[PerryPlatypus]

The OR Manual section 8.1.1 states:

"The program calculates
rolling resistance, or friction, based on the Friction parameters in the Wagon section of .wag/.eng file.
Open Rails identifies whether the .wag file uses the FCalc utility or other friction data. If FCalc was used
to determine the Friction variables within the .wag file, Open Rails compares that data to the Open Rails
Davis equations to identify the closest match with the Open Rails Davis equation. If no-FCalc Friction
parameters are used in the .wag file, Open Rails ignores those values, substituting its actual Davis equation
values for the train car."

And then goes on to describe the use of the Davis equation But then the manual states the following:

"When a car is pulled from steady state, an additional force is needed due to higher bearing forces. The
situation is simplified by using a different calculation at low speed (5 mph and lower). Empirical static
friction forces are used for different classes of mass (under 10 tons, 10 to 100 tons and above 100 tons).
In addition, if weather conditions are poor (snowing is set), the static friction is increased."

Could someone in the know (or who can look and decipher the code) please say what empirical data is being used? The reason I ask is that the starting friction forces seem to be extremely high at starting. I obviously am expecting somewhat higher forces to break the static friction within the bearings, but I am finding friction forces at 0 MPH that are on the order of 10 to 20 times the friction force at 5 MPH, which seems very questionable. We are seeing friction forces in the neighborhood if 3,000 lbs force, quickly reducing to about 300 lbs force at 5 mph. I would love to know what the empirical data suggests for the classes of mass mentioned in the manual (under 10 tons, 10 to 100 tons and above 100 tons) and see if they agree or disagree with data I have seen.

I would think that using a few empirical numbers rather than equations based on a more reliable spreadof data is highly subject to problems.

For what it's worth, the cars we are testing are 136 ton loaded coal cars using the following in the WAG file, with David values calculated from Fcalc:

ORTSBearingType ( Roller )
ORTSDavis_A ( 1231.28 )
ORTSDavis_B ( 20.3245 )
ORTSDavis_C ( 1.16519 )

See this link: https://www.arema.or.../PGChapter2.pdf

Again, just some empirical data, but it indicates a force of 5 lb/ton for roller bearing starting resistance. This would equate to a force of 680 lbs on a 136 ton coal car, rather than 2,000-3,000+ lbs.

Another source stated values as low as 3 lb/ton for roller bearing starting resistance.

Does anyone have access to a bit greater spread of data on starting resistance of freight cars?

[PerryPlatypus]

WAG file attached for reference. Hopefully someone who knows the ins and outs of the physics code can help with addressing the extremely high starting resistance force on these cars. The Davis coefficients were calculated from FCalc.

Attached File(s)

•  BN_BETHGON_533969_LD.wag (3.58K)
  Number of downloads: 369

[Coolhand101]

On my MGR HAA Hopper coal wagons, I set the "ORTSBearingType" to "low". This allows realistic low speed operations for under 5 mph!

Thanks

[PerryPlatypus]

Coolhand101, on 27 October 2019 - 11:45 AM, said:

On my MGR HAA Hopper coal wagons, I set the "ORTSBearingType" to "low". This allows realistic low speed operations for under 5 mph!

Thanks

We did try to use the "Low" friction, but in our testing this also produced friction forces the same or just marginally lower than "Roller". For now, we have resorted to using the legacy MSTS-style friction block in our WAG files until this can be looked at more closely. I would still like to know what equation ORTS is using under 5 MPH to calculate friction when the Davis friction coefficients are used.

[Coolhand101]

PerryPlatypus, on 05 November 2019 - 11:02 AM, said:

We did try to use the "Low" friction, but in our testing this also produced friction forces the same or just marginally lower than "Roller". For now, we have resorted to using the legacy MSTS-style friction block in our WAG files until this can be looked at more closely. I would still like to know what equation ORTS is using under 5 MPH to calculate friction when the Davis friction coefficients are used.

Hmmm, for my loaded 45t coal hopper wagon, my standing friction is 479ibf and 248ibf for "roller" and "low" respectively. Once above 5mph, the friction is the same as roller throughout. No idea about the ORTS friction equation under 5 mph.

Thanks

[steamer_ctn]

PerryPlatypus, on 24 October 2019 - 11:32 AM, said:

Does anyone have access to a bit greater spread of data on starting resistance of freight cars?

Sadly the starting resistance of a train does not appear to be well documented, and what does exist is based upon mostly empirical data. (Though in theory it might be possible to use published formulas provided by bearing manufacturers to calculate the starting torque required. However each wheel design and bearing design would need to be considered.)

Starting resistance appears to be impacted by ambient temperature, bearing design and type (especially lubrication), as well as load. The Open Rails model attempts to model these factors as best as it can.

I have also seen various values quoted as a guide for starting resistance for a train, but often they are very generic figures without any reference parameters. For example the AREMA guide quoted, states that starting resistance can vary from 5 to 50lbf/ton. It quotes a value for above and below freezing. This does not provide a satisfactory reference temperature, and implies that there is potentially a 10lbf variation as the temperature passes through the freezing point. It also doesn't state what type of bearing it is quoting, ie whether it is a standard roller bearing or a low friction bearing.

Looking at this site (under resistance link) a graph can be seen at the bottom of the page which shows some "typical resistance curves", some of which have reasonably high starting resistance values.

In regards to the stock that you are running, when I run OR, I get 2415lbf starting resistance for standard roller bearings, and 1320lbf for low friction bearings, so I am not sure how you are getting very little variation between the two values.

As AREMA suggests:

Quote

Starting resistance is generally not much of a problem with the very large tractive effort available with modern diesel locomotives, except on steeper grades. If necessary, the locomotive engineer can bunch up the train first, then start the train one car at a time. The cars already moving will help start the ones to the rear. This is called 'taking slack' to start.

Open Rails does not currently allow the 'taking of slack" to start, and therefore some trains may not be able to start. The loads may need to be reduced.

As you have requested, if anybody comes across a well referenced (in a research paper, etc) and defined set of starting resistance values that provide support for the OR model I would be interested in reading it.

Thanks

[Coolhand101]

NickonWheels, on 18 December 2019 - 07:00 AM, said:

Point A is the starting friction, the real problem is that you can´t influence it.

I must say, I can change this starting value by the bearing type, between Roller and Low. However, I am still using the latest XP version of OR.

NickonWheels, on 18 December 2019 - 07:00 AM, said:

By the way when slowing down the vehicle friction follows the blue line down to zero speed Point B, at which friction jumps back up to Point A... This may be problematic too.

Yes I also see that problem when the speed does goes past 5mph then below that speed. If the speed is under 5 mph, there is a gradual rise to the starting resistance.

Thanks

[steamer_ctn]

NickonWheels, on 18 December 2019 - 07:00 AM, said:

By the way when slowing down the vehicle friction follows the blue line down to zero speed Point B, at which friction jumps back up to Point A... This may be problematic too.

I will have a look at this.

[steamer_ctn]

steamer_ctn, on 19 December 2019 - 11:09 AM, said:

I will have a look at this.

I have corrected this issue in the latest unstable version.

[engmod]

thanks Peter.

[steamer_ctn]

NickonWheels, on 20 December 2019 - 01:53 AM, said:

But the real problem for many complainig users out here is being able to start a given train, at least not realisticly.

I don't believe that it is appropriate to do any more code changes until fuller research is done on appropriate starting friction values, and especially how they vary with temperature, load, etc. So far with all my code changes I have tried to base them on solid research information that I have found, and realistic test results rather then using my "gut feelings".

I would also want to see realistic test scenarios established to confirm whether or not we are meeting realistic performance expectations and standards.

For example, based upon appropriate tonnage ratings for a section of track, is OR performing realistically or not? You have quoted a couple of examples (Timken and Big Boy, etc), if these are set correctly, what load would we expect to be able to start? Can we start these published loads?

If you are interested in assisting in the above tasks (under my direction - it will take some effort on your part), I would be happy to consider reviewing the starting resistance model.

Thanks

[steamer_ctn]

NickonWheels, on 20 December 2019 - 05:16 AM, said:

I´m indeed interested in helping you as ORTS needs all help it can get, but I´m little more than an ordinary user, having only complated writing an assisting program (calculating diesel engine power, fuel consumption, traction curves etc.) with massive outside help;

I am not interested in somebody to assist with coding.

I am more interested in somebody to help to do research, set up tests, perform tests and feedback the results, etc.

NickonWheels, on 20 December 2019 - 05:16 AM, said:

otherwise I can only express myself by words and drawings.

You have expressed a view as to what you think the starting resistance values should be. What research information are you using to form your view in terms of the appropriate values for points A and B?

I have done some reasonably extensive research to form a view on the starting resistance, and any new information would be beneficial to me to confirm or challenge the model currently in place.

This is but one of the areas that I am interested in getting some assistance.

If you are still interested, then contact me through the Coals to Newcastle contact page so that we can establish an email dialogue.

[ATW]

It’s not only the bearings that heat up from resistance but wheels heat up too by brake resistance forces, brake rubbing based on distance/speeds etc. Handbrakes left on a wagon dragging high speeds or wagon brakes not releasing would intentionally heat up the wheels.

[ATW]

For satisfaction to reach goals on what's mentioned here about bearings I think a special parameter like Brake Shoe friction or even the traction/dynamic curve should be made from scratch just for Bearings or/and frictions. Does that make sense? So you can plant various weather adhesion a like curves in addition to Davis equation totaling to more friction resistance. ORTS is the way to go when making variables an new features for the laws of physics and intentional errors due to poor bad ordered builds or just reaching goals for cars?

[scottb613]

Hi Folks,

RE: Friction - https://archive.org/...wyuoft/page/n99

I'm not sure if anyone has read the article I posted - but it's an interesting discussion on friction from 1922...

I've noted that that the starting forces seemed a bit hard at times while working on my locomotive physics files...

One thing I found interesting in the article - is starting friction is decreased by using draft gear with slack - so the locomotive doesn't have to overcome the starting forces of the entire train just those of each car in turn... It appears from the F5 display that ORTS totals the forces for the entire train ?

Regards,
Scott