Science desk
< May 22	<< Apr \| May \| Jun >>	May 24 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

May 23

Agatha Christie's fireman

Suppose a large passenger steam locomotive (I'm thinking a 4-8-4, like our FEF-2 or the Russian P-36) was stuck between stations for some reason (maybe snowed in somewhere on the Trans-Siberian Railway or in the Donner Pass on the Central Pacific line) -- assuming that it's not stuck in a tunnel (which would be an immediately life-threatening emergency), how long can a typical locomotive of this sort be kept in steam before it runs out of either coal and/or water? 2601:646:9882:46E0:D9BE:FAE8:F7ED:9973 (talk) 03:25, 23 May 2023 (UTC)[reply]

Did the locomotive come with a full tender? Then the answer depends on the capacity of its fuel bunker – if it is snowed in, the snow can be used as a virtually unlimited water supply. The next unknown is how well the barrel is thermally isolated. The water has to be kept at about 100 °C while the outside temperature is about 0 °C. The heat loss has to be compensated by the heat of the burning coal. --Lambiam 04:34, 23 May 2023 (UTC)[reply]

Yes, assume that the tender is (initially) almost full. 2601:646:9882:46E0:D9BE:FAE8:F7ED:9973 (talk) 06:39, 23 May 2023 (UTC)[reply]

Sorry to disagree with Lambiam but it only has to be kept warm enough to stop the boiler freezing. 10°C ought to be enough for long term survival. However this is throughout the boiler, so at the firebox end it would need to be warmer. Once the fuel runs out then the boiler would need to be emptied via the blow-down valve to avoid damage, but then you have a "dead" locomotive needing a live one to pull it. Of course if you do drop some or all of the boiler water, then the running gear gets frozen solid, which it will do anyhow. See Snowdrift at Bleath Gill. Martin of Sheffield (talk) 08:07, 23 May 2023 (UTC)[reply]

We appear to have a different understanding of the concept of a locomotive (or actually its engine) being kept in steam. --Lambiam 16:25, 23 May 2023 (UTC)[reply]

OK, fair point, I will admit I was thinking about simply ensuring the loco could be brought back to a working condition ASAP. If you want to ensure there is steam available, then next question will be at what pressure? Enough to move? Enough to work the brakes? Martin of Sheffield (talk) 21:31, 23 May 2023 (UTC)[reply]

Well, since you put it this way, "in steam" here means with enough steam to work the blower, mechanical stoker (any engine as big as the one I described would certainly have one), injector, brake pump, dynamo and train heat, but not necessarily enough to move. 73.162.86.152 (talk) 04:25, 24 May 2023 (UTC)[reply]

Depending on how much hope you have to get unstuck soon, you may want to keep some pressure in the boiler. If you can operate the blower, restart time can be cut from half a day to an hour or so, it may be nice to be able to run the steam heating of the carriages (I think that on Hercule Poirot's Orient Express the heating was shut off) and if you don't regularly operate the injector, the water in the hose connecting the tender to the locomotive may freeze, giving you a dead locomotive too. The water in the tender will freeze anyway, but this may take a few days. PiusImpavidus (talk) 09:24, 23 May 2023 (UTC)[reply]

No, I read the book fairly recently and there's nothing about the heating being off. Not in the movie either. You might be thinking of The Lady Vanishes, where all the characters move into an overcrowded hotel for one night for that reason. --142.112.220.184 (talk) 23:53, 23 May 2023 (UTC)[reply]

A different version maybe. I've seen two movie versions, one tv version and read the book a long time ago; I may be confusing some versions. PiusImpavidus (talk) 08:52, 24 May 2023 (UTC)[reply]

Yes, but there's only one the movie version. --142.112.220.184 (talk) 15:45, 26 May 2023 (UTC)[reply]

OK, I've done a back-of-the-envelope calculation. Assume the boiler+firebox is a cilinder, 10 m long, 2 m diameter, excluding the smokebox. That gives a surface area around 65 m². With an insulation 15 cm think with thermal conductivity of 0.04 W/(m·K) and a temperature difference around 140 K, you loose about 2.5 kW of heat. Burning 1 kg of coal gives about 34 MJ of energy, a bit less if the coal is wet, and at least 5 MJ of that will leave the chimney as smoke, so let's say you get 25 MJ/kg burning coal. Conclusion: if you disable all steam-powered appliances and only have to counter heat loss through the walls of the boiler, you need less than kilogramme of coal per hour. With several tonnes of it in the tender, you can hold out for ages. I wouldn't be surprised if the insulation is actually thinner, given that this is such a minor heat loss. The limit will be your ability to prevent the water in your tender from freezing. Has the tender been equipped with tender water heating?

BTW, such big passenger locomotives were very rare in western Europe, where almost all steam locomotives were manually fired until the most demanding lines switched to electric traction in the interbellum. In eastern Europe, steam was used longer and in America, everything is bigger (as we say in Europe). PiusImpavidus (talk) 16:56, 24 May 2023 (UTC)[reply]

OK, so I guess we can safely call conductive losses negligible! However, shutting down all steam-powered appliances would not be an option in this scenario -- at the very least you need to run the mechanical stoker to keep the fire burning, and (at least occasionally) the injector to keep the boiler topped off, plus the blower to maintain a draft on the fire in the absence of exhaust steam from the cylinders! Plus, you would probably want to run the train heat so as not to make things miserable (or possibly outright deadly in the case of the Trans-Siberian) for the passengers, the dynamo (also for the sake of the passengers, as well as for safety at night and for any communications with the outside), the tender water heating to keep the water from freezing (and to melt the snow if that is to be used as a water supply) and the brake pump (which might be going at full blast if the train had been stopped by a passenger pulling the cord -- one possible scenario being that a passenger stopped the train because he/she had witnessed a murder aboard, and the train cannot be moved again because the killer had thrown some of the evidence outside so the murder has to be solved first). So, how much energy would it take to keep these things going? 73.162.86.152 (talk) 03:39, 25 May 2023 (UTC)[reply]

You don't really need the mechanical fireman. Every locomotive can be manually fired too (how else would you perform a cold start?) and if you really burn coal so rapidly that you need the mechanical fireman, you'll make so much steam that you'll run out of water after one, maybe two hours and out of coal within ten hours. This changes a bit on an oil burner, but even then the oil pump would consume a very small fraction of the energy released by the oil it pumps. The injector is no user of energy as all energy needed to operate the injector is forced back into the boiler along with the feedwater. It lowers the temperature of the boiler, but increases the thermal energy as it increases the mass.

I'm not aware of any tenders with tender water heating, but wouldn't be surprised if that had been used in cold countries. Or you could send the exhaust steam from the dynamo or mechanical fireman into the tender water tank to condense, heating the tank. Safety lights may take a few hundred watts, a portable telegraph connected to the telegraph wires along the track would be battery powered.

The amount of steam used by the blower depends on the amount of air you want to pull through the firebox, which isn't so much with a small fire. How much power does it take to draw 500 litres of air through the boiler every second? Not that much. The brake compressor is needed when releasing the brakes while preparing for departure, not while the train is stationary. And for a longer stop, you would apply the handbrakes. Which leaves the dynamo (if your locomotive is fitted with one; most only had an axle-mounted dynamo and used battery power when stationary) and steam heating. For ordinary homes in winter, heating takes far more energy than lights and for a train, which due to its small width and height has a very large surface area compared to its volume, that will be even worse. So let's assume the energy use is dominated by heating.

Suppose your passenger train is very long, 350 m, or why else would you use such a big locomotive? 60% of the length of the train consists of windows, on both sides, 0.7 m high, for a total surface area of 294 m². Let's use old-fashioned windows: 6 mm of single glass, with a conductivity of 0.2 W/(m·K), and we want to keep the interior temperature 20 K above the outside temperature. That means that about 200 kW of heat is lost by conduction through the windows. There's also some loss through the walls and roof, but those would be much better insulated, and some loss through ventilation. Still assuming 25 MJ/kg of coal, you have to burn about 30 kg of coal per hour to run the heating – which is indeed easily doable with manual firing. It could be a bit worse, as some steam and hot water will leak out of the pipes. With about 10 tonnes of coal in the tender, it'll take some time to burn it all.

Consider collecting the warm water dripping out of your radiators to feed it back into the tender. Also consider moving all passengers to the front half of the train, so you can shut off heating in the rear.

200 kW of heating isn't much for a locomotive designed to provide more than 1 MW of mechanical output at a thermal efficiency less than 15%. PiusImpavidus (talk) 19:31, 25 May 2023 (UTC)[reply]

Thanks! So, 200 kW for heating (or 2-3 times that that if the weather is really cold, as on the Trans-Siberian -- why does everyone assume that the outside temperature is just below freezing?), plus tack on a maximum of 1 kW for the dynamo (I looked up the figure for the P-36, and I believe it would be typical -- in any case it's not much by comparison) and however much it takes to run the blower and heat the water in the tender (probably also not more than a couple kW), and that would be it -- right? (Are you sure that the brake pump won't be running continuously if a passenger had pulled the air just before the stop, though?) And that would mean that barring truly extreme cold (not all that likely even on the Trans-Siberian), the coal supply should hold out for maybe 10-15 days -- more than long enough for the snowdrifts to be cleared, or for the murder to be solved, or for whatever other holdup had taken place to be rectified, and for the train to get on its way again! 73.162.86.152 (talk) 05:04, 26 May 2023 (UTC)[reply]

Your reference to the Murder on the Orient Express suggested the Balkans and it doesn't get very cold there. Places where it does get very cold are in the (Ant-)Arctic or the deep interiors of continents, which tend to be sparsely populated, so not so many trains there. Granted, you mentioned the Donner Pass and the Transsib and the Orient Express didn't use such big locomotives. I'm just more familiar with Europe.

I don't know what services are supposed to run on that 1 kW dynamo. I suppose the headlights. To power the lights in the passenger carriages, it's not enough – at least not with old-style incandescent lights. The carriages may have had their own axle-mounted dynamos and batteries to use when stationary. Nowadays they get power from the locomotive, although backup batteries are still there.

Not all railway air brakes (the German article is better) are the same, but the basic idea is always similar for indirect, automatic air brakes. When pulling the emergency brake, all air is released from the brake pipe and much air is released from the equalising reservoirs in each carriage. When the brake is released later, air is fed from the main reservoir on the locomotive, via the driver's brake valve and the brake pipe, to the equalising reservoirs. The compressor recharges the main reservoir. It's only needed after the driver orders the brakes to release. The brakes on the locomotive and tender may still be applied at that time, as they can be controlled separately.

On some trains, there's an additional air pipe running the length of the train, filling additional main reservoirs on the carriages, so that the equalising reservoirs can be filled from there, instead of from the brake pipe. This allows for faster cycling of the brakes, in particular useful if the brakes don't allow gradual release. Many American trains have such brakes. European trains pretty much always allow gradual release of the brakes. But even then, refilling happens after release of the brakes, not after application.

But the conclusion is indeed yes, coal should last for several days when the train is stationary, assuming it has an express passenger locomotive with a full tender. In most cases, freezing of the tender would be a bigger problem, if you have no tender water heating. The lights in the carriages will go out too, as they rely on batteries (unless you've got a generator car). PiusImpavidus (talk) 18:17, 26 May 2023 (UTC)[reply]

Well, I have a small confession to make -- although I spoke of the Murder on the Orient Express, my question was actually inspired by Criminal Cases (specifically, the 16th case of the 3rd season, "Fast Track to Murder", which does indeed take place on the Trans-Siberian, somewhere between Mysovaya station and Ulan-Ude), so extreme cold could be a factor (but even so, the coal should hold out long enough for "The Bureau" to finish their case and clear the train to proceed!) Electricity would be more of a problem, though -- not only is it needed for lighting (and probably also for cooking), but the detectives also need it to run their HPLC-MS instrument, computers, Starlink connection, electrophoresis kit and the light for the microscope (although in principle all of these could be battery-powered, right?), as well as needing some light to continue working after sundown. But regarding the brakes: if the emergency brake is pulled from one of the passenger cars (as opposed to from the engine), wouldn't the main reservoir still be connected to the brake pipe, therefore draining the air pressure from that as well? 73.162.86.152 (talk) 22:50, 26 May 2023 (UTC)[reply]

Only if the driver's brake valve is still at the release setting. As soon as the brake pipe pressure reaches the normal value, the driver would move it to the hold setting (unless the brake pipe is too leaky). And those driver's brake valves would normally be designed such that if there's a sudden drop in pressure, it would move automatically to hold. Using a pneumatic actuator, that's 1920s technology. Without that, the brakes near the front of the train would apply very slow if the emergency brake was pulled at the rear of the train.

I smell a technology dichotomy: the last bit of the Trans-Siberian Railway was electrified 17 years before the first Starlink satellite was launched. Electric cooking on trains became common with electrification, although there's no one-to-one correspondence. Before that, they would have used gas, oil or coal. Bottled propane/butane works quite well. PiusImpavidus (talk) 11:48, 27 May 2023 (UTC)[reply]

Thanks for the explanation, and I stand corrected re. the brakes -- as for the technical dichotomy, there's a simple explanation, and that is that the train in question is a tourist special using steam traction for the sake of nostalgia (as it is indeed depicted in the "Trans-Siberian Train" crime scene in Criminal Cases -- although in that scene the overhead wires are missing (IRL they would still be present regardless of the particular train being steam-powered), and the engine looks like a 2-8-0 Consolidation of some kind (most likely a Japanese 9600) -- which were typically used for heavy freight, not passenger service, and weren't popular in Russia anyway -- rather than a P-36 4-8-4 or S^u-class 2-6-2 which would probably be used on a tourist special). 73.162.86.152 (talk) 22:01, 27 May 2023 (UTC)[reply]

Some thermal nitpicking (though none of it should affect the conclusion by more than an order of magnitude):

With an insulation 15 cm think with thermal conductivity of 0.04 W/(m·K)... That’s plausible (after all, 40mW/m/K is the rating for cheap glass wool or insulating foam), but optimistic (the materials involved usually age poorly; achieving anywhere close to the rated performance heavily depends on proper installation, which is, let’s say, rare; and even if all is done well, non-insulated parts can ).
Burning 1 kg of coal gives about 34 MJ of energy, a bit less if the coal is wet, and at least 5 MJ of that will leave the chimney as smoke Where did you get that figure? I suspect that comes from locomotive performance, bound by second law considerations of getting work out of heat. However we are talking about heating water, where the lower bound of heating losses is zero.
Let's use old-fashioned windows: 6 mm of single glass, with a conductivity of 0.2 W/(m·K)... If your train is using single-glazing, I’m going to guess the windows are not tightly-sealed either. In that case air leakage in the installation will dominate thermal losses, not conduction through the glass. Also, neglecting the thermal resistance of the air/solid interface (convection) is not valid in that case: if we take a convection coefficient for still air of roughly 5W/m2/K, that is equivalent to 25mm of glass for your value of 0.2W/m/K.

Tigraan^{Click here for my talk page ("private" contact)} 09:44, 30 May 2023 (UTC)[reply]

1 kg of coal needs about 2 kg of oxygen to combust. Taking a somewhat lean mixture, that's about 20–30 kg of air. I forgot the exact number I used in my calculation. I assumed that the smoke as it comes out of the chimney has the same temperature as the water in the boiler. As I said, a crude back-of-the-envelope calculation.

Yes, and I also neglected the effect of wind on cooling. Or rather, I assumed the wind was strong enough to keep the outside of the train at ambient temperature. PiusImpavidus (talk) 10:59, 30 May 2023 (UTC)[reply]

Garden plants' lifespan ?

We have had the same rhubarb, peonies and day lily plants growing in our garden for 30 years. They originated from cuttings taken from plants my mother had. How long can they live? 24.72.82.173 (talk) 15:44, 23 May 2023 (UTC)[reply]

Looks like pretty much forever. This says peonies can last 50 years, but that suggests that they can only be expected to die due to a new disease, or a bad frost or something. The common houseplant Epipremnum aureum is propagated throughout the world by cuttings, and was last observed by anyone to flower naturally was in 1964. Other clonally reproducing plants have been going since the end of the last ice age. Abductive (reasoning) 15:43, 24 May 2023 (UTC)[reply]

Same with rhubarb, I believe. I have known many a gardener to find that rhubarb pretty much comes back every year, forever, whether you want it or not, unless you aggressively dig it out. This piece says that it can last 20 years or more. Daylily#Cultivation also basically says they are a "plant them and forget them" type plant, needing very little maintenance and that they come back every year more or less indefinitely. They are part of the plant grouping known as Asparagales, many of which are hardy perennial bulb-based plants. I have daffodils that I've had for 20 years; they bloom the last week of February/first week of March every year, and have shown no signs of letting up. Other plants from that clade include things like Allium (onion, garlic, leek, etc) also known for their hardiness. I've got wild chives in my lawn that stubbornly grow in the same patches year after year. I don't mind much, because they're green and look like grass, but you definitely smell them when you mow over them. --Jayron 32 15:57, 25 May 2023 (UTC)[reply]

Nested virtual reality in virtual reality

Virtual reality is when a VR player interacts with a virtual environment.

But what if the virtual environment itself contains VR players interacting with a nested virtual environment? Would the VR player then also be able to interact with the nested virtual environment? What about further levels of nesting, where a nested virtual environment in turn contains VR players interacting with another nested virtual environment that may itself contain another virtual environment nested inside, etc.? GeoffreyT2000 (talk) 16:25, 23 May 2023 (UTC)[reply]

If the simulation is sufficiently realistic, players should be able to play any number of games: operate a pinball machine, play Space Invaders in an amusement arcade, and play Half-Life: Alyx using a simulated VR headset. None of this is automatically the case; if the game developers have not catered for one of these potential possibilities, it will not be actually possible. This applies likewise for multiplayer games such as Population: One. It is theoretically possible, and I think practically feasible, to design a multiplayer VR game, say Droste, in which the players can start playing, among several games, a nested instance of Droste. --Lambiam 16:45, 23 May 2023 (UTC)[reply]

There is a serious scientific hypothesis that our own apparent reality is actually a virtual reality existing in a simulation created, it is presumed, by some form of advanced beings. It is further conjectured that those advanced beings' 'reality' (including them) is itself a simulation created by some higher order of advanced entities. This "turtles all the way up" conjecture has no theoretical limits. Some people look for 'glitches in our matrix.' Good luck to them, I say. {The poster formerly known as 87.81.230.195} 90.199.210.77 (talk) 23:24, 23 May 2023 (UTC)[reply]

"No theoretical limits" includes the possibility that every turtle level has its uber turtle level, so that there is no prime mover. --Lambiam 13:21, 24 May 2023 (UTC)[reply]

Could be done in robotic simulators like V-REP where you can have multiple robots running around and seeing each other and behaving accordingly. I don't suppose it would be easy to create a version of V-REP that runs inside V-REP, but in theory it could be done. https://www.coppeliarobotics.com/downloads Greglocock (talk) 00:39, 24 May 2023 (UTC)[reply]