Scaling brilliance

August 13, 2016

Bombe detailIn July I visited Bletchley Park and was impressed, as so many before me, by the bombes (mechanical calculators – not computers) and the brilliance of the code-breakers, Turing among them.  I was also massively impressed by the community of volunteers who have rebuilt demonstration bombes in the absence of drawings or of parts – all were destroyed after the war.

But actually, I was most impressed by the insights of a man called Gordon Welchman.  He did have his moment of cryptoanalytic brilliance in designing the diagonal board – a key part of the system for decoding Enigma messages.

But his greater contribution, in my eyes anyway, was his realization of the need for scale and for joined-up operations.  Linking together the intercept stations, the understanding of the radio networks and what they said about organizational structures, the multiplicity of bombes, their manufacture and operation, the deciphering of messages, and the extraction of insight from the results.  And he got his head around this early enough to have the infrastructure in place before the inevitable explosion in volume and complexity of traffic.  3000 motorcycle dispatch riders per day, plus teleprinters delivering messages.

A system capable of managing this much data, decoding it, indexing it and then disseminating the information was a remarkable achievement that was hidden in plain sight of all involved.

(And I enjoyed his book “The Hut Six Story” in which it is all detailed. However it is a shame that the publishers saw fit to remove, from the 2016 edition, Welchman’s call to action for the future, replacing it with pages of minor bickering about details of history.)

Advertisements

Last weekend, another brilliant Goodwood Revival. And in what has become a pattern of the last few years, I happened upon some technical marvels from the past.

Peugeot radial 150920Like the 1930s Peugeot motorcycle with the (modern model)Radial cyclinder 150920 radial engine, 100cc and 9 cylinders, each with its tiny separate exhaust.

.

.

.

.

Hilborn 150920And the Hilborn mechanical fuel injection on the Sadler Chev from 1957. Fuel pump driven from engine speed so you get the variation in fuel pressure that you need – and you use the size of the orifice in the return pipe to the fuel tank to manage your metering. Very neat. But best left for race applications where most of the time you’ll be at wide throttle openings!

.

Lycoming 150920The Lycoming aero-engined race car from the mid-50s? Now there’s a unique car, with more information at: http://classicdriver.co.nz/the-lycoming/ explaining that there was more innovation than just the engine.

.

.

So, another weekend not only of wonderful racing, but also of innovation in action.

A couple of days ago visited a brilliant museum celebrating a maritime dynasty of commercially energetic Scots and a wonderful walk back through time (www.scottishmaritimemuseum.org/sites/dumbarton/). William Denny, in 1882, created the first commercial towing tank for his shipyard with exquisite clarity of purpose; “to determine with commercially acceptable accuracy the power required to achieve the contract speed, and to reduce that power for any installation to a minimum.”

Associated with feathering paddle wheels and 20 kt paddle steamers, helicopters (development of airscrews in 1900 to a flying tethered prototype in 1913), America’s Cup yachts (Shamrock II, 1901), the Irrawaddy Flotilla Company (with ~600 river boats in Burma in the 1930s), fin-stabilisers from the 1930s to the 1950s, and hovercraft in the 1960s, this commercially enterprising family company and its employees must have been one of the world’s most innovative companies in the UK before succumbing to commercial pressures in the early 1960s.

Integrator In addition to the test tank, the museum includes precision machinery, such as the model cutting machine invented by Froude,  and instrumentation such as the all-mechanical dynamometer to measure tow-tank model forces, and a pallograph (a mechanical device to mechanically measure vertical structural vibration, also compensating for ship roll).  The  drawing office includes tools of the trade such as cylindrical slide rules and a mechanical integrator, based on the principles of a planimeter, for calculating moments of area from drawings (see photo).

The whole place is a wondeful insight into engineering and commercial innovation from a world before electronics.

Who needs sparkplugs?

July 24, 2015

P1070701 (Who needs sparkplugs)

Have a look at the photo above – a 1908 Mors Grand Prix car that was at the 2015 Goodwood Festival of Speed. Spot anything odd? Well, you can see the insulated conductors that obviously are attached to the spark plugs (far left and far right), right? But what are the funny little lever arms just to the right of the left hand “spark plug” and to the left of right hand “spark plug” and why do they both have a connecting arm downwards?

This, believe it or not, is a make-and-break that is actually inside each cylinder to create a spark from the low-tension ignition system. Each driven by a cam below and a push/pull rod. Actually there are two camshafts, one for starting and one for running, so the moment it starts you jump out and run to the side of the engine to pull a lever that slides in the operating cam to replace the starting cam. Who needs advance / retard?

Oh yes, and the little cup visible on the top of the right hand cylinder? Well, you fill these cups on each cylinder with fuel and turn the tap, so dispensing a known quantity of fuel in before starting. Four cups for cold and 1.5 for a hot start. Crank it – and stand well back.

Don’t you love the experiments in early automotive engineering? (BTW – the spark plug had already been invented and I’m not sure why Emile Mors didn’t adopt it.)

And many thanks to the friendly and tolerant team from the Revs Institute (https://revsinstitute.org/the-collection/1908-mors-grand-prix-car/) who were working on the car and happy to explain obscure technical details.

Condensing the message

October 26, 2014

How do you summarise the message into the one-line essence that will get attention? While still having all the detail to support your proposition?

“The Pyramid Principle” by Barbara Minto was the original bible on this – but last week I saw an advertising poster on the London Underground which turned it into an art form.

From left to right was a series of messages –

  • “My train’s practically here, make it snappy” – a one line pitch.
  • ”My train’s a couple of minutes away, tell me more. But don’t bore me.” – two sentences about how it works.
  • “My train’s who know where, hit me with all you’ve got” – more detail of the benefits

And so it went on, each column adding more detail. And culminating in a call to action “OK, I’m intrigued, what now?”

And clever to acknowledge the context of place and reader .

A nice ad and a nice illustration of getting a message summarised.

Hive ad 141013

(Sorry about the quality of the photo – but the Tube during rush hour isn’t a place to dawdle)

Early IP management

September 17, 2014

P1070375Now here was a company that was clear about the value of its intellectual property.

P1070376

(Spotted at the Goodwood Revival last weekend)

I’m intrigued by early ‘automatic systems’ and this one seems particularly anachronistic. It is a water-level control system on the Crinan Canal (visited during our last holiday). In essence, a lever arm that raises and lowers a ‘plug’ from a hole in the floor of the canal – it’s been operating for more than 120 years. (See diagram below). The fact that the ‘sensing’ point is 30 metres downstream doesn’t seem to me to significantly influence the dynamics of water level. But I still can’t see why a weir wouldn’t be simpler, cheaper, more reliable and fail-safe. (There is in fact a weir discharging from the canal to the sea a short distance away). Maybe I’m missing something? On the other hand, despite a web search I can’t find evidence of another one of these devices so perhaps nobody else could see the point either.

Water waster poster 140809

 

BTW – a good video about the history of the canal can be found at www.islayblog.com/2014entries/20140106-interesting-crinan-canal-video.shtml

 

CruachanOn holiday last week, visited the world’s first high-head pumped-storage hydroelectric power plant at Cruachan in Scotland. Proposed by Edward MacColl in the late 1930s and built between 1959 and 1965, it’s a seriously impressive bit of kit. 440 MW used for peak-lopping and can go from standby to full power in 28 seconds (with turbines already spun up). What impresses me most though is their claim of a pumping efficiency of 90% of generating efficiency. That speaks to some interesting fluid and electrical machinery designs to meet such different requirements. 250 tonnes of rotating machinery in each of the four units. And very high levels of environmental impact management given their proximity to a fish farm. Deeply impressive

(Image from Argyll and the Isles Tourism – cameras not allowed on site.)

If you think that pure frequentist stats are only a starting point and that Bayes was right (and I do) and if you struggle to remember Bayes’ Theorem represented as a set of probabilities (and I do) then you’ll really appreciate the messages laid out very clearly in Gerd Gigerenzer’s book “Reckoning with Risk”.

He makes and supports a number of compelling points:

  • We live in a world of uncertainty – get used to it.
  • In such a world it is important to communicate effectively (and to understand accurately so that you don’t get exploited) what these uncertainties really mean.
  • There are a number of different ways of representing (to understand and to communicate) uncertainty – and natural frequencies have a natural and intuitive immediacy that gives them a huge advantage
  • And, by the way, Bayes becomes obvious when you use natural frequencies.

Brilliant book, just finished it and highly recommend it.

When fear blocks value

March 5, 2014

Went to the Innovation Forum’s “Innovation Leaders Conference” (sic) (www.inno-forum.org/#!conference/c1u75) last week.  One of the most striking things I heard, from Derek Jones, CEO of Babraham Bioscience Technologies (www.babraham.co.uk/), crystallised something I’ve seen several times.  Indeed, only earlier that week I’d seen one technology stuck in academia while another launched successfully due solely to their owners’ different views and behaviours.

In essence, Derek’s point was, isn’t it a pity when a fear that a technology may not be valued correctly leads to delays, lack of timely action, or more aggressive blocks that then prevent it being used at all.