A How to for Train Physics for RS

[RavenX]

A How-To for Train Physics for Rail Simulator... By Members of Triansim.com

Thankyou Bob Boudin,

Let's see, as per your suggestion;
I do have a suggestion. I think a MUCH more positive and constructive approach would be to give some specific instruction and direction on how to do some of the things mentioned. For example, how about a DETAILED description of how to edit the physics for a GP38-2. I would be happy to provide a complete set of specs and any data needed to the project. I think such a tutorial would be a super welcome addition to the community knowledge base. That would do so much more to foster good will for RS than a shouting match as to who has the better opinion. Show us how to do it, don't just say you know how to do it!

And
GP38-2 specs in brief.
weight 255,400 pounds, typical working weight.
clean dry rail adhesion limit 26.5%
Max continuous force 54,800 pounds
minimum continuouis speed 10.8 mph
Power 2,000 HP.
Transmission efficiency 84%
top speed 70 mph
max brake force 66 kN.
90 pound brake system pressure.

May I work on this tonight, (it's 8:28pm here) and see if I can post something up tomorrow, or Monday, and see where we might get stuck through the conversion process.

Regards

RavenX

[RavenX]

Let's try to move in a positve direction with this! I noted the thread on a GP38 (probably a GP38-2 but I can give specs for a GP38 as well?). That is why I suggested the change in direction with specific goals!

Bob, sure, let's do that one as well, can you post the specs for that one too.

Then members can have both to try out and compare.

First I'll need a model to attach/refence the files to ~ I'd use any 3d model hmm, perhaps the BR 294. or one of the other 3D models, so that gauges can be read other than the F5 Hud.

Thanks.

RavenX

[plainsman]

GP38 (the version with generators) brief specs.
252,900 pounds typical working weight.
clean dry rail adhesion limit 25%
Max continuous force 54,700 pounds.
Minimum continuous speed 10.8 mph
Maximum speed 65 mph.
Power 2,000 HP
Transmission efficiency 83%
Max brake force 64.5 kN
80 pound brake system pressure.
The GP38 has transition SP to P (about 35-37 mph).
Shunt before transition 40% and 57%
Shunt after transition 40%
Note GP38-2 does not use transition or shunting, is in constant P.
BTW, note the spelling of the last name, that is the French version, hehe.

[plainsman]

GP38 and GP38-2 throttle progression:
Notch1___4.5%
Notch2__10.0%
Notch3__25.0%
Notch4__33.0%
Notch5__50.0%
Notch6__66.7%
Notch7__85.0%
Notch8_100.0%
The GP38-2 will essentially form a pure power function TE curve.
TE= (2000x375x0.84/speed(mph)) above 15 mph and 54,800 pounds at 10.8 mph.
The GP38 is not quite so straightforward. It will have a slight TE sag as it nears transition speed of 35-37mph. It will be a pure power function from 37-60mph. TE=(2000x375x0.83)/speed(mph)). Stated another way, you will see a slight drop in transmission efficiency from about 32-37 mph as back emf fights to overcome output voltage(down to about 80-81% at transition). Therefore, I would use a datapoint of TE=(2000x375x0.805/35) and another of TE=(2000x375x0.83/38) to define this. (you could use that instead of 40 mph point). Using 5 mph points otherwise, should give a nice curve. Transmission efficiency is not constant with speed, but above is pretty close over speeds defined above. Again, GP38 TE is down to 54,700 pounds TE at 10.8 mph. Also, you might want to drop TE=(2000x375x0.805)/65 for the last 65 mph point, as that is what limits max speed on a GP38 with generators. As back emf builds again, excitation is steeply increased to prevent flashover of the traction motors. This limits top speed to 65 mph.

[OTTODAD]

Hi Bob !

Good to see you getting in on the act at last, trying to improve some of RS too. ;)

It would help if you could make the amended by you RS files available in the KRS section of the File Library to prevent users messing up their files, many not backing up as much as they should ! :(

Take care, O t t o.

[plainsman]

Hi Gary,
Also note, even with sand, starting tractive effort for a 255,400 GP38-2 will only be about 70,000 pounds. Even if transmission efficiency falls to 75%, by 8 mph, TE becomes adhesion limited even using sand.
Hi Otto,
I have been lurking all along, just not much I could contribute at the time.

[OTTODAD]

Hi Bob !

Finding the brakes of most of the RS locos rather sluggish and not responding very well, Jean-Louis Chauvin taught me how to improve their effectiveness.

http://forums.flightsim.com/vbts/showpost.php?p=1383859&postcount=41

The notching of the UK Class 166 combined throttle/brakes missing all stages between 0 and 17 % I then had a go at applying the improved by Jim Ward notching of his PON294US diesel's <RestInNotchedPosition and associated parameters into the Class166_dmocl.bin and it now performs beautifully just like Jim's diesel. :D

Sniffing around these filled to the brim with parameters *.bin files I noticed a sProjectedLight and other related to headlights parameters, but trying to figure out which settings do what in all these countless parameter lines is a nightmare ! :mad:

Anybody out there who can make sense out of what these files do and how ?

O t t o

[JLChauvin]

Hi Bob !

Finding the brakes of most of the RS locos rather sluggish and not responding very well, Jean-Louis Chauvin taught me how to improve their effectiveness.

http://forums.flightsim.com/vbts/showpost.php?p=1383859&postcount=41

The notching of the UK Class 166 combined throttle/brakes missing all stages between 0 and 17 % I then had a go at applying the improved by Jim Ward notching of his PON294US diesel's <RestInNotchedPosition and associated parameters into the Class166_dmocl.bin and it now performs beautifully just like Jim's diesel. :D

Sniffing around these filled to the brim with parameters *.bin files I noticed a sProjectedLight and other related to headlights parameters, but trying to figure out which settings do what in all these countless parameter lines is a nightmare ! :mad:

Anybody out there who can make sense out of what these files do and how ?

O t t o

Hi Otto, this is a more complete modification:

[OTTODAD]

Thanks, Jean-Louis !

Still a bit erratic, cycling it's notches.

Try my version of the DMOCL, using Jim Ward's settings and your brakes modification. ;)

O t t o

[JLChauvin]

Thanks, Jean-Louis !

Still a bit erratic, cycling it's notches.

Try my version of the DMOCL, using Jim Ward's settings and your brakes modification. ;)

O t t o

Thanks Otto, I will try it... as soon as I re-install RS on my HD;)

[OTTODAD]

Welcome back on board, Jean-Louis ! ;)

RS has it's drawbacks but some novel ways of doing things too, some of which need some more polishing. ;)

But to guys like you, Jim Ward, myself and others like us it's a challenge, providing many hours of satisfaction getting results.

Try my CAJON-OTTO-MINOR, Jim Ward's routes and my test routes on my web site and you will realize that there is more to RS than what the out-of-the box version would appear to supply ! :D

Take care, O t t o.

[plainsman]

Hi Gary,
Do you need any more info or data on the GP38 or GP38-2?
Thanks,

[plainsman]

Is the silence here making anyone else uncomfortable?:confused: Please, just ask questions or tell me what you need, I can probably translate it into a workable format to fit the parameters.

[Dan1]

Bob - surely NorfolkSouthern37 is the person who needs the data as he is the person who is building the GP38-2 model.

[plainsman]

Hi Dan,
I sent Justin a message offering to provide any data that would be helpful for him to use, or to translate it into whatever format the game requires.

[RavenX]

Bob,
Hi Gary,
Do you need any more info or data on the GP38 or GP38-2?
Thanks,
If you have any additional info/data, post it up... the more info/data the better.

And...
Is the silence here making anyone else uncomfortable?:confused: Please, just ask questions or tell me what you need, I can probably translate it into a workable format to fit the parameters.
Sorry, ~ I live in a different Time Zone to the US & UK, plus I don't normally surf the web on weekends, (to many other things to occupy my time & energies)...

Well, in the the mean time, so far... I've decided that there is a need to write up a doc. to list all available parameters within the "Engine Simulation Blueprint" as it appears within the "Asset (or Blueprint.exe) Editor" (Even though RSDL have released a pretty good introductory guide for them).~ There are far more parameters & Brake settings than compared to the old ".eng" files of MSTS days, to take note of.

The reason for a doc. or manual is:~ it is not as simple as changing a line of script here, and line of script there... it's a little more involved, as those of us, who have looked through the "Blueprint" editor, are aware. Fortunately, we do have a lot of accurate information and a wealth of experience to hand.

By the way, it took 2 years of testing for Rudolf Richter to get a final edition to his "Manual for .eng & .wag files" together. Hopefully it won't take us that long. :cool:

Regards

RavenX
PS: Just as an aside, I'm employed within the Architect Camp on the design & build of a multi-billion Rand stadium in Durban, South Africa, for the 2010 FIFA Soccer World Cup, so my time is limited where dedication is concerned to RS. ~ Just a bit of background info. (even if it is a bit long winded ;) )

[plainsman]

Hi Gary,
Sorry about the silence comment. I am around off and on most days. I understand you enjoy your weekends.
The reason I need direction, I have way too much data to just post everything. I also have ability to derive a lot of stuff. You may have seen the graphs in the other thread. Here is the data behind them:
Now if you are particularly a perfectionist, you will note some minor generalization, for example I used 8 mph for each unit getting to adhesion limited TE, that would vary with a number of factors, and even on perfect conditions, each unit will be slightly different, but I wanted to simply illustrate adhesion limited tractive effort. Also, another example, transmission efficiency is not constant with speed, but to account for it in this project would be too complicated and would not really change the performance of the program for our purposes.
BTW do you know if weight is used as metric tons or US short tons? The reason I ask is the SD40-2 uses metric tons, the ES44AC uses US Short tons, but probably for another RRs version.

[OwainGlyndwr]

Whatever happened to imperial tons. ;)

regards,
Mike.

[plainsman]

Hi Mike,
I didn't know the emperor was that heavy?

[OwainGlyndwr]

Hahahahahaha, very droll Bob. ;)

regards,
Mike.

[RavenX]

Bob
BTW do you know if weight is used as metric tons or US short tons?
" Mass of the Locomotive in Imperial Tons "
(According to 2.06 Engine Blueprint Doc.)

The Blueprint Editor doesn't state what the units are. Hmm... :rolleyes: Even though in some cases it does. Another "Hmm..." :rolleyes:

It does however, produce a default value of "117" (This is under Blueprint "Engine Blueprint" -> "Rail Vehicle Component" -> "Mass")

I would like to mention that without seeing the actual equations of the "units of measure", that is present in the code, it is an assumption on our part, and an acceptance of truth (or a statement of fact) on the part of RSDL, by ourselves. ~ The reason why I say this, is we are all human, and even with the utmost care, errors can be made. Fortunately, we do have direct contact with the folks at RSDL. Which is another big "thumbs-up" to them.

BTW, I've written up to page 11 on that Blueprint Manual I spoke about doing, earlier. ~ It's looking good.

Regards

RavenX
(By the way, it's good to see some light-heartedness here, once again :) )

[plainsman]

Hi Gary,
Thanks that is a big help. But both US locomotives are set up incorrect.
The SD40-2 for a UP version of that number is ~393,500 pounds. That would be 175.7 Imperial tons. The 177 was a metric weight for a generic SD40-2 from my V4.5 or V4.8 files, that was for a 390,200 pound unit (about average for most SD40-2 units run around North America).
The ES44AC has a broad potential weight range as well. BNSF runs them at 415,000 pounds. I think most of the UP versions run from 420-425,000 pounds. CSX I think has two versions, a 420,000 pound and a heavy 432,000 pound. CP may have some very heavy units as well, that is reported, but I have not verified that with CP folks that really know. So for a BNSF ES44AC, the weight would be 185.3 Imperial tons. For a 424,000 pound UP version it would be 189.3 Imperial tons, for a 420,000 pound it would be 187.5 Imperial tons, and for the heavy 432,000 pound versions, it would be 192.9 Imperial tons.
I think both BNSF and UP software limit TE to 180-182,000 pounds on their units, CP probably close to that as well. I think CSX has some of the heavy units able to get 198,000 pounds of TE. I think the light weight ones run 180,000 pounds TE.
Same with dynamic braking, I think standard equipment is ~98,500 pounds of max dynamic braking force, with an optional 112,000 or 117,000 pound DB.
BTW, will the TE vs speed data I supplied (see jpg chart above) fit into the RS graphical format for that data?

[RavenX]

Hi, Bob,
BTW, will the TE vs speed data I supplied (see jpg chart above) fit into the RS graphical format for that data?
The Charts should fit into the .CSV / .DSCV file. Thus,

The .csv file must be in the format with [ Speed in Column A ] & [ Tractive Effort in Column B ]

according to the ... "Diesel (Engine) Simulation Doc." from RSDL.

The example graph given by RSDL shows Tractive Effort (Lb.f.) vs. Speed (mph) ~ I would expect the .csv file could be made of either Imperial or SI Units (as long as it is consistant, ie: Imperial Units only or SI Units only) should / would give the same result within RS. (NOTE: I'm not certain about that though, as Tractive Effort (Lb.f.) vs. Speed (km/h) would produce the same graph ~ Please advise).

Regards

RavenX

[Bill Hobbs]

You need to be careful about working with the proper units. While the shape of the graph will be unchanged, the difference in values will certainly have different impacts on the simulator. So look at the docs and also look carefully at some of the default equipment.. There are a couple of errors in the docs for steam locos, the first one that comes to mind is that Max Force is described as kN but is actually klbf. It is safest to look carefully at one of the included locos and compre what you see in the Engine and Simulation files with independent data.
I'm happy to see this discussion progressing. It has inspired me to get back to work on my spreadsheet for setting up steam locos. I am making additions to the one I created for MSTS to include as many RS Engine and Simulation variables as I can. I'm not looking forward to doing the documentation, though. That's a lot harder than the calculations.
Bill

[plainsman]

Hi Bill,
The problem with using the default equipment, is the two examples don't agree on some things. A good example is weight (mass), The SD40-2 is in metric tons, the ES44AC is in US Short tons, and supposedly the docs say it should be imperial tons???:confused:
Both defaults have the TE vs Speed graphs in kN but supposed to be klbf?
Power is also in disagreement?:confused:

[RavenX]

RSDLadam...

May our esteemed Technical Director of RSDL, please advise and assist w.r.t. the SI & Imperial Units used within the inner workings of your sim.

Much appreciated,

Regards

RavenX

[Bill Hobbs]

Bob,
No, for the UK steam locos, the docs say the Max Force should be in kN but the number showing in the Simulation file agrees with my calculation of the Force in klbf. If it is in kN, then the loco would barely be able to move itself. The Black 5 shows 25.440 as Max Force. As kN that would make the lbf about 5700!! The steam csv files that use TE do it as a percentage of max so not a problem there.
As far as the steam evaporation vs coal burnage file goes, I interpreted the units to be lbs of steam evaporated vs lbs of coal combusted., though the ratios would still look ok if both were in kg. However, in kgthe range of evaporation would be far in excess of the required evaporation and grate limits.
But it would be nice for the gurus to offer the inside story.
Bill

[ENTrainman]

Very good thread, this has helped quite a bit with my modifications to the engine simulation files :)

does anyone have access to the requried information to verify (or change) the SD40-2's engine csv's?

[ohanlon]

Tractive effort in the Source files clearly requests pounds foot LBF

Regards
Ohanlon

[ENTrainman]

it requests Pound feet, but it is actually asking for a number in 1000's of pound feet

[plainsman]

UP SD40-2
weight (mass) 393,500 pounds (175.7 imperial tons).
Max tractive effort ~105,000 lbf
Continuous Tractive effort 83,100 lbf
Gross power 3,000 hp
Power at the rail 2520 hp
dynamic braking max force 60,000 lbf @ 25 mph
minimum continuous speed 11.1 mph.
transmission efficiency 84%
max brake force ~105-106 kN
no power limiting or transition, so TE curve is a pure power function above 15 mph.

[ENTrainman]

UP SD40-2
weight (mass) 393,500 pounds (175.7 imperial tons).
Max tractive effort ~105,000 lbf
Continuous Tractive effort 83,100 lbf
Gross power 3,000 hp
Power at the rail 2520 hp
dynamic braking max force 60,000 lbf @ 25 mph
minimum continuous speed 11.1 mph.
transmission efficiency 84%
max brake force ~105-106 kN
no power limiting or transition, so TE curve is a pure power function above 15 mph.


perfect! thanks :)

[ENTrainman]

dont know if anyone here has downloaded this yet... but document 2.11 Diesel Simulation details ALL units used in the blueprint :)

[Dan1]

Bob - although it would require separate eng files, would it be possible to create eng files for the same type of unit but reflecting differing mechanical conditions. In the sense that a unit that is ex-works will produce a better performance than a unit that is coming up for overhaul, etc. I am not sure if this makes sense - but if you take a route like KHP2 - it came with a number of SD40-2s, but as anyone knows in real life performance varies from unit to unit. Would it be possible to create eng file templates for units in ex-works, good, run down and very poor condition?

[plainsman]

If we can figure out how to get the data we have into the program correctly, then sure it can be done. Sone things will detereorate more than others. mass and adhesion will stay about the same, so starting tractive effort will be about the same. As wiring ages, it may become less able to conduct without loss, so power at the rail would drop. If the prime mover is not fully producing its rated output that can reduce power as well. These things would shift the TE curve then as well (TE for any given speed above the adhesion limit would be less).