Smoke and Regulators

[SquidvT]

I recently got Brian Toppings book "The Engine Drivers Manual" which is a fantastic thing. Great detail and some amazing information on actaully how to tackle a kettle.

Couple of things though that stuck me about RWs.

1) Smoke colour. BT mentions keeping an eye on the colour of smoke as an indicator as to how the fire is doing. Is this modelled in RW? Looks to me as if the smoke stays one colour regardless.

2) Whats this 1st valve, 2nd valve thing? Is the regulator a fine adjustment or does it have certain settings?

Let me know so i can drive properly

[Kariban]

RW doesn't do the smoke colour by default. The fire is one homogenous mass & only has temperature, and the fireman just adds mass. It's scriptable, if someone likes messing with emitters.

Regulator valves are just that, once you run out of travel in the first valve it opens a second one. I assume they're mostly so you can shut it halfway or wherever without wire-drawing the steam ( also without having to hammer it shut with a shovel! ), generally you always want the regulator valves fully open or closed if you can help it. I should go find out exactly why there's multiple valves, I guess. RW has some blueprint entry for extra valves, but I have no idea if they work.

[Kromaatikse]

Railworks' modelling of smoke and steam emission is actually rather primitive. Some loco builders have made the extra effort to put in multiple emitters, simply to get variable amounts of steam (or diesel fumes) for different levels of power. Similarly the modelling of the fire is greatly simplified from real life.

The typical British regulator valve has two sliders. The first part (maybe a quarter) of the regulator's movement opens only the first slider, which permits a small amount of steam to pass - this is simply a device to get better control at low power levels, eg. when moving off (which requires high tractive effort but not much power). The rest of the movement leaves the first slider open but also opens the second slider, which permits the full valve capacity of steam to pass. Obviously if you have the regulator at half open, the second valve will still only be partly open and the available power will be less than full.

Railworks does not correctly model this, either. The Regulator is treated as a limit on tractive effort, not on power, and the gradual filling and emptying of the steam chest that real locos have is completely absent. Indeed where a steam-chest gauge is fitted, Railworks instead seems to show some analogue of exhaust back-pressure. It should be fixable by heavy scripting (in the same way as for the more advanced diesel and electric types currently available), but nobody has done this yet.

[Kariban]

( at least some ) GWR engines open the 2nd valve at half regulator, which turns it into a more useful general purpose system. The way RW limits different parts of a steam engine is considerably more restrictive than a diesel, so I'm not really sure how much can be done. I don't think I want to consider scripting here or I'll go off on a rant, so maybe someone else can give it a few initial thoughs

[ckmemphis]

I think if one has to reprogram the way steamers are scripted, it would be too complicated to handle both fireman and driver in the same sense. Being a firelighter on our local steamer, I now how challenging it is just working with the fire, as there are different types of fires, when moving or not. Smoke colour is directly linked to how the fire is doing. Should you have a 'dirty fire'...: Long, dark orange flames, you get a rather darkish colour smoke; should you shovel you fire dead, your smoke goes yellowish..... Then again, the burning properties of the coal itself is something on it's own, you usually give fresh coal about 4-6minutes to 'çome alive' and actually start burning.... before that time, it's just dust that's burning, not really having an impact. Then there's a fact that you can't just throw in coal all willy nilly, you have to read the fire and throw in where the fire is either dying or in certain places for certain results. You have back corners, under the brick arch, light sprinkle, horse shoe, tower fire, grave shoveling.... and all this is just the fire. You still have Primary Injectors, Blowers and many more to worry about.... So as you can see, steamers are more complicated than what RW says it is....

I find the simulation of steam in RW about 30-40% of real life physics and operation. I guss you could simulate most of it's opreation, but that would push the sim behond it's current limit, and of coarse, the limit of the players abilities and understanding..... You need to go on a coarse to understand the opreation of the loco, which I'm sure many of us might not really want to..... It's rather difficault for me, as I work with these things I'm taking about quite often, and not having them in the sim is a liitle strange for me... Either way, it gives more more understanding of what to do on the footplate.....

[Kromaatikse]

I suspect that the driving side could be improved a lot without touching the fireman side or increasing workload. As I said, it shouldn't be any more complicated than some of the advanced-controls modern traction now available. Simple things like modelling the steam chest and the regulator valve properly, and perhaps considering back-pressure more sanely, and then remapping all that to the underlying model.

I also suspect that the fireman side could be improved without increasing workload too, and with several different approaches to meet different needs. Modelling the shape of the fire is probably going a bit too far for single-player mode though. I would personally be happy with the "shovelling" control remapped to "target fire mass" and the virtual fireman then automatically handling the details and applying some minimal intelligence to the injectors, blower and dampers - and that essentially captures the communication between driver and fireman on the footplate. Other people are happier taking on the fireman's entire job.

Maybe if I get some more big lumps of free time, I can look into actually doing that.

[Kariban]

If I'm on a branch line the fire/water management is something to do while the passengers are milling around...if it's a mainline run I just set everything and leave the fireman alone, because once you hit his sweet spot you don't need to touch anything until you stop again. That is quite amazingly off, might as well not have a fireman . But one thing I hate about the auto-fireman is that he'll decide you need more boiler water just when what you actually want is more steam and not to have it sent to the injectors; while you could make guesses with that, like "don't open the injectors while we're going uphill unless we're about to pop the plugs" or "make use of us being stopped and the pressure rising", I'm not sure you can really script the fireman well enough for him not to be a bother. Also don't appear to be able to get the fire mass at all

[SquidvT]

Thanks for the replys guys, that at least clears some things up, and also some intresting info there too!

I am quite happy to simulate stuff myself, trying to get "As real as it gets". Its a pity about the smoke and regulator being quite "arcade though".

Perhaps in the future we will see more authentic stuff due to some clever scripting? There are plenty of examples of FSX addons that rewrite the book when it comes to default behaviour. With some pretty drastic departures from default behaviour. Of course they tend to command a suitible price too compared to other "lite" addons, but they are very popular.

The smoke colour would be awesome. I appreciate that the fire mass is grossly simplified, but just something like smoke colour indicating the state of the fire (too much coal, too little) would be cool. Again it would be amazing to for the regulator to be more accurate too.

[ckmemphis]

There was a discussion at one point about simulating the regular in a more realistic way, not sure on the title of that thread. The basic conclusion was that in order to do that, you would need to correctly simulate the Steam Chest, seeing that the Regular and Steam Chest would work hand-in-hand.... i.e. Regulator fills Steam Chest with steam out of the Boiler Dome, and the Steam Chest gets drained with each fill of the Cylinder's piston stroke. This would in turn give you the "mooshy" initial lever travel, simulating real life starting process....

Mayvbe someone would point us in that direction for further reading......

[Kromaatikse]

That would most likely be in the Physics forum somewhere. It's not the first time I've suggested that technique, I just haven't got around to actually trying it.

[Kariban]

I think that would probably be the first time smoke emitters have ever been an integral part of a physics simulation...

[bdy26]

I have noticed that there does seem to be a more marked change in generation and consumption on the default locos when moving from 49% to 50% regulator - this suggests some modelling of 2 regulator valves to me. I only have very limited experience of real kettle driving but the locos I have driven were fairly noticable when you went into second valve.

B

[Kariban]

Some of the older ones like the B5 go off the consumption scale and start over, which has generally suggested that there's a broken two-valve implementation somewhere :p. There are blueprint slots for valves, can just try it.

[crumplezone]

Variable smoke and colour change is already there with the 56xx when you change the regulator, so its not actually impossible in its current format, and smoke could be quite realistic should further research be put into it. Meshtools 0-6-0 has different smokes with multi-emitters and it responds to regulator levels so I don't see why it couldn't be done. Incidentally, the King seems to have below 50% regulator smoke which pulses out rather than one stream of smoke, well on my system anyway, I've noticed that after they fixed the smoke jittering problem recently.

But on a more simplistic front, the smoke texture could do with being updated and looking abit more "smokey" yet not to heavy on performance, since hte main issue is if you make it to, how to put it, dusty and particle like it causes considerable performance drop, however it done right you can get a pretty nice effect.

[ttjph]

I have noticed that there does seem to be a more marked change in generation and consumption on the default locos when moving from 49% to 50% regulator - this suggests some modelling of 2 regulator valves to me. I only have very limited experience of real kettle driving but the locos I have driven were fairly noticable when you went into second valve.

B

Maybe this is going back to EA RailSim, but ISTR a mention somewhere a of a simpling valve being in operation below 50% regulator - which would tie in with the sudden drop in steam consumption above this point.