Reliable remote coupling and uncoupling

[Robin2]

I am building a small OO Gauge layout which I hope will be run by a PC program and will include programmed shunting. For that to work a very high level of coupling and un-coupling reliability is needed.

 

Delayed action un-coupling and symmetrical coupling (i.e. both ends of a wagon must have the same coupling) is essential for what I have in mind.

 

I would also like a system that allows a wagon to be lifted from a rake without derailing everything else - but I might sacrifice that in preference to high reliability.

 

I have not been able to find any Thread on the Forum where people have commented on the reliability of any of the coupling systems - for example, out of 1000 couplings and uncouplings how many need the attention of the hand of god?

 

DG Couplings are attractive because they couple without disturbing the stationary wagon. However if there are loops on both couplings there is a great risk of the loops colliding and failing to couple. I have been using N-Gauge couplings simply because I have some in stock.

 

I got some Bachman EZMate couplings and they tended to push the stationary wagon rather than couple to it.

 

I have been experimenting with an idea inspired partly by the Alex Jackson coupling and partly by the Peco simplex coupling and I think I can get it to work. I have never had the patience to get the regular AJ couplings to work.

 

However if someone has experience of a very reliable existing system I would like to have the option of not reinventing the wheel.

 

Thanks

 

...R

[Guest 40-something]

You really can't go wrong with Kadee's. Incredibly reliable and robust

[TEAMYAKIMA]

You really can't go wrong with Kadee's. Incredibly reliable and robust

 

I have used Kadees for years ........... I used to sell/promote them for a living.

 

Compared to the Bachmann EZMate they are far more robust and are less prone to push the wagon without coupling because

 

1) the knuckle spring is lighter so the knuckles open with less force.

 

2) Also as the coupler face is metal they slide across each other with less resistance than the plastic EZmate couplers

 

However I would strongly recommend that you do NOT mix the traditional (US) style couplers with the European style (NEM box) style ... for extra reliability use one style or the other. Of course the two styles match and will couple and uncouple but under test conditions it was clear that mixing the two designs does in extreme cases cause issues. So if you are looking for 100% reliability ( rather than 99%) I suggest you choose US style Kadees or NEM box Kadees and stick to that design.

[Robin2]

From what I have read the Kadees need a complex manoeuvre to get the delayed action to work -

1. stop over the magnet,
2. push back 2mm to provide slack to allow the magnets to work
3. pull forward 1 mm to allow the coupling to open wider

Is that correct?

 

I don't think my computer program could have that level of fine control.

 

...R

[TEAMYAKIMA]

From what I have read the Kadees need a complex manoeuvre to get the delayed action to work -

1. stop over the magnet,
2. push back 2mm to provide slack to allow the magnets to work
3. pull forward 1 mm to allow the coupling to open wider

Is that correct?

 

Yes, that is correct

[jcm@gwr]

I think you'll find that most auto-couplers will need

a bit of 'setting back', as they are designed to not

uncouple accidently.

The tension needs to be off to uncouple, the closest

to what you want might be Lincs, as need very little

slack to uncouple, but they do need care in setting up.

They also allow you to lift out wagons without problems.

[Robin2]

I believe I can create the slack without difficulty but I don't think I can guarantee the 3rd element of the Kadee "dance" because if the loco moves a little too far it could allow the coupling to return to the normal position.

 

The Lincs coupling looks interesting, and it should not be difficult to make a DIY version. But from what I can discover it does not allow delayed action.

 

Has anyone got a link to the dimensions for the Lincs coupling - I have not been able to find one.

 

...R

[34theletterbetweenB&D]

sd

 

From what I have read the Kadees need a complex manoeuvre to get the delayed action to work -

1. stop over the magnet,
2. push back 2mm to provide slack to allow the magnets to work
3. pull forward 1 mm to allow the coupling to open wider

Is that correct?

 

I don't think my computer program could have that level of fine control.

 Does it help that some decoders offer a 'Kadee shuffle' as a facility? That only works though for loco directly coupled to vehicle, Will not reliably perform an uncouple between a pair of vehicles.

 

Really though, I think you are up against the fact that existing autocouplers were all designed when human operation of the loco was pretty much the only method of control. So there was always a human monitor to assess whether the couple or uncouple function had succeeded.

 

A system that appeals to me is magnetic latch with electromagnet cancellation of field. (This was a latching system designed for camera shutter release, far less error than the mechanical latches that were previously used.) In a model railway context it would mean all vehicles having power (battery or track) and acting on commands (broadcast or via track). Not unthinkable for a small layout with limited stock, and offers complete freedom to uncouple anything from anything, at any location; reduces placement precision requirements as no fixed track magnets have to be located successfully, let alone a manoeuvre performed, for an uncouple to occur.

[Robin2]

I plan to use battery powered radio control rather than DCC.

 

I have now made a pair of crude Lincs couplings and I can see how they work. Unfortunately they definitely do not have a delayed action capability - so that rules them out.

 

Thanks for all the comments, and more will be very welcome.

 

...R

[Grovenor]

I believe I can create the slack without difficulty but I don't think I can guarantee the 3rd element of the Kadee "dance" because if the loco moves a little too far it could allow the coupling to return to the normal position.

There seems to be a bit of confusion here, if you are using the delayed uncouple then your next move is to push the uncoupled vehicle, so you don't need to stop the loco just go ahead with the push. The critical issue is that the vehicles must not seperate while pushing so the contact must be slow enough to avoid bounce and the wagon must not be able to run away from the loco, as it might if the track has a slight downgrade. For reliable operation the wagons need a bit of friction in the bearings.

Regards

[Robin2]

I got the following quote from this Kadee web page

 

 

1. Stop with the couplers over an uncoupler and back up slightly with the couplers still over the uncoupler, allowing slack to occur between couplers.
2. Pull forward bringing coupler off the uncoupler. Couplers will snap to the delayed position.

 The first line corresponds with 1 and 2 in my Reply #4 and the second line correspons with 3 in that Reply.

 

If I understand you correctly, line 2 in the Kadee instruction is not necessary ?

 

...R

[TEAMYAKIMA]

I got the following quote from this Kadee web page

 

 The first line corresponds with 1 and 2 in my Reply #4 and the second line correspons with 3 in that Reply.

 

If I understand you correctly, line 2 in the Kadee instruction is not necessary ?

 

...R

 

Confusion is bucket loads!  The little demo on the Kadee webpage tells you all you need to know.

 

Line 2 is necessary although in fact you don't need to move the loco off the magnet just move it forward sufficiently for the couplers to snap into delay then move the loco back onto the train/wagon

[Robin2]

just move it forward sufficiently for the couplers to snap into delay

That's the part that I don't think I can get my computer control to do.

 

...R

[Grovenor]

Part of the problem here is that, in the original post, forward and back was not defined, neither was it specified if the first arrival on the uncoupler is pushing or pulling.

The Kadee reference assumes arrival pulling, so first action is to ease back so that the tension on the couplers is lost and the magnet can do its thing, instruction 2 is then to pull forward slightly so the couplers move into delayed position. If you move off the uncoupler the coupling that is now off the uncoupler will go back to centre, the one left behind will remain in the delayed position, when you move back over the uncoupler the moving coupling should be swung to the delay position before making contact with the standing one. Better not to move off the uncoupler in the first place, IMHO.

If you arrive pushing then action 2 becomes the first action.

Best just to get a pair of couplers and a magnet and play with them.

Regards

 

 

That's the part that I don't think I can get my computer control to do.

Why not, that's what the "kadee shuffle" in decoders does.

[Robin2]

I had been assuming "pulling" as in the Kadee reference.

 

 

Why not, that's what the "kadee shuffle" in decoders does.

As I said in Reply #9 I won't be using DCC.

 

...R

[phixer64]

Have you tried the Brian Kirby option

 

http://www.rmweb.co.uk/forum/viewtopic.php?f=8&t=35605

 

Regards Jeff

[Robin2]

No. That does not have a delayed action AFAIK and it also does nothing for the awful tangle if you try to remove a wagon from a rake.

 

I have spent a bit more time developing my own idea and I think it is now satisfactory. The 11 thou guitar string was too strong to allow coupling without moving the stationary wagon but 9 thou guitar string seems to work fine. And a small (2mm wide) piece of steel from a marmalade lid seems to make a suitable dropper that is not grabbed too strongly by the magnet.

 

My plan is to use a servo to lift the permanent magnet into position when uncoupling is required.

 

...R