Elvas Tower: US Steam Locomotive Performance - Elvas Tower

Jump to content

  • 2 Pages +
  • 1
  • 2
  • You cannot start a new topic
  • You cannot reply to this topic

US Steam Locomotive Performance

#11 User is offline   scottb613 

  • Superintendant
  • Group: First Class
  • Posts: 1,150
  • Joined: 06-July 09
  • Gender:Male
  • Location:"Upstate" New York
  • Simulator:ORTS
  • Country:

Posted 27 April 2018 - 09:55 AM

:discuss_gathering:

View Postcopperpen, on 27 April 2018 - 08:47 AM, said:

Yes you can. I don't know what the maximum is, but I did set one up with 4 exhaust emitters and all worked.


Hi...

Thanks...

I'll give it a shot with a test...

Regards,
Scott

#12 User is offline   steamer_ctn 

  • Open Rails Developer
  • Group: Contributing Member
  • Posts: 850
  • Joined: 24-June 11
  • Gender:Male
  • Country:

Posted 27 April 2018 - 11:44 PM

View Postlongiron, on 27 April 2018 - 04:45 AM, said:

The unique thing about MSTS/OR is the physics is completely separate from the content - i.e. any ENG file can be associated with any S file. So we don't need models for each and every configuration to be tested, just appropriate ENG files with just one S file.

Whilst I understand the ability to use one locomotive shape, and then apply different ENG files to it, I believe that this approach would somewhat defeat the outcome that I was trying to achieve.

One of my aims was to demonstrate the realism of OR and its capabilities to configure a locomotive to perform within reasonable tolerances of the real life locomotive. I feel that, for example, picking a shape such as an Atlantic locomotive (4-4-2) would provide an unrealistic perception if it has a 900 ton load behind it, or alternatively picking a T1 shape and then only being able to pull a Atlantic type load would also give a unrealistic perception.

These demonstration modelsshould be designed to run with representative loads. By this means I also hope to demonstrate that smaller locomotives can't necessarily haul the same load as larger locomotives. My gut feel is that this is not always understood clearly, and then bugs are reported if an overloaded train won't start or get up to a "fast enough" speed. Having a number of different "sized" models might help to demonstrate this.

View Postlongiron, on 27 April 2018 - 04:45 AM, said:

Regarding American steam locomotives - I would try to find data for a 2-8-2 configuration, as this was ubiquitous and good performance baseline.

I don't quite understand this. My thinking is that there will be a number of different (size) locomotives would be set up as examples.

The page reffered to in my earlier post shows reports that are available for some of these different locomotives, including a 2-8-2.

View Postlongiron, on 27 April 2018 - 04:45 AM, said:

Two ENG files are generally required for all articulated models (Player and AI) - and we don't have good guidelines for OR ENG setup.

As the Q2 and T1 are duplex locomotives, I don't think that they need separate ENG files, but hopefully this will be demonstrated in attempting to set them up.

Could you explain why there needs to be separate Player and AI models?

View Postscottb613, on 27 April 2018 - 06:14 AM, said:

I think Articulated engines need remote control of the second set of cylinder cocks

Bearing in mind that we are only discussing duplex locomotives at the moment, can you provide a link to a reference that discusses these sorts of controls for duplex locomotives?

Thanks

#13 User is offline   longiron 

  • Open Rails Developer
  • Group: Elite Member
  • Posts: 2,914
  • Joined: 25-July 08
  • Gender:Male
  • Location:Manasquan, NJ
  • Simulator:Open Rails, MSTS editors
  • Country:

Posted 28 April 2018 - 05:20 AM

View Poststeamer_ctn, on 27 April 2018 - 11:44 PM, said:

Whilst I understand the ability to use one locomotive shape, and then apply different ENG files to it, I believe that this approach would somewhat defeat the outcome that I was trying to achieve.

One of my aims was to demonstrate the realism of OR and its capabilities to configure a locomotive to perform within reasonable tolerances of the real life locomotive. I feel that, for example, picking a shape such as an Atlantic locomotive (4-4-2) would provide an unrealistic perception if it has a 900 ton load behind it, or alternatively picking a T1 shape and then only being able to pull a Atlantic type load would also give a unrealistic perception.

These demonstration models should be designed to run with representative loads. By this means I also hope to demonstrate that smaller locomotives can't necessarily haul the same load as larger locomotives. My gut feel is that this is not always understood clearly, and then bugs are reported if an overloaded train won't start or get up to a "fast enough" speed. Having a number of different "sized" models might help to demonstrate this.

Thanks

My only objective was to point out that we don't have to wait until models are available to 'test' ENG parameters and performance. I'm totally with you on the longer term objective.

The 2-8-2 comment was designed to point out that I'd recommend focusing on getting the physics right for the workhorse American locomotives, rather than exotic limited edition ones like the T1 and Q.

Big US steam was almost exclusively articulated designs. Because of the way MSTS/OR handles models in game, these beasts must be made up of two models and ENG files. As a result, they are treated as player and AI even though in reality it is a "single" ENG from a water and coal usage perspective.





#14 User is offline   steamer_ctn 

  • Open Rails Developer
  • Group: Contributing Member
  • Posts: 850
  • Joined: 24-June 11
  • Gender:Male
  • Country:

Posted 28 April 2018 - 02:20 PM

Hi Chris,

Thanks for the clarification.

View Postlongiron, on 28 April 2018 - 05:20 AM, said:

My only objective was to point out that we don't have to wait until models are available to 'test' ENG parameters and performance. I'm totally with you on the longer term objective.

Fair enough. I am happy if we start work on this basis.

View Postlongiron, on 28 April 2018 - 05:20 AM, said:

The 2-8-2 comment was designed to point out that I'd recommend focusing on getting the physics right for the workhorse American locomotives, rather than exotic limited edition ones like the T1 and Q.

Ok, there is a test report for a PRR L1s (2-8-2) that we could start with, and there is a "L1s" model in TrainSim that we could use as a shape file.

View Postlongiron, on 28 April 2018 - 05:20 AM, said:

Big US steam was almost exclusively articulated designs. Because of the way MSTS/OR handles models in game, these beasts must be made up of two models and ENG files. As a result, they are treated as player and AI even though in reality it is a "single" ENG from a water and coal usage perspective.

I agree that at the moment true articulated models appear to need to be separate ENG files. However my understanding is that both the T1 and Q2 are duplex locomotives and not articulated ones. So I am thinking that they will be able to be handled as a "standard" ENG file configuration.

Now the question, who would like to take on the lead for setting up the physics for the L1s? Who else would like to be also involved, and support the group activities through research, testing and other assistance?

I would provide guidance and assistance with the process.

Thanks

#15 User is offline   Traindude 

  • Fireman
  • Group: Active Member
  • Posts: 101
  • Joined: 17-November 13
  • Gender:Male
  • Location:Seattle, WA
  • Simulator:MSTS
  • Country:

Posted 28 April 2018 - 07:31 PM

Here's some info on another 2-8-2, the New York Central H10. This was recently posted at Elvas Tower. Skip to pages 497 to 502.

I'll get more info on this later.

#16 User is offline   longiron 

  • Open Rails Developer
  • Group: Elite Member
  • Posts: 2,914
  • Joined: 25-July 08
  • Gender:Male
  • Location:Manasquan, NJ
  • Simulator:Open Rails, MSTS editors
  • Country:

Posted 29 April 2018 - 06:33 AM

The fact is that many designs were closely related:

PRR L-1 is essentially the same as USRA Heavy 2-8-2

USRA Light 2-8-2 (1266 were made) used the same boiler as USRA Heavy Pacific with a slightly different firebox - so steaming characteristics would be very similar.

PRR L-1 has the same boiler and Belpaire firebox as the PRR K-4s, which were very similar to B&O P-7s and Southern's famous Ps4s. These were essentially USRA Heavy Pacifics with slightly smaller fireboxes.

My point is that a few well tested versions can be the foundation to build physics for a broad portfolio of locomotives.




#17 User is offline   Genma Saotome 

  • Owner and Admin
  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Owner
  • Posts: 12,448
  • Joined: 11-January 04
  • Gender:Male
  • Location:United States
  • Simulator:Open Rails
  • Country:

Posted 29 April 2018 - 11:10 AM

AFAIK the driving wheel arraignment has very little to do with what is needed. While fixed measurements, such as driver diameter, evaporative areas, etc., are required, variables like boiler pressure are equally important -- railroads did change the maximum making their own USRA locomotives different from all others.

Baldwin used a ratio of the the total evaporative area (including superheat) divided by total tractive power. The result is normally in the range of 6 to 18. They created speed curves for each round number therein and they look very, very different; A lower number meant a very large ability to produce steam for a rather low power locomotive -- an ideal mix for fast, light trains. A high number was the opposite, ideal for drag freights.

The spreadsheet I have that replicates these curves works extremely well -- for some locomotives. The WP 251 class 2-8-8-2 is a perfect match. Others, well, no. I don't know why. I suppose the answer of "this is all theoretical anyway" might be pretty close to the truth.

The point is knowing what kind of power the locomotive can produce does you no good if you can't match any given amount of power to a specific speed and just eyeballing the driving wheels won't let you copy parameter values from one railroad to another.

#18 User is offline   steamer_ctn 

  • Open Rails Developer
  • Group: Contributing Member
  • Posts: 850
  • Joined: 24-June 11
  • Gender:Male
  • Country:

Posted 10 May 2018 - 02:09 PM

I have now sourced a PRR T1 and Q2 test report, so I feel that there is a good set of test reports available to tune and test some US steamers with.
I have had one offer from a US modeller to work on tuning the physics for one of the locomotives.
Are there any more community members who wish to offer their services to tune another locomotive?
Thanks


  • 2 Pages +
  • 1
  • 2
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users