[openheaders]

I've been watching this topic for years and well I've come to this conclusion regarding heavy system resource usage holding back people from creating puffing and chuffing effects... Make it an optional beta feature so that those with decent rigs can use it and deal with the lag if it happens. Now this might come as a bit harsh what I'm about to say but.... This is 2019, we have people at home with personal VR headsets emulating physics, we have games getting nearer and nearer photorealism and yet the train sim community cant make smoke puff out of a steam engine? Personally I think a beta test should be made and tried out to see what happens, did the first steamer to hit the rails run flawlessly? No it didn't, same goes for scripting and programing. The first product will be a little rough around the edges... in time though pioneers come along improving it simply because the groundwork is there. Personally I think it should be done as a beta option or experimental feature option, but thats just me.

[darwins]

With 32 pages of input and 6 years of working on this it seems that someone should perhaps go through and summarise what has been suggested, what has been achieved and what yet remains to be done.

I suspect that it comes out something like this:

Graphics:

Smoke and steam rendered separately:
Smoke depends on fire/combustion.
Steam depends on steam chest pressure and speed.

Puffing/chuffing effect:
Needs one puff for each cylinder exhausting up the chimney / Can be worked out from driving wheel diameter.
Might need to add a parameter for exhausts per revolution to get this right:
Default 2-cylinder / 4-cylinder 2x90 degree cranks - four exhausts per revolution [also works for most 3-cylinder (Smith-Johnson) compounds and most 4-cylinder compounds]
3-cylinder 3x120 degree cranks - six exhausts per revolution
4-cylinder 4x90 degree cranks - eight exhausts per revolution
Geared locos - (?) - based on engine speed and number of cylinders?
Turbine locos - probably what we see at present is about right!
Puffs of steam (steam+smoke) should not just look like individual balls.

Cylinder cocks - there is a card on Trello that asks for cylinder cocks to puff too - presumably whatever logic allows puffs from the chimney will allow for this.

Height of the smoke column
Depending on exhaust pressure and orifice size.
Somewhere along the way a swirly exhaust was mentioned.
What about things like double chimney, jumper, Kylala or Kylchap, Giesel ejector, spark arresters?
Did any of these make sufficient visual difference to exhaust to be worth modelling? (Probably not!)

Other things exhausting steam up the chimney
Blower was mentioned and is currently in the physics so might be added to steam in the graphics.
Large ejector / Small ejector also exhausted up the chimney.
Tube cleaning of various kinds - throwing sand through or using a steam jet.
Self-cleaning smoke boxes.

Steam Physics

Steam production and usage:
I believe the present model should provide almost all the necessary information for the steam needed in the graphics for all the types of steam loco currently supported.
Possible exception might be to do with blast pipe/chimney orifice to work out how high the exhaust goes - but not sure if it needs to get that complicated to keep most folk happy with the graphics.

Smoke and combustion:

Amount of smoke:
This we have as it is directly related to the rate of combustion.

Colour of smoke:
Even discounting the summer/winter appearance (atmospheric temperature) this seems to consist of four factors none of which is yet modelled individually.
Primary air - opening or closing dampers - plus clinker building up on fire bed or ash building up in ash pan on a long run.
Secondary air - fire doors open, half open or closed - very important for manual firing - too little = black smoke / okay - light haze / too much - clear
Fuel added to fire - very important for manual firing
When each shovelful of coal is added - unburned fuel on top of fire - needs more secondary air - too much fuel added cools top of fire and black smoke produced.
If no fuel is added the reverse that as fuel is burned through less secondary air is needed - lighter smoke - also needs to consider stoker being switched off.
Type of fuel - oil firing has been mentioned but is not yet supported, wood is similar to coal. Manual firing can use large coal, whereas stoker fired coal is generally more pulverised/even sized.

Sparks or embers:
Crude graphic effect can be added by MSTS lights effect, but this is also mentioned on Trello - presumably with a thought to make them depend on some of the factors mentioned in considering smoke.

[ATSF3751]

Thank you so much for summarizing all of this! As far as I am aware the only one working on this is Peter but this should really help him out with what others are looking at when it comes to smoke and steam. I have tried to mention a lot of these things in the past but they have never come to light. Smoke really needs to be worked on and needs to be done right. We are in a world now where the graphics can handle this type of thing and Open Rails is WAY BEHIND when it comes to graphics compared to other rail simulators that are out now.

Brandon

[ATSF3751]

Since it has been awhile since any of us have talked on this thread I thought I would try and start it back up once again. As far as I am aware Peter is still the only one working on steam locomotive coding so I totally get if he can not get to everything when it comes to steam locomotives and and has other areas of the Open Rails project he is also working on. That being said here are a few things I would still like to see in the future when it comes to steam locomotives.

More realistic smoke- The balls of smoke need to eventually be looked at and figure out the best way to go about to eliminate it.
Height of the smoke column- This should depend on how hard the locomotive is working and how much draft is going to the fire. It should also depend on how much the blower is on.
Smoke and Steam separate- Smoke and steam should have there own code separate from each other.
Puffing/chuffing effect- Smoke linked to the drivers so when the locomotive gets to the stroke the smoke puffs out of the stack with force depending on how much you have the throttle open and how heavy your train is. This could also be used for the cylinder smoke as well.

As Darwins posted above this is how it should be linked
Might need to add a parameter for exhausts per revolution to get this right
Default 2-cylinder / 4-cylinder 2x90 degree cranks - four exhausts per revolution [also works for most 3-cylinder (Smith-Johnson) compounds and most 4-cylinder compounds]
3-cylinder 3x120 degree cranks - six exhausts per revolution
4-cylinder 4x90 degree cranks - eight exhausts per revolution
Geared locos - (?) - based on engine speed and number of cylinders?
Turbine locos - probably what we see at present is about right!
Puffs of steam (steam+smoke) should not just look like individual balls.

[copperpen]

Dont forget the early 2 cylinder compounds, only 2 beats per revolution. We already have code for the burn rate of fuel, the smoke colour could be linked to that. cylinder effects should also be tied to the exhaust with all other effects delinked to enable a steady stream, except for steam driven pumps which would need a different set of code to the exhaust.

The present 'balls of smoke' are caused by the way that OR is coded to handle the exhaust ace file. MSTS did a great job of expanding the file segments realistically with height partly controlled by the exhaust nozzle diameter. I suspect that there was also an element of draft air speed and cylinder exhaust speed added. Both items can be approximated in code. The nozzle diameter has zero effect in OR beyond changing the size of the puff balls.

Colour was also handled quite well by the Autofireman in MSTS and I would like to see the same in OR at some point.

[atsf37l]

As to the Height of the smoke column, it should also be influenced by the speed of the train over the tracks. A high speed passenger locomotive's exhaust is going to lay down across the top of the boiler and down the train. A slow speed freight slogging up a hill is going to have a near vertical column.

At speed:


Slogging:

[ATSF3751]

Yes I agree with you Herb and believe that is already put in for Smoke.

Brandon

[atsf37l]

Super. I was just reading the above regarding the height and the variable nature due to train speed didn't seem to be mentioned as a factor.

[Traindude]

 copperpen, on 06 March 2022 - 01:01 PM, said:

cylinder effects should also be tied to the exhaust with all other effects delinked to enable a steady stream, except for steam driven pumps which would need a different set of code to the exhaust.

Speaking of steam-driven pumps, it should be noted that some steam-driven air compressors exhaust directly into the air, while others exhaust into either the smokebox or the exhaust passages in the cylinder saddle. here is a video of what the typical exhaust from a steam-driven air compressor would be in both instances, courtesy of the Mid-Continent Railway Museum in Wisconsin, on the C&NW 1385 restoration project:


In these tests, with the air compressors mounted on the forklift, demonstrate what the compressor exhaust would look like when discharged directly into the air, with each compressor stroke causing the steam exhaust flow to "shoot" out, while sort of "drifting" between strokes. In the archival clip of the 1385 in service in the 90s at 3:52-4:01, you can see what happens when the compressor exhaust discharges into the smokebox or cylinder saddle. The compressor operation influences the velocity of the main stack smoke (even though the exhaust blasts themselves are much more subtle), but not the density or color of the smoke.

[Weter]

Ejection effect: thin jet with high speed may share its energy with large amount of smoke, passing conical chimney.
This helps utilize partially such energy, wasted outside in other case.