Captainowie

You've already got 4 different 24t gears there. If that argument was valid you'd only have one entry for all 24t gears.

They're very small, and they're very light. And apparently they make good minifig-scale generators. I've seen one used to very good effect making a dynamic brick-badge at a convention. Owen.

I don't understand - how does picking up the phone make you crash? Does that switch disable the link between the steering wheel and the car? If so, that's not a very fair demonstration of the difficulties of concentrating on two different complex things at the same time. If not, what does the switch under the phone control? Does the road go faster when the phone is picked up? Owen.

That seems plausible. You learn something new every day!

Really? That would surprise me. It would mean that I didn't understand how either or both of the PF variable speed system or the speed regulator worked. As I understand it, the speed regulator works by varying the voltage supplied to the motor, whereas the power functions system uses coded messages to tell the motor how fast (or in the case of the servo motor, how far around) to go, all at the full 9 volts. I don't see how using the speed regulator can help here.

For a LEGO solution you'll need a train remote or a rechargeable battery. Or an RCX programmed to speak the Power Functions protocol over its IR port. There are also non-LEGO solutions too. Arduino has been mentioned, but presumably there are others. Heck, even a universal remote control would probably work.

I am going to put the whole design up, one section at a time as I get there, in a thread on the greatballcontraption.com forum, here http://greatballcontraption.com/wiki/Special:WikiForum/Table-spanning_module if anyone is interested in following my progress.

Yeah, it looks nice, doesn't it! Thanks for your help. If I decide I want a 2L thick continuous edge, I can always add a couple more frames in the gaps, but I don't think that'll add enough strength for the extra weight and cost to be worth it. I haven't decided what the final colour scheme will be. I made them yellow only really to differentiate them from the red connectors and grey frames. Owen.

Wow, Eric, that's quite the truss! The only thing that could possibly strengthen it further would be some 9L beams connecting the black bricks to each other transversely (over the studs). At the moment, you've only got pins holding the sides on. Nonetheless, this is perhaps too complex for my needs - I can't imagine my structure needing to support more than twice its own weight in conveyor belt and other bits. I appreciate the thought though, and the effort you've put in. Owen.

Some good suggestions here, I'll play around and see what I can do. I did try putting the beams on the inside of the frame, but I did it with both beams together. I like the idea of only having one beam inside and having only a 5-stud overlap - it gets me more truss for my dollar. I read a report the other day (http://www.eurobrick...showtopic=58343) that found that adding a third (and fourth, and fifth) pin to a connection did add significantly to the strength of that connection. Though extending the overlap didn't help overmuch. I think I'd better allow for the fact that the entire structure might get pulled over, in which case I'd much rather have the parts separate than break. Edit: Pretty happy with this.

Thanks Jeroen - that's exactly the kind of reply I was looking for. With that in mind, I had another go. I've got one truss module that I'm hoping would go OK either vertically or horizontally. There are now 5 studs between rails rather than 3, but only two pins connecting each rail to the frame boxes rather than 4. I wonder if that's a good compromise. The interior is only 5x3 now, instead of 5x5, but I can live with that. The only torsion I was worried about was indeed in the case of someone bumping the tower, which might cause the other tower to twist given the big long lever between them. I am certainly planning to brace the joins between the horizontal/vertical segments. I went with the red parts because I figured that one 3-long connector would be stronger than 2 1-long connectors with a space between them. They also make for an easy way to attach beams with the holes in the other direction. But I concede your point about looseness. Thanks again Owen.

Thanks all for your comments so far. I had hoped that the rigidity provided by the 5x7 frames would reduce the need for diagonal bracing. That's going to be tough, because there's no corresponding hole in the 5x7 frame to put the other side of the pin! Oh, and one design constraint I forgot to mention: I anticipate most of the parts coming from next year's LUGBULK program, so I want to minimise the variety of parts used. Owen.

Hi folks. I need some advice from people who have experience building large structures. I'm looking at building a GBC module that takes balls across a gap between two tables, that's high enough to walk under. I need a vertical lift of about 1.3m, and a horizontal travel of about 1m. Now, this would be by far the largest model I've ever undertaken, so my intention is to design the thing in LDraw before I buy anything. This means I can't see how sturdy the structure actually is, so I'm asking you all whether what I've designed so far is sensible. I've got two slightly different truss designs. One that I think would go better for the vertical parts, and one for the horizontal. Each is a single unit 16 studs in length, with the idea that you join as many units together as you need to get the length/height required. First, the vertical segment: And what three of them together look like: I feel that this truss is going to be strong in compression/tension, but all that space in the middle makes me nervous - is the 5x7 frame box going to be solid enough to make it strong in torsion as well? Next, the horizontal segment: And three of them together: This one I think would be quite resistant to bending in the up/down direction (exactly what you want for a horizontal beam supported at its ends), less so in the side/side direction, but also strong in compression/tension. This one would be somewhat heavier than the other - overkill? or is the extra strength warranted? I'd appreciate your thoughts on these two truss designs. If you want to see the .ldr file that these images came from, it's here. Regards Owen. P.S. If anybody wants to take these and make them into a tower crane for [TC8], go right ahead. EDIT: Yes, I know there are no triangles. I am relying on the rigidity provided by the 5x7 frames in place of diagonal bracing.

Trouble is, they all sound pretty much the same.

How is the main mast connected to the superstructure? Surely all that weight isn't resting on one axle (or possibly one each side). Nice work! Owen.

Why not? It's been long enough since the last argument, about time it rolled around again ;-)

10: 10 13: 6 9: 4 11: 3 16: 2 1:1 A few too many variations on "car -> car with small stubby wings" for my liking, but a lot of good work overall. Owen.

There can't be much ground clearance at the low points of the loop. I'd hate to be a slightly taller than average fig!

I also liked the smooth pan across the top of the module at the end. A nice camera technique.

I have a pair of rubber-tipped pliers that I got from a hobby store. They're pretty crappy, so I'm on the lookout for a replacement. I also have a couple of paperclips that I've bent into interesting shapes (useful for helping to thread a chain through a bunch of gears, or hook a rubber band onto its pulley in a tight spot), and a 12L axle with its end sharpened just a little (great for pushing axles through cross-holes - you don't have to get it in the exact orientation first).

Wow. Just wow. Excuse me a moment while I pick my jaw up off the floor.

I've no doubt that someone here will take the 3D model and produce adequate instructions. Owen.

Dagnabbit, I came to this thread looking for tachometer designs so that I could measure a GBC ball rate. Seems I've been beaten to the punch! Even if, as aeh mentions, it's only an approximation. Nicely done. If nothing else, I now know the following things: Even though it's only an approximation it's certainly good enough; and to do it properly you'd probably want a centrifugal tachometer, which requires much speed to be effective, which you can't really get with the start-stop motion you have with the 6-arm wheel. Owen.

That's in the first four pages of this thread!

Yes, fraud. That's totally the best way to go.