Elvas Tower: Diesel Mechanical Transmission - Elvas Tower

Jump to content

  • 2 Pages +
  • 1
  • 2
  • You cannot start a new topic
  • You cannot reply to this topic

Diesel Mechanical Transmission Some problems with having a gearbox in the eng file Rate Topic: -----

#1 User is offline   darwins 

  • Superintendant
  • Group: Status: Elite Member
  • Posts: 1,222
  • Joined: 25-September 17
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 19 July 2019 - 12:42 AM

Strange things are happening with a train that has a mechanical gearbox.
Including certain ORTS parameters cause OR to crash - DieselPowerTab and DieselTorqueTab

Without these parameters included the train works, but with other issues.

If the gearbox is set to manual, then although the engine stops at max rpm with 100% of throttle when in neutral, if the train is in gear the engine will overspeed to a much higher rpm (and a much higher speed of train) quite quickly and quite happily.

With some variations tried even with gearbox set to manual the train itself seems to change gears automatically.

With gearbox set to automatic thus:

GearBoxUpGearProportion( 0.98 )
GearBoxDownGearProportion( 0.58 )
GearBoxOperation( automatic )

Then gears seemed to change at more or less the correct speeds. Given GearBoxMaxSpeedForGears( 15.3 27 41 65.5 ) and assumed units in mph.

Strange things still happened with the speedometer needle disappearing in the cvf view and gear numbers not always corresponding to what they were expected to be.
With some variations of information gear change happened at a lower speed and I had tried changing the above numbers to km/h.
There were some trial where even though automatic was specified I needed to put the train into first gear using the E key before it would set off.

A copy of the train is attached. Together with some thoughts on other parameters in a the text file.

Attached File(s)



#2 User is offline   darwins 

  • Superintendant
  • Group: Status: Elite Member
  • Posts: 1,222
  • Joined: 25-September 17
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 19 July 2019 - 12:44 AM

Should add to that on one occasion OR went into earthquake mode.
Diesel engine racing, no forward movement apparent, speed shown as NaN and the whole world shaking violently through as seen through the cab window.


#3 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 20 July 2019 - 03:36 PM

Hello !
Just discover your post. I'll try "online" later, but I'm sure that a mechanical gearbox config in MSTS file is not compatible with a complete "include" OR eng. file.
It could be interesting for you to test my latest config for an hydromechanic transmission, which is available here : http://www.bb25187.e...800_YJ_v300.exe
If you test it, you'll see it works well, but that only parameters supported by OR in the "include files" are idle/max/starting/rate of change RPM, max power and "smoke"/ All are effective in the game, but log.txt mentions that OR eng. file is incomplete (no values of power in f(RPM), no torque values, no power curves). If you introduce these values, OR crashes or becomes erratic.
In fact, there's only two solutions :
- First is to treat an automatic gearbox like an electric transmission, with variations of power corresponding to different ratio speeds. In this case, you loose variations of RPM with changing gear (i.e. if Throttle is set on 100 %, you'll stay continuously to maximal RPM, even if there's a changing of speed ratio)
-Second is to treat classically by using old MSTS scripts : you'll get variations of RPM following variations of speed and gear ratio, but you can't introduce power curves, torque variations, nor fuel consumptions values in fonction of RPM, because those values remain fixed following old MSTS codes.
In the case of French X2800, I have chosen the second way.
In fact, the problem is simple : OR Power Curves, or non-linear variations of power are specific to Electric transmissions. Adaptation to mechanical or semi-automatic transmission would assume that the notion of a variable gear ratio is introduced in the logics, which is not the case at the moment.
I'll try tomorrow to configure BR 104, following the same scheme that the one I used for X2800.
Best Regards,
Jean-Paul

#4 User is offline   darwins 

  • Superintendant
  • Group: Status: Elite Member
  • Posts: 1,222
  • Joined: 25-September 17
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 20 July 2019 - 10:08 PM

Thanks for that Jean-Paul
What you have said was not clear from the OR manual. I think you will find that the parameters that are included are much the same as with X2800 model.
I can understand that TE curves would not be needed for a regular geared model. Though as we discussed elsewhere they can help to model gears or stages in hydraulic or hydromechanical transmission.
I am less clear why the power tab or torque tab would not be needed for a gear transmission - these are properties of the diesel engine, not of the final drive. Now starting to see that OR caclulates both of these from the gears information.If you delete these lines from the MSTS part of the eng file you will find they are not needed and do not change performance:
MaxPower ( 270hp ) Comment(* max power at rails - calculated from mean speed and tractive effort *)
MaxForce ( 6570lb ) Comment(* first gear @ 15.3 mph *)
MaxContinuousForce ( 6570lb )
The next thing that makes me think is about GearBoxMaxSpeedForGears( 15.3 27 41 65.5 )
GearBoxMaxTractiveForceForGears(29200N 16500N 10700N 7150N )
It was not clear if the speed should be in mph or km/h - there seemed to be some benefits of each, but mph looks generally better.
Still the behaviour is not correct. I think these are bugs that a developer needs to look at: (1) Although manual gearbox is specified - OR seems to add an automatic gear change half way through first gear
(2) As said before Max rpm in neutral is 1800 as it should be, but with a manual gear box OR allows the engine speed to rapidly rise much higher if you do not change gear - that should not be possible.


#5 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 21 July 2019 - 12:38 AM

View Postdarwins, on 20 July 2019 - 10:08 PM, said:

Thanks for that Jean-Paul
What you have said was not clear from the OR manual. I think you will find that the parameters that are included are much the same as with X2800 model.
I can understand that TE curves would not be needed for a regular geared model. Though as we discussed elsewhere they can help to model gears or stages in hydraulic or hydromechanical transmission.
I am less clear why the power tab or torque tab would not be needed for a gear transmission - these are properties of the diesel engine, not of the final drive. Now starting to see that OR caclulates both of these from the gears information.If you delete these lines from the MSTS part of the eng file you will find they are not needed and do not change performance:
MaxPower ( 270hp ) Comment(* max power at rails - calculated from mean speed and tractive effort *)
MaxForce ( 6570lb ) Comment(* first gear @ 15.3 mph *)
MaxContinuousForce ( 6570lb )
The next thing that makes me think is about GearBoxMaxSpeedForGears( 15.3 27 41 65.5 )
GearBoxMaxTractiveForceForGears(29200N 16500N 10700N 7150N )
It was not clear if the speed should be in mph or km/h - there seemed to be some benefits of each, but mph looks generally better.
Still the behaviour is not correct. I think these are bugs that a developer needs to look at: (1) Although manual gearbox is specified - OR seems to add an automatic gear change half way through first gear
(2) As said before Max rpm in neutral is 1800 as it should be, but with a manual gear box OR allows the engine speed to rapidly rise much higher if you do not change gear - that should not be possible.


Hello !
With mechanic transmission, ORTS allows, on first speed, to stay at max RPM during starting (approx. up to 2/3 of max speed on first speed), to avoid "zero power" which could result of a too low RPM. It's even possible to start directly, like in reality, on second speed ! In fact - and this is a good and realistic thing - OR reproduces gliding of converter or clutch : motor speed doesn't correspond to wheel speed * ratio gear during starting procedure.
Concerning speed as written in GearBoxMaxTractiveForceForGears, it really seems that mph is the only adequate unit, although ORTS is virtually able to integrate km/h or m/s. By opposite with TractionEffortCurves in which m/s is the only convenient unit ! Similarly, in OR eng.files, torque values must be expressed in N*m. I agree with you concerning the fact that torque values are characteristic of the diesel motor and not of the whole engine : Torque (N*m) = Power (Watt)*60 /(2*Pi(=3.141592)*RPM (rev/mn). In fact, to define a complete mechanic transmission, it would be sufficient to introduce in OR files a "gear ratio" concept, which is an ensemble of coefficients, variable for every gear speed, which could calculate wheel forces by multiplying torque (motor) by this coefficient. For instance, a (motor) torque of 5000 N*m with a "demultiplication" or "gear ratio" of 4 would give an effort at wheel of 20 kN (Beware, this coefficient is not only calculated starting of gear ratios, but also taking in account wheel diameter, exactly like for a motorcar). It doesn't seem as complicated it may look ! (and it would avoid to mix old MSTS logics with newer ORTS).
Well, I'm going to try to do my best with DMU 104 !
Best Regards,
Jean-Paul

#6 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 21 July 2019 - 04:24 AM

Hello !
Some news : I have tested to introduce Torque and power values in an "include" eng file. Then, in both cases, OR crashes, because given values are different of those calculated by Openrails for a mechanic Diesel engine (An "include" file has a priority on classical .eng file, so, if anything is unconvenient, OR crashes - With just a classical .eng file, in case of inappropriate values, OR "forgets" them and substitute its own logics if they exist). When I suppress Torque and power values, it works with the incomplete .eng include. In this case, only fuel consumption and smoke parameters are taken in account.
To be continued...
Jean-Paul

#7 User is offline   darwins 

  • Superintendant
  • Group: Status: Elite Member
  • Posts: 1,222
  • Joined: 25-September 17
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 21 July 2019 - 06:11 AM

View PostJean-Paul, on 21 July 2019 - 04:24 AM, said:

Hello !
Some news : I have tested to introduce Torque and power values in an "include" eng file. Then, in both cases, OR crashes, because given values are different of those calculated by Openrails for a mechanic Diesel engine


I understand why now - but I hope this will be explained in the next update of the OR manual, so that others will not spend time doing what is not needed.


#8 User is offline   darwins 

  • Superintendant
  • Group: Status: Elite Member
  • Posts: 1,222
  • Joined: 25-September 17
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 23 July 2019 - 08:32 AM

I just found an older thread related to this.
http://www.elvastowe...pped%22__st__30

It suggested that rate of change of rpm should be greater than or equal to 1/3 max rpm. But using:

ChangeUpRPMpS ( 600 )
ChangeDownRPMpS ( 600 )

or even
ChangeUpRPMpS ( 1200 )
ChangeDownRPMpS ( 1200 )
does not change the strange behaviour at all.


#9 User is offline   darwins 

  • Superintendant
  • Group: Status: Elite Member
  • Posts: 1,222
  • Joined: 25-September 17
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 27 July 2019 - 01:41 AM

Some further observations.
Having just realised that another power car with MSTS eng file was pushing mine http://www.elvastower.com/forums/public/style_emoticons/default/sign_sorry.gif So now just experimenting with a single power car.
( 1 ) Using ORTS diesel engines only diesel mechanical transmission does not work. The engines rev up and make lots of smoke but there is no power output. The engines rev up and make lots of smoke but the gears remain firmly in neutral and there is no power output at the wheels.
( 2 ) If you add only the following lines of MSTS code (outside of the ORTS diesel engine definition) then the manual gears will work:

DieselEngineIdleRPM( 400 )
DieselEngineMaxRPM( 1800 )

regardless of those factors being defined within the OR diesel engine section or not.
However the power output from the engines fluctuates wildly in a sawtooth fashion in first gear.

https://i.imgur.com/iqt5OPl.jpg

There is some fluctuation in higher gears but it is fairly minimal and not like the wild situation in first gear.
( 3 ) On the other hand if there is only an MSTS definition for the diesel engine an no ORTS diesel engines are defined then power output looks reasonable
https://i.imgur.com/J9Leaeh.png
Not sure if this helps solving the problem, but it may mean that for the moment it is best to remain with the MSTS diesel engine definitions when dealing with mechanical transmissions.


#10 User is offline   R H Steele 

  • Executive Vice President
  • PipPipPipPipPipPipPipPipPip
  • Group: ET Admin
  • Posts: 3,403
  • Joined: 14-March 13
  • Gender:Male
  • Location:known universe
  • Simulator:Open Rails
  • Country:

Posted 27 July 2019 - 07:14 AM

I would think that if the prime mover was a diesel engine, the ORTSDieselEngines block ( in its entirety ) would apply -- no matter what the method of power transmission was. While the block does not use the MSTS style dieselenginerpm values it does contain IdleRpm and MaxRpm which should apply to any diesel engine.
I don't know if OR is properly coded for mechanical, hydraulic, or other types of power transmissions other than electrical generator or alternator.
If you PM me the diesel specifications I'll make up an OR diesel engine block as a template for you to experiment with. Interesting thread, thanks for the work.

  • 2 Pages +
  • 1
  • 2
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users