in France. As an aside for an engine with 6ft 8in drivers, 86 mph is close to the 6rps/ 360rpm figure suggested in an earlier post as a target maximum speed.

I had a rummage through some of the books on my shelf, looking for evidence of speeds above 90 mph from anything other than City of Truro, but found nothing before 1930, when a step-change began with GW Castles and the acceleration of the Cheltenham Flyer. It seems that although trains speeded up at the end of the 19th Century, 90 mph remained a barrier for some time in the early 20th Century for anything other than GW Cities. I did find quoted speeds of 92-93 mph for GN Atlantics and GW Stars, but only post-1930.

Perhaps I am being misled by my particular book availability. But if maximum speeds did indeed stall for some decades in the early 20th Century, why should that be? WW1 may of course have played a part, as did some accidents such as at Salisbury. Another factor may have been that as gangwayed corridor coaches came into wider use, trains became much heavier and the focus had to be on dealing with the increased train weight. Superheating arrived during this period and of course facilitated both increased efficiency and increased power, but possibly not any speed increase.

 

Thanks both - I thought I was going mad last night! Too much time spent reading the books and not enough time digesting it.

 

This is the point I'm trying to get over with my sketch showing bands. Because times can only decrement/increment in in multiples of 0.2, and because an actual decrement of 0.3 will be measured as either 0.2 or 0.4, the actual speed can be reconstructed in a number of ways, steady, reducing or even increasing acceleration. And the actual acceleration will also be affected by gradient, track curvature, cross winds and so on. A steady acceleration and increasing gradient will result in increasing acceleration and so on.
It's all part of the mystery. If you accept Rous Marten`s published figures then the train certainly exceeded about 98 or 99mph and most likely 100. If they had continued they would almost certainly have exceeded 100mph if the gradient continued to fall — or else piled the train up on a bend, broken rail, or something else.

 

Churchward's rumoured quote of “Withhold any attempt at a maximum speed till I give the word — then you can go and break your b— necks“ is perhaps relevant. Maximum speed was dangerous, bad for the track, bad for the locomotive, bad for coal consumption, very bad for the fireman's back, broke the speed limits...
Thinking of all the motor vehicles I've driven, I think I've attempted maximum speed on very few and on very few occasions. Apart from the first, which had a maximum speed of around 60mph and was thrashed unmercifully on motorways, I don't think I can recall a single situation where I actually hit maximum speed without backing off first, for example when I discovered the M3 has corners...
Maximum speed isn't really useful for anything except maximum speed records...

 

Sort of between the points I was trying to make. The GWR were aware of Rous-Martin's claim (to the point they tried to prevent the specific details being published), at that point they then had the opportunity to do something to authenticate it. Nothing was specially prepared for CoT's run on that day, if they accepted 100 happened on that event with no prior planning, I would be wondering what we could do with an empty train and a track clear of permanent way staff and proper preparation. Still, there probably are many valid reasons why they might chose not to pursue it. Excessive speed was/is bad, but the railway companies at the time were still selling 'speed' in the form of express services to the public. Someone clearly wanted to get somewhere quicker then they had done before, or there wouldn't be the continuous investment in faster services right up to the second world war. July 1904 also saw the first running of the Cornish Riviera Express, the GWR didn't hide the fact that it arrived nearly half an hour quicker than the option before.

Yet to this day individuals and companies sink literally millions into pursuing them. For someone, it's always going to be a target worth aiming for.

 

According to R.A. Riddles and others Coronation's 114mph was recorded from the locomotive speedometer not the dynamomoter car which I don't think was on the train. Note however the comment about all the stopwatches getting 112.5; not much variation there.

 

You're right, the dyno wasn't in the train formation. Just got a bit carried away demonstrating that technology can also be deceiving.