Flying Scotsman Speed

In the book Speed on the East Coast Main Line by P Semmens there are two graphs showing the speed achieved by 4472. Both would have been created by hand from the time and distance dynamometer car roll (marks made every second on paper passing at 12" per mile). Oddly they show different speeds during the descent from Stoke and as far as I can see neither shows a speed of 100 mph. One clearly shows 98 mph, the other 99 mph when you look closely.

As far as I know the original dynamometer roll no longer exists. Would be interested to know if anyone knows better or if anyone has a better quality copy of the speed graph.

 

Hence some people think Papyrus was the first loco to unequivocally do over 100 mph.

 

C. J. Allen did not accept the 100mph figure in 1934 with Flying Scotsman. 'The Dynamometer car chart showed a marked but rather un-natural peak of exactly 100mph', according to O.S. Nock, based on the Allen write up in Railway Magazine at the time.

One could argue Flying Scotsman's claim to be first at 100mph with a dynamometer car is as uncertain as the stop watches used by Rous-Marten and post van timekeeper with City of Truro 30 years earlier.

I could go into this in a lot more detail, but suspect it would get very boring.

Cheers,
Julian

 

It seems we can never know!

Personally I think it extremely unlikely that 4472 was the first loco to reach 100. Papyrus seems to be the first to unequivocally reach 100 while it would seems very probable that some loco had previously reached 100.

Possible candidates include PRR no 460 and Lady of Lyons. City of Truro is perhaps slightly less likely to have reached 100 but more likely to have been the first than Flying Scotsman as there are fewer earlier claims (the earliest of all, though very unlikely, actually being on the GWR!).

 

Locomotive maximum speeds are pretty pointless in most cases. There is no minimum loading requirement so just stick a single, light, free running carriage on, and also no requirement to repeat the speed in the opposite direction.

That said it is fascinating, especially with steam locos where the crew must use their skills to push the engines and perhaps more importantly hold back in preparation.

It'd be interesting to work out locomotive speed vs loading and taking into account gradients and curves, if somehow possible. I suppose that'd be how much power at speed there is...

 

I think one of the favourable features of our national character is our interest in pretty pointless things - speed records, climbing Everest etc.

Power at 100mph as a % of power at 60mph would be an interesting way of comparing locomotives of different sizes for speedworthiness.

 

I know it has been been long thought that 4472 was unlikely to have reached 100mph based on CJA's stopwatch timings and the peculiar bump in the dynamometer speed curve (also seen with Papyrus and Mallard). However the point I'm trying to make is that (as far as I can see) the speed curve also shows <100mph - ie there is no evidence from any source of 100mph being achieved.

It is unfair to bracket 3440 and 4472. The evidence for 3340 still exists and points at 100mph, the evidence for 4472 doesn't exist (we don't have the original dyno roll and I don't think we have CJA's stopwatch timings) and what there is points at <100mph.

 

Cecil J. Allen was invited to record the 1934 run. He was on the payroll of the LNER yet disputed the Dynamometor results with a slight 'blip' of 100mph. He also disputed the later LMS record Coronation run. The parallels with City of Truro are well founded.

Modern computer simulations add a great deal to all these debates.

Cheers,
Julian

 

Engineering formulae in the hands of railway enthusiasts - what could possibly go wrong?

I've never seen a convincing disproof of 3440's claimed speed - if you have one it would be interesting to see.

best regards

David

 

Hi David,

As a GWR fan, I am unlikely to disprove the City of Truro claim of 100mph. I think I have read everything ever written on the City of Truro record run, including, thanks to the internet 'The Engineer' magazine report from Graces Guide of the original Rous-Marten report (which is quite revealing and fills in lots of gaps, and creates more problems than discussed by later commentators). I think I have all the primary source info including the original newspaper articles and the later stuff from Collett and Inspector Flewellen.

There does not seem to be any doubt that City of Truro topped 100mph, though reading Rous-Marten's original reports one is left with a considerable feeling of inaccuracy plus error as to when the train was braked after the descent.

Flewellen was also on Lady of Lyons (Saint class) in the 1906 allegedly high speed light engine run, with no less than Collett with him on the footplate. Collett's own revelations many years later in print about both runs are very important , as were Flewellen's.

Cheers,
Julian

 

just in case you haven't seen it before - this is Rous Marten's first published account of the speed - a year before he mentioned it in the Railway Magazine

 

Must say I'm inclined to agree, the LNER were very good at publicity. I do not doubt Mallard was fastest, by 1/2mph, the original Dynanometer recording shows 125mph for a period, but a 'momentary maximum of 126mph on the speedo' ? Pull the other one!

 

Gresley according to a number of sources recognised 125mph as the top speed, not 126mph. That was what was given on the plaques when Mallard was partially overhauled and repainted into garter blue for the exchange trials. If 125mph was good enough for Gresley, I'm satisfied with that - but driver Joe Duddington always recognised the higher figure of 126mph and the recording of him describing the run is still awe inspiring.

I feel we must have some balance - there's been a number of very reasonable articles in the railway press over the years and every single one to a man has said City of Truro went very fast - probably mid to high nineties - but didn't reach 100mph. Even accounting for human error with Rous-Marten, there was the question of whether she could physically be capable of doing it. I'll try and dig the article out I remember best - it was a great read. I remember coming away very impressed with the engineering behind the City Class loco but not convinced it was capable of 100mph. It looked to me to be a close run thing though!

Flying Scotsman is a different kettle of fish. The dynamometer car results have a lower margin for error (if correctly calibrated) and at the time Scotsman achieving this result was accepted because of the dynamometer records. We no longer have the original roll I believe but if we're happy to accept Mallard's record where we do have the same type of roll used (being used in the same vehicle as it happens) then I think for consistency you have to accept Scotsman probably did touch 100mph. For how long - not very long it looks like.

Either way I think we can all agree both CoT and FS were two very fast steam locomotives and deserve to be remembered for that - the questions over both of their claimed records is part of their overall myth and allure and in the case of CoT it is why the LNER felt she was worth saving for her York museum.

 

I think people are surprised that such an antique locomotive such as 3440 could get any where close to 100mph, but when she returned for the 100 year anniversary I understand drivers were quite surprised as to how free running she remains. I think it is clear she achieved a remarkable speed on that day. Thank goodness the lner had the forethought to save her.

As an aside get a free running locomotive and a decent crew it is a amazing what is possible. I went on 6024 over shap in 1999 and we got stuck, on the return we hammered up the bank and flew down, we reached around 85mph when the brakes were applied.

 

Don't forget though that boty City of Truro & 4472 are now significantly different to what they were when they did 'The Ton'

 

I am hoping that I live long enough to see Mallard re-enact the 100th anniversary of its famous run in 2038. Given the will, the money could be found.

 

CoT may be an antique, but, I assure you, from personal experience on a couple of photo charters, on the SVR, she is a delight to drive, very free - running, smooth - riding, and has very long legs, (not surprising with 6' 8" drivers.)

100mph? No idea, but it would not surprise me. A lot of engines of that era, and earlier, had surprising turns of speed. Read up on the LNWR George the Fifth class's exploits. We look at them now, as antiques, but they were strongly and beautifully - engineeredand, as an ex - Stourbridge fireman once said to me, (about 45110), "Ar, these was built to goo."

They was.

 

We wil never know for sure how near City of Truro got to 100mph, but one run from the same era which makes me think she could have got pretty close was the record-breaking London-Brighton run behind LB&SCR B4 4-4-0 No. 70 Holyrood in July 1903, which touched 90mph at Horley. The B4s were not one of the greatest 4-4-0s, by any means and not a patch on Wainwright's D class which the neigbouring SE&CR were building at the time, but no one has ever questioned this top speed.

If a rather indifferent 4-4-0 could get up to 90mph on a 1 in 264 downhill gradient, surely one of the most technically advanced 4-4-0s of the time could go somewhat faster down a 1 in 80. How fast, of course is another matter, although thanks to Courier for reproducing Mr Rous-Marten's account of the run, which makes for fascinating reading.

 

Some of the deconstructions of CoTS run are very strange indeed. There's one on line one in which the writer takes C R-Ms figures up to the top of the hill as gospel, and then completely and utterly discards the rest of them to produce a figure in the low 90s based almost completely on a whole host of assumptions, apparently utterly unaware of how much changing any one of them would throw his calculations out... The logical flaw in taking some CR-M's figures as gospel, and completely discarding others seems to evade that writer completely! There are enough other timings of locomotives of that period in the 90s that you must either assume that all timings are worthless, and discard them completely, or else consider CR-Ms numbers as being what his stopwatch said.

A more rational analysis of any record, COTS or FS' starts with understanding the limitations of the timing apparatus. CRM's stopwatches were effectively digital devices, with an absolute limit of accuracy. An 8.8 second quarter mile was not a speed of 102.3mph, but a speed of between 101.1mph and 103.4mph. Those are the limitations of the device. You then have to consider what potential there was for error in those timings, but I've written this before. But any analysis that starts with the assumption that CR-M, who had done this many many times before don't forget, effectively made up his figures, is not in my opinion particularly worthy of serious consideration.

FS, on the other hand, was a dynamometer car trace. This is almost certainly genuinely an analogue device, recording an instaneous speed (assuming its a wheel connected mechanically to the needle). However you do need to look at the original traces, the redrawn version for a magazine won't help at all. Then you have to understand how well the device was calibrated. We can be pretty confident that the LNER had not repeatedly calibrated the car at 100mph, so there are all sorts of potential problems there. Just because it was accurate at 50mph does not mean it was necessarily accurate at 100mph, and of course under reading is just as likely as over reading (could it have been skidding on the track a little?). Same is true of Mallard and the German locomotive: how well was their instrumentation calibrated? Its ironic to consider that the only device that could be readily calibrated for its (limited) performance at a new record speed was CR-Ms stopwatch!

 

I gather there were also some pre-grouping era allegations of "the ton" had been achieved by an L&YR Atlantic (courtesy of WA Tuplin, I think?) to add the various other claims.

As has been commented elsewhere, the chances of 3440 being preserved without this "claim to fame" would be probably nil, so I am just glad it served as an excuse to preserve the loco.