[pschlik]

Alright, the whole brake charging slowdown thing bugged me, so I did some of my own testing. Using the built-in, unmodified MSTS consist 2d950chm.con (2 dash 9s, 50 tankers) on path noon mixed goods .pat on Marias Pass, I recorded some times to recharge the brake pipe after a full service application. I used 21 psi/s brake pipe charging rate, but turned on correct questionable braking parameters and ran the locomotives at full throttle in neutral as the MSTS Dash 9 has pathetically useless compressors otherwise. I did notice some oddities, but it's nothing I didn't already see.

Stable 1.5.1:
Took 22 seconds for all brake cylinders to empty, 7 minutes and 26 seconds for the brake pipe at the end to reach 90 psi.

Unrealistically slow already, but that's limitations of MSTS content.

Latest Testing (Same brake simulation updates as New Year 145):
Took 37 seconds for some of the brake cylinders to empty as a few wagons got tripped into emergency, the ones that went to emergency released at 3 minutes 34 seconds, and the pipe was fully charged after 8 minutes.

I know exactly what's going on here. The slower brake cylinder release time is down to a change (in triple valve improvements) in the way the emergency reservoir is used during release; for content not set up correctly (MSTS) this will be observed as an increase in the time taken to release the brakes but has no effect on the total time to recharge, so I don't see the problem there. The thing that's actually slowing down recharge, even if by 30 seconds, is a quirk of the emergency brake update and the triple valve improvements combined resulting in these false emergency brake applications. I already noticed that and this is fixed in the fixes for undesired emergency applications.

Latest Unstable (Has emergency fixes, which are NOT in New Year):
Took 44 seconds for all of the brake cylinders to release, took 8 minutes 59 seconds to fully recharge.

The slower recharge time is mostly in the last couple of psi due to changes to make brake pipe flow more realistic (ie: slower). Up until ~88 psi it recharges just as quickly as before any brake updates were made.


Before continuing the witch hunt on brake regressions that have already been looked at, I'd recommend people try the latest unstable to see what's going on there. At this point, there should be nothing causing major differences in recharge time between unstable and stable except when the brake pipe pressure is close (within 5 psi) of the equalizing res pressure. I do know of a way to return the speedup for getting all the triple valves to go to release without breaking the new features, so I'll get that in, but that will not change the time required to fully recharge (it does speed up the initial pressure increase, but slows down the last few psi).

This does highlight that it's quite easy to input nonsense, making even the slightest change to physics have weird consequences, so maybe some more work on correcting questionable brake parameters is due to help avoid these sorts of 'garbage in, garbage out' problems.

[charland]

Putting the engine in neutral and running it at full throttle is something I have not been doing, I'll give it a try and let you know what kind of time I get with the same 11 car consist I've been using. The engine I'm using is an Alco S-2, so wouldn't be surprised at being on the slow side but not an hour to charge 11 cars!

Paul :-)

[Csantucci]

Phillip,
thank you very much for your analysis. However, after some exchange with César, I'd like Paul to perform two further tests, to see if the origin of the problem could be identified.
Paul, if you are willing to, could you try running this official testing version https://github.com/o...-g55f992dcd.zip (generated before the suspect change) and this other one https://github.com/o...-gebafd5dad.zip (generated after)? The difference is PR#829 by the way.

[Luis]

Hello everyone.
Can it be that the curves sound on the vehicle (not on the route) no longer works in the MG 145? With OR 1.5 the sound works perfectly. can this be checked?
Many thanks for the help

Greetings from Zurich

[charland]

OK Carlo, I'm just firing up my computer to get started and I'll download and test both.

I've been around shortlines and local freights when they have been made up. I can't remember ever hearing an S-2 opening up it's throttle while pumping the air, usually it's just been in idle. The 12 minutes I it was taking in 141 was pretty much what I would expect. I haven't been using "Correct questionable braking parameters" because everything had been acting as expected, but I will give that a try as well.

Paul :-)

[charland]

Test results from OR T1.5.1-495

I didn't check the box beside Correct questionable braking parameters because I hadn't for all the other tests. I used the same engine, 11 cars, and caboose and the engine and caboose had their brakes fully charges as in the other tests.

After 12 minutes of charging the train was up to 23 psi.

It took 47:18 to charge the train to 40 psi at which time I suspended the test.

I'll download the next one now and do the test.

Pail :-)

[charland]

OR T1.5.1-507 Test Results.

Used the same rollingstock and the same engine and caboose with full air before connecting the air hoses like all the other tests.

After 12 minutes of pumping air the train was at 21 psi.

To get to 40 psi it took 27:23.

The noticeable difference was that it only too 1:30 for all cars ro reach 10 psi, then it slowed down.

Next two tests in ORNYMG 145 with 'correct questionable braking parameters and then in idle with full throttle.

Paul :-)

[Csantucci]

Paul, thanks again for your patience. Could you please test with Stable 1.5.1? Just to have a counterproof that once things were OK?

[charland]

Just finished testing ORNYGM 145.

Same train and everything the same as previous tests.

Engine in neutral and IDLING

Flow started at 25 L/s

At 36 psi Flow dropped to 17

6:10 to reach 40 psi.

At 12 minutes train was at 62 psi.

At 74 psi Flow dropped to 13.

At 17:20 all cars were at 80 psi.

Second test.

Engine in neutral and in notch 8.

Flow quickly got up to 32 L/s, maybe three seconds from 0 to 32.

At 4:38 all but one car was at 40 psi.

At 76 psi the flow rate dropped from 32 and by the time the train hit 80 psi the flow rate was down to 8 L/s.

At 11:09 all cars were at 80 psi.

End

I'll do the test in the Stable 1.5.1 now Carlo.

Paul :-)

[Csantucci]

Luis, on 24 August 2023 - 02:23 AM, said:

Hello everyone.
Can it be that the curves sound on the vehicle (not on the route) no longer works in the MG 145? With OR 1.5 the sound works perfectly. can this be checked?
Many thanks for the help

Greetings from Zurich

Nothing has changed in that code. I just tested the motor of an electric locomotive and heared its pitch changing with speed.