Alstom and Eversholt to hydrogenate class 321s

[whole diary]

[stuving]

It occurs to me that I do have personal experience of a hydrogen fire - the largest ever in Britain, I'm sure - and explosion. Not very close experience, I'm happy to record, it just woke me up at 7 am one Sunday (22 March 1987, in fact). I was 5 km away, so it was a loud bang - it was calculated as equivalent to 90 kg of TNT - but was purely mechanical (i.e. no chemical reactions involved).  A large pressure vessel (20 t, holding 66 m³) designed to operate at 9 bar, and withstand 20 bar, was filled to 50 bar and so burst - one piece weighting 3 t was hurled 1 km.

The hydrogen and petroleum liquids in the tank all escaped, obviously, and caught fire. Burning the hydrogen alone gives about 60 times as much energy as its explosive depressurisation, yet the HSE^▸ (Health and Safety Executive) describe this as a "fireball" and do not reckon it caused much of the damage. That it, basically, because it has to mix with air before it can burn, that takes time. And, crucially, hydrogen is so light (7% as dense as air) that it shoots upwards at once and so burns at height. The same is true of the continuation of that fire fed by all the hydrogen in the hydrocracker which rapidly escaped. Five hours later, the escaped liquids sitting on top of water flooding the site caught fire, and that did do a lot of damage.

I would at this point refer you to an HSE summary report on this, covering also two other accidents (both fires) at BP^» (Beyond Petroleum (Former name - British Petroleum) - home page) Grangemouth in 1987. It was produced as a warning about complacency, as BP were thought (by themselves as well) to have a good safety culture. But HSE have removed it from their web site, and I can't find it anywhere else - PM me if you must have a copy (I know we have some accident report aficionados on the forum!).

[Red Squirrel]

Would this be a good time to reiterate that hydrogen tanks in vehicles tend to run at about 550 bar? For comparison, the boiler pressure of a 'King' is around 17 bar. I wouldn't want to be sat on top of either if it went off...

[Bmblbzzz]

Going off on a tangent for a minute, this thread has prompted me to look up the relationship between bars and pascals, something I've never really known (or had much reason to know). The answer makes me wonder why?

[stuving]

Going off on a tangent for a minute, this thread has prompted me to look up the relationship between bars and pascals, something I've never really known (or had much reason to know). The answer makes me wonder why?

The base SI unit of pressure is N/m² - it just is - and got named a Pascal. The name has a rather restricted, use, however, and kN/m² and MN/m² are still more common in some fields (e.g. materials and structures). The bar is just an atmosphere, rounded to a handy multiple of the Pascal.

Bar and mbar are not proper SI units, and are frowned upon by purists. Thus while our Met Office use mbar for atmospheric pressures, Météo France insist on hPa - hectopascals. Same difference, obviously, and in this case neither is really instinctive; you need to know what it is. For rainfall there is a similar cross-channel difference of usage, where we always use mm for total rainfall (except in drainage calculations). But in France you are as likely to hear "litres par metre carré" (l/m²). Again, it's the same size, but I don't think this one is as intuitive - unless maybe if you are a hydrogéographe.

[Red Squirrel]

Well you could just as well ask yourself 'why pascals', when Nm^-2 is exactly the same thing?

Just to add to the confusion, a bar (at 1x10⁶ Pa) is just a little bit less than a standard atmosphere (101325 Pa), which suggests it is neither fish, fowl nor good red squirrel herring. Even meteorologists, for whom you might have thought the unit was invented, prefer to use it in its reduced form as the millibar - which is the same as a centipascal...

Apparently, in the olden days they had units of measurement where things didn't all multiply by 10 - imagine how confusing that must have been!

(post crossed with stuving...)

[Red Squirrel]

For rainfall there is a similar cross-channel difference of usage...

...and for fuel economy; in Britain we have a unit based on distance/amount of fuel used (e.g. miles/gallon or miles/kWh), whereas in most other places they talk in terms of the amount of fuel needed to travel a certain distance (e.g. litres/100km or kWh/100km).

[stuving]

Well you could just as well ask yourself 'why pascals', when Nm^-2 is exactly the same thing?

Just to add to the confusion, a bar (at 1x10⁶ Pa) is just a little bit less than a standard atmosphere (101325 Pa), which suggests it is neither fish, fowl nor good red squirrel herring. Even meteorologists, for whom you might have thought the unit was invented, prefer to use it in its reduced form as the millibar - which is the same as a centipascal...

Apparently, in the olden days they had units of measurement where things didn't all multiply by 10 - imagine how confusing that must have been!

Well, loads of derived units have names - I've never been sure if it's really a worthwhile simplification (it does also conceal the relationships between units) or if it's just "unit names for the boys". But if you are going to be writing pressures by hand on charts, transferring them by phone, and quoting them over the radio, avoiding decimal points is a very good idea. As an aside, I've always written statistical index numbers such as RPI^▸ (Revenue Protection Inspector (or Retail Price Index, depending on the context))/CPI (that come with a single decimal place as a percentage) as "per mil" just so as to lose that annoying (and literally pointless) little decimal point.

In the case of atmospheric pressure, the true average value really doesn't matter at all. For one thing it's not the same everywhere, if you measure it accurately enough. And it's not what you observe on any day. It's not even directly observable - you have to take loads of readings and process the numbers. So in this case having a nominal value that's within a few percent is all all you need.

Oh, and it's 10⁵, isn't it - which is why a common SI multiple doesn't work, and people either resort to non-tridecimal hecto or a new unit.

[Red Squirrel]

Oh, and it's 10⁵, isn't it...

Yes it is. Well done stuving; I was just checking that you were listening at the back there!

[Bmblbzzz]

Oh, and it's 10⁵, isn't it...

Yes it is. Well done stuving; I was just checking that you were listening at the back there!

And isn't a millibar the same as a hectopascal, not a centipascal?

[Bmblbzzz]

I'm not a scientist, so I don't use SI*, and I'm not an engineer, so I don't use many of the more involved or derived metric units. In fact I don't even use bar – I see them marked on tyres but I think of tyre pressure in terms of psi and my most used weather forecast (yr.no) gives pressure in hectopascals. I just thought if 1 bar = x number of pascals, why not do it all in (hecto/kilo/etc)pascals?

*The specific SI units as distinct from general metric.

[Red Squirrel]

And isn't a millibar the same as a hectopascal, not a centipascal?

Yes. I'm just off to the garden to see how my cabbages are doing...

[Bmblbzzz]

And isn't a millibar the same as a hectopascal, not a centipascal?

Yes. I'm just off to the garden to see how my cabbages are doing...

Those aren't cabbages, they're sprouts! 

[CyclingSid]

An academic view:
https://theconversation.com/hydrogen-trains-are-coming-can-they-get-rid-of-diesel-for-good-110450

[bradshaw]

A new report by the Institute of Mechanical Engineers suggest that the way forward in a rolling electrification programme. Hydrogen more suited to branch lines where electrification is a problem.
Hydrogen less efficient than electrification.

https://www.railwaygazette.com/news/traction-rolling-stock/single-view/view/imeche-recommends-electrification-instead-of-hydrogen-trains.html

[stuving]

A new report by the Institute of Mechanical Engineers suggest that the way forward in a rolling electrification programme. Hydrogen more suited to branch lines where electrification is a problem.
Hydrogen less efficient than electrification.

https://www.railwaygazette.com/news/traction-rolling-stock/single-view/view/imeche-recommends-electrification-instead-of-hydrogen-trains.html

Of course that should be "A report by Railway Gazette of a new report by the Institution of Mechanical Engineers...". When you look at the headline in RG:

IMechE recommends electrification instead of hydrogen trains

and compare it with the headline IMechE have on their site:

New report: Hydrogen trains needed to eliminate harmful emissions on non-electrified lines

Institution calls for urgent action to introduce hydrogen trains in areas outside the electrified rail network

So, is that misleading reporting by RG? Maybe not, in this case. Obviously the IMechE is pushing both electrification and hydrogen, as the objective is to replace diesel. The question is whether their stance on hydrogen (vs electrification) is a half-full or half-empty endorsement.

In that report itself, the conclusions are:

The Institution of Mechanical Engineers recommends:

1. That the UK^▸ (United Kingdom) Government rethinks the cancellation of electrification programmes and moves forward with a more innovative, and long-term approach, electrification rolling programme, that can create skills and careers, develop supply chains, and work with existing rail networks to manage projects.

2. That the industry encourages the development and deployment of hydrogen trains and their fuelling and servicing facilities. Creating and supporting demonstration lines and trains will help to de-risk the technologies and servicing relating to hydrogen fuels and trains.

3. That hydrogen train technology is developed in industrial areas where hydrogen production already occurs, and can support the wider transport system. For example, as well as local trains, local hydrogen buses could be refuelled at an industrial site, and hydrogen could also be pumped into the gas grid to help decarbonise heat. Both the North West and the North East could support test beds. These test beds will support knowledge sharing across sectors, providing cost reductions in hydrogen fuel.

And, on the specific question of how much to electrify, they say:

Nevertheless, there is a concern that hydrogen trains will be used by funders as a reason to avoid future electrification. This fear should be eliminated at source, by ensuring a universal understanding that fuel cell traction should be viewed as an option only where long-term technical, environmental and/or economic factors make electrification a poor option. These relate to frequency of use, remoteness from electrical supply and physical constraints (including those of freight yards). On this basis, most lines in the UK should be electrified. Even the frequency of use measure is a grey area, specifically with respect to infrequently served branches of electrified mainlines.

PS: One thing which does look odd in the report is the pie chart of uses of hydrogen - 90% goes on "food industry" and "others", and only 1% on making ammonia or oil refining.