[cesarbl]

darwins, on 13 November 2020 - 09:24 AM, said:

This works for brake controllers that use either RELEASE - LAP - APPLY or CONTINUOUS SERVICE. It does not work with stepped brake controllers like the Westcode.

I said that TrainBrakesControllerGraduatedReleaseStart wasn't necessary, but the other tokens with graduated release, e.g. TrainBrakesControllerEPContinuousServiceStart are.

darwins, on 13 November 2020 - 09:24 AM, said:

I found that none of these work. They all behave as far as I can see in exactly the same way as TrainBrakesControllerFullServiceStart

In fact, FullService behaves the same as Apply, but these tokens behave like ContinuousService (should ContinuousService have graduated release too?). If there's no one opposed to what I said in the previous post, I'll modify it.

[darwins]

Quote

In fact, FullService behaves the same as Apply

I did find a difference in testing - FullService will reduce brake pipe pressure at service rate until it reaches MaxSystemPressure - FullServicePressureDrop, whereas Apply will reduce brake pipe pressure at the service rate until it reaches zero.
As far as I can see ContinuousService and EPContinuousService must always have graduated release. At present neither can be used as a notched brake - both again behaving in the same way as FullService when you try to make notches with them.

Perhaps there is a case to add both TrainBrakesControllerBrakeNotchStart and TrainBrakesControllerEPBrakeNotchStart - both having graduated release.

[Weter]

cesarbl, on 12 November 2020 - 08:45 AM, said:

• Neutral: can be modeled with TrainBrakesControllerNeutralHandleOffStart, but OR code must be changed in order to isolate EQ from BP. There's already TrainBrakesControllerHoldStart which can be used to keep pressure constant against leakage (the EQ remains connected in this case).

Has sence: now I use SelfLap, as Hold doesn't work, but it's looks incorrect.

Quote

Notched brakes (i. e. first notch makes a reduction of 0.3 bar, second of 0.6, etc.): probably the Dummy/default notch type should behave like this, but the way it is implemented now is strange.

Yes, as MSTS-style with value, followed bu ApplyStart causes under ORTS effect of continious increasing, not self-lap, so I have to change it to Dummy

Quote

For uncommon features of brake controllers, like some kinds of dynamic brake blending or valves which apply EP brakes without reducing EQ res, the scripting interface is a powerful tool. For example, this script https://github.com/S...keController.cs models a sprung brake valve with Release, Hold and Apply positions, commonly used in Spanish locomotives.

Quote

Thanks for that. I should probably learn how to script things - or find someone who can do that. It would help to cover the unusual cases.

I can say the same: it's long time, I'm going to test and see, what the brake and circuit breaker scripts are, and wanted to ask about that.

[Weter]

Let's say about EP brakes purpose and advantages:
Fast and accurate response when apply
Fast and graduated releasing-and for that, in case of single pipe system the absence of BP discharge is essential- the BP must be ready to resupply ARs on each car as quick as it possible.
I often use an ЭТ or ЭД series EMUs to get to my job in the morning and from time to time observe the testing of air brake during approach to pre-terminal stop (you know, what it's doing for) the sequence of test as follows: the driver turns ЭПТ(EPT)* off and performing the brake apply by service discharge of BP, using V position of 395 valve instead VЭ. The difference is significant: I hear the escaping air hiss from the cab's side (valve itself), then 1-2 seconds needed to such drop, when distributors could "feel" it and open their own ports to atmosphere-so hiss under wagon's floor added; more 1-2 seconds of discharge and finally the brake effect increases gradually-I feel increasing deceleration, becoming significant after another 3-4 seconds. Then, if desired effect is observed (speed decreases by 10-20 kph), the driver releases the brakes, it costs 3-5 seconds, then turns EPT on and starts targeting to platform's end: the brake applied instantly and faster 2-3 times, without sayed 3-5sec lag, and, in case of over-brake and as well to avoid of digging at stop, the brake effect surely can be quickly and accurate reduced by partially releasing of brakes by short-time placing the handle of 395 from III to II position and back again to III.
_________
* electro-pneumatic brake (controller) электропневматический тормоз

[darwins]

Quote

As far as I can see ContinuousService and EPContinuousService must always have graduated release.

I should have realised that there is one (sort of) exception to that - the Westinghouse 26-L in Freight mode. Probably the most common brake in North America for the last 70 years! It works now with ECP as well as with regular air brakes.


Like most 'modern' brakes it has a PASSENGER / FREIGHT switch. In Passenger mode ContinuousService works normally - that is it has graduated release. In freight mode it does not have graduated release. It may be that TrainBrakesControllerFreightContinuousService might be needed as a special case for eng files in Freight mode.

I have come to accept that for the moment I need to have different eng files for different braking modes - when the new eng file format is put together I hope it will allow me to included switches that can change modes - that is instead of having four or five different eng files for one locomotive it will allow me to select the braking mode. A typical selector switch in a dual braked loco might have five modes: AIR GOODS AIR PASSENGER Light Engine VACUUM BRAKED VACUUM UNBRAKED Something similar could also be used for changing performance when changing voltage on dual voltage trains - or even perhaps for changing from diesel to electric on bi-mode trains.

[Weter]

Maybe I'm not right, but it seems to me, that FullServiceStart releases all the air from BP of my MSTS models too-it not stops at...
Oh, maybe the MinSystemPressure was referred equal to zero...

[cesarbl]

I attach here a .dll replacement for unstable U2020.11.14-1121 with some of the new brake tokens. It includes the following parameters:

• ORTSTrainBrakesControllerMaxOverchargePressure: to be put in the Engine() section of .eng file. It controls the maximum pressure the brake pipe reaches when overcharged.
  
• ORTSTrainBrakesControllerOverchargeEliminationRate: for Engine() section. Rate in bar/s at which pressure in EQ reservoir is reduced to normal pressure after overcharge. A good value is 0.0025 bar/s.
  
• ORTSBrakeInsensitivity: for .wag files. If pressure in BP is reduced at a rate lower than this value, the triple valves will stay in Release position. Recommended value: 0.005 bar/s (0.3 bar in 60 seconds).

These brake notches have been modified or included:

• TrainBrakesControllerOverchargeStart: notch to overcharge the brake pipe.
  
• TrainBrakesControllerNeutralHandleOffStart: notch to isolate EQ reservoir from BP.
  
• TrainBrakesControllerEPApplyStart: to apply EP brakes. Can be used as a non smooth notch.
  
• TrainBrakesControllerGraduatedEPApplyStart: same as above but with graduated release.
  
• TrainBrakesControllerGraduatedSelfLapLimitedStart: to gradually apply and release brakes.
  
• Dummy or nothing specified: similar to GraduatedSelfLapLimitedStart, but can be used as a non smooth notch.

I modified EPApplyStart so air brakes aren't applied, but I will revert that to keep compatibility with models where EPApply was used because Dummy wasn't available.
Note that this is just a experimental version, so many of the tokens will change.

Attached File(s)

•  Orts.Simulation.zip (896.99K)
  Number of downloads: 426

[Robo_TR]

Hello,

can someone check following situation in source code, please ?

http://www.elvastowe...668#entry263668

Thank you.

[Weter]

Quote

ORTSBrakeInsensitivity

YYYES!
Thank You, will try.

[darwins]

Many thanks for that. I am very pleased with the overcharge facility. As far as I can see this works perfectly. I have tested this with an old single pipe Westinghouse system with triple valves and with a modern twin pipe air brake system with distributors. (There are one or two other problems I would like to solve with each system, but the overcharge function is a major step forward.) You may need to put ORTSBrakeInsensitivity ( 0.005bar/s ) in a as a default value to prevent any problems with overcharging MSTS legacy stock or OR stock that has not been set up to allow overcharge. I will try out the other brake tokens tomorrow.