[railguy]

Just yesterday, I did some adjusting of the DieselExhaust for an old GE unit that I wanted to simulate running "dirty" under load. I changed it to this:

DieselSmokeEffectInitialMagnitude( 0.0 )
DieselSmokeEffectMaxMagnitude( 50.0 )
DieselSmokeEffectInitialSmokeRate( 0.0 )
DieselSmokeEffectMaxSmokeRate( 11.0 )

The effect I achieved was a unit (like the prototypes often would do) that would have a clean stack at idle, smoke heavily when throttling up and then smoke moderate to heavy under load. In the prior version to X1931, this worked near perfectly. I just started my activity with X1931. The unit idled clean, smoked heavily when throttling up, but ran with a perfectly clear stack under load--as did all of the diesel units in the train. 90% of the time this is more prototypical of diesels, but it would still be nice to have an old smoking unit in a consist. How can this be achieved with the X1931 changes in smoke behavior?

Thanks.

[rdamurphy]

Interesting comment, thank you. Basically, I made a small change, when accelerating, the exhaust output doubles, then drops back to the same level it would have been at before, and when decelerating, it drops by half, then resumes the previous output level.

Eventually, when OR engine files are available, I was planning on having those as user settings so you could decide how much increase or decrease you wanted, not to mention, there's also provision in the code for changing the default smoke colors, but not, of course, until post 1.0.

Can you attach your engine files so I can take a look? Perhaps doubling the output was too much and I should tweak that a bit...

[EDIT] Duh! My apologies, I see you DID attach the numbers. Let me check those out and get back to you...

Thanks,

Robert

[railguy]

Also, not to beat a dead horse, but the loose consist diesel units continue to smoke at full throttle. In my opinion, it would be preferable as at least a temporary fix for them not to smoke at all. Thanks.

[eric from trainsim]

 railguy, on 11 January 2014 - 06:39 AM, said:

Also, not to beat a dead horse, but the loose consist diesel units continue to smoke at full throttle. In my opinion, it would be preferable as at least a temporary fix for them not to smoke at all. Thanks.

I agree with that... at the very least, they should be at idle.

Better yet, create an ENG parameter for what they do when in a loose consist?...

[Matej Pacha]

I'd rather make it a part of the consist definition. We will definitely need some kind of consist definition extension. IMO it will be after the v1.0 release...
In the mean time, it's reasonable to change the diesel behavior to "Idle" when in a loose consist.

[rdamurphy]

OK, fixed the exhaust problem introduced in X1931. It was the result of the infamous floating point bug that's been in every processor ever made, all the way back to the abacus.

Basically, when you run a long series of floating point operations, rounding errors creep in. So, if you raise the RPM in steps from 250 to 675, and then back, the float errors cause the idle to end up at 249.999994. Basically the exact same thing that happens when you try to trisect a circle.

Take a look, it should work the way it did before, except with the increase in volume during throttle up, and the decrease during throttle down.

Robert

[PA1930]

I have yet to try X1940 to see that... :sign_thanks: But I have a question related to the smoke: Is it really possible to make a loco, lets say without much noticeable smoke when being with "high RPMs" but having like a bigger cloud of smoke when increasing the throttle? That'd be quite interesting thing to see... Because so far I can notice there's more smoke upon throttling up, the thing is that the the smoke is still rather very noticeable while the loco is, lets imagine, almost in full power...
[That kinda reminds me... maybe in the future one could develop a ORTS feature of oil leaking so that the loco make a huge smoke cloud? :D ]

[Matej Pacha]

I think we are still looking for the way of the smoke feature definition (or description). Now it's just a general way how to show we can do something. I'd like someone from the graphic team to make the particle position relative to the ground or the exhaust point - in other words: each particle should have a position and speed vector. That would allow us to compute some physics - movement of the particles would be dependent on RPM, ambient air conditions, etc. There is also some difference between the steam smoke and the diesel smoke, what should be considered in the model.
Now the only thing affecting the particle color is the engine speed. I'd like to see more detailed diesel engine model with temperature computations, etc. Exhaust color should vary from light gray to black for cold or broken engine to invisible for healthy engine running at full power. The other thing is, that it would need a lot of parameters to set up...

[PA1930]

 Matej Pacha, on 15 January 2014 - 07:29 AM, said:

I think we are still looking for the way of the smoke feature definition (or description). Now it's just a general way how to show we can do something. I'd like someone from the graphic team to make the particle position relative to the ground or the exhaust point - in other words: each particle should have a position and speed vector. That would allow us to compute some physics - movement of the particles would be dependent on RPM, ambient air conditions, etc. There is also some difference between the steam smoke and the diesel smoke, what should be considered in the model.
Now the only thing affecting the particle color is the engine speed. I'd like to see more detailed diesel engine model with temperature computations, etc. Exhaust color should vary from light gray to black for cold or broken engine to invisible for healthy engine running at full power. The other thing is, that it would need a lot of parameters to set up...

That would be priceless awesome though! :)

[railguy]

I fiddled with this in X1931. Whether it will work in the upcoming release, I don't know. I used this on three locomotives:

#1 is an GE ES44. This latest generation locos almost never smoke. Here is the parameter for them:

DieselSmokeEffectInitialMagnitude( 0.0 )
DieselSmokeEffectMaxMagnitude( 10.5 )
DieselSmokeEffectInitialSmokeRate( 0.0 )
DieselSmokeEffectMaxSmokeRate( 1.0 )

Since it is a 4-cycle prime mover, I use and RPM change rate of 17. The significance of this, I'll explain in a minute. The effect that I get in X1931 is no smoke at idle, very little smoke at throttle-up and no smoke at full throttle.

#2 is an older GE Dash 9. These were known to smoke a lot at throttle up, with moderate smoke at full throttle. My parameter for it is:

DieselSmokeEffectInitialMagnitude( 0.0 )
DieselSmokeEffectMaxMagnitude( 10.5 )
DieselSmokeEffectInitialSmokeRate( 0.0 )
DieselSmokeEffectMaxSmokeRate( 100.0 )

RPM change rate is 18 on this locomotive. This setting produced no smoke at idle, dark black smoke at throttle up, but no smoke at full throttle. The lack of smoke at full throttle on all of these is what X1931 introduced into the picture.

Finally, there is an EMD SD-75I. The locomotives were not known for smoking a whole lot on throttle up, but would smoke some at full throttle. Unlike the GE's, they have the EMD 710 2-cycle prime mover with a combination low-RPM supercharger and higher RPM turbocharger (so, too, does the older 645 prime mover used the 40 and 45 series EMD locomotives). For the 710, I used this parameter:

DieselSmokeEffectInitialMagnitude( 0.0 )
DieselSmokeEffectMaxMagnitude( 8.0 )
DieselSmokeEffectInitialSmokeRate( 0.0 )
DieselSmokeEffectMaxSmokeRate( 8.0 )

The RPM change rate for the 710 is 30. This is because the 2-cycle 710 will "spool up" in RPM much quicker than the 4-cycle GE. the smoke effect is a shorter duration "smoke puff" on throttle-up with the 710. An important note here: many model builders send out their engine files with RPM change rates not suited to the particular prime mover--many times I don't even think that they check them. When I modify an .eng file, that it one of the first things that I check. It also affects how the sound file will play.

The issue with the X1931 smoke is that it precludes smoke once a level throttle setting is achieved. That is prototypical of the new GE GEVO and EMD 710 prime movers, but not of the older prime movers. Also, under X1931, the engines emit gray smoke on throttle-down, which is not prototypical.

My suggested fix for all of this is to add an additional optional smoke parameter--a throttle-up smoke rate ("DieselSmokeThrottleUpSmokeRate"). If the DieselSmokeThrottleUpSmokeRate is absent in the .eng (as it would be in the current MSTS .eng file), the DieselSmokeThrottleUpSmokeRate would default to the smoke computation made by OR using the DieselSmokeEffectMaxSmokeRate, meaning that the smoke rate would be the same during throttle-up as at level RPM. Note that the DieselSmokeEffectMaxSmokeRate is RPM-dependent, but the new DieselSmokeThrottleUpSmokeRate would not be.

Now, if we wanted a GEVO to puff a little smoke at throttle up, the DieselSmokeThrottleUpSmokeRate could be set at, say 2.0, with the DieselSmokeEffectMaxSmokeRate set at 0.0. That would mean that the unit would smoke a little at throttle up, but none at level RPM .

With, say a GE Dash 9, the DieselSmokeThrottleUpSmokeRate could be set at 100.0, with a DieselSmokeEffectMaxSmokeRate set at, say 25 . This would allow the loco to smoke profusely at throttle up, but clear to moderate smoke at level RPM.

With a EMD 710, the DieselSmokeThrottleUpSmokeRate could be set at, say, 30, with a DieselSmokeEffectMaxSmokeRate also set at 30. Because the DieselSmokeEffectMaxSmokeRate varies with the throttle setting, but the DieselSmokeThrottleUpSmokeRate does not, the unit would smoke a little more at throttle up than the level RPM smoke level, with the difference between the two narrowing to the same smoke rate at full RPM. That would be pretty prototypical of an EMD 710.

What say ye? Thanks.