Thinking about another thread <http://forums.uktrainsim.com/viewtopic.php?t=61808> I thought I'd experiment with 'real world' loads on the steep inclines of the test track from TrainSim.com (I think).
After reading round the subject and trying adjustments to different parameters, it seems to boil down to the "Adheasion" ones. Basically the required adhesion values for realistic uphill haulage make the engine grip like velcro on the flat - not a trace of wheelslip.
Does anyone know how of an .eng file I could look at that manages both?
Thanks,
asharte
PS In the steam era, were loads reduced in snow - ie is the Adheasion factor of 0.55 in default.wag realistic?
Steam & realistic loads
Moderator: Moderators
-
supergoods
- Very Active Forum Member
- Posts: 2752
- Joined: Fri Jun 07, 2002 12:00 am
- Location: Sugar Land, Texas
I feel there are still a lot of open questions on the adhesion parameter.
It would appear that this could be another case where a different programmer at Kuju wrote the internal physics for steam compared to the programmer who did diesels and/or electrics.
The major missing link is establishing the adhesive weight from the data provided.
The mass is stated, the factor of adhesion can be calculated and the steam chest pressure at which the engine should slip on dry rail can be calculated from this and the general engine details.
Conn McCarty and Bill Hobs posting on the US forums and some posts in 2004 here on this subject have made some important contributions on this subject, however there still appear to be questions on how the internal workings of steam work.
There are two NumWheels parameters in an .eng file, one in the wagon section which appears to work on friction not adhesion and one in the engine part which appears to work on the adhesion parameter.
It has been stated that the NumWheels in the .eng file does not have to be a whole number, my experiments at present have a value of 2.01 here which represents total weight divided by adhesive weight as being about the only place where adhesive weight could be factored in.
In the US the axle loading was such that you could almost always get the factor of adhesion in the range of 4.0 to 4.25 without too much difficulty.
Elsewhere this was not the case, for example a certain 4-8-4 from Australia has a factor of adhesion of only 2.97 which means that it should begin to slip at a tractive effort of 37984lbf, despite haveing a conventionally potential tractive effort of 55000lbf.
The calculations used by Con McCarty and Bill Hobs assume the US prototype of a two cylinder engine with full equalization between the drivers and the trailing bogie.
This particular engine is a three cylinder loco and in my opinion should be a little less likely to slip because of the more even torque exerted by multi cylinder engines. This was claimed to be fact by Alco when marketing three cylinder engines before the war and for the purposes of experimentation I used 10%.
The equalization issue would reduce the tractive effort necessary to slip by, for the purposes of experimentation, 5% if equalization is not used. This does not apply here.
Thhe end result is that the engine should slip at a steamchest pressure of 144psi.
Again although it does not apply here locos such as 4-6-0's with no trailing wheels were less likely to slip as the rear drivers did not transfer weight to the trailing wheels on starting.
The driving wheel diameter also effects adhesion. Contrary to what you would think larger drivers have better rail contact, however as there is less space for them under a given locomotive the greater adhesive weight possible with smaller wheels allows more effective use of the tractive effort.
This is a work in progress and more will be posted later. At the moment I can get a loco tuned by observation of the steamchest pressure until I get this balanced, but I would like a more universal calculation than anything I have seen so far.
More to come on this
PS, the Gresley Pacifics tendency to slip on starting, which seems to refute much of the above was more the fault of a poorley designed regulator than adhesion itself.
Ian
It would appear that this could be another case where a different programmer at Kuju wrote the internal physics for steam compared to the programmer who did diesels and/or electrics.
The major missing link is establishing the adhesive weight from the data provided.
The mass is stated, the factor of adhesion can be calculated and the steam chest pressure at which the engine should slip on dry rail can be calculated from this and the general engine details.
Conn McCarty and Bill Hobs posting on the US forums and some posts in 2004 here on this subject have made some important contributions on this subject, however there still appear to be questions on how the internal workings of steam work.
There are two NumWheels parameters in an .eng file, one in the wagon section which appears to work on friction not adhesion and one in the engine part which appears to work on the adhesion parameter.
It has been stated that the NumWheels in the .eng file does not have to be a whole number, my experiments at present have a value of 2.01 here which represents total weight divided by adhesive weight as being about the only place where adhesive weight could be factored in.
In the US the axle loading was such that you could almost always get the factor of adhesion in the range of 4.0 to 4.25 without too much difficulty.
Elsewhere this was not the case, for example a certain 4-8-4 from Australia has a factor of adhesion of only 2.97 which means that it should begin to slip at a tractive effort of 37984lbf, despite haveing a conventionally potential tractive effort of 55000lbf.
The calculations used by Con McCarty and Bill Hobs assume the US prototype of a two cylinder engine with full equalization between the drivers and the trailing bogie.
This particular engine is a three cylinder loco and in my opinion should be a little less likely to slip because of the more even torque exerted by multi cylinder engines. This was claimed to be fact by Alco when marketing three cylinder engines before the war and for the purposes of experimentation I used 10%.
The equalization issue would reduce the tractive effort necessary to slip by, for the purposes of experimentation, 5% if equalization is not used. This does not apply here.
Thhe end result is that the engine should slip at a steamchest pressure of 144psi.
Again although it does not apply here locos such as 4-6-0's with no trailing wheels were less likely to slip as the rear drivers did not transfer weight to the trailing wheels on starting.
The driving wheel diameter also effects adhesion. Contrary to what you would think larger drivers have better rail contact, however as there is less space for them under a given locomotive the greater adhesive weight possible with smaller wheels allows more effective use of the tractive effort.
This is a work in progress and more will be posted later. At the moment I can get a loco tuned by observation of the steamchest pressure until I get this balanced, but I would like a more universal calculation than anything I have seen so far.
More to come on this
PS, the Gresley Pacifics tendency to slip on starting, which seems to refute much of the above was more the fault of a poorley designed regulator than adhesion itself.
Ian
-
supergoods
- Very Active Forum Member
- Posts: 2752
- Joined: Fri Jun 07, 2002 12:00 am
- Location: Sugar Land, Texas
The second NumWheels parameter in the .eng part of the file does have an effect on adhesion but not in the way I expected.
Doubling the mass from 148t to 296t would have been expected to produce a stickier engine, to compensate for this I increased the NumWheels from the 2.01 I had calculated for the relationship between total weight and adhesive weight to the new calculated value of 4.01.
The effect of this was to reduce the tractive effort at the start of slipping to about 50% of the original figure.
Not what I expected.
I later reset the value to 1 and recalculated the adhesion parameter to bring the values back to obtain slipping at 41783lbf tractive effort which is where I calculated it should begin to slip.
So the conclusion is that the mass does not enter into the calculation of adhesion since I get the same results regardless of the mass used in the tests
This will make the variations for tender first running easier to configure without addressing the adhesion parameter.
The next step will be to calculate the adhesion parameter for a different loco to see if the calculation is sound.
More to come on this
Ian
Doubling the mass from 148t to 296t would have been expected to produce a stickier engine, to compensate for this I increased the NumWheels from the 2.01 I had calculated for the relationship between total weight and adhesive weight to the new calculated value of 4.01.
The effect of this was to reduce the tractive effort at the start of slipping to about 50% of the original figure.
Not what I expected.
I later reset the value to 1 and recalculated the adhesion parameter to bring the values back to obtain slipping at 41783lbf tractive effort which is where I calculated it should begin to slip.
So the conclusion is that the mass does not enter into the calculation of adhesion since I get the same results regardless of the mass used in the tests
This will make the variations for tender first running easier to configure without addressing the adhesion parameter.
The next step will be to calculate the adhesion parameter for a different loco to see if the calculation is sound.
More to come on this
Ian
Ian
-
supergoods
- Very Active Forum Member
- Posts: 2752
- Joined: Fri Jun 07, 2002 12:00 am
- Location: Sugar Land, Texas
When attempting the test of an alternative loco I noticed that I had made a mistake when entering the adheasive weight for the 4-8-4
This should have given a factor of adhesion of 3.79.
I reduced the enhancement for multi cylinder to 5%.
Slipping commences at about 175psi steam chest pressure
Interestingly a side effect is that once slipping occurs the regulator must be virtually closed before the slipping stops which is much more like the prototype.
Ian
This should have given a factor of adhesion of 3.79.
I reduced the enhancement for multi cylinder to 5%.
Slipping commences at about 175psi steam chest pressure
Interestingly a side effect is that once slipping occurs the regulator must be virtually closed before the slipping stops which is much more like the prototype.
Ian
-
jbilton
- Very Active Forum Member
- Posts: 19267
- Joined: Fri Oct 10, 2003 12:08 pm
- Location: At home ..waiting to go to Work.
- Contact:
Hi Ian
I have not done much work with adhesion figures( or steam engine files), and generally stick with the default 0.2 and 0.4.
However I believe the first number of wheels are used for adhesion calculations, where as the second number of wheels is a multiplier for traction.
So basically if you put 1 in the first but 6 in the second it'll slip like hell.
The reverse it would hardly slip at all.
For diesels its easy. I use actual number of wheels contacting the rail for the first, and number of traction motors for the second.
Unfortunately light locos with six motors cause trouble....so I usually reduce to 5 or 4.
Otherwise I've found you have to nearly double the actual true mass.
Cheers
Jon
I have not done much work with adhesion figures( or steam engine files), and generally stick with the default 0.2 and 0.4.
However I believe the first number of wheels are used for adhesion calculations, where as the second number of wheels is a multiplier for traction.
So basically if you put 1 in the first but 6 in the second it'll slip like hell.
The reverse it would hardly slip at all.
For diesels its easy. I use actual number of wheels contacting the rail for the first, and number of traction motors for the second.
Unfortunately light locos with six motors cause trouble....so I usually reduce to 5 or 4.
Otherwise I've found you have to nearly double the actual true mass.
Cheers
Jon
------------------------Supporting whats good in the British community------------------------
-
supergoods
- Very Active Forum Member
- Posts: 2752
- Joined: Fri Jun 07, 2002 12:00 am
- Location: Sugar Land, Texas
-
supergoods
- Very Active Forum Member
- Posts: 2752
- Joined: Fri Jun 07, 2002 12:00 am
- Location: Sugar Land, Texas
A series of tests tonight on what works with the adhesion parameter and what might work but does not
Max Power Does not work with Adhesion
MaxForce Works with Adhesion Reducing Max Power increases steam chest pressure before slipping
MaxVelocity Does not work with Adhesion
WheelRadius Does not work with Adhesion
Sanding Does not work with Adhesion
NumWheels Works with Adhesion but if left at "1" has no effect on calculation
CutOffMaxForward Does not work with Adhesion
CylinderStroke Does not work with Adhesion
CylinderDiameter Does not work with Adhesion
CylinderVolume Does not work with Adhesion
RegulatorPilotValve Exponent Does not work with Adhesion
Eliminating some of the possible variables may get me back to a constant
More later
Ian
Max Power Does not work with Adhesion
MaxForce Works with Adhesion Reducing Max Power increases steam chest pressure before slipping
MaxVelocity Does not work with Adhesion
WheelRadius Does not work with Adhesion
Sanding Does not work with Adhesion
NumWheels Works with Adhesion but if left at "1" has no effect on calculation
CutOffMaxForward Does not work with Adhesion
CylinderStroke Does not work with Adhesion
CylinderDiameter Does not work with Adhesion
CylinderVolume Does not work with Adhesion
RegulatorPilotValve Exponent Does not work with Adhesion
Eliminating some of the possible variables may get me back to a constant
More later
Ian