Making Lego Tracks — A Kickstarter Story

Peter Naulls
13 min readMar 1, 2019


Everyone knows what Lego is, and everyone knows what trains are, and some people even realize that Lego make trains! In fact, the first train was made by Lego way back in 1966. Those early trains though the 1970s were very chunky, and by today’s standards, pretty ugly.

It wasn’t until 1980, where Lego adopted a more realistic setting to many of its lines including town and trains that things really got going.

I’m an AFOL (Adult Fan of Lego) and I collect Lego trains from 1980 to the present. I have perhaps 30, with additions like stations and other paraphernalia and a lot of track. Not an extensive collection, but not small potatoes either.

The problem with inter-operation is this — although the track gauge is identical in all cases at 6 Lego studs apart or “L-Gauge”, the systems of power and the track itself has varied over nearly 4 decades, and so the challenge to the hobbyist is how to run all those trains on the same track.

Back in 1980, or the “grey era”, there were 3 types of Lego trains — push trains, 4.5V trains, which were powered by a battery box in a tender and also 12V trains. All of these ran on the same grey plastic track, but the 12V trains also needed a powered center rail. The train motors took power from the center rails via sprung bumpers. You can also see blue track from previous decades, but it’s much the same.

In 1991, Lego switched to a more integrated 9V system, where individual track pieces didn’t need assembly and included sleepers. the track was mostly plastic, but the top part was metal, more like a traditional model train. These motors had metal wheels, and took power directly from the outside metal track. Also of note, these motors had 4 metal wheels, not 6 plastic like the 12V system, but apart from appearance, most or all of the 12V trains can be swapped out with 9V motors.

Lego kept the 9V system through to 2004, then switched to entirely plastic tracks. There are advantages to an all plastic system — reversing loops are possible without tricks, and because the trains now all had onboard power and control systems (Lego would use various RF technologies), trains could be controlled independently. Also it allowed the introduction of short, flexible track, which would have been difficult with metal.

And so, all the Lego trains and track since 2005 to present have been with this plastic track. It’s easy to get hold of, and cheap, and you have a far greater choice of layout options due to more flexible geometries than with metal track.

But the problem for us hobbyists is a 9V and 12V trains which can’t run on this. It’s possibly certainly to buy second hand track, but it’s become increasingly expensive. Certainly some clubs have hundreds or even thousands of feet, and without modification, you’re limited to the single curve radius provided by Lego.


It’s worth talking here for a moment about track geometry. In short, it’s the ways you can physically assemble the track, to make say, a loop, figure eight, or which pieces you might need to match every up when you use a point (that’s a switching track, or just “switch”).

For standard Lego curves, you need 16 to make a full circle of radius 40 studs or R40, which is the contraction used by the Lego community. A straight is 16 studs long. If you stick to 90 degree turns (4 curves in the same direction) and correctly place straights in the right places, then it’s easy to match everything up. However, Lego track collections usually have way more curves than straights, or if don’t follow the requirement of matching a left handed point with a right curve, then you end up with some mismatches.

Now, there’s some degree of play in the track, and there’s a number of known approximations in the geometry, so it’s possible to make it work if you are careful and don’t mind a little flex.

For non-9V plastic track, there’s the later flexible pieces. These are very short lengths that curve and can be snapped together to make any length. This makes it easy to join up that very last section that’s off by a few studs and a weird angle, and again, this is good for some setups, but of course 9V trains can’t run on it.

ME Models

So, a demand existed in some quarters for more both more 9V track and also just the plastic track in different geometries. (For completeness, I want to mention here that many Lego train hobbyists have substantial 12V track collections which they maintain. The focus of this article is mainly on “9V people” like myself, but I have one note about 12V a bit later)

And so ME Models launched this KickStarter project:

As it happened, I missed the original Kickstarter dates for reasons I don’t precisely recall, but managed to pick up support on BackerKit. My receipt shows that my credit card was changed in June 2014.

How much?

What is slightly embarrassing is perhaps how much I spent. I opted for the full $1000 deal, which was advertised as:

10 ABS FULL loops including:
2 X R56 16 sections,
2 X R72 16 sections,
3 X R88 16 sections,
3 X R104 32 sections,
16 ½ length ABS rails,
64 sections of standard length ABS rails,
16 sections of double length ABS rails; Color: Customer Choice; Free shipping in the USA.

Stated differently, this was $1000 worth of track plus $200 extra plus the engine. In what proved to be the first confusion (mainly my part), I realized I didn’t want the ABS (plain plastic) track, but the metal track — which of course was more expensive. Some frantic emails followed, and it appeared to corrected, although I did get asked on at least 2 more occasions exactly what my order was. In the end, I did get the correct track, but we’re getting ahead of ourselves. Here’s my order:

  • Half Straight — 4 packs(total 40)
  • Single Straight — 4 packs (total 40)
  • Double Straight — 22 packs (total 88)
  • R56 — four packs (total 32)
  • R72 — four packs (total 32)
  • R88 — four packs (total 64)
  • R104 — four packs (total 64)

Yes, that’s a lot of track. How much? Well, again, we’re ahead of ourselves, but this is 360 sections.

From the Kickstarter page, we can see that 8 people went for this. From BackerKit, who knows, but it seems likely the number was small. Is $1000 a lot to spend on an unknown product for a hobby? In hindsight, it was. On the other hand, for something which could be used for decades (remember that some of my trains are nearly 40 years old), it’s pretty cheap as hobbies go, and there’s still no good alternative.

Updates and Progress

It would be charitable to say that updates were spotty and infrequent. There were large periods where nothing at all was said.

From the communication that was received via Kickstarter updates (which I often didn’t get directly, but had to chase down), I got the impression that who ever was leading the project was someone to charge ahead and not really sanity check anything. Don’t get me wrong — this was always going to be a big undertaking and could only be pulled off by someone very determined — but as evidenced by the sometimes weird grammar, previous confusion over the track I was going to get, and continued insistence on referring to BackerKit as “Kickbacker”, as well as the content of the updates, there was a slightly uneasy feeling pervading the whole deal.


Did I seriously consider asking for a refund? Yes, on several occasions. However, there were just enough updates to decide me otherwise, and it wasn’t really clear that I’d get one either given Kickstarter policies. I had of course known that anything like this was a risk. It’s still a lot of money to lose though.

Further Update

In an April 2015 Update some of the financial problems had become clear — there had been a significant short fall in the collections. No doubt this helped lead to the eventual downfall of the company. From the numbers here, we can also make some guesses about the overall number of BackerKit backers too.

The Metal Rails

The update contained something else insidious too — problems with making the metal rails. If we look at the numerous images on the original pages and earlier updates, we can see something like this:

New and old

The end pieces of the metal track are a mechanical fit for the original pieces.

In some updates before shipping, something changed. The rail being used was switch to standard “Code 100”, which is used extensively in “regular” model railroading. This switch implies that the original plan to use mechanically identical track had been abandoned as well as the presumably expensive tooling for the metal parts. It’s unclear if the plastic needed rework too to hold the track, but the conclusion is that the metal extrusion just wasn’t practical in the end, and it would use completely stock (and unbent) metal track, plus standard track joiners.

The Shipment

After yet more confusion about the exact order and address, my shipment finally did arrive in a beat-up box of 27lbs.

It included the fully 360 pieces of track over 46 bags including 22 bags of double length straights. Also included were 6 bags of half-length track joiners, which I’ll get to in a moment. That’s a total of 384 track pieces, in 52 bags with 3072 individual components.

Bagged up, ready for assembly


As I noted, the track has to be assembled, unlike regular Lego track, which is a single piece. It took me a long time to assemble my first double length — the clutching (or to non-Lego people, how well it sticks together), is fantastically bad and not at all like regular Lego pieces. It takes a while to press everything together, but if you happen to drop it, it will quite literally explode due to the contained stresses. After some assembly and then later disaster when the box of partly assembled pieces was dropped, meaning almost all the pieces came apart, I realized it would have to be glued. The pieces then sat for the best part of 2 years.

The Lego company makes a big deal about the tolerances of its pieces and how they last for decades — and it’s true, it’s not unusual to have collections mixing pieces 40 years apart. In this case, the tolerances would have had to be off by only a small amount — perhaps only a 1/10th of a millimeter to make it not work well. What a shame though — seems like the kind of thing that ought to come out in test molds, but maybe it’s just too hard to get right.

Other Assembly Problems

There’s other problems during assembly too — there’s no instructions! For straight pieces, there’s 3 ways to assemble the directions of the tracks, and only one way will match up with the curves.

Double lengths being glued

The second problem is more aesthetic — what do to with the sleepers? There’s a mismatch between packs of sleepers with and without holes, and some packs have spares. I opted for a solid sleepers on the half lengths, one on the single length, and 2 on the double lengths and each of the curves, not knowing if I would have exactly enough at the end. I in the end I didn’t — out of 384 pieces of track, I was just 5 holed sleepers short, but ended up with 20 solid pieces spare. Unfortunately, due to clutching problems above, these can’t be used as regular 2x8 Lego plates.

Track Joiners

The track joiners are the solution employed by ME Models to match up the code 100 track with the Lego track. They are half length sections with code 100 joiners on one end, and stepped ends on the other. Do they work? Well, yes, but they impose certain limits on the all important geometry.

First of all, there’s only a limited number of joiners — well, 24 in my case, since I had so much track. But for practical purposes, you can’t just interchange with regular Lego track — instead, you need to plan long runs of ME Models track and then use a train joiner when you need to use regular track — as in the case particularly of switches. The further irony is that despite the vast availability of regular Lego 9V curves (R40), you can’t immediately interact with it without the joiner. Given the 4 other curve geometries, this is hardly a limitation, but it’s something to be aware of in track planning.

Track joiners up close

The Call

Sometime after receiving my shipment, I finally arranged a call with one of the “Mike”s. The call wasn’t entirely pleasant — he’s the type of guy to run over the top of you, and it was hard to get a word in edgeways.

I did manage to complain that the track was “explosive”, to which he remonstrated that it worked just fine. Of course it didn’t. After some more back and forth, some glue was recommend, which I carefully noted down. When I went back through my note books some time later (several times in fact), I was unable to find the exact glue that I should buy.

Mike explained that there were more geometries coming, and everyone would get their track, and said my engine was on its way.

The Engine

Well, the engine wasn’t on its way, and it took several months of queries for it to finally show up, in several pieces in dubiously wrapped bubble wrap. But it was the promised engine.

Definitely not a bullet train

One thing that struck me about the E44 is that it’s actually pig ugly for a train — both the original and as a model. It’s really not the most stunning piece of railway engineering by a very long way. Also the model was mechanically quite poor — this was largely due to the mismatch of parts at one end where they seemed to have run out. I did figure out how to reassemble it, sans instructions.

I later asked for instructions, which again were promised, but never showed up. That was the last I personally heard from ME Models.

The Great Gluing

Finally, around Christmas 2018 and one more look through my notebooks, I gave up and went and purchased Loctite brand glue from the hardware store for about $5, which promised to be OK on plastic. Really good plastic glue actually melts the plastic slightly making for a very strong bond. These tend to be 2 part epoxies, which are great if you need that, but are also potentially expensive and messy.

Double length drying

Realizing just how much gluing I might need to do, I opted for an easier approach and just used basic glue. After about 20% of the way through, I ran out the first tube, and ended up buying “Gorilla clear grip”. This turned out to be a slightly weaker and much slower drying glue, but set overnight was just fine — and it came in a much larger tube which ended up being enough for almost all the remaining track.

Different sizes drying

Assembly Problems

Due to the clutching, assembly for gluing wasn’t entirely trivial. I came up with these rules:

  • For straights, start with the pieces at either end, and move
    evenly inward.
  • For curves, put the middle piece first, and match up the sleeper
    with the indents on the bottom.
  • The correct orientation, with a piece of track vertically, is for
    the rail joiners to be on the top right and bottom left
    (they can pull off, but better to get it right first time around).

And so it went. It took me a couple of weeks to get to all of it. Some of the pieces got dropped and the metal rails came loose — I ended up gluing those.

When it had all finally dried, I dumped it out on the floor and went through everything — twice. In about 15% of cases, I found problems — either I had forgotten to glue a sleeper, or had put insufficient glue or one of the metal pieces had come loose. For these final fixups, I ended up using the strong Loctite to be sure.

How Much?

Along with the joiners, here’s the final tally of pieces:

All done!


Does it work? Yes, it does. The flexibility of the track is certainly different than regular track, but it’s OK. What I wouldn’t call this track is a “toy”. In the sense, that you can’t just snap it together like Lego track — each piece has to be carefully laid with the metal track joiners and you have to take care how the run works with the track joining pieces. But for people like me who actually got their track, it’s still pretty satisfactory. I’ve yet to do a very large layout with it, but we’ll see then how it fares.

Some Other Experiences

A review of all the currently available parts. Me Models is (was) of course the
only maker of metal parts.

Some pictures:

And this guy had problems with the metal parts keeping a curve:

In the notes at the bottom of this, one guy is talking about replacing
all the code 100 with longer pieces to make continuous rails.

Other 3rd Party Track

These are all the people are present producing Lego-compatible track that I could find.

  • TrixBrix — Extensive range of plastic track.
  • 4DBrix — Also many kinds of track, and especially control systems.
  • BrickTracks — Yet more plastic tracks.
  • Sprite World — Amazon Seller of compatible plastic tracks.
  • OKBrickWorks — Only present producer of metal curved track. Special shout out for research helpful to this article and comments about the ME Models track.
  • This guy, on eBay — Making standard length straight metal track only. Note, this is copper tape, which to me doesn’t seem like it would last.



Peter Naulls

Vegan, IoT, Coffee, Lego and DeLoreans