Type E Coupler for LEGO Trains

[coaster]

While work continues on additional track, I've had a number of people ask about developing a new coupler for trains.  US trucks don't typically have bumpers, and the old style coupler and magnet are becoming more rare and expensive on Bricklink.  Furthermore, the pulling power of the magnets is limited, so long freight trains can be very problematic.

As such, I've started poking around a design for a traditional Type E knuckle coupler that could retrofit onto existing car in place of the magnet.  Here's what I've been thinking:

View from above, coupled and uncoupled:

The knuckles are spring loaded, so uncoupled, would pop to that approx. 45° open position.  On closing, a pawl locks into a catch on the back of the knuckle.  This can be released by pressing up on the leg of the pawl, hidden beneath the knuckle.  The only real weakness is I have no room to integrate a magnet into the coupler for automatic alignment, so you probably wouldn't be able to just ram into some freight cars and shunt them around. 

Before I dive headlong into this though, I'd like to get some feedback and gauge interest in this.  If it's a DOA project, I'm not going to waste any more time on it.

[Jeffinslaw]

I'm throwing money at the screen but nothing is happening! I have also been one of those individuals buying large quantities of the train couplers and magnets...

 

-Jeffinslaw

[Hod Carrier]

A lot of real world couplers have structures that help to align the coupler heads as they come together. For example, Tightlock couplers (which are a development of the Buckeye knuckle coupler) have a “nose and pocket” arrangement that assist with alignment. Whether such measures can be replicated at LEGO scales and on LEGO curves is another question altogether.

One thing I would advocate though is that any coupler design should be capable of self-centring so that they can at least operate correctly when used on straight track.

Rather than considering the Type E design DOA, have you considered other alternative designs? The Scharfenburg coupler could be an option with a sprung catch, as could the conceptually similar BSI coupler.

**EDIT**

On reflection...

I’ll be honest and say that I think the existing LEGO magnet couplers are probably the best solution we have at present, although that’s not to say that they couldn’t be improved upon. A free-swinging coupler that can self-align, even on tight curves and on a variety of different stock types, and make a reasonably strong coupling, all without the need for springs, latches or other mechanical aids, seems to be a good starting point. I wonder if all that’s really needed is a development along the existing theme, perhaps using stronger neodymium magnets and perhaps adding a weak sprung catch that can be easily made and broken, would be a good solution. Such a development might also have the advantage of being backwards compatible with the existing coupling standard.

As a railwayman with a dozen years experience I can see that there are real-world solutions that have applications for LEGO train construction, but equally I can appreciate that some real-world technologies could not easily be applied without introducing further complications. I can certainly see a use for a mechanical coupler such as the one you propose, but I don’t see it as versatile or user-friendly enough to be a universal replacement for the LEGO magnetic couplers.

At the end of the day it all comes down to how well they work. For long semi-permanently coupled trains I can definitely see a use, but for frequent coupling and uncoupling it could be more problematic. However, you will only know for certain by making some and trying them out in a variety of scenarios, and I’m interested to know too.

Edited November 13, 2017 by Hod Carrier

[zephyr1934]

Thought #1, ___ that's cool

Thought #2, I have how many existing cars?

Thought #3, those could be a hassle to work with both in terms of having to manually center the couplers and how do you release the couplers?

Thought #4, as a fan of stupid long trains, I have contemplated alternatives to the magnets, e.g., in the style of old lionel couplers- look like knuckles, but they are one piece and don't open. Those could be nice for assembling long trains easier and (hopefully) cheaper than magnets. You would have to manually lift one car to make the couple, but it would be a STRONG couple. Points of concern, see thought #2 and how well would only having one degree of rotation do on a show layout with bumpy tables? I could see one knuckle slip below its mate and come undone, or if they somehow locked in place so they don't slide vertically that it could lift the flanges on one of the cars out of the rails.

I am fortunate enough to still have a bunch of the old style magnets (and am not above cutting the the buffers off of the new style magnets), so while I think the design looks really cool, I doubt I would be an early adopter.

Thought #5, On the other hand, I could see those as being a piece for a display train that is also capable of running. I know that if I had a show piece on my shelf that I would much prefer those couplers over the lego magnets.

 

[ALCO]

I think this is a really interesting idea, and it will be fun to see where this goes. I have seen these couplers with a (for lack of better words and in complete ignorance) "thing" on the end of the pawl on top and bottom that interfaces with the pawl from the other car in such a way to prevent them from becoming disconnected on uneven track. Maybe a bit more design research will uncover what it is called. As you said, this is just the beginning of this idea.

Another thought is that the magnet system is so well established that and that is what everyone has and uses. While I agree a more realistic (to North American railroads) and stronger alternative is a great idea, the trouble is compatibility with the magnet system. IF it were possible to make it "backwards" compatible then I would think it should be more acceptable.

[Rotundus]

From an operational view a self centering device would be better, but I build only for display and these look better than anything I've seen so far.  I can see myself using these on my projects.

[Hod Carrier]

How large or powerful would the magnets need to be simply to align the coupler heads? I mean, all they’d need to do is pull the coupler faces towards each other just before they make physical contact, and provided that the heads swing freely enough I can’t imagine that the force required to do this would be very large.

I was just idly musing while looking at the proposed design whether you could add two small magnets, one each side of the knuckle. With the poles correctly arranged they *should* attract each other as the coupler heads approach each other aligning them so that they are square to each other and removing the need for them to self-centre.

I know this is likely to detract slightly from the prototypical look of the design, but it might help to make them more functional. Thoughts...?

Edited November 13, 2017 by Hod Carrier
Syntax (it’s too early in the morning for me)

[coaster]

I'll see what I can do about adding a small magnet into the base.  A small ceramic bar magnet inserted transversely where, say, North is to knuckle and South to the receiving end would be sufficient to auto-align the parts.  Just need to figure out how the heck to place it.  I'll fiddle with it a bit, and then maybe print a few to test it out.

[dr_spock]

Instead of a molded technic pin to connect to train body, would it be better to have hole to receive a frictionless technic pin or axle?  I think it might give more mounting options.

 

[coaster]

10 hours ago, dr_spock said:

Instead of a molded technic pin to connect to train body, would it be better to have hole to receive a frictionless technic pin or axle?  I think it might give more mounting options.

 

That is something I have been considering.  Certainly would simplify the tooling.  @Cale, I know you're lurking here, would that solve your mounting issue, using a 1/2 pin?

[knotian]

The lack of a remotely controlled coupler in Lego Trains, or at least the availability of easily installed decoupling points is a major limitation for the use of Lego in an industrial setting. The capability of spotting cars, making consists, and dropping off cars would be a fantastic increase in Lego utilization. The motor operated 'push' decoupler is the only viable option at this time and it make as engine too long to be used as a switcher, or yard engine.

[coaster]

@knotian This could help in that regard.  Since the pawl release is on the bottom, something as simple as a sliding cheese slope could pop the release and open up the coupler. 

[McWaffel]

When I was running the new trains in the office today I noticed how the new couplers have studs on the top and bottom. So could you not just clip a 1x2 plate there and you'd be done? No more issues with long trains decoupling.

[CaL]

I was just wondering why not just simply revive the old magnets as 3d-printed parts?

[Daedalus304]

I really love this idea and would be totally down for these if the price isn't prohibitive. I've looked into trying out Kadee O-Scale couplers which are about $7-$8 I think, which is right about the most I could see paying for these. Certainly would not mind if they turned out cheaper than that, though.

Two thoughts I have regarding the design -

1. Instead of using magnets to line the couplers up, is there any way you could fit some sort of centering spring instead? That seems like it would be a simpler and possibly more reliable option. The auto-coupling wouldn't line up the best on sharp turns I'd wager, but that doesn't seem like too big of an issue and I wouldn't care at all.

2. Thinking a little bit ahead, it might be cool to put some sort of very small hole through the pawl release. With a hole there, someone could theoretically loop a bit of string through it and have it connected to a tiny motor inside the engine for a hands-off uncoupling without needing some sort of mechanism in the track bed.

Or, perhaps, offer an option for a slightly more expensive pawl release that's magnetized. In that case, we could hide magnets in the yard track bed and it could release when the coupler is stopped over the magnet, like the way the Kadee option works.

[Cale]

On 11/14/2017 at 9:18 AM, coaster said:

That is something I have been considering.  Certainly would simplify the tooling.  @Cale, I know you're lurking here, would that solve your mounting issue, using a 1/2 pin?

I think that might be a solution with pursuing.

Cale

Edited November 16, 2017 by Cale

[coaster]

4 hours ago, McWaffel said:

When I was running the new trains in the office today I noticed how the new couplers have studs on the top and bottom. So could you not just clip a 1x2 plate there and you'd be done? No more issues with long trains decoupling.

It makes it much more difficult to uncouple each car, but yes, for long, continuous displays, that's what's typically done.

18 minutes ago, CaL said:

I was just wondering why not just simply revive the old magnets as 3d-printed parts?

Couple reasons:  First, you can't 3D print magnets, and even if you could the cost to 3D print copies of the existing part wouldn't gain you anything.  Second though, there are still the issues of them decoupling under longer/heavier loads. 

6 minutes ago, Daedalus304 said:

I really love this idea and would be totally down for these if the price isn't prohibitive. I've looked into trying out Kadee O-Scale couplers which are about $7-$8 I think, which is right about the most I could see paying for these. Certainly would not mind if they turned out cheaper than that, though.

Two thoughts I have regarding the design -

1. Instead of using magnets to line the couplers up, is there any way you could fit some sort of centering spring instead? That seems like it would be a simpler and possibly more reliable option. The auto-coupling wouldn't line up the best on sharp turns I'd wager, but that doesn't seem like too big of an issue and I wouldn't care at all.

2. Thinking a little bit ahead, it might be cool to put some sort of very small hole through the pawl release. With a hole there, someone could theoretically loop a bit of string through it and have it connected to a tiny motor inside the engine for a hands-off uncoupling without needing some sort of mechanism in the track bed.

Or, perhaps, offer an option for a slightly more expensive pawl release that's magnetized. In that case, we could hide magnets in the yard track bed and it could release when the coupler is stopped over the magnet, like the way the Kadee option works.

Target price would be around $4-5.  I've been looking at a centering spring, but it becomes dependent on you setting the spring properly when you snap it into the plate with the hole.  Could be problematic.  Plus, I couldn't work out an elegant way to install said spring.

One of the problems with the magnet though is keeping it retained.  As these are technically toys, there are safety regulations I must adhere to.  This is one of the reasons LEGO changed the coupler and has moved to an encased magnet.

[legoboy3998]

@Daedalus304, I have used Kadee O scale couplers in my LEGO trains (Kadee #806 to be exact).  A pack of 2 couplers is about $4.50 plus buying backs of nuts, bolts, and washers (12 each per pack).  These couplers "bolt on" to LEGO 2x3 pates modified with technic hole.  

@coaster, Kadee couplers use centering springs to keep them centered, (coil springs in the O scale couplers I use, and Whisker springs in many HO scale couplers).  The coil springs can be a bit of a pain to install, and the whisker springs are built into the coupler shank.  The Kadee couplers also use magnetic "trip pins" for "automatic" uncoupling via magnets mounted under the track or can be uncoupled manually using a small pointed object (skewer, pencil, etc).  

Here are some pictures of the instalation process: https://www.flickr.com/photos/73180761@N02/sets/72157687181305176

Sal

WFB, WI

 

 

 

[zephyr1934]

5 hours ago, coaster said:

It makes it much more difficult to uncouple each car, but yes, for long, continuous displays, that's what's typically done.

Couple reasons:  First, you can't 3D print magnets, and even if you could the cost to 3D print copies of the existing part wouldn't gain you anything.  Second though, there are still the issues of them decoupling under longer/heavier loads. 

Target price would be around $4-5.  I've been looking at a centering spring, but it becomes dependent on you setting the spring properly when you snap it into the plate with the hole.  Could be problematic.  Plus, I couldn't work out an elegant way to install said spring.

One of the problems with the magnet though is keeping it retained.  As these are technically toys, there are safety regulations I must adhere to.  This is one of the reasons LEGO changed the coupler and has moved to an encased magnet.

A few more thoughts, as I recall from the horn-hook HO couplers of my youth, they had an integrated "spring" in the coupler. That is to say, molded in with the coupler was a plastic tab that served as a centering spring... let's see if I can get a linked image to work. (presumably these are the "whisker springs" legoboy was talking about)

These springs are designed to be mounted with sides for centering, but you could probably figure out how to do it from the back, perhaps requiring a 1x4 plate right behind the coupler. Another alternative would be rubber bands instead of springs, but then that becomes a hassle for the user.

 

[dr_spock]

Perhaps a box that can make use of Kadee couplers rather than reinventing the spring centering coupler for LEGO applications.  It can have studs and Technic holes to mount to your LEGO carriages.

 

 

[Bricksmith]

I like it! Would look great on builds that aim to look more like the real thing. 

[zephyr1934]

9 hours ago, dr_spock said:

Perhaps a box that can make use of Kadee couplers rather than reinventing the spring centering coupler for LEGO applications.  It can have studs and Technic holes to mount to your LEGO carriages.

If coaster is talking $4-$5 per coupler and legoboy has a pair of Kaydee at the same price that might make a lot of sense (not clear if coaster's target price is per coupler or per pair though).

@legoboy3998 have you ever attempted putting your Kaydee couplers over uneven track? If so, how bad can the track get before problems start?

[coaster]

I'm not all that wild about the Kadee ones.  I've used them on other trains, and they fit well into that prototypical look, but they also look totally out of place on a LEGO build.  It would also cost less to start from the ground up rather than try to shoe-horn in a Kadee coupler.  From a design standpoint, my objective is to always "LEGO-ize" the design, such that custom parts need to fit seamlessly into the LEGO environment.  "How would LEGO do this?" is my design mantra.  For this coupler, I've based the design not only on actual couplings, but was also using the mini-blaster assembly as a form-and-function reference:

My target price point would involve a quantity break, such that a pack of just (2) couplers would be around $8, but a box of 50 or 100 or so would work out to $5 or so/pair.  Something along those lines; cost dependent, naturally. 

As mentioned, thus far, this has just been a design exercise. They obviously would have be robust enough to hold heavy loads, but also tolerant of going over very uneven track surfaces. 

[BurkusCircus]

If this could work, I'd buy it.

[legoboy3998]

10 hours ago, zephyr1934 said:

If coaster is talking $4-$5 per coupler and legoboy has a pair of Kaydee at the same price that might make a lot of sense (not clear if coaster's target price is per coupler or per pair though).

@legoboy3998 have you ever attempted putting your Kaydee couplers over uneven track? If so, how bad can the track get before problems start?

Yes, I have run them on uneven track, it can be problematic if the "bumps" are to extreme.  I believe I ran my mostly Kadee coupler train at Brikworld this year.  Our layout had a handful of 1-2 brick elevation changes (aiming for 1-2 plates per track section), plus general unevenness due to poor rental banquet tables, but we tried to make those "bumps" as gradual as possible by using shims and spreading it out.  I have all my trains packed from doing a show this weekend.  When I get around to unpacking it (by this weekend) I'll set up a loop of uneven track and shoot a video.

i did a quick calculation.  A pack of 2 couplers is $4.55, a pack of 12 bolts is $3.90, a pack of 12 nuts is $3.90 and a pack of 12 washers is $3.75 (you need 2 washers per coupler.).  Breaking it all down, it costs about $5.80 per car to install Kadee couplers.

@coaster, I would suggest getting a couple pairs of Kadee couplers (one with coil centering springs and one with whisker centering springs).  Since they already have solutions for many of the issues discussed above, I think they will offer great insight into what is needed to make your LEGO-ized couplers function as desired.  

As as a side note.  Several years ago, I talked to a Kadee rep at a train show and asked if they would have interest in doing a LEGO compatible coupler.  He said a few years prior they had had discussions with some LEGO train fans, and I believe even had a prototype made.  But interest from the LEGO train community died.

i would love to see what you end up coming up with.  It would be nice to have a ready made solution that simply snaps on to LEGO trains.  

Sal