[steamer_ctn]

ErickC, on 06 March 2020 - 05:37 AM, said:

Do the couplers rotate as well as stretch? If so, I'm quite excited about this feature.

I am not sure what you mean.

ErickC, on 06 March 2020 - 05:37 AM, said:

Is there a possibility of animated knuckles? Or perhaps different models for coupled v. uncoupled? That might be the simplest way, instead of

I think that animated knuckles would be fairly complex to code, however perhaps an open and closed coupler as you have suggested might be possible.

I would need somebody to provide a set of "open source" open and closed coupler shapes (as well as a chain version as well) to try and set it up. The connecting air hoses would also need to be added to the coupler.

darwins, on 06 March 2020 - 12:44 PM, said:

It seems that I should divide these distances by two.

The displacement distances should be for a "single" coupler only. When two wagons are coupled together, there is in effect two couplers in play, and hence the distances will be doubled.

For example, if we get 114mm of movement in compression on one coupler, we will get 228mm of total movement for the two couplers between cars.

Also there are three zones of operation in the model, typically the two displacement diagrams do not show the Z1 distances, ie when the couplers are not in contact with each other. So these distances need to be allowed for. I am also wondering if the "rest" position is correct, would the buffers both be in contact with each other at rest?

See this article, as it tends to suggest that buffers were not in close contact when at rest.

Perhaps the tension and compression slack Z1 values need to be a bit more evenly spread?

darwins, on 06 March 2020 - 12:44 PM, said:

(Also if I understand correctly then the length of each zone is measured from the start of each zone and not from zero).

Correct.

darwins, on 06 March 2020 - 12:44 PM, said:

Does this look correct?

Is the drawhook fixed "solidly" to the wagon, or does it have a spring? If it has a spring then Z2 Tension may have a bit more play.

The article referenced above tends to suggest that springs were fitted to the draw hooks in "more" recent times.

Otherwise it is probably a good start.

darwins, on 06 March 2020 - 12:44 PM, said:

Looking now at your two charts the friction wedge has separate curves for loading and unloading for both tension and compression.

The OR model currently (and probably won't) take into account the unloading parts of the curves.

[ebnertra000]

I think what Erick's referring to is a rotary coupler, found most often on coal gondolas here in the US. These can rotate, allowing a car to be tipped in a rotary dumper to unload its contents without uncoupling. Typically only one coupler on each car can rotate, though

[steamer_ctn]

ebnertra000, on 06 March 2020 - 03:58 PM, said:

I think what Erick's referring to is a rotary coupler, found most often on coal gondolas here in the US. These can rotate, allowing a car to be tipped in a rotary dumper to unload its contents without uncoupling. Typically only one coupler on each car can rotate, though

Whilst I haven't tried it, I suspect that the couplers will rotate with the vehicle. However this may mean that they don't line up correctly with the vehicles either side of them. I am not sure.

[darwins]

Quote

Also there are three zones of operation in the model, typically the two displacement diagrams do not show the Z1 distances, ie when the couplers are not in contact with each other. So these distances need to be allowed for. I am also wondering if the "rest" position is correct, would the buffers both be in contact with each other at rest?

See this article, as it tends to suggest that buffers were not in close contact when at rest.

Perhaps the tension and compression slack Z1 values need to be a bit more evenly spread?

That is a good point, the answer is complicated - I had assumed that the zero values relate the OR model - and most if not all modellers would give the size of their wagon measured over the ends of the buffers (without compression). If we want to build in a gap between buffers at rest I assume we would need to increase the length in the size statements of our models.


For the real train if there is a gap or not at rest and how big the gap was would depend on the previous events in the history of the train.


If wagons had been shunted together then there might be little or no gap between buffers - probably a small gap as wagons bounced off each other.

If the train had been shunted into a siding to wait for another train to pass there might be a large gap as the wagons would continue to roll after the loco had applied the brakes - having said that at the rear of the train wagons would be buffer to buffer against the brake van with zero gap due to the guard applying the brake.

When stopped on the level or on a hill the reverse might be true - wagons at the front bunched up against the engine - with zero gap and those at the rear stretched out by the brake van. At the bottom of a steep descent they might be bunched in groups of two or three or four depending on the number of handbrakes applied for the descent.

Starting from an uphill situation it is possible they might all be fully stretched out.

Quote

Is the drawhook fixed "solidly" to the wagon, or does it have a spring? If it has a spring then Z2 Tension may have a bit more play.

The article referenced above tends to suggest that springs were fitted to the draw hooks in "more" recent times.

Indeed there were springs on such wagons - though I do not know how effective they were. Details of them are on the RCH page.


Logically then if zone 2 is about springs - an older wagon with a draw hook fixed to the headstock and dumb (wooden) buffers should have very small zone 2 values for both compression and extension and very steep zone 3 curves for both.

I will find a model of an older wagon...

[steamer_ctn]

darwins, on 06 March 2020 - 10:25 PM, said:

That is a good point, the answer is complicated - I had assumed that the zero values relate the OR model - and most if not all modellers would give the size of their wagon measured over the ends of the buffers (without compression). If we want to build in a gap between buffers at rest I assume we would need to increase the length in the size statements of our models. ......

I think that we are going to need to define an "at rest point" so that we can correctly define the required parameters fro the coupler. This point is probably the same as the condition that OR starts in.

So for the hook and chain type coupler I would suggest that both coupler faces would be touching each other, and the chain would have some slack (ie hanging down).

As the train starts to move forward, Z1 would apply until the chain is taut, Z2 would apply until drawhook spring is fully compressed, and Z3 would apply beyond that point.

Z1 would be calculated by the amount of travel distance that the chain needs to move before it becomes taut (see that there are measurements in one of the plans that you referenced for the three link chain, so I think that we have enough to calculate this distance). It would need to be halved as it impacts both cars.

Z2 can be identified from the reference site again as there are dimensions for the different drawhook spring states.
One point with the hook and chain coupler, it will possibly need a different animation shape, or alternatively it will need to be left out. The buffers may merge into each other during compression.

darwins, on 06 March 2020 - 10:25 PM, said:

Indeed there were springs on such wagons - though I do not know how effective they were. Details of them are on the RCH page.

Thanks for the reference, a really good site.

It appears that there were two different varieties, one type had individual springs at each end, and one had a single spring for one of the drawhooks only. The OR model doesn't support only one spring, so I would suggest that the total spring movement value is proportioned, ie divided by 2, and a bit allocated to each wagon.

darwins, on 06 March 2020 - 10:25 PM, said:

Logically then if zone 2 is about springs - an older wagon with a draw hook fixed to the headstock and dumb (wooden) buffers should have very small zone 2 values for both compression and extension and very steep zone 3 curves for both.

Sounds reasonable.

See my comments about using the reference site to find the travel distances for the springs.

Given the dimensions of the spring, I wonder if we can calculate the stiffness forces?

[darwins]

Quote

It appears that there were two different varieties, one type had individual springs at each end, and one had a single spring for one of the drawhooks only. The OR model doesn't support only one spring, so I would suggest that the total spring movement value is proportioned, ie divided by 2, and a bit allocated to each wagon.

The single springs are shared by both drawhooks - if you check the diagrams you will see a box around the spring - the aim being to link all the draw hooks along the train with a spring in between each coupling. This was the preferred arrangement.

[steamer_ctn]

darwins, on 07 March 2020 - 12:53 AM, said:

The single springs are shared by both drawhooks - if you check the diagrams you will see a box around the spring - the aim being to link all the draw hooks along the train with a spring in between each coupling. This was the preferred arrangement.

Ok, fair enough.

Then this distance amount will need to be proportioned between two cars.


EDIT: If you look at the Dec 1944 diagram for a metal underframe wagon, it shows a drawhook at either end of the car with an individual spring, rather then a central single spring. At the end of the day, it will just become a point as to how it is handled in defining the spring movement distance.

[ErickC]

ebnertra000, on 06 March 2020 - 03:58 PM, said:

I think what Erick's referring to is a rotary coupler, found most often on coal gondolas here in the US. These can rotate, allowing a car to be tipped in a rotary dumper to unload its contents without uncoupling. Typically only one coupler on each car can rotate, though

I am not. I am referring to the fact that couplers need to be able to rotate from side to side as the train goes around curves. Otherwise, things get really awkward really quickly, which is why my couplers have always been part of the truck assembly.

steamer_ctn, on 06 March 2020 - 02:41 PM, said:

I am not sure what you mean.

See above.

Quote

I think that animated knuckles would be fairly complex to code, however perhaps an open and closed coupler as you have suggested might be possible.

I would need somebody to provide a set of "open source" open and closed coupler shapes (as well as a chain version as well) to try and set it up. The connecting air hoses would also need to be added to the coupler.

Open/closed models would be good enough, I think. I don't think that even the TS models really have animated knuckles. I can provide open and closed versions of my standard AAR Type-E couplers sometime this week. I don't think it's too necessary to have the air hoses part of the coupler, though. I've always had the air line positioned on a chain, as is often done in real life. I would imagine that we wouldn't want the side effect of a moving air line, since it'd be fixed to the body at one end.

Unless you're proposing separate models for the air hoses, too, which have coupled and uncoupled positions, and which otherwise don't move. That would be excellent, and I can absolutely provide that material as well, if someone who builds UK models can provide chain couplers. I could actually send a full car or locomotive that has had the relevant parts split off, ready for positioning and testing. I think the PS1 would be a good choice.

Anyway, I've got to try this whole thing out sometime today.

[scottb613]

Hi Folks,

I'm just starting to look at this...

Just curious - would it be a great deal of work to give us the ability to associate a sound [wav] file with the event of a coupler changing state - such as a clank when it goes from neutral to expansion or contraction ? Trainz had this and I always thought it was pretty neat to hear the cascading effect upon start up... Some randomization of the sounds would be gravy...

We should have a standardized coupler template available for all when building models...

Thanks...

Regards,
Scott

[steamer_ctn]

ErickC, on 07 March 2020 - 04:30 AM, said:

I am referring to the fact that couplers need to be able to rotate from side to side as the train goes around curves.

In theory the shapes should "follow" the car shape, so whether this means that they rotate in a reasonable fashion I haven't really checked at this stage. Without more coding effort this is probably as good as it gets. You will need to give it a go to see how effective the animation is.
My main intention with the animation was to overcome any gaps between couplers which appear with larger slack values as the cars move further apart.

ErickC, on 07 March 2020 - 04:30 AM, said:

Open/closed models would be good enough, I think. I don't think that even the TS models really have animated knuckles. I can provide open and closed versions of my standard AAR Type-E couplers sometime this week.

That would be good. Thanks.

ErickC, on 07 March 2020 - 04:30 AM, said:

I don't think it's too necessary to have the air hoses part of the coupler, though. I've always had the air line positioned on a chain, as is often done in real life. I would imagine that we wouldn't want the side effect of a moving air line, since it'd be fixed to the body at one end.

Again it was mainly for appearances as it may look funny with "disconnected" air hoses.

Perhaps we could give it a go. Would you be able to provide an additional shape with the air hose in the "connected" position (coupler and air hose in one shape file), and we will see what it looks like?

ErickC, on 07 March 2020 - 04:30 AM, said:

I could actually send a full car or locomotive that has had the relevant parts split off, ready for positioning and testing. I think the PS1 would be a good choice.

Thanks for the offer. I will be in contact re this.

scottb613, on 07 March 2020 - 08:51 AM, said:

Just curious - would it be a great deal of work to give us the ability to associate a sound [wav] file with the event of a coupler changing state - such as a clank when it goes from neutral to expansion or contraction ? Trainz had this and I always thought it was pretty neat to hear the cascading effect upon start up... Some randomization of the sounds would be gravy...

It maybe possible, but I would need a sound file for it. Are you able to help with this?

scottb613, on 07 March 2020 - 08:51 AM, said:

We should have a standardized coupler template available for all when building models...

If you are referring to the shape files, these actually could be "embedded" in OR as content. People could call their own or use the default shapes. I am hoping that the shapes that Eric provides will become the default shapes.