[Laci1959]

 steamer_ctn, on 31 May 2021 - 08:31 PM, said:

Other countries other then America use "knuckle" type couplers. These modifications should suit any knuckle type coupler which has a single moving part.

The knuckle coupler was done as it is the easiest to model, as it only has a single moving part. The hook, chain and buffer coupler described in the posts above is much more complicated as the buffers would need to move independently of the chain. The chain also adds a lot of potential complexity as each individual link would need to move to really do it justice. If you can think of a simple way to model it, then perhaps it could be added.


I am not sure what you mean by this, perhaps you could provide some more detail.

Hello
The video clearly shows the movement of the corner stops and the coupler. On this type, it is sprung, but there are also vehicles with a rigid coupling.
Regards Laci 1959

My link

[steamer_ctn]

 Laci1959, on 03 June 2021 - 07:17 AM, said:

The video clearly shows the movement of the corner stops and the coupler. On this type, it is sprung, but there are also vehicles with a rigid coupling.

Thanks for that.

I can't see both cars, but to clarify, it appears that there is a rigid chain between two sprung hooks. Is this correct?

If this is the case the two sprung hooks could be easily modeled to replace the knuckle coupler, however the issue then becomes the "floating chain" between them, as this would require a special shape and appropriate code to render it in the correct position. Not a simple addition.

You could possibly cut the chain in half, and use that. If you are a modeller then perhaps you could build two couplers to experiment with.

Let me know if you are interested and I will describe the co-ordinate system that we might have to use to get them to work.

[Rj Zondervan]

Not quite, the 'chain', is sprung between a pivot just behind the hook on the carriage, and the hook on the locomotive.

A screw/spring coupler typically exists from a hook, a shackle that pivots just behind the hook, and connects to a thread, on which a second shackle mounts that can be connected to the hook of another vehicle. When the shackle is placed on the hook, the thread is used to tension the coupler until the buffers of both vehicles are slightly pressed (if I recall correctly), creating a (semi-)rigid connection between the vehicles.
The spring thus should never loose tension during operation, ensured by the buffers keeping as most distance between the vehicles as allowed by the coupler.

The use of this type of coupler is (at least here in the Netherlands) bound to a number of rules, of which most peculiar is that when an unmotorised vehicle, a carriage or a wagon, is coupled to a motorised vehicle (locomotive or multiple unit, in which the tender of a steam engine counts as part of the engine), always the spring of the unmotorised vehicle is used, except when it is damaged.

I think this poses four challenges to the programming to model correctly:

• Buffers: As mentioned, the buffers play an active part in tensioning the coupler. However, this is quite predictable, and for normal gauge the position is standardised on 875mm from the center of the vehicle (1750 mm apart), and between 1040 and 1060 mm from the ground^[2]. This however varies when looking at different gauges.^[1]
  
• Back-up-couplers: In the past, some countries enforced the presence of a back-up coupler underneath the main coupler that was not tensioned, so the train would not break when the main coupler broke. This practise was abandoned after the continued brake line became compulsory on all railway vehicles in the 1930s^[3], but the presence of back-up couplers persisted until the end of the steam era in some countries, as some companies didn't bother to remove them.^[1]
  
• Directionality: As mentioned before, there are some rules about which spring, and which hook is used. In many cases, but not by regulation, a consist of carriages or waggons is coupled in one direction, so if the spring of the carriage is used on one side, the hook is used on the other, except for the coupler to the motorised vehicle: then always the spring of the unmotorised vehicle is used, regardless of which way the other side of the vehicle is coupled.
  
• Brake hoses and other hoses: Although the couplers themselves are tensioned rigidly, the brake hoses, auxiliary air hoses and for example steam heating hoses were not. This might be the biggest challenge of all, as it would run down to flexible hoses, and that will be quite the challenge to implement in models.

All in all it will be quite the challenge to get this type of coupler working realistically in the simulator, but I think most of us would be already quite pleased if a way is found around point 3, point 1 would be very nice and 4 spectacular. Point 2 would only be of interest to the rather small number of people in continental Europe that have studied steam era thoroughly, unless this was also in use in the UK, as the community of steam connaisseurs over there is a lot larger than on this side of the pond.

If it is of interest, I have got some drawings at my disposal in which the working of a spring/screw coupler is shown more detailed, as well as a drawing of a steam heating hose, but I cannot share them publically unfortunately.

^[1]: I think that flexibility in the coupler itself can be solved mostly by having the shackles and thread as separate sub-objects to the coupler with correct pivots, this however leaves the question of tensioning, especially as not all buffers have the same length.
At rest most buffers were 610 mm in the Netherlands, but 650 was quite standard in Germany, and I have seen buffers of 597 mm on older drawings of Dutch steam engines.
This can be solved too, but that would require advanced logic in the shackles sliding up and down the thread.
^[2]: The bandwith is a design bandwith, in practise the difference can be larger due to wear and tear to the wheels, and the load of the vehicle. My friends in the railway world tell me that regulations state that at all times two-thirds of the buffer area has to be in touch.
^[3]: The last railway vehicles in the Netherlands were obliged to have a continued brake line in 1934, but I'm not sure if this was a UIC regulation, or if the Netherlands were downright late to the party, as quite often in railway history

[hroch]

Hello

Maybe it might help.



Sorry for the czech language


Buffer



Draw-Gear


more here
https://axtoneglobal...er-locomotives/

[steamer_ctn]

Hi @Rj Zondervan & @hroch,

Thanks for the information on the different types of couplers.

Whilst I agree with your thoughts and suggestions on how to possibly implement them, I am not sure how to write the required code easily in OR to achieve the functionality desired.

For the current version of coupler animation (with the knuckle coupler) OR has been coded to manage two coupler pieces (one on each car, ie front and rear). So whilst this was a bit complex, it was somewhat manageable.

Whilst the types of couplers that you have suggested might require the management of up to eight different coupler pieces between the two wagons (two buffers, two tensioning hook, air hoses and the chain pieces, etc ), so the coding I think will be a lot more complicated.

However I will continue to reflect upon it, and see if I can come up with an easy way to do it (although I am unable to provide any timeline).

[Laci1959]

Hello

Quote

I can't see both cars, but to clarify, it appears that there is a rigid chain between two sprung hooks. Is this correct?

Yes, although it used to be left loose on freight trains for easier starting. But let’s take the case that’s easier to implement.
I think it would be enough to use the front couplings on every vehicle. In reality, only one is used.
Thanks in advance.
Sincerely, Laci 1959

[Laci1959]

 Weter, on 08 June 2021 - 07:41 AM, said:

Hello, Laci1959.
I have a question about MAV and sent you PM, but it looks like you turned notifications off.
Would You see it, please.

Sorry, no.

[Aldarion]

Hi Peter.

Wanted to ask if you are considering the couplings between a loco and a passenger coach.
Since I discoverd this I am thinking on remodeling one set of passenger coaches I have with not less than 51 shapes in wish I will need ae the very least three pairs of couplers for all of them and I wanted to include the tension hooks, the air hoses and the gangawys between the coaches.
I'm thinking on how to have a differente behavior for the coupling between the first coach and the locomotive... my first thought would be having a diferente wag file for the first coach in the consists, but i think that will put me in trouble if I flip the coach or if the loco goes to the other end in the middle of the journey (i use a lot of timetable mode)... so perhaps tha tproblem is only solved by coding and resorting to the old "CouplingUniqueType" were we could define unique type options that are not compatible or compatible with certain coupler shapes...

[steamer_ctn]

 Aldarion, on 26 June 2021 - 01:49 PM, said:

Wanted to ask if you are considering the couplings between a loco and a passenger coach.
Since I discoverd this I am thinking on remodeling one set of passenger coaches I have with not less than 51 shapes in wish I will need ae the very least three pairs of couplers for all of them and I wanted to include the tension hooks, the air hoses and the gangawys between the coaches.

At the moment I have not contemplated any other couplers that use more then one animation object, because I still have an issue with flipped cars, hence I need some assistance from another developer who is more competent with Quaternion rotation before I can go much further.

Currently, if you can model the coupler you desire as a single object then any type of coupler could be built using the existing code.

[Aldarion]

Yes. I will do that. It will take some time for me to implement it but I will certainly share the experience in this thread if it's alright!

Cheers.