Another Physics Thread

[ENR3005]

Good evening All,

I don't pretend to understand the physics side of this sim however being a longtime former MSTS user it became pretty clear from day one when I started using this sim that there was some room for improvement. Over the past few years we have seen the addition of scripted braking which has been a welcomed improvement to newer products and I look forward to seeing further improvements which will surely come. Older products however are just painful to run and watch when compared to the recent advances with braking scripts. I have tried adding the newer braking scripts to older engines by way of trickery such as cloning a scripted engine such as an SD40-2 and making it a entirely new engine such as a GP15 by way of aliasing the shapes. This works however you have to use the cloned locomotive's cab in order for the braking script to function and requires a considerable amount of modification to work well. I am however not content with using a SD40-2 cab for a RS11 just for improved braking. I am also not content with several nice looking engines setting sidelined with default physics files and my railroad having to lease power with scripted brakes to get the type of operation we have waited years for. Wanting to fix my older power, I spent some recent time building upon the findings of others in some older threads who I am very appreciative for their time trying to make this sim more realistic. Again I will emphasis I am not a physics guy but I am pretty logical when it comes to experimentation. I have built upon the findings of other such as Kali along with a desire to have my older power function like my newer purchases. I have also borrowed values from some of my older MSTS equipment along with comparing my results to equipment in Open Rails. What I have been able to do this far is change the brake recharge and release rates and make them mimic their scripted counterparts. This is a start and I am hoping someone will build upon what I have provided. Attached to this message you will find two different brake value modifications, one for freight cars and one for engines for anyone who wants to experiment. In order for this to work you "MUST" modify both your rolling stock and engines. This is a simple cut and paste process for your Engine Simulation.bin file and wagon.bin file which can be done in less than 30 seconds with RWTools. I would suggest taking one freight car and an engine and building a test train of course being sure to back up the originals before modifying. So far I have modified my SD70s, SD40-2s, GP40-2s, GP15s, SW1500s along with thirty pieces of rolling stock. I consider this an improvement over what I was running up till last week however I will let everyone else try and judge for themselves.

Chris

[ENR3005]

Just a little note for anyone who downloaded the engine braking values, I had the loco brake setup wrong. I had it set at 159 and should have been 59. I uploaded a corrected file.

[GreatNortherner]

Hi,

Thanks for sharing these. Very interesting stuff -- and awesome to see a renewed push with RW physics development.

Did I understand you correctly that these are intended for "old style" cars and locos, without using the scripted brakes simulation from recent releases by DTG or Smokebox? I intend to give these a try. I have no clue how a real world train behaves, but I can compare to how I remember it from that refined MSTS equipment where everybody said it was so accurate (and fun/challenging to run in the game).

I've got a couple more questions about the technical values, to try and better understand them and the differences between your data set and the default settings.

37.5</MaxCylinderPressure> and 37.5</PressureForMaxForce>
That is 50/50 on some default or even 90/70 on some 3rd party models -- quite a difference. What purpose do the lower figures serve?

Emergency braking 10</TriggerRate>, is 2 in your set. What does value that do?

Lastly, not a question: 0.05</MaxChargingRate> -- wow, that is 1.69 on the Kuju cars. No wonder that the old default trains could stop & start again immediately.

I'm going to give this a try later today.

Cheers!
Michael

[ENR3005]

Hi Michael,

I actually started my files by using numbers others had used from an older physics thread. viewtopic.php?f=9&t=6291 The values you have asked about were setup on the similar test files provided by simer4 at that time which I have also used and are in that thread. I haven't experimented with those numbers very much as my first goal was to change the release and charge rate in the engine files which I think I am on the right track with. The release and charge rate was a bit of a critical milestone in whether or not this experimental project would have a future at all. Happy with the results thus far and the brakes taking more than a minute or more to charge and release I will start looking at some of the other values.

Chris

[ENR3005]

Good evening all,

Well it seems my above noted modification is a lot easier to do then cut and paste a whole brake section like I noted above. After some more experimenting today, all you need to do is edit the following two lines in your simulation file. The following two lines are from RSC's UP United Way SD40-2, an older DLC item without scripted brakes. Now here is the interesting part which I mentioned earlier, I used values straight from an MSTS SD40-2.eng file with enhanced physics. The original lines have been pasted below. Anyone care to try this and confirm my findings. Why would the default setting be set so high, they appear to be off by a decimal point. 0.93 and 0.45 also work for any experimenters.

Default TS2016 file
<MaxReleaseRate d:type="sFloat32" d:alt_encoding="000000A099992240" d:precision="string">9.3</MaxReleaseRate>
<MaxApplicationRate d:type="sFloat32" d:alt_encoding="0000000000001240" d:precision="string">4.5</MaxApplicationRate>


Original MSTS SD40-2 Engine File, Values Copied

MaxReleaseRate( 0.95 )
MaxApplicationRate( 0.51 )


Modified TS2016 file

<MaxReleaseRate d:type="sFloat32" d:alt_encoding="000000A099992240" d:precision="string">0.95</MaxReleaseRate>
<MaxApplicationRate d:type="sFloat32" d:alt_encoding="0000000000001240" d:precision="string">0.51</MaxApplicationRate>

[ENR3005]

Good day,

A little more experimenting and clarification to my last post. Only the train brake assembly section is to be modified, not the loco brake assembly. You can watch the pressure release and application rates by pressing F5 twice in the sim and watching the results. A image is below of what you should see. I've had a few folks have emailed me advising their train won't move or start braking instantly, that is the way it is supposed to work. We need to un-think the way we have been operating, as this is closer to the way the brakes should behave. Instant releases and applications just don't happen in the real world. More experimenting and work still needs to be done though.

Chris

Default TS2016 File Train Brake Assembly

<MaxReleaseRate d:type="sFloat32" d:alt_encoding="000000A099992240" d:precision="string">9.3</MaxReleaseRate>
<MaxApplicationRate d:type="sFloat32" d:alt_encoding="0000000000001240" d:precision="string">4.5</MaxApplicationRate>


Original MSTS SD40-2 Engine File, Values Copied

MaxReleaseRate( 0.95 )
MaxApplicationRate( 0.51 )


Modified TS2016 File Train Brake Assembly

<MaxReleaseRate d:type="sFloat32" d:alt_encoding="000000A099992240" d:precision="string">0.95</MaxReleaseRate>
<MaxApplicationRate d:type="sFloat32" d:alt_encoding="0000000000001240" d:precision="string">0.51</MaxApplicationRate>

[GreatNortherner]

Hi,

Thanks for the further testing. I don't fully understand your two latest additions though: wouldn't only changing the set/release times in the loco's brakes slow down the brake application and release process, yet once applied the braking would still be too powerful? In other words, I still think that the brakes on the cars should be modified too (and if it's only for the MaxBrakeForcePercent value).

I quite like the results I got from your first post's attachments. I compared them to stock Kuju cars and then also to the slightly modded brakes on the RWA route's equipment and found your settings to offer the most "MSTS-like" braking.

Has anybody tried comparing these experimental brake settings with the equipment from Feather River Canyon? Being the most recent route DLC I suppose the rail vehicle physics should be DTG's latest & greatest (but I don't have the DLC to give it a try).

Cheers
Michael

[mrennie]

Has anybody tried comparing these experimental brake settings with the equipment from Feather River Canyon? Being the most recent route DLC I suppose the rail vehicle physics should be DTG's latest & greatest (but I don't have the DLC to give it a try).

The problem there is that the values in the brake setups for the cars in FRC will have been chosen to work well with the scripted braking for the FRC locos. If you use them without the scripting, you won't get the same effect.


There's one thing about brake release that you need to take into account. In real life, the way the triple valves on freight cars works is that the brakes on a car release when the brake pipe pressure at the car's branch pipe rises by at least 1.5psi. At that moment, the valve allows the air in the brake pistons to vent to atmosphere completely. That happens very quickly and there's no stopping it (unless the car is fitted with retainers to hold a minimum set amount of pressure in the brake cylinders). The reason that the brakes can appear to come off slowly is that (a) it takes a certain amount of time to raise the pressure in the brake pipe by 1.5psi (and in the meantime the brakes don't release at all), and (b) the further away the car is from the head end, the longer it takes for that pressure rise to reach that car - so the brakes don't all release at the same time, but instead it's like a ripple going from the head end to the tail end. The longer the consist, the longer it takes for the pressure to rise by 1.5psi and the longer the ripple takes. The upshot of all that is that on a long consist, it would be right to expect the brakes to take longer to release, but on a short consist, it's much quicker and depends mostly on how quickly the locomotive can increase the brake pipe pressure (which is typically quite fast). Therefore, if you're going to fiddle with the brake release times in the cars' blueprints, it should only be for when you're running them in a long consist.

[ENR3005]

Hi Michael,

My first post noted my primary goal was to fix the release and application rates. You are correct the freight cars should be adjusted too with proper brake forces however it is not necessary for the release and application recharge rates. The freight cars should still be modified per my original post for enhanced braking and do work with better results as you noted. I am still doing some reading on the calculations others have used and testing of my own. I am also curious if anything can be tweaked in the car files to change the application and recharge rates too.

Chris

Hi,

Thanks for the further testing. I don't fully understand your two latest additions though: wouldn't only changing the set/release times in the loco's brakes slow down the brake application and release process, yet once applied the braking would still be too powerful? In other words, I still think that the brakes on the cars should be modified too (and if it's only for the MaxBrakeForcePercent value).

I quite like the results I got from your first post's attachments. I compared them to stock Kuju cars and then also to the slightly modded brakes on the RWA route's equipment and found your settings to offer the most "MSTS-like" braking.

Has anybody tried comparing these experimental brake settings with the equipment from Feather River Canyon? Being the most recent route DLC I suppose the rail vehicle physics should be DTG's latest & greatest (but I don't have the DLC to give it a try).

Cheers
Michael