[steamer_ctn]

After watching the thread titled "Perpetual stall Problem", and providing information and suggestions to a number of different people in recent months, I felt that it was time to develop a consistent way to test the performance of steam locomotive so that it could be determined whether a performance issue was due to the Open Rails program, a poor physics definition in the ENG file of the locomotive, or an incorrect perception of how the locomotive should be performing.

Hence I have started to develop a standard process to test and compare steam locomotive performances. Currently the process is still in development, and is described on the Steam Locomotive Performance Testing page.

It is proposed that the testing process is done in a common environment, so therefore it will allow people to swap information and duplicate test results more easily on different computers. This will also allow more specific suggestions, as well as allowing advice to be shared to identify and resolve problems.

It suggests that the initial step is to define a realistic performance benchmark. I have used an English Hall class locomotive to demonstrate a methodology that I believe can be used to establish a realistic performance benchmark. Once the benchmark is defined, tests can then be run to compare the performance of the locomotive against the benchmark.

Currently the test process proposes a load test, and as a future development, possibly a steam performance test might be developed.

I have set a Hall locomotive up with a number of Open Rails and the original MSTS configurations, thus allowing an easy direct comparison of Open Rails and MSTS.

After running some comparison test it has become very clear that Open Rails and MSTS calculate performance based upon very different values in the ENG file.

Interestingly this is seen very clearly demonstrated by the difference in the displayed tractive effort value for MSTS and OR files. The MSTS file shows a value of approximately twice the actual value of the Hall Class locomotive. The error is caused by an incorrect value in the drive wheel radius parameter in the original MSTS file. It appears that MSTS doesn't use this value to calculate the tractive effort, and therefore is not affected. Whereas OR uses this value and other key steaming parameters to calculate tractive effort. Thus, as a first step, it is important to check the values in an ENG file, when a locomotive does not appear to be operating correctly.

This process should be able to be used on all steam locomotive once appropriate research is done by the tester.

As suggested the process is still under development, so any suggestions will be considered and factored into the process as appropriate.

[captain_bazza]

Great Idea, Peter. There are so many variables come into play in the ENG and with the differences between MSTS and OR, a test workshop forum is a great idea and an ideal place to swap theory and practice.

Cheers Bazza.

[Gehe]

Hi All,



I just started to perform some tests with steam locomotives and as a first result do not understand the behaviour of the model with respect to the Parameter ORTSBoilerEavporationRate ( x ).
According to the documentation I understood the purpose of this parameter is to adjust the maximum output of a boiler because boliers with the same evaporation area may have different outputs due to differences in the construction which invreases or decreases their capability to prduce steam.
When I change the evaporation area ORTS adapts the max boiler output accordingly as expected. But if I change the ORTSBoilerEavporationRate parameter I do not see any visible effect at least not in the F5 HUD.

Maybe my understanding of this parameter is not correct, what is your experience here?

Regards
Gehe

[Csantucci]

Two points after a fast OR code inspection:
first a stupid question: you write ORTSBoilerEavporationRate . It is instead ORTSBoilerEvaporationRate (a and v exchanged), but maybe this is only a typo in your post.
Second: the value you insert is clamped between 10 and 15. So it cannot be beyond these values.

[mopacfan]

Hi,

What an excellent proposal to have a steam locomotive performance test ideally using control variables shared between testers!

My only suggestion to this proposal is to ideally use a model of a type of steam locomotive or wheel class that regularly saw service in Europe, the United States, and elsewhere globally. I raise this suggestion to use a model of a type of steam locomotive that saw international service because there would be more data available to testers around the world.

[copperpen]

ORTSBoilerEvaporationRate is not a parameter I have ever tested and it does not even exist in the OR advanced eng files I am using. Having done a quick static test, it makes no difference that I can see.

[steamer_ctn]

Gehe, on 18 October 2015 - 07:35 AM, said:

I just started to perform some tests with steam locomotives and as a first result do not understand the behaviour of the model with respect to the Parameter ORTSBoilerEavporationRate ( x ).

As correctly pointed out below, this parameter is incorrectly spelt. Having checked the manual again, it appears that it is due to an incorrect spelling which occurred when I submitted the content to the manual. The manual will be corrected. Thanks for the feedback on the error.

For clarity, the correct spelling is ORTSBoilerEvaporationRate.

As also pointed out, the parameter will only accept values between 10 and 15, and it currently defaults to the maximum value of 15 when no value is entered, so in effect this parameters is only used to reduce the boiler evaporation rate.

mopacfan, on 18 October 2015 - 09:00 AM, said:

What an excellent proposal to have a steam locomotive performance test ideally using control variables shared between testers!

My only suggestion to this proposal is to ideally use a model of a type of steam locomotive or wheel class that regularly saw service in Europe, the United States, and elsewhere globally. I raise this suggestion to use a model of a type of steam locomotive that saw international service because there would be more data available to testers around the world.

Whilst I generally agree with the comment, and it would certainly apply for confirming the correctness of the code, this test process is also about trying to have a consistent method of tuning locomotives when detailed test data and reports are not readily available. This can be done by using operational information.

Having said that, I am happy to receive contributions to enhance and develop the testing process further.

[copperpen]

copperpen, on 18 October 2015 - 11:55 AM, said:

ORTSBoilerEvaporationRate is not a parameter I have ever tested and it does not even exist in the OR advanced eng files I am using. Having done a quick static test, it makes no difference that I can see.

I am correcting the last part of the above statement. I have just run a test again, and there is a marked difference in steam production between the two extremes allowed. If used it will of course make a steam locomotive just that little bit harder to operate, which is a good thing. The old MSTS made it far too easy, never ran out of steam for one thing.

[Gehe]

Thanks for the quick reaction.

The "ORTSBoilerEavporationRate" is just a typing error.
In my .eng the spelling is correct.

Does your last answer mean that there is a difference in steam production but the max value in the F5 HUD remains unchanged or let's say the flexibility of the boiler within the limit given by the evaporation area can be adjusted by this Parameter? With flexibility I mean the agility of the boiler to react on changing demands for steam.


Regards
Gehe

[copperpen]

If you look at the locomotive information page, where all of the data is displayed, you will see a maximum that the boiler can produce which does not change. On the line above is the figure showing the actual production figure. It is this one that is altered by using the ORTSBoilerEvaporationRate parameter. I have not tested so far to see how fast the boiler reacts, or how high it will go in production.