[captain_bazza]

Herb, if you are performing multiple brake operations you would eventually run out of air - until the res's were recharged?

With MSTS, you could tweak the brake controller to 'insert' a dead[?] step, or 'notch' in the sequence. Remember my infamous sticking throttle trick? BB C16, was it? I recollect it caused you to mutter some technical term, or other, under your breath, at the time of discovery ! ;-)

Cheers Bazza. http://www.elvastower.com/forums/public/style_emoticons/default/pleasantry.gifhttp://www.elvastower.com/forums/public/style_emoticons/default/girl_devil.gif

[gpz]

atsf37l, on 14 November 2014 - 12:35 PM, said:

OK, bug reports submitted on this issue as well as the problem of not having a detent between "Lap" and "Running" which makes it entirely too easy to lose your brake set.

We'll see what the gurus can come up with. :good2:

I have tried a fix in r2656, please try.

[atsf37l]

Cap'n: That's kind of what I had in mind with a detent built into the 'sweep' of the brake 'handle.'

Thanks Peter, I'll give it a go. :good2:

[atsf37l]

OK Peter, I checked it out comparing 2656 with 2618. There seems like there might be a little more resistance between Lap and Running, but one thing I noticed that there is a definite detent between Running and Release. If there were a similar detent between Lap and Running that would make a big difference. It might be as simple a fix as increasing the dwell time in Lap like it seems to be in Running.

Also, to repeat an issue in my original bug post, Running should not release the air but merely reflect that last application or release. If Running didn't dump the brakes this whole thing wouldn't be that big an issue. :good2:

[gpz]

It is really strange what happens on your side. Could you please copy here your train brake notch configuration from the eng file you have problem with? Please note I added a stop only for "Lap" type notch, but not for "SelfLap", as you asked in your bug report. The detent between Running and Release is not because of some special programming, that must be because of the eng config.

About Running behavior you wrote in the bug report:

Quote

This is actually a dual bug in that "Running" should not immediately release the air brake set. That should only happen in "Release." "Running" is like a second "Lap" except that the set will gradually release in "Running" where it won't in "Lap."

Now you write something different. Any behavior can be programmed, but exact responses to conditions should be defined. But I am confused here. Could you please write some lines in a form:

If Condition 1, then Response 1 should happen.
If Condition 2, then Response 2 should happen.
If Condition 3, then Response 3 should happen.
...
...
If Condition n, then Response n should happen.

Programs work this way, general instructions cannot be feeded in.

[atsf37l]

If the brake control is moved from Release through Running and through Rap into Apply then the brake pipe pressure should be reduced until the control* is moved back to Lap, or Running or Release.

If the control is moved from Apply to Lap (status quo) the reduction in brake pipe pressure should cease and the progress of the control* should stop at lap and not be allowed to move to running or release for a brief but notable period of time OR (preferably) without another tap of the ";" key.

If the control* is moved from Lap or Release to Running there should be no change in the brake status - i.e. the brakes should not set or release.

I hope that clarifies this somewhat. It should be noted that the detent at Lap and the releasing air in Running are two separate but related issues.

These are the train brake settings:
Brake_Train ( 0 1 0.1 0
NumNotches( 6
Notch(0 0 TrainBrakesControllerReleaseStart )
Notch(0.2 0 TrainBrakesControllerRunningStart )
Notch(0.4 0 TrainBrakesControllerSelfLapStart )
Notch(0.5 1 TrainBrakesControllerApplyStart )
Notch(0.9 0 TrainBrakesControllerFullServiceStart )
Notch(1.0 0 TrainBrakesControllerEmergencyStart )
)
)


* Bearing in mind that in the real world the control is a lever with physical stop points built into the cast quadrant upon which it rotates (kind of like a shifting gate in an automobile with paths not allowing you to shift from drive into reverse without conscious effort) whereas in the sim the controls are up/down push buttons.

[gpz]

Aha! Please note you have the configuration

Notch(0.4 0 TrainBrakesControllerSelfLapStart )

You created the bug report mentioning non-self-lapping brakes, while you are testing a self-lapping one now! I made the code change to have effect only on non-self-lapping brakes. What to do here? Do you amend your proposal to have the detent stop effect on self-lapping brakes too?

About the "Running" issue: Do I understand correctly, that you amended your other statement (in bug report) too? It was: "Running" is like a second "Lap" except that the set will gradually release in "Running" where it won't in "Lap."

[copperpen]

The same principle applies to both self lap and non-self lap. The last digit in each controller line states if it is smooth, or notched.

Notch(0.4 0 TrainBrakesControllerSelfLapStart ) is a notched control and therefore should only move to that position when a key is pressed.
Notch(0.5 1 TrainBrakesControllerApplyStart ) is a smooth control so the movement from lap to braking and back again has no break in it, but when the Lap position is reached, the controller lever should stop.

However, to be fair, the OR behaviour in this is the same as MSTS. The notched controller does work like that if single key presses are used. If the key is held down it acts more like a smooth controller. So perhaps what is needed is a time delay for a brake notch so that the controller stops at that point, then requires a further key press to move again. I know that the notched reverser works properly so maybe take a look at that for comparison.

[gpz]

copperpen, on 21 November 2014 - 04:12 AM, said:

So perhaps what is needed is a time delay for a brake notch so that the controller stops at that point, then requires a further key press to move again.

This is exactly what was implemented. (The latter one, the stop at that point, not the time delay.) This way the behavior diverges from MSTS, because there the brake controller notches continuously, as you have observed correctly. Any behavior can be implemented, just an agreement is needed by the user community. A stop at self-lapping brake too can also be easily added, but the original request was to do this only for non-self-lapping brakes.

[steved]

There are a couple ways to deal with a cars air brakes when preforming switching moves.
The first is to "bottle" the air. This means to close the air cock on the cut of cars to prevent loss of train line pressure thus keeping the brakes from setting up. You can also use this method to stop a cut of cars when you need to by opening the air cock, kind of like using the brake valve on the locomotive.
The other involves bleeding off the cars which basically drains the air in the brake system. There is a small rod that extends from side to side on the car that is connected to a lever on the triple valve that you can pull or push to release the brakes on a single car.
I've used both methods (it's quite common) and I kind of miss the option in ORTS.

Steve

Oh yeah, don't forget to set the handbrakes when you're done.