Josephiah

That is a very good question - I just grabbed a handful of reel ends I had lying around, so I don't actually know, but my stock is nearly all PLA or PETG. It seemed to work with various different types - went for this one as the closest colour match. One I know doesn't work is older stuff that has gone brittle - this becomes apparent the moment you try to bend it! So I suspect the longevity of these parts may not be great, as they will inevitably be exposed to moisture in the air. But they are cheap and now relatively simple to make, so I'm not too worried about that. As I said, the hardest part is getting the length right!

Resurrecting an old thread, as my youngest kids have just discovered "Daddy's big Lego sets!" and are desperate to build them. So I again started looking at how to repair/replace the flex system parts using 3D printer filament. And I've found another method of forming the ends which is quite simple, and seems to work reliably with a bit of practice: 1. Firstly, when measuring out the raw filament, we need cut it slightly short. If one end of your raw piece is aligned with the end of a beam, the other end should align with the edge of the stud (so we are a couple of mm shorter than a full module). It took me a bit of trial and error to get the length right, but this should get you close. 2. Next, grip the end 2-3mm of the filament in the end of some pliers, and bend it round nearly 90deg. Flex it back and forth a few times. 3. Now, rotate the filament in the grip of your pliers by 90deg and repeat step 2, so we are now bending on the other axis. 4. We have now locally necked down (stretched/thinned) the filament to form a nice notch, with no cutting or moulding required. With a bit of practice, it fits nicely into the end pieces every time. 5. That's it! If you like, warming it in hot water can help soften a bit for straightening it out. Hope this is helpful! Now, back to that 8485 T-Rex...

Good work. It never occurred to me to try raw filament, as I'd assumed that the diameter was too large to fit the sleeve, and had never got round to measuring. I guess we just need a reliable way to notch the ends... (goes off to design a tiny mould to squish a warmed filament end into... 😉)

Loving the concept, the complexity, and the execution of this! Some lovely mechanisms in the, combined with some nice sculpting. If I may make one tiny suggestion: I think your entry video would be much improved by showing the transformation up front - at present the viewer has to search through 12 mins of gear/mechanism close-ups to find the real highlight!

I've heard that ABS becoming one of the leading filaments in the industry was largely an accident, i.e. it just happened to be first, rather than being inherently great - and it's enduring popularity is a bit of a mystery to me. At my work we basically skipped ABS entirely and went straight to PETG - much better all round. Good strength, easy to work with, and less brittle than PLA. Others I know also swear by ASA for ABS-like properties with less fuss, but I haven't tried that myself.

Now we're talking! I much prefer your approach to quite a few of the other entries which largely just substitute the studded components for studless ones one-for-one.

Laws on battery accessibility are definitely tightening up - at my engineering consultancy we've had clients asking us for product design adaptations, looking towards upcoming regulation changes. In those cases, though, it has tended to be around product which use the small "button" or "coin" lithium batteries, which can cause horrific injuries if swallowed.

I guess I was thinking if you were going to program it's movement at all, you won't get a reliable result because the direct connection between input and output is broken. But I'm not too au fait with the Control+ system, so I may be overestimating its capabilities!

I can't see them doing Control+ with pneumatics, as you would then need position sensing to tell when an actuator stroke had been completed (unlike a geared mechanism or linear actuator where the number of turns is known).

We have a couple of Prusa i3 filament printers at work (used a lot for several years), and a new SL1 resin printer (not used much yet). I haven't tried doing technic parts - as you say, the tolerances will be a challenge. I have had reasonable success with printing Duplo parts (well, actually, adaptor pieces from Duplo to Brio train tracks - worked really nicely with a bit of trial and error). I don't know if you might want to try a finer nozzle than the 0.4mm one many printers come with by default? Don't bother with ABS (it's ubiquity in the 3d printing world was largely a matter of just being there first) - PETG is my go-to material at the moment - it's stronger than PLA, and less brittle - it has a bit of flex to it. Friends recommend ASA as well, though I haven't tried it myself. One of the biggest challenges with 3d printing is getting the directionality right for any given part - you really have to think through what direction the stresses will be going through the part - if the forces on it are in a direction that makes the layers want to separate, then consider printing in with the layers in a different direction. And sometimes the best solution for a particular part is to print in two halves (to get a nice flat plane to put on the printed) then glue together. With a bit of trial and error you can even peg/glue parts together with short pieces of raw filament. My limited experience with resin printing so far suggests great surface finish, detail and appearance, but with a tendency to sag/warp in the printer if not supported well enough (so we've often found the dimensional accuracy worse than our filament printers), and the materials tend to be a little more brittle. In terms of the mail-order services, we've had good results from Materialise (both the full on commercial (expensive!) service for work, and the more accessibly-priced consumer version); haven't tried Shapeways, so would be interested to hear people's impressions. Have fun experimenting - will be interested to see how you get on.

Thanks @Rudivdk, appreciate the effort. I hadn't thought of the issue of the cable just buckling in compression - I guess the plastic ones have enough stiffness to prevent them from doing this over the short distance they protrude from the sleeve.

Thanks @Rudivdk, my forum-searching skills are obviously lacking, as I didn't find that one - will take a look.

Hi guys, This week my boys had a special treat as we brought my old 8485 set out of the garage so they could build the T-rex. Everything is in surprisingly good condition, except for a couple of the flex system cables, which are a lot more brittle than they were 25 years ago and have not survived the building process. (These were always fairly weak parts, I know, but they are definitely worse than I remember!) So the question is: has anyone discovered any way of mending them? Or replacing them in some way? (I have ordered a couple of spares from bricklink, but I don't expect them to last particularly long either, as they are just as old...) Thanks in advance!

Just stumbled across this project, and I love it! Particularly like the way the switch track flexes as it moves. Do the train and the switch tracks interact with each other? In other words, does the train "know" the status of the switch, and vice versa? I also like the versatility of the tiled track system with the colour sensor - adding a slow down area or a station is as easy as swapping a couple of tiles.

Depends whether it has a crane on it or not!

We use it for prototyping stuff fairly often at work, especially for fast iteration around mechanism design. Complex hand tool mechanisms, an adjustable wind turbine blade, bits of a wheelchair, a folding table mechanism, even a new type of horse saddle... (Unfortunately not stuff I can readily share as pictures, as they now form parts of our clients' IP!)

As a few have commented further up the thread, the seemed to be a change of insulation material for the sets around 1996/7 (shuttle/bar code truck) to the matte/rubbery textured stuff which was particularly bad. I have plenty of earlier wires (shiny finish, like in 1995's control centre II) which are in much better condition.

What about 8862? https://brickset.com/sets/8862-1/Backhoe-Grader

Brilliantly executed - accurate motions, and love the folding rails. There used to be one of these that would come with the fairground every year (complete with the independently moving arms with hinge, as in your version), which I was always too scared to go on. Can't remember what it was called, but it had a purple colour-scheme, with lots of spotlights and the words "hasta la vista baby" on the underside of the carriage.

Reminds me a bit of a project my company (a small engineering/r&d consultancy) has worked on now and then over the last couple of years, which uses the same principle for turning crane loads, but uses water as the reaction mass: http://torquer.com/services/ (In fact, my latest bit of work for them was making a small working model version for them to take to trade shows with them - one of those rare projects that's almost too much fun to count as real work!)

Yes! For this kind of simple model I'd say 30-60 seconds is plenty. (I find even with large and complex MOCs, 2-3mins is usually plenty, and much better to watch than a drawn-out 6-8min epic!)

Also, this rather fine 42053 telehandler by @Hogwartus:

This looks promising! Looking forward to seeing how the pneumatics get incorporated. And, no, definitely not copying, just inspiration

Noting @Leonard Goldstein's excellent model down for future reference:

Snapped up a 42043 for my birthday before they all disappeared - my first decent-sized set since 42009. THIS IS EXCITING!!