[roeter]

Using the present version (2796), I have problems with an engine for which the brakes will not release.
This problem is introduced with the recent changes in brakes logic. The engine never used have these problems. I have not run this engine for some time so I cannot be sure at which version the problems were introduced.

The picture shows that the brake controller is at full release, the state of all brakes is shown as release, but nothing is happening - the pressure in the brake cylinder of engine and all wagons stays at the same value as was reached on brake application, and the brakes clearly are not released.



The engine is from a payware route so I can not upload the files here, but these are the relevant values :

	BrakeEquipmentType( "Triple_valve, Auxilary_reservoir, Emergency_brake_reservoir" )
	BrakeSystemType( "Air_single_pipe" )
	MaxBrakeForce( 84kN )

	EmergencyBrakeResMaxPressure( 145.038 )
	TripleValveRatio( 2.5 )
	MaxApplicationRate( 25 )
	MaxReleaseRate( 30 )
	MaxAuxilaryChargingRate( 20 )
	EmergencyResCapacity( 10 )
	EmergencyResChargingRate( 25 )
	BrakeCylinderPressureForMaxBrakeBrakeForce( 50.763 )

	AirBrakesAirCompressorPowerRating( 1.2 )
	AirBrakesMainMinResAirPressure( 0 )
	AirBrakesMainMaxAirPressure( 145.038 )
	AirBrakesMainResVolume( 8 )
	AirBrakesAirCompressorWattage( 10000 )
	AirBrakesCompressorRestartPressure( 120 )
	AirBrakesAirUsedPerPoundsOfBrakePipePressure( 0.22 )
	AirBrakesIsCompressorElectricOrMechanical( 1 )

	EngineBrakesControllerMinPressureReduction( 6 )
	EngineBrakesControllerHasProportionalBrake( 1 )
	EngineBrakesProportionalBrakeLag( 0.9 )
	EngineBrakesControllerMaxApplicationRate( 20 )
	EngineBrakesControllerMaxReleaseRate( 15 )
	EngineBrakesControllerEmergencyApplicationRate( 50 )
	EngineBrakesControllerMinSystemPressure( 0 )
	EngineBrakesControllerMaxSystemPressure( 72.52 )
	EngineBrakesControllerEmergencyBrakeTimePenalty( 0 )
	EngineBrakesControllerFullServicePressureDrop( 26 )
	EngineBrakesControllerDirectControlExponent( 1 )

	TrainBrakesControllerMinPressureReduction( 6 )
	TrainBrakesControllerMaxApplicationRate( 20 )
	TrainBrakesControllerMaxReleaseRate( 4 )
	TrainBrakesControllerEmergencyApplicationRate( 50 )
	TrainBrakesControllerMinSystemPressure( 0 )
	TrainBrakesControllerMaxSystemPressure( 72.52 )
	TrainBrakesControllerEmergencyBrakeTimePenalty( 0 )
	TrainBrakesControllerFullServicePressureDrop( 26 )


      BrakesEngineBrakeType( "Air_single_pipe"  )
      BrakesTrainBrakeType( "Air_single_pipe" )
      BrakesEngineControllers( "Independent, Train, Dynamic" )
    
             Brake_Train ( 0 1 0.0125 0.28 
                NumNotches ( 6
                    Notch ( 0    0 TrainBrakesControllerReleaseStart )
                    Notch ( 0.2  0 TrainBrakesControllerHoldLappedStart )
                    Notch ( 0.25 1 TrainBrakesControllerGraduatedSelfLapLimitedHoldingStart )
                    Notch ( 0.85 0 TrainBrakesControllerSuppressionStart )
                    Notch ( 0.9  0 TrainBrakesControllerContinuousServiceStart )
                    Notch ( 0.95 0 TrainBrakesControllerEmergencyStart )
                   )
              )

Only the train brake was used.

Any idea what is causing this problem?

Regards,
Rob Roeterdink

[gpz]

As I see, the main reservoir pressure is 3.8 bar only which is the problem here. Main reservoir will supply the air to pump up the BP to 5 bar, but at its current state it is unable to. I think at this time you are waiting for the compressor to fill the main reservoir, thus fill the BP too.

Is this state at startup? I noticed the main reservoir not always initializing to its full pressure. I made some changes about it in r2799. Could you please try that?

[hroch]

Hi
AirBrakesMainResVolume is very small. Put for exaple 20. Write up parameter ORTSMainResChargingRate ( 0.6 ) in compressor.

[hroch]

Or MainResChargingRate ( 0.6 ). I don't know which write parameter is functional.

[Csantucci]

Apparently the immediate brake release does not work well any more for vacuum brakes. Using the default Scotsman, if one uses the immediate brake release (Shift-etcetera, depending on the international settings of the keyboard), BP is immediately decreased instead of increased, and then slowly returns increasing to reach the final value of 21.

[roeter]

I'm not in the habit of commenting on programming errors - I make plenty enough myself.
But this one really has me :good: :furious: :censored2: :diablo: .

Let me show you why.
The problem is in the AbstractPowerSupply class, and in particular in this bit of code :

        public PowerSupplyState AuxiliaryState
        {
            get
            {
                return auxiliaryState;
            }

            protected set
            {
                if (auxiliaryState != value)
                {
                    auxiliaryState = value;

                    if (Locomotive.Train != null && Locomotive.IsLeadLocomotive())
                    {
                        foreach (TrainCar car in Locomotive.Train.Cars)
                        {
                            MSTSWagon wagon = car as MSTSWagon;

                            if (wagon != null)
                            {
                                wagon.AuxPowerOn = AuxPowerOn;
                            }
                        }
                    }
                }
            }
        }

What it does : on change in auxiliary power, if the change is for the leadlocomotive, it distributes that change to all wagons in the train.
So why is this so wrong and why is it making me so angry?

Well, here it is.
This is what happens on restore :

• This code is part of the restore for cars - but cars are restored before the leadlocomotiveindex is restored.
  So the check for IsLeadLocomotive will always fail.
  
• Even if the leadlocomotive index was restored, it still would not have worked, for the car is not yet added to the train - that is done when restore is completed.
  
• As each car is restored in turn, the state can never be passed from the leadlocomotive through the train as the other cars have not been restored yet.

By the way, this does not work for AI trains at all as for AI trains the leadlocomotive index is not set.

So why does that make me so angry? Well, obviously this method can never have worked on restore - which means this has never been tested when introduced. And that is :censored2: :clapping: .

For now, I have set a patch (version 2839) which, for each wagon, always sets AuxPowerOn to true on restore. So at least the brakes work properly again.

Regards,
Rob Roeterdink

[vince]

roeter, on 06 February 2015 - 03:02 AM, said:

I'm not in the habit of commenting on programming errors - I make plenty enough myself . . . . . . . . . . . . SNIP. . . . . . . . . . . .
So why does that make me so angry? Well, obviously this method can never have worked on restore - which means this has never been tested when introduced. And that is :censored2: :thumbdown3: .

For now, I have set a patch (version 2839) which, for each wagon, always sets AuxPowerOn to true on restore. So at least the brakes work properly again.
Regards,
Rob Roeterdink

Wow! That sort of behavour needs to be stopped! It still causes trouble in MSTS add-on content and it's sad to see it creeping in here. Downright disrespectful not to mention damaging to the brand. Thankfully there is a growing army to catch this internal spam.

I hope the developers are aware that there are some in the Sim world that do not like ANY open source sim as it is contrary to their financial interests. Not saying that this is the case here but it's best to be watchful.

Best,
Vince
:clapping:

[markus_GE]

Vince, could you please clarify what your last post has to do with this discussion?

Cheers, Markus

[vince]

markus_GE, on 06 February 2015 - 09:07 AM, said:

Vince, could you please clarify what your last post has to do with this discussion?

Cheers, Markus

I am suspicious why code was inserted without testing.

vince

[markus_GE]

AFAIK, the code in questions was meant to improve the program, and make it more realistic. However, IDK if it is currently understood to be finished for the time being (IE, completely implemented in its first incarnation) or still work in progress.

Cheers, Markus