Elvas Tower: Wheelslip on steam engines - Elvas Tower

Jump to content

  • 3 Pages +
  • 1
  • 2
  • 3
  • You cannot start a new topic
  • You cannot reply to this topic

Wheelslip on steam engines Wheelslip occurs on wet rail with cutoff at 0 % Rate Topic: -----

#11 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 28 February 2018 - 10:07 PM

 ErickC, on 28 February 2018 - 12:11 PM, said:

Oh, I deleted the clip. I also deleted all of the default track sound clips because they're garbage.

More general answer :
In Europe, most of steam engines had a FoA < 4 for a simple reason : axle load was generally limited to 20 metric tons/axle (rarely 22 or 23 on a few lines ). The tested engine : 141 R was, in fact, an adaptated and improved version of light 141 USRA, delivered in France between 1945 and 1947 by Alco, Baldwin, Lima and MLW. So, in calculation of adhesion, isn't FoA influency widely overestimated, essentially because most of conceptors worked only on American locomotives, on which axle load may reach 30 t/axle ? In France, 141 R hauled coal trains from North to Paris suburbs with a load of around 1000 to 1200 met.tons on an itinerary with some steps approaching 1 %. Slowly, but surely, and I can ensure that North of France climate is not as dry as Nevada's one !! :p . As a child, I saw so many of these trains to affirm it's a fact !! On the French Riviera, where it also rains sometimes, they hauled the famous "Mistral" with an admitted max load of 800 metric tons, with a perfect regularity ! And what to say about our "Pacifics" whose FoA was around 3, authorized tu haul express up to 550 met.tons on Paris to Cherbourg line (Normandy... where it doesn't rain very much, but very often...::). So, I did an .eng giving, "on the sheet of paper", a reasonable adhesion. Curtius-Kniffler parameters are adapted to adherent mass (in fact, "ORTSDriveWheelWeight" is totally inoperative, didn't you notice it ?). MaxTractiveEffort is calculated on a basis of 75% cutoff and regulator wide open. In these conditions, answer is double :
- First hypthesis is that there is a mistake in calculation mode (wrong factor, unit conversion errors...)
- Second hypothesis is that a too rigourous approach of adhesion problems dwells to such complex calculations that wheelslip is the "default solution", by overcharge of processor...
The way by which steam modelzation has been approached is what we call in French "une usine à gaz" (no possible translation, but it means that anecdotic parameters have been taken in consideration with a priority against more important factors..).
When a railway director had to determinate a maximal load for an engine - let's take our 141 R - he started from its adhesion mass : 80 metric tons, and estimated that a wet rail has tipically an adhesion around 0.14. 80*0.14 = 11.2 metric tons, let's say around 110 kN. With 110 kN, taking in account inerty of friction wheelboxes and mechanism, on a step of 1 %, you can start with around 950 metric tons load without sanding, even if it's better to sand rail to avoid little wheelslips. Calculation is as simple as this ! And it's not me who explains this, but specialists like André Chapelon or Nigel Gresley...
Cylinder cocks were open, to allow not to apply the whole effort on wheels (they are completely unefficient, except for esthetic reasons, under OR...), ans I must precise that sanding is applied just BEFORE starting, because there is no interest to sand the rail behind the wheel you try to move...!!!
It remains that starting an heavy train by poor adhesion remains - and must remain - difficult.. But not impossible in the limits of reasonable physical parameters.
I also think, and that was a good idea, indeed, that acceleration of wheel is taken in account to minimize tractive effort. Well, but I agree with many to find its influence is excessive, this remark being valuable for all types of engines... With modern railcars, it may grow up to almost 20 % of initial effort, even with high coefficient of adhesion... Isn't it a bit too much ? Must say that having a little experience on driving railway engines, I know that you never must apply full power roughly.. but only at low speeds ! At 120 km/h (75 mph), a "loss by slip" of 10 % is clearly excessive on dry rail.
Best regards,
Jean-Paul

#12 User is online   steamer_ctn 

  • Open Rails Developer
  • Group: Status: Elite Member
  • Posts: 1,879
  • Joined: 24-June 11
  • Gender:Male
  • Country:

Posted 01 March 2018 - 02:29 AM

Firstly to be crystal clear, in OR the diesel/electric adhesion model is completely different to the steam model, so it is not appropriate to make comparisons between them.

 Jean-Paul, on 28 February 2018 - 10:07 PM, said:

More general answer :
The tested engine : 141 R was, in fact, an adaptated and improved version of light 141 USRA, delivered in France between 1945 and 1947 by Alco, Baldwin, Lima and MLW. So, in calculation of adhesion, isn't FoA influency widely overestimated, essentially because most of conceptors worked only on American locomotives, on which axle load may reach 30 t/axle ? ...........

You have provided quite a few comments, some of which, I currently don't understand, nor in some instances agree with.

I think that the best way to resolve this is to build a common testing environment so that we can both test with the same model so that results can be compared. This process will soon demonstrate whether calculations are flawed.

Is the SNCF R141 a freeware model? Can you build an appropriate consist up to the tonnage that you believe should be possible to start on a 1% slope? Ideally it would be easier to only have a single wagon in the consist to reduce the size of various transmissions.

Can you please provide the test consist, and the relevant download links for the locomotive and wagon.

Thanks

#13 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 01 March 2018 - 11:14 AM

 steamer_ctn, on 01 March 2018 - 02:29 AM, said:

Firstly to be crystal clear, in OR the diesel/electric adhesion model is completely different to the steam model, so it is not appropriate to make comparisons between them.


You have provided quite a few comments, some of which, I currently don't understand, nor in some instances agree with.

I think that the best way to resolve this is to build a common testing environment so that we can both test with the same model so that results can be compared. This process will soon demonstrate whether calculations are flawed.

Is the SNCF R141 a freeware model? Can you build an appropriate consist up to the tonnage that you believe should be possible to start on a 1% slope? Ideally it would be easier to only have a single wagon in the consist to reduce the size of various transmissions.

Can you please provide the test consist, and the relevant download links for the locomotive and wagon.

Thanks


Thanks for your answer. I did notice model wa quite different, but still have not understood how it is calculated. I attach a view from 141 R on flat ground, cutoff @ 0 %, wheelsliping while no effort is applied on the wheels, as described by HUD F5. Before testing anything, don't you think there is clearly a problem ? Which could occure with another engine, this one beeing not particular. How to slip without any effort applied, and all brakes liberated.. I don't know, and that's the problem !!
Best regards,
Jean-PaulAttached Image: 141_R_511.jpg

#14 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 01 March 2018 - 11:27 AM

 Jean-Paul, on 01 March 2018 - 11:14 AM, said:

Thanks for your answer. I did notice model wa quite different, but still have not understood how it is calculated. I attach a view from 141 R on flat ground, cutoff @ 0 %, wheelsliping while no effort is applied on the wheels, as described by HUD F5. Before testing anything, don't you think there is clearly a problem ? Which could occure with another engine, this one beeing not particular. How to slip without any effort applied, and all brakes liberated.. I don't know, and that's the problem !!
Best regards,
Jean-Paul 141_R_511.jpg

Oh... I add this view with regulator @ 55 % - before "wheelslipping" - just to explain that no force was applied before slipping, which would be = 0 because of wheelslipping... How to explain the two forces (tangential and static), whilst the only force existing at this instant is... gravity !??
Thanks,
Jean-Paul

Attached thumbnail(s)

  • Attached Image: 141_R_511 (2).jpg


#15 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 01 March 2018 - 12:18 PM

Lasr snapshot :
Just before wheelslipping, in a step of 1.0 %. Resulting forces are widely positive, tractive effort is around 60 % of amxi allowed by adhesion coefficient, but can't go further without slipping... An explanation, please !
Thanks,
Jean-Paul

Attached thumbnail(s)

  • Attached Image: Step_141_R_511.jpg


#16 User is online   steamer_ctn 

  • Open Rails Developer
  • Group: Status: Elite Member
  • Posts: 1,879
  • Joined: 24-June 11
  • Gender:Male
  • Country:

Posted 01 March 2018 - 01:02 PM

 Jean-Paul, on 01 March 2018 - 11:14 AM, said:

Before testing anything, don't you think there is clearly a problem ?

Before I consider a problem with the OR calculations, I want to make sure that we have a well defined test scenario.

 Jean-Paul, on 28 February 2018 - 10:07 PM, said:

So, I did an .eng giving, "on the sheet of paper", a reasonable adhesion. Curtius-Kniffler parameters are adapted to adherent mass (in fact, "ORTSDriveWheelWeight" is totally inoperative, didn't you notice it ?).

For example, you have suggested that ORTSDriveWheelWeight is "totally inoperative", however in the ENG file that you supplied at the start of the thread, ORTSDriveWheelWeight is included in the file in both the wagon and engine sections. This duplication of parameters could cause issues. Remove the parameter in the wagon section, and then vary the value in the engine section. You should notice the value for WheelM in the locomotive HUD varies as you change the value.

I strongly suggest that you format your files more clearly to make them easier to check, and read. Have a look at this page for some suggested section headings, etc.

If you are interested in looking at the basis for the OR model, then have a look at this page. In particular you can read the reference at the bottom of the page for references to rotation forces, and slip.

When you have supplied the test models, and scenario, we can start looking at any potential issues and explanations.

Thanks

#17 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 04 March 2018 - 09:54 AM

Hi,
Just to remark that, regarding usual formula, calculation of FoA (factor of Adhesion) is apparently wrong.
In the case of a 141 R, total weight on drivers wheels is 80 metric tons. Max Tractive effort is 175.144 kN ; i.e. 17.853 metric tons. That gives a FoA = (weight on drivers / Max Tractive effort ) = 4.48. Declared value by OR is only 3.9. Could you explain why ?
I've perhaps an idea... Is MaxTractive Effort calculated on basis of 75 or 85 % cutoff ? (Usually, cutoff max is 75 %, at least in Europa, and not 85 %). .eng file limits this max cutoff @ 75 % - which is realistic.
Thanks !
Jean-Paul

#18 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 04 March 2018 - 01:21 PM

Good evening (in France...)
I really must apologize : working under Windows 7 Premium, I discovered yesterday evening that a System file had been damaged by a wrong update from Microsoft (precisely nvapi.dll - being fitted with a NVidia GeForce card). After restoration, and download of a new Microsoft update, I could discover that this problem of illogical wheelslip had completly disappeared. Now, sanding is efficient, and it's possible to open completly regulator with 0 % cutoff without any wheelslip. Similarly, cylinders cocks are normally efficient. I think this M'soft temporar "bug" affected, in fact, ability of my computer to execute multitask works.
I don't if other fans have known the same problem, but I have to recognize it was nearly impossible to detect it. It was just a regular visit on NVidia actualization site which allowed this detection...
With my renewed excuses, and also great thanks to whole Elvastower members who gave me answers ! :sign_sorry: :sign_thanks:
Jean-Paul

#19 User is offline   copperpen 

  • Executive Vice President
  • Group: Status: Elite Member
  • Posts: 3,139
  • Joined: 08-August 05
  • Gender:Male
  • Simulator:MSTS & OR
  • Country:

Posted 04 March 2018 - 01:27 PM

 Jean-Paul, on 04 March 2018 - 09:54 AM, said:

Hi,
Just to remark that, regarding usual formula, calculation of FoA (factor of Adhesion) is apparently wrong.
In the case of a 141 R, total weight on drivers wheels is 80 metric tons. Max Tractive effort is 175.144 kN ; i.e. 17.853 metric tons. That gives a FoA = (weight on drivers / Max Tractive effort ) = 4.48. Declared value by OR is only 3.9. Could you explain why ?
I've perhaps an idea... Is MaxTractive Effort calculated on basis of 75 or 85 % cutoff ? (Usually, cutoff max is 75 %, at least in Europa, and not 85 %). .eng file limits this max cutoff @ 75 % - which is realistic.
Thanks !
Jean-Paul


Hi
I have taken your eng file and put it into a 141R that I had in stock. With just under 800 metric tons behind the tender, I cannot reproduce your problem in rain or snow on a 1% grade, even tried on level track as well. Throttle closed, reverser put forward to 40%, 50% 60% driving wheels do not move until throttle is opened far enough. I have advanced adhesion on.

#20 User is offline   Jean-Paul 

  • Fireman
  • Group: Status: Active Member
  • Posts: 178
  • Joined: 28-October 14
  • Gender:Male
  • Simulator:Open Rails
  • Country:

Posted 04 March 2018 - 02:56 PM

 copperpen, on 04 March 2018 - 01:27 PM, said:

Hi
I have taken your eng file and put it into a 141R that I had in stock. With just under 800 metric tons behind the tender, I cannot reproduce your problem in rain or snow on a 1% grade, even tried on level track as well. Throttle closed, reverser put forward to 40%, 50% 60% driving wheels do not move until throttle is opened far enough. I have advanced adhesion on.

Thank you very much for your help. Previous message gives, I think, the explanation of my problem... Let's say "Thank you" to Microsoft (or NVidia ?) not to test inter-compatibility of their updates... With surprising consequences, at least !
Thank you again !
Jean-Paul

  • 3 Pages +
  • 1
  • 2
  • 3
  • 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