E.M.U trains running in OR

[coolhand101]

Hi
Awhile ago, i started a thread about E.M.Us ( Electric Multiple Unit Trains ) having prototypical traction performance curves in MSTS.

This was achieved by having all the consist WAGs files as ENG files, to allow tractive curve drop vs speed.
Also, using the Hi-Lo acceleration parameter for the low speed range.

With valuable help from members JTR1962 and Bavli, i got the real life peformance of these EMUs almost spot on.

I then went a step further, to allow the brakes to perform differently at certains speeds. This was to simulate the cast iron tread brakes for these EMUs. The "Blending" option of the air and dynamic brake, simulated the retardation rate. Higher at low speeds, lower at high speeds.

I then simulated "air resistance" of these EMUs. This was achieve by having one of the ENG files of an 8 and 12 car trainset
having a lower fricton values at higher speeds. This gave me the different balancing speeds on level track for the same EMU.
ie
4 car trainset = 82mph
8 car trainset = 84mph
12 car trainset = 85mph

For a few years, this setup has performed well. Then OR v1 got released, and i read about the extra functions and real physics, so i tried one of my EMUs in OR -

Noooooo. Nothing worked correctly as they did in MSTS. The Hi-Lo acceleration function did not work, the brakes did not work because of the "blended" function. And the power in the ENG files were so low that in OR, i had trouble getting over 30mph, its low because of how MSTS simulates certain aspects to get the correct performance of these trains.

Back to the drawing board, or so i thought.

I then use one of the default unmodified EMU trainset (One ENG and four WAGs) using real world figures for horsepower and force and was astonished in how they perform to my abovel set up ( power curves between 0 and 70mph and rolling resistance ).
Above 70mph they went too fast. Again using WAGs as ENG files, and adjusting the tractive curve above 70mph to balance at 83 mph on level track. The peformance was still about 75% to my above setup

The brakes had a constant brake force and i could not simulated the retardation rate, as OR for some reason, won't read the blended parameter for brakes.

Also, the balancing speed made no difference if using an 4, 8 or 12 car train set, i still get to 83 mph, so it looks like OR does not simulated air resistance for trains with more traction motors(EMUs).

Another thing with the Hi LO acceleration parameter, was i got the same initial starting acceleration in the four power notches. In OR, the power is absolute at 25% 50 % 75% 100%.

Then i stumbled upon this parameter "ORTSTractionCharacteristics" I can now get all the performances curves spot on to real life values, simulated "camshalf resistance notches" and get the same rate of initial starting acceleration in any power notch. Even better, i can now give the four power notches their design balancing speed of 10, 35, 70 and 82 mph. What a welcome feature this is.

I can live without the blending brakes for the time being, but im loss in how to adjust the fricton in WAG/ENG files in OR for air resistance.
I read that OR will adjust the frictons values to its own internal physics code. I then notice another OR parameter :-

ORTSDavis_A ( x )
ORTSDavis_B ( x )
ORTSDavis_C ( x )
ORTSBearingType ( Roller )

Does this override OR default fricton values ? If so what parameter needs to be change to allow lower fricton at high speeds?

Thanks

[baldwin]

Been a while since I played with these figures, but air resistance increases with speed. However, using the Davis calculations from Fcalc2 as a starting point I arrived at a set of figures that match the resistance curve for Mark 1 stock published by BR.

ORTSDavis_A ( 482.68 )
ORTSDavis_B ( 11.1877 )
ORTSDavis_C ( 0.4952 )
ORTSBearingType ( Roller )

[coolhand101]

Been a while since I played with these figures, but air resistance increases with speed. However, using the Davis calculations from Fcalc2 as a starting point I arrived at a set of figures that match the resistance curve for Mark 1 stock published by BR.

ORTSDavis_A ( 482.68 )
ORTSDavis_B ( 11.1877 )
ORTSDavis_C ( 0.4952 )
ORTSBearingType ( Roller )

Many thanks for this data. I will try this on my 4 car EMU.

Will the resistance still be exactly the same for a 4, 8 or 12 car EMU? I ask this because im very happy with the 4 car EMU resistance using the default OR friction physics.
If the resistance is still the same at high speed for an 8 or 12 car EMU then i will modify a single coach per 4 car, to lower its resistance at high speed. I just need to know what Davis parameter to change. The first parameter "A" looks like the same value as the first MSTS friction line ie xxxN/m/s.

Of course air resistance, based on multiple units, could be added in future OR releases.

Thanks

[baldwin]

What you need first of all is a proven curve to plot results against. Then use Fcalc2 to generate the Davis figures and work from that point. Take a run with the base figures in place and plot the differences. The C line is supposed to include an element of air resistance and for lead and trailing cars should be higher than those inside. As always with friction figures it is a case of adjust and test until you arrive at a set that fits the situation.

[coolhand101]

Ive been using my emu curves for 6 years, so I know what to look for. I was unsure of the third value in the Davis parameters. In msts, it was the last value in the friction line that affected the resistance at high speed. I was looking for the OR Davis parameter equivalent.

Thanks

[baldwin]

The MSTS friction line contains 5 elements and works in a different way to the Open Rails Davis formula so there is no direct equivalent. If there is no ORTSDavis data in the eng file Open Rails will take the existing friction line and apply its own internal Davis conversion.

As the Davis formula only has three sections one cannot say which of the three would have the same effect as manipulating the last figure set in the MSTS friction line. I can only say that the last element of the Davis formula contains the car air resistance figure. All you can do is like I have already said, set up a base Fcalc Davis set for each car in the set. Using Fcalc should produce different figures for the lead car if requested. Then it is simply a case of running the train to see how it fits your curve. Then adjust A,B or C and run again. Also make sure you have the ORTSBearing line in using Roller or Low.

The higher number of cars per train, the higher total friction becomes. Assuming you have the correct power levels per car then performance should be broadly similar across an emu/dmu family assuming each set has the same number of powered cars relative to unpowered.

[coolhand101]

Hi
Yes, i hoping for easier solution. I explain why.

In the MSBin extended HUD, there was an acceleration meter for M/S/S. this very useful for checking predictable balancing speed on grades.
It was also very useful to check my progress on an 8 or 12 car EMU for air resistance at high speed, also coasting on the level.

In OR extended HUDs, there is no acceleration meter so this is very time consuming, and cannot check figures directly. The odometer in OR, gives me some indication that an 8 and 12 EMU is peforming better at high speeds. ( i was using this temporary method with different power levels on the mention power cars ). But its more trial and error without an acceleometer as i know exact figures in m/s/s to look for.

On another matter. Gravity. No matter what the weight of the EMU, being 4 , 8 or 12 car, coasting downhill. They all accelerate (under gravity) on a 1% ( 1 in 100 ) grade at the same speed, and reach the same distance at the same time. This was also the case in MSTS.

The OR extended HUD, shows the train weight, ranging from 150 to 450 tons. But the physics don't seem to apply for this.

Thanks

[baldwin]

MSTS physics seemed to be lacking in some areas for whatever reason. Open Rails physics for diesel and electric has been designed by rail engineers far as I am aware and so should match real world performance if everything is correct. Far as coasting downhill goes, a heavier train should accelerate faster and thus arrive at the end point earlier even allowing for the extra air and rolling resistance.

The acceleration meter is something you could ask to be added at some point, but would need someone to be interested enough to code it.

I know all about time consuming. Took best part of a day to get a MN with 13 on to perform to real world figures, now got to do it again because the goalposts have moved slightly.

[markus1996]

[...]Far as coasting downhill goes, a heavier train should accelerate faster and thus arrive at the end point earlier even allowing for the extra air and rolling resistance. [...]

No.

To explain the physics of this, however, I need to do go for a small digression from topic: Acceleration by the Earth´s gravity on any object in free fall always is 9.8 m/s/s, and thus objects in free fall all accelerate a the same rate - this is without air resistance, of course, like seen in these videos:

• Dropping a feather and a hammer on moon: https://www.youtube.com/watch?v=5C5_dOEyAfk
• Dropping a feather and a hammer in the world´s largest vacuum chamber: https://www.youtube.com/watch?featur...Ct9JobjOg#t=31
• The same is also described here in theory: http://www.physicsclassroom.com/clas...-Misconception

The same holds true for objects rolling down an incline: Absent rolling and air resistance, they will accelerate at the same rate, irregardless of their weight, as they are still subject to the gravity-acceleration of the earth, which is fixed at said 9.8 m/s/s.

Thus, the only thing that will make heavier cars accelerate differently downhill than lighter cars on a railway in reality is their rolling resistance / wheel-to-rail friction and their air resistance.

In ORTS, things are even easier, because to the best of my knowledge, air resistance is currently not simulated in the program. Thus, only the friction parameters of two cars should influence how fast different cars accelerate down an incline by gravity - and at the relatively slow speeds we simulate, that influence should not really make a visible difference.

Cheers, Markus

[baldwin]

Yes indeed with no air resistance the hammer and feather arrive at the same time. However do that in the real world with zero wind, and I guarantee the hammer will hit the ground first. Air resistance is modelled in Open Rails as a component of the Davis equation. The 4 car EMU will reach its terminal velocity sooner than the 12 car EMU because it is lighter.

[coolhand101]

Hi

What i was hoping to do for the 8 car EMU in OR : Between Point A and Point B, the distance is 5 miles:-

4-car EMU : 150 tons -

Point A - 1 in 100 - power shut off at 65 mph and coasting
Point B - 1 in 100 - coasting at 75mph


8-car EMU : 300 tons( same emu but 2 x 4 cars) -

Point A - 1 in 100 - power shut off at 55 mph and coasting
Point B - 1 in 100 - coasting at 75 mph.

The 8 car reached the same point, starting at 10 mph lower.

Thanks

[baldwin]

OK, we are looking at the same problem from different points of view. Assuming the EMU is one of those using stock similar in size to the BR Mk1 stock use my Davis figures as a start point as they fit the resistance curve all the way out to 80mph when loco hauled. Having been thinking on this, if that does not fit your curve properly, try reducing the third element in stages for the carriages behind the lead unit. That has the effect of reducing air resistance which is a component of Davis_C. Do the 4 car unit first. When that fits the curve and speed requirements of the downhill roll, the 8 and 12 cars should behave in the same way with the same Davis settings.

[coolhand101]

OK, we are looking at the same problem from different points of view. Assuming the EMU is one of those using stock similar in size to the BR Mk1 stock use my Davis figures as a start point as they fit the resistance curve all the way out to 80mph when loco hauled. Having been thinking on this, if that does not fit your curve properly, try reducing the third element in stages for the carriages behind the lead unit. That has the effect of reducing air resistance which is a component of Davis_C. Do the 4 car unit first. When that fits the curve and speed requirements of the downhill roll, the 8 and 12 cars should behave in the same way with the same Davis settings.

I was thinking almost the same method!

The Stock is SR EMU Mk1. I will try the above method out.

Thanks

[markus1996]

Yes indeed with no air resistance the hammer and feather arrive at the same time. However do that in the real world with zero wind, and I guarantee the hammer will hit the ground first. Air resistance is modelled in Open Rails as a component of the Davis equation. The 4 car EMU will reach its terminal velocity sooner than the 12 car EMU because it is lighter.

Sorry, forgot about the Davis parameters, was only thinking of friction as ORTS derives it from a car´s mass.

Yes, exactly. If there is a difference, the heavier consist would likely be the slower one, due to increased friction between wheel and rail - but it would in no event be the one to accelerate faster, as you originally stated.

Cheers, Markus

[coolhand101]

Been a while since I played with these figures, but air resistance increases with speed. However, using the Davis calculations from Fcalc2 as a starting point I arrived at a set of figures that match the resistance curve for Mark 1 stock published by BR.

ORTSDavis_A ( 482.68 )
ORTSDavis_B ( 11.1877 )
ORTSDavis_C ( 0.4952 )
ORTSBearingType ( Roller )

Hi

When using the above Davis data, A 35 ton coach, travelling at 90 mph, has a rolling resistance of 390 ibf or 1.735kN.

This seems too low ? That is just my observation. As my figures have always been higher on all my stock for the last 8 years.

With regards to Fcalc2, i always use axles of four for an eight wheel bogie train at around 35 tons. So the first parameter with N /m/s, always have a figure around 750N. It looks like your DAVIS A parameter of 482, is base only on two axles when using Fcalc2! Is that correct ?

Thanks