TrainSimulations Davis numbers are incorrect

[R. Steele]

I purchased the new sets and the ORTS Davis numbers for the locomotive & wagons are really way off.

UP Heritage SD70ACe Engine Pack-Vol1
MILW Road Train Pack
NS SD70Ace Heritage Trainset - the coilcars used as the example in this post.

Now this is not a criticism of whoever worked these numbers up, I'm sure it's just not knowing how OR handles Units of Measure, we are all trying to learn something new here, most of all myself. I only noticed this because of a post I made at ET a year ago in November.

http://www.elvastower.com/forums/index.php?/topic/31108-orts-davis-number-uom/

Here's the a wagon file from the newly released NS Aces Heritage set, it is a coilcar

SIMISA@@@@@@@@@@JINX0D0t______

Wagon ( CR_COIL_CAR_628115
	Type ( Freight ) 
	WagonShape ( CR_COIL_CAR_628115.s ) 
	Name ("CONRAIL STEEL COIL CAR 628115 (LD)") 
	Size ( 3.142m 4.483m 18.221m ) 
	Mass ( 128t-us ) 
	WheelRadius ( 36in/2 ) 
	InertiaTensor ( 
		Box ( 3.142m 4.483m 17.621m ) 
	) 
	Coupling ( 
		Type ( Automatic ) 
		Spring ( 
			Stiffness ( 8e4N/m 4.5e6N/m ) 
			Damping ( 2e6N/m/s 2e6N/m/s ) 
			Break ( 350000lb 350000lb) 
			r0 ( 12cm 12cm ) 
		) 
		 Comment ( CouplingHasRigidConnection ( 1 ) )
		Velocity ( 0.2m/s ) 
	) 
	Coupling ( 
		Type ( Automatic ) 
		Spring ( 
			Stiffness ( 8e4N/m 4.5e6N/m ) 
			Damping ( 2e6N/m/s 2e6N/m/s ) 
			Break ( 350000lb 350000lb) 
			r0 ( 12cm 12cm ) 
		) 
		CouplingHasRigidConnection ( 1 ) 
		Velocity ( -0.2m/s ) 
	) 
	DerailRailHeight ( 5cm ) 
	DerailRailForce ( 2.5*128t-us ) 
	DerailBufferForce ( 1142kN ) 
	NumWheels ( 8 ) 
	ORTSBearingType ( Roller )
	ORTSDavis_A ( 264.50 )
	ORTSDavis_B ( 1.9200 )
	ORTSDavis_C ( 0.005000 )

Notice the A,B,&C Davis numbers and the weight of the car in us tons >> 128t-us

Using FCalc2 I got the same numbers by choosing Us Units as the basic UoM for calculations.



The problem arises when those numbers are used in OR without specifying the units of measure. OR will assume the metric defaults for the Davis Numbers...A is Newtons (N), B is Newtons/Meter/Second (N/m/s) and C is N/m/s^2

The file above should be using the US units which have to be written with no space after the number as follows:
ORTSDavis_A ( 264.50lbf )
ORTSDavis_B ( 1.9200lbf/mph )
ORTSDavis_C ( 0.005000lbf/mph^2 )

Because the US units are not in place in any of these wagon files the physics are incoherent.
It's fine to choose to use US units, you just gotta make sure they are written in the file, please.

Can someone who has contact with TS, please let them know, this mistake keeps appearing in their, otherwise, very fine work.

Lastly,since the chosen figure for front area is 10.0 feet squared, which equals 0.93 meters squared, I suspect the frontal area has been much too under estimated to be useful. The instructions in FCalc2 recommend using 10.0 Meters Squared if a more accurate measurement cannot be obtained, so that's where I think they went wrong with this figure. (10 m2 would be 108 ft2)

Here's the example I used to illustrate my point in the ET post a years, different wagon, but you can see the what the difference is.


One last thing, by using 105mph and not specifying mph, OR is using kph, so the calculation is made at speed of 169Kph.
Man, that is one fast freight.

[Guest]

"One last thing, by using 105mph and not specifying mph, OR is using kph, so the calculation is made at speed of 169Kph.
Man, that is one fast freight."

Not only fast, but almost unstoppable as well.

[TrainSimulations]

Hi Gerry,

I think you had mentioned in an earlier post that it might be a good idea to just stick with the metric numbers. If you did say that, then you were right then, and it's what I should have done. Thanks for pointing out this inaccuracy. I'll need to revisit these files and those in development to make that correction. Clearly, leaving everything in metric units is the safe way to go.

Cheers,
Jason@TrainSimulations

[R. Steele]

Hi Gerry,

I think you had mentioned in an earlier post that it might be a good idea to just stick with the metric numbers. If you did say that, then you were right then, and it's what I should have done. Thanks for pointing out this inaccuracy. I'll need to revisit these files and those in development to make that correction. Clearly, leaving everything in metric units is the safe way to go.

Cheers,
Jason@TrainSimulations

Hello Jason,
I did mention my preference for using the OR default UoM, which are metric. It causes me confusion sometimes, having used the US system all my life, but when dealing with files and lots of data, not having to specify or make sure the UoM are correct is one less thing to worry about. That's my primary reason for going metric with OR.

I also used the 10.0 square meter default for a while, but then read through the threads and saw that most folks considered achieving some accuracy with frontal area to be important. The best way is to actually measure the cars in the yard...not possible for most of us!
Next best is to use shapeviewer, the bounding box, and the X&Y axis product to get a gross measurement in m2 then multiply by some factor to reduce that to, hopefully, something in the ballpark. I usually use 85% to 95% of the gross measurement, depending upon how tight or loose the bounding box is. A problem arises with models that use dds and lack ace files. Then I usually default to using 85% of the gross m2 measurement. Seems better than every car being 10.0 m2. There are defaults for all cars in the FCalc2 readmes, some are in the docs listed under View, others in the docs listed under Help.
If you or anyone reading this knows of a more accurate measurement method, please let me know, I would love to learn it.

By the way....excellent work, very fine models, indeed. Love the coil cars!!

Couple of illustrations pertaining to the coil cars, I'm using a frontal area of 11.9m2, one with roller bearings, the other with low bearings.
11.9 seems a little large to me? I am always waffling between Roller and low, I usually use low for all passenger cars; friction for BLW_ZT models (even though some of them are in the roller era) and roller for everything else, that can't be correct in all cases. I wish I could lay my hands on some definitive information concerning bearings used in different types of freight cars. I do know that using Low bearing does noticeably decrease the starting resistance.


Best Regards, Gerry