Eng & Wag files for OR

[R. Steele]

Before deciding on a set of figures and giving test results, what are your settings for adhesion in the Experimental section?. I always use 100%, no random factor change, and tick the weather box. My Numwheels in the engine section of the eng file is always set to 2.

Thank YOU, I had totally forgotten to post those...very relevant...
I did not change the NumWheels, it has been left alone...default for each loco
For the locomotives used in this thread the parameter is found only in engine section of the ENG file and reads ( 6 ) for all locos.

After reading your post I looked at the OR test stock Diesel and found this:

Wagon ( DEE_10203_or_adv

Comment  ( *****************  Known Required Parameters for OR - Wagon Section ************************)

Comment ( *************************** General Wagon Information ************************** )
	Type ( Engine )
	WagonShape ( DEE_10203.s )
	Size ( 2.82759m 4.02032m 19.6234m )  
	Mass ( 132.8t-uk ) Comment(Docu PDF p36)
	WheelRadius ( 0.4699m ) Comment(Docu PDF p10: 3'1'' -> 0.4699m)
	[B]NumWheels (	2 ) Comment(Docu PDF p36: 1Co Co1 -> 8 axles)[/B]
	
	Comment( ****************** Coupler Information ******************* )		

	Comment( Coupling Information )	
	Coupling (

How bad is my mistake, baldwin??
Also looks like I have to track down this >>> Docu PDF p36: 1Co Co1
Here's a screenshot of the experimental tab...
bandicam 2017-02-10 13-58-39-038.jpg

[R. Steele]

From post#10
"If you find a mistake in one of the files...NOT POSSIBLE - Yeh Right, I was careful, and checked my work, but something might have slipped past me."

Evidently a lot slipped past me. It's late and I've been rereading some stuff and reviewing other files, also comments.
I'll post another revised set of ENG files tomorrow...tonight just the highlights.

First the wagon section of the ENG file (these examples are from the lead, SP7400) - but apply to all other locos.

        DerailRailHeight ( 4cm )
	DerailRailForce ( 2.5*184.6t )
	DerailBufferForce ( 375kN )
	ORTSBearingType ( Roller )
	ORTSDavis_A ( 1692.65 )
	ORTSDavis_B ( 60.7426 )
	ORTSDavis_C ( 9.257661 )
        [B]ORTSAdhesion ( ORTSCurtius_Kniffler ( 7.5 185 0.315 0.7 ) )
        ORTSAdhesion ( ORTSSlipWarningThreshold ( 70 ) )[/B]
        Comment( == Assumptions -LocoDiesel-Electric, speed - 65mph (105km/h), Roller Bearing, 6 axles, frontal area - 16.1m2, Weight - 184.6 ton (metric) == )
        Comment ( Brake parameters not prototypical - these are fast brake values )
	BrakeEquipmentType ( "Triple_valve, Auxilary_reservoir, Emergency_brake_reservoir" )
	BrakeSystemType ( Air_single_pipe )
	MaxBrakeForce ( 140kN )
	TripleValveRatio ( 2.5 )
	MaxReleaseRate ( 40 )
	MaxApplicationRate ( 40 )
	MaxAuxilaryChargingRate ( 35 )
	EmergencyResCapacity ( 10 )
	EmergencyResChargingRate ( 30 )
        EmergencyBrakeResMaxPressure ( 110 )
        EmergencyResVolumeMultiplier ( 1.461 )
	BrakeCylinderPressureForMaxBrakeBrakeForce ( 70 )

First: note that the line Antislip ( 1 ) has been removed, it belongs in the engine section of the ENG file (the example is below)
Second: ORTSSlipWarningThreshold is now set to 70, set too low before.
Third: Both ORTSAdhesion lines are now a single line, easier to type, OR accepts this formating.
Fourth: Curtius Kniffler values have been changed...these may require further fine tuning!


Next is the engine section of the ENG file:

        MaxOilPressure ( 90 )
        ORTS ( ORTSWheelSlipCausesThrottleDown ( 1 ) )
        Antislip ( 1 )
        
	CabView ( DW_SD40.cvf )
	HeadOut ( 1.7 3.35 5.75 )

	AirBrakesAirCompressorPowerRating( 4 )

Both the ORTSWheelSlipCausesThrottleDown and Antislip belong in the engine section of the ENG file.

I hope I've caught the major goofs, we"ll see.

[baldwin]

With a random factor change of 10% I would expect to see wheelslip occur. The best way to test is have zero% change which eliminates any random effects generated by the OR code. Any wheelslip you find then is down to a combination of the weather, train weight, grade and curve resistance plus the eng file data. To remove any random factor caused by the route used, it is also best to use the test route from the Coalstonewcastle route. That has a selection of grades and curves to test on.

[R. Steele]

Locos.zip

Here are the final ENG files.
I'm not confident of some of the parameters, the Curtius_Kniffler numbers and the Wheel Slip Warning Threshold.

        ORTSBearingType ( Roller )
	ORTSDavis_A ( 1692.65 )
	ORTSDavis_B ( 60.7426 )
	ORTSDavis_C ( 9.257661 )
        [B]ORTSAdhesion ( ORTSCurtius_Kniffler ( 7.5 185 0.291 0.7 ) )[/B]
        [B]ORTSAdhesion ( ORTSSlipWarningThreshold ( 70 ) )[/B]
        Comment( == Assumptions -LocoDiesel-Electric, speed - 65mph (105km/h), Roller Bearing, 6 axles, frontal area - 16.1m2, Weight - 184.6 ton (metric) == )

ORTSCurtius_Kniffler
I'm reasonably confident of the 1st (7.5) and last (0.7) figures, all the sources* I found do not change these numbers.
It's the 2nd (185 - "weight"? of loco) and 3rd (0.291 - ?, some sort of adhesion coefficient?) that are puzzling.
Decreasing the 2nd number - 185 to 44 produces no change in the simulation that I can detect, this does not mean there is no change.
I have found these numbers in sources* given as ORTSCurtius_Kniffler ( 7.5 44 0.161 0.7 ), for either ENG or WAG files.
*sources: Anyone interested in a list of these sources, please PM.

ORTSSlipWarningThreshold set at 70, which I take is the default (from the manual):
"The Wheel slip value is displayed as a value relative to the best adhesion
conditions for actual speed and weather. The value of 63% means very good force transition.
For values higher than ( ORTSadhesion ( ORTSSlipWarningThreshold ) ) or 70% by default,
the Wheel slip warning is displayed, but the force transition is still very good. This indication
should warn you to use the throttle very carefully. Exceeding 100%, the Wheel slip message
is displayed and the wheels are starting to speed up, which can be seen on the speedometer or
in external view 2. To reduce the wheel slip, use throttle down, sanding or the locomotive brake."

When using the above parameters and also changing Options/Experimental/Adhesion Factor Correction from Very Hard to Hard to Normal you will see changes, so perhaps the numbers are close to the correct values??

Also some changes were made to the coupler values:

        Coupling (
		Type ( Automatic )
		Spring (
			Stiffness ( 1e6N/m 5e6N/m )
			Damping ( 1e6N/m/s 1e6N/m/s )
			Break ( 950kN 950kN )
			r0 ( 12cm 15cm )
		)
		Comment ( CouplingHasRigidConnection ( 1 ) )
		Velocity ( 0.12m/s )
	)
	Coupling (
		Type ( Automatic )
		Spring (
			Stiffness ( 1e6N/m 5e6N/m )
			Damping ( 1e6N/m/s 1e6N/m/s )
			Break ( 950kN 950kN )
			r0 ( 12cm 15cm )
		)
		Comment ( CouplingHasRigidConnection ( 1 ) )
		Velocity ( -0.12m/s )
	)
	Buffers (
		Spring (
			Stiffness ( 1e6N/m 5e6N/m )
			Damping ( 1.3e6N/m/s 3.8e6N/m/s )
			r0 ( 0m 1e9 )
		)
		Centre ( 0.5 )
		Radius ( 1 )
		Angle ( 0.5deg )
	)

I'm using these parameters for all of my ENG files. They will break in OR with poor handling.

Which brings me to Rio Grande BGWCC Part 2

With these parameters (Couplers; all the ORTS Adhesion; Davis; Curtius_Kniffler ) I cannot move out of Helper in Part 2 without breaking the couplers. Uncle E (creator of the activity) was able to do it, but he is much more experienced in train handling than I am. I had to move the helper locos from mid-train to the rear before I could move the train, and it crawled out of Helper. It was explained to me that mid-train helpers were used for this time period. So chalk it up to my inexperienced train handling.
Try raising the Break numbers to these: Locos: Break ( 1250kN 1250kN ) Wags: Break ( 950kN 950kN ) or same as locos.

I finished part 2 with no problems once I made it out of Helper using the modified consist (helpers at rear).

That brings me to the end of this little exercise. I learned some, but still have questions. The development of OR is going to take time ( all-volunteer, remember? THANK YOU to the whole team, official & unofficial, who have contributed ). I'm sure things will become clearer...honestly, probably 80+% is due to operator inexperience...(me).

[R. Steele]

An Open Rails developer, steamer_ctn >> http://www.coalstonewcastle.com.au/, was kind enough to read through this thread and offered the following observations - regarding ORTSCurtius_Kniffler:

My understanding, and research suggests that it is an empirical formula. That is, a series of tests were done on locomotives, the results plotted on a graph, and then a curve developed to fit the points as best fit as possible. A number of other researchers have followed the same approach, and found similar results.

( See section 2.3.1 (pg 19) of this document for a brief description of some tests done - http://archives.njit.edu/vol01/etd/2...td2010-023.pdf )

As far as I have been able to tell the numbers do not relate to any particular parameters.

I have attached a spreadsheet with two Curtius-Kniffler curves* that I have been able to find for wet and dry conditions. I have included the values from the statement that you are using. Note the difference in the results. You can try replacing the red numbers in the spreadsheet to vary the curve. I think that the user curve does not reflect the "true" value very well.

My personal feelings are that there is not a great deal of value in varying the Curtius-Kniffler values, as the "default" values are pretty close, unless there are values provided by the railway company. The adhesive weight of the locomotive and the acceleration are probably the main areas to be considered when looking at the performance of the locomotive under load.

*Curtius-Kniffler curves
bandicam 2017-03-01 16-47-17-935.jpg

As can be seen examining the above curves, green representing the Curtius-Kniffler numbers I was using for the tests, changing the numbers can result in a curve that is way out of tolerance. At this point I'm convinced (unless someone provides documented compelling evidence) that using values other than default is not beneficial.

Currently, I'm using these values for both ENG&WAG files: ORTSAdhesion ( ORTSCurtius_Kniffler ( 7.5 44 0.161 0.7 ) )

[muskokaandtahoe]

Gerry,

what is needed is an appendix to the manual that is a list of every parameter that includes what it is for, a Y/N for whether it is a legacy MSTS parameter that is no longer being used, a Y/N for whether OR expects rather different values as compared to MSTS, and perhaps something to say about the kind of unit of measure it uses (e.g., distance, volume, pressure).

What is provided is several hundred pages of manual you can hunt thru and maybe, if you are lucky, find an explanation of some parameter you've never heard of before.

[Don6218]

Gerry,

what is needed is an appendix to the manual that is a list of every parameter that includes what it is for, a Y/N for whether it is a legacy MSTS parameter that is no longer being used, a Y/N for whether OR expects rather different values as compared to MSTS, and perhaps something to say about the kind of unit of measure it uses (e.g., distance, volume, pressure).

What is provided is several hundred pages of manual you can hunt thru and maybe, if you are lucky, find an explanation of some parameter you've never heard of before.

Yes a very basic tuning to OR standards of generic files would be a great help to those of us who are Tech-Speak challenged. Something along the lines of a better file for a range of common road locomotives. What it would do is allow a somewhat realistic Consist to exist, for example, being able to have 2 SD40-2s pull a 60 car bare table train along a mountainous/hilly route without stalling, or 2 P42ACs pull a 7 car passenger over the same terrain without stalling.

It would not be difficult to pick out maybe 12 or so, do up proper files and instruct where to put the files. Then the new files for more specific Locomotives/Wagons would have the time to be developed. Feedback could be gained from those using these basic files to enhance the more specific files.

What comes to mind would be basic files for GP units, SDs, E44s, P42s, P32s, C40s, AC44s, F40PHs, and some newer SD75s, SD90s etc.
Then the more specific files could be developed with the input from a very broad base of users. Much like using the default GP40 for almost all early MSTS diesel files and the Scotsman for steam files.

[baldwin]

The diesel and electric eng files for OR are broadly the same as for MSTS. The only tweak I make is to the Maxpower line which should be the actual max of the unit, not the reduced figure used by MSTS. Couplings are optional, but I am working through mine reducing the break forces to real world values. There are more things that can be done such as weight on drivers, diesel engine torque curves etc should anyone want to go deeper.

[cjakeman]

what is needed is an appendix to the manual that is a list of every parameter that includes what it is for, a Y/N for whether it is a legacy MSTS parameter that is no longer being used, a Y/N for whether OR expects rather different values as compared to MSTS, and perhaps something to say about the kind of unit of measure it uses (e.g., distance, volume, pressure).

Dave is quite right.

I'm already working on another improvement he has suggested but, if anyone would like to write an appendix along the lines Dave suggests, I will include it in the manual.