[darwins]

In theory according to the Manual, hydraulic transmission is not yet supported.

Two alternatives suggested were:

(i) Mechanical transmission with an automatic gearbox. - Which did not work very well for large fast locomotives, definitely not with only 3 speeds, although it is possible to build in some rise and fall curvature (like the power curves for hydraulic transmission) by setting gear change up % to something above 1.0, say 1.22, giving something better than the sawtooth of a geared transmission.

(ii) Electric transmission using ORTSMaxTractiveForceCurves.

In fact I was very happy to see that this second option works well with locos defined as DieselEngineType ( hydraulic ) and having no reference to Maxcurrent or anything else electrical.

Having taken some time to look at them these curves are brilliant for hydraulic transmission. Any straight line between two points on a tractive effort v speed graph becomes a beautiful parabola when you plot power against speed. Hydraulic transmissions give you exactly that - a series of (very close to) straight lines for TE and a series of curves for power, one for each set of gears or converter stage.

This is a set of curves representing a Voith transmission with 3-converter stages:

https://imgur.com/iFwQyaW.jpg

Another set representing a Maybach Mekydro transmission having a single converter with a four speed automatic gearbox:

https://imgur.com/wodYYJb.jpg

One significant difference here, the Voith changes over from one converter to another at fixed speeds, whereas gear change in the Mekydro transmission depends on the ratio of input speed to output speed.

In terms of y=ax^2+b then for the Voith when the power input is changed both a and b are multiplied by %power to get new curves. For the Mekydro then a remains constant and b is multiplied by the % of maximum rpm of the engine. The gear change speeds for the Mekydro are then also determined by dividing the % of maximum rpm of the engine by different constants.

Some examples of TE curves that I have now attempted to model are:

DieselEngineType ( hydraulic )

Comment (* Tractive Force Curves for 3 speed Voith LT 360r transmission *)
Comment (* Torque Converter changes at 25mph and 44mph *)

ORTSMaxTractiveForceCurves(
0 ( 0 0 40.2 0 )

0.08 (
0.0 22453
10.7 9677
11.6 8891
19.2 4645
20.1 4321
40.2 1900
)

0.44 (
0.0 128304
10.7 55300
11.6 50806
19.2 26545
20.1 24689
40.2 10854
)
... and so on for the Voith.

Also something like this for the Mekydro

DieselEngineType ( hydraulic )

Comment (* Tractive Force Curves for 4 speed Mekydro transmission *)
Comment (* Gear change ratios Engine rpm : Speed mph / 58.3:1 32.9:1 and 20.9:1 *)

ORTSMaxTractiveForceCurves(
0 ( 0 0 20 0 )

0.08 (
0.0 127450
4.9 54009
5.8 45733
9.1 27786
10.0 26481
14.5 18045
15.4 17333
26.8 1250
)

0.44 (
0.0 187534
7.5 76325
8.3 68457
13.5 39721
14.4 39354
21.6 26163
22.5 25681
40.2 10841
)

I could have done the above with only four pairs of numbers for the Voith and five pairs for the Mekydro, but put an extra one either side of the gear change / converter change speeds.

Now that is the only thing that seems to missing from the OR model - the impact of a gear or converter change. You hear it when you travel on trains with such transmission systems.
The engine rpm drops of very rapidly. I would say in something like 2 to 5 seconds it drops to about half of the previous power level, and then rpm and power build up again taking perhaps 10 to 30 seconds or maybe more to get back to their previous level.

Not sure that there is any data around to support that, but that seems to be all that stands between OR now and having a realistic hydraulic transmission.

Perhaps something like the mechanical gearbox input could be adapted to tell OpenRails when converter changes happen in the Voith transmission.

GearBoxNumberOfGears( 3 )
GearBoxMaxSpeedForGears( 22 44 90 )

But something different would have to be input for the Mekydro, so that user could give ratio of engine rpm to train speed (or axle rpm) rather than fixed speeds for each gear change.

[Jean-Paul]

Hello !
Very interesting !
Using for 2-3 years these calculation modes for different hydraulic Diesel engines (railcars and locomotives), I confirm it works perfectly with an excellent restitution of real performances. And I agree it would be great to have such abilities for mechanic transmission with gearboxes, which still depend on "pure MSTS" adjustments.
Best Regards,
Jean-Paul

[R H Steele]

Interesting work. What program did you use to plot the curves?

[darwins]

I did notice the nice OR Caravelle with hydraulic transmission recently. :D
Quite amused by the thought that those things could get to 140 km/h!

The curves are just graphs inserted in Open Office Calc.

Two more thoughts towards wanting OR to know when gears or converters are changed both found in cab -

1. Tachometer

2. Some locomotives had a "transmission changing" light to tell the driver when the change was taking place.

[Jean-Paul]

darwins, on 16 July 2019 - 12:13 PM, said:

I did notice the nice OR Caravelle with hydraulic transmission recently. :D
Quite amused by the thought that those things could get to 140 km/h!

The curves are just graphs inserted in Open Office Calc.

Two more thoughts towards wanting OR to know when gears or converters are changed both found in cab -

1. Tachometer

2. Some locomotives had a "transmission changing" light to tell the driver when the change was taking place.

You're right ! "Caravelles" reached 140 km/h only in downwind...or in downhill !! :lol: The two models by "Compagnons du Rail" represent last series, fitted with a 2-stage converter and, then, "fully hydraulic". First series had an hydromechanic transmission, with a 8-speed gearbox. They were undermotorized railcars, with poor aerodynamics (in fact, their real cruise speed was around 100-110 km/h), but they had a very smooth and comfortable ride, and for the era, were not noisy.
On the oldest railcars, in France, fitted with a manual gearbox, shifting was generally made "à l'oreille" (with ears), or -for neophytes ? - by observing RPM dial(s). It was easy because, very often, driver was sitting just beside motor, or above in the "kiosk" ! Youthful memories : on 4-speed gearbox railcars, starting were often done directly on second speed, and the most skilled drivers were shifting without using clutch, only by playing with throttle control to synchronize gears. Must say that deflection of clutch command was around one foot, with the hardness we can imagine !
"Transmission changing light" is very interesting, but not so easy to realize, because - I'm perhaps wrong - there is no command line in cvf files allowing to correlate a rpm value with a color change (I'm afraid that RPM Dial could only be defined as a dial with a needle). I don't think that "RPM Digital" exists (if so, changing color with value is very easy).
Best Regards,
Jean-Paul

[darwins]

"Transmission changing light" is very interesting, but not so easy to realize, because - I'm perhaps wrong - there is no command line in cvf files allowing to correlate a rpm value with a color change (I'm afraid that RPM Dial could only be defined as a dial with a needle). I don't think that "RPM Digital" exists (if so, changing color with value is very easy).
[/quote]

Both are requests. They are things I would like to see in the future.

RPM should be easier if one of the developers can find a way to read it into a cvf file. Although I can't find it in any of the F5 HuD displays, I assume that it is already calculated somewhere in OR. (There are lots of input references to it.)

Transmission change would need the things I mentioned above to be added into the OR model. You are quite correct, at present it has no way to tell when this is happening. Though I was wondering about putting some sort of flag into the ORTS TE curves as an alternative to giving speeds or ratios for transmission change. Something like this:

ORTSMaxTractiveForceCurves(
0 ( 0 0 20 0 )

0.08 (
0.0 127450
4.9 54009
5.2 0
5.8 45733
9.1 27786
9.4 0
10.0 26481
14.5 18045
14.8 0
15.4 17333
26.8 1250
)

Where the zero is another way to say this is transmission change.

There is another challenge I have just found - although I am not sure if there were many examples - mixed hydraulic / mechanical transmission. There was a strange sort of shunting and trip freight loco in UK (class 14) that had a two stage converter, followed by a direct drive gear.

[Jean-Paul]

Hello !
I've just downloaded your files on uktrainsim. Thanks for OR's diesel fans ! :good:
Best Regards,
Jean-Paul