[copperpen]

steamer_ctn, on 07 July 2019 - 11:59 PM, said:

braking the train (using the train brake) may not cause compression either as the cars may not change speed relative to each other, and thus all slow down at the same rate.

This may well be the case if running a short train as the propogation time along the brakepipe will be minimal. A long train will behave differently, propogation along the brakepipe will take longer so the cars at the front of the train will be slowing down before those at the rear start to brake, thus there will be varying degrees of compression along the length of the train.


Modern trains will behave differently where fitted with an EOT device that can activate braking from the rear of the train as well. Not currently modelled in OR.

[steamer_ctn]

Advanced coupler functionality has now been added to the Unstable version of OR (from 05-03-2020).

The functionality allows slack movement of couplers, and also animation of couplers (big thanks to Chris Jakeman for his assistance in working out the world co-ordinate positions for placement of the coupler shapes).

A demonstration activity has been created and is available from here.

This activity and associated tasks demonstrates the following:

i) Operation of slack movement on a train (The test train is over 200 cars long, and each set of couplers can expand up to approx 8 inches). The couplers will expand to a slack distance consistent with the force on the coupler and the stiffness of the coupler
ii) The interaction between slack and starting resistance can be seen. In other words a train with "sufficient" slack movement requires less starting effort then one without.
iii) Poor driving when starting a long heavy train may cause coupler failures, so some added realism is now in play.

For more information on setting up Advanced Couplers see here.

Any issues identified should be referenced back to the test models as this will assist in any code fault finding.

Over time it would be good to establish a small set of standard coupler configurations as this would make settings consistent within OR.

[darwins]

Sounds good. Let me know when someone has worked out the correct numbers for three link couplings.

First of course as the links go from drooping to stretched - the force needed to stretch them would be very minimal. No force at all passed on to the next wagon.

Second as they reach the point of being stretched there is the 'snatch' when force begins to be transmitted through the three links to the next vehicle.

At this point force is also on the leaf springs (if any) at either end of the draw hooks on each wagon - so there may be some coupler stretch depending on train load and springs.

With no train brakes then when the locomotive brakes the wagons are retarded by colliding with the vehicle in front.

At this point there is no compression force as the coupling links can freely change from being stretched to being in various stages of not stretched at all.

As the chain coupling reaches the not stretched at all state then the side buffers collide and start to bear the retarding force.

The side buffer may have springs that can absorb the impact and compress - or on very old wagons could be just blocks of wood with no more ability to absorb compression forces than a block of wood!

[steamer_ctn]

darwins, on 04 March 2020 - 10:14 PM, said:

Sounds good. Let me know when someone has worked out the correct numbers for three link couplings. .............................................................

The description you have described I believe can be modeled with the Advanced coupler features added to OR.

If you look at the introduction section for couplers on my website you will see a force/displacement diagram for two coupler types. One is a drawhook/buffer type of coupler so perhaps the figures suggested on this diagram can be used to construct a coupler as you suggest.

I would be interested in looking at this type of coupler if you are interested. If so please contact me.


Thanks

[darwins]

I have been able to get dimensions from RCH drawings.
The amount of slack with totally free movement in either direction (where no force is passed on to the next wagon) is 4.125 inches (105 mm).
With the chain extended and tension force being passed on to the next wagon the distance between headstocks is 40.125 inches.
With buffers touching and compression force being passed on to the next wagon the distance between headstocks is 36 inches.
I am going to estimate from the drawings that buffers will compress up to about 6 inches maximum (150 mm)
The draw hook and buffer diagram you have on your website seems to be for a screw coupling with buffers (UIC standard).
It seems reasonable to assume that the forces and distances for a three link coupling with buffers will be the same as for the screw coupling, but with the extra 105 mm of free movement on the tension side.
** That adds up to about 26 feet (8m) of slack in an 80 wagon mineral train (or about 2% of train length) **

https://i.imgur.com/Zb6eCJa.jpg

[steamer_ctn]

darwins, on 06 March 2020 - 01:38 AM, said:

I have been able to get dimensions from RCH drawings. ....................................................................

Ok, seems some reasonable thoughts to start with. Though this document (towards the back) suggests a lower compression distance.

Can you set up a consist using appropriate CTN stock with your thoughts on proposed coupler settings for this type of coupler?

We will also need to set a realistic set of "Break" values, and I suspect with the amount of slack in the coupling system this might cause challenges for drivers who start too quickly (depending on train length, etc). This document has some values that might be helpful.

[ErickC]

Do the couplers rotate as well as stretch? If so, I'm quite excited about this feature. Is there a possibility of animated knuckles? Or perhaps different models for coupled v. uncoupled? That might be the simplest way, instead of

FrontCouplerAnim ( name x y z )

You could have

FrontCouplerAnimOpen ( name x y z )

And

FrontCouplerAnimClosed ( name x y z )

Cool stuff either way! I am presently studying the information on your page, and will try out this new feature on my GP9 prototypes over the weekend, I think.

[scottb613]

Hi Folks,

Yeah - the physics changes are what interest me - I'll try out the features over the weekend as well, I think...
:p

Regards,
Scott

[darwins]

Quote

Can you set up a consist using appropriate CTN stock with your thoughts on proposed coupler settings for this type of coupler?

Will send one during next week.

Quote

We will also need to set a realistic set of "Break" values, and I suspect with the amount of slack in the coupling system this might cause challenges for drivers who start too quickly (depending on train length, etc).

I am not sure if the chains were any weaker than the standard UIC screw coupling. I doubt that they were any stronger, so 1200 kN should be the max break value.

The slack in those trains was challenging for drivers. You had to start very gently.... very gently picking up one wagon at a time or risk a broken coupling... stop also very gently or risk damaged goods, injury to the guard or even derailed wagons.

Quote

This document has some values that might be helpful.

Thanks. I should have checked that one. It gives 228 mm for maximum compression of buffers (114 either side of the coupling), slightly more than my estimate of 150 to 180 mm - although the diagram on CTN seems to give 150 mm for compression...

[darwins]

I am thinking that I can adapt your diagram to show this coupling (between two wagons) like this:https://imgur.com/QsvGH7F.jpgNow reading the guidance on CTN it says:

Quote

OR typically models the coupler based upon the coupler defined for the rear of the car, and it is assumed that the coupler on the front of the following car is the same type, and hence, coupler slack between the two cars will be double (2x) the value specified in the following coupling definition statements.

It seems that I should divide these distances by two. (Also if I understand correctly then the length of each zone is measured from the start of each zone and not from zero).
So I am now assuming that the entry for this should be something like:

Comment ( Open Rails Advanced Coupler )
Coupling (
Spring (
ORTSTensionStiffness ( 200kN 1200kN )
ORTSTensionR0 ( 0mm 53mm )
ORTSTensionSlack ( 3mm 10mm )

ORTSCompressionStiffness ( 1200kN 2400kN )
ORTSCompressionR0 ( 0mm 2mm )
ORTSCompressionlack ( 60mm 15mm )

ORTSBreak ( 700kN 1200kN )
)
CouplingHasRigidConnection ( 0 )
)

Does this look correct?


Looking now at your two charts the friction wedge has separate curves for loading and unloading for both tension and compression. The chain and buffers has separate loading and unloading curves for compression, but only one curve for tension. Should there be some way to make this distinction in the model?