So last year, our family went down to Southern California, did the Disney-world, Seaworld, and Legoland triumvirate around San Diego using the awesome house at Pelican Cove, Caarlsbad as a home-base. While we were there, our 5-year-old boy loved the build-your-own-boat game that's just by the entrance. We could barely get him off it to see the rest of the park.
As is its wont, time rolls on, and in a few months it's his 6th birthday. The Boss-lady has decided this year's party theme will be Lego-based and suggests that it would be "really cool" (this is code for "you need to figure this out pronto") if we could do something similar in the back yard. Thanksgiving week is coming up, so I'll have some time to work on it, so this is a record of the (hopeful) success in making a back-yard version of the Lego-land boat race.
One of the first things to decide was just how close to the original we wanted to get. We could have just gone for the easy option (putting a bunch of rain-gutters at an angle, and piping some water down them all); but … if I was going to do this, I wanted it to be as faithful an implementation as possible - my son had already played on the "real thing", I didn't want the home version to be a disappointment, fun though the rain-gutter approach would undoubtably be.
The basic design is as below. I'm going to use standard 4' x 8' plywood sheets to make the "ground" level, with the first two of them being slightly angled so the water will flow downhill, to a flat corner piece, which lets me make a right-angle bend to another angled-downwards full-size sheet, and then onto a flat retrieval area where the boats will collect.
The plan is to paint the pieces basically darker blue as a base colour, and then (once the obstacles and sides are in place) use lighter blues and white to make water-like background. Once the water is running over it, it'll hopefully look pretty realistic, especially as it'll follow the curves of the water raceways.
Each section will be stand-alone for ease of storage, and they'll rest on purpose-built 2x4 trestles (which will be stackable, again for ease of storage) of different heights to get the downhill effect. I'll put 2x4 strengtheners underneath the plywood to keep them rigid, and have a 2x4 beam protruding down underneath each side where the pieces connect, right at the edge. The idea is that I can then use G-clamps to lock the pieces together and keep them attached to each other.
To keep the seals between the pieces watertight, I'll be glueing a rubber bead along each adjoining edge, which the clamps can then compress. I'm not overly worried about a *bit* of leakage (this is an outside-only activity :) but I'll try and make it as watertight as possible.
To form the 6"-tall sides and the (probably 4" tall) obstacles within the water-stream, I'm planning on using 2" tall boards of smoothfoam. I ought to be able to cut this into curvy strips that mate together well at the piece-boundaries, and still keep the modular design of the pieces. I'll use the same procedure to make the islands etc. past which the water will flow.
I'm not overly worried about the sides not having the same rubber strip to make a watertight seal. I'm planning on making the edges of the foam protrude by about 1mm on each side, and I'm hoping the resulting compression will make it "good enough" for a day's boat-racing.
To handle fixing the foam to the plywood, the main support will be glue but at the plywood-piece-borders, I plan to screw the foam to the plywood, using a 2"x4" "washer" to make sure the pressure of the screw doesn't rip through the foam. That way, the foam will be strongly anchored at each junction, and the glue can take care of the interior-to-the-piece adhesion. This is only on the lowest 2" layer, of course, subsequent layers will just be glued to the lower one.
Finally, we have the water circuit itself. The idea here is to have a reservoir of water at the bottom end of the run (a sump), and pump up water to the top-end, letting it flow down to the bottom and thus the cycle is complete. I happen to have a 400 gallon saltwater fishtank, so plumbing isn't new to me, and I have some of the parts already.
The sump will be made from a Rubbermaid 100-gallon tub that comes with a 1.5" exit drain at the bottom of the tank . I'll use that to plumb in one of the spare superDart Gold pumps that I have lying around (one really doesn't want to have to order in a new pump if your main aquarium return pump fails…). These pumps can push about 3600 gallons per hour at the expected head-height (the difference in height between the top of the sump and the location we're pumping water to), or about a gallon per second. If it turns out that that's not enough, I'll plumb in two of them, although hopefully that won't be required.
The bases are constructed as a simple 2x4 rectangle, measured out to match the size of the table they are supporting - so either 8'x4' or 4'x4'.
To help make the corners square, I was using a corner-vise. There are various permutations of these available, and it was lying around in the garage… It really helps to get that first corner straight!
The 2x4's are connected together by two 3" deck screws at each corner, which doesn't provide for the strongest of joints, but the plywood that will be tacked down on top will provide the rigidity.
Once the frames have been built, the next task was to attach the plywood to the top. For the 4'x4' pieces, the 8'x4' sheet had to be cut in half. I wasn't really looking forward to trying to do that on the table-saw (my table-saw is a portable-type construction-site one, so there's not much support for large pieces of wood).
Instead, I used the circular saw, after supporting the plywood over many pieces of 2x4. I could safely kneel on the plywood, and just push the saw through along the center-point line. Et voila! Two 4'x4' pieces.
Repeating the above a few times leads to a vertically-challenged layout similar to the plan.
The whole thing measures about 16' square, a 12' run, followed by a 4' turn, followed by another 8' run, followed by a 4' flat pool to drain into.
I started off by waterproofing 3 of the boards I'd need. The goal was to do an initial test of the first run, just to make sure everything was ok before I started committing too much effort. I wanted to make sure the project was viable, basically.
I used Blue Max Liquid Rubber to waterproof the plywood and the 2x4's. I put a single coat onto the 2x4's (in theory, these ought not ever actually get wet…) but 3 coats onto the plywood. It goes on a bright blue, but slowly dries to a more sea-green colour, which is still semi-transparent.
The grain of the wood (as well as my painting "technique") created what would seem to be pretty true-to-life watery effects once the water is running over them - once I'd realised the second coat wouldn't get rid of the brush-stroke effect, I tried to make a necessity into a virtue and make it look "watery"…
The initial plan called for using composite shims to set the angle for each of the 'run' sections, with the thinking being that they'd make it easy to keep the angle the same between sections. In the end, I abandoned this idea, because I didn't think they'd supply sufficient gradient for the 'run' sections.
There were three main constraints I wanted to make sure we handled … The water basin everything drains into is 12" high, so the very lowest point of the table should be higher than this to allow plumbing
I had three constraints to take into account:
I did a very basic fluid-dynamics calculation, assumed zero friction and a drop of 4" over every 8', and this gave me the requirement for about 4000 gallons per hour. So I went for 4 pumps, each of which can provide that. I figured that would be sufficient.
Worth noting is that pumps are generally quoted under optimum conditions, and mine are far from optimal - they have a reasonably long run from pump to outlet, and that run is through 1" pipe, not the 1.5" pipe that the manufacturer would have used. In plumbing, Bernouilli says half an inch makes a big difference…
With the initial goal being to test out the system as a viable project, the first module to be completed was the start of the boat-race.
Looking at the sizes of Lego boats (mainly 2" and 3" wide) and how wide I wanted the water-flow to be, I chose to make 4 starting points, each of which would have a water source at the top, and a starting raceway to provide the initial momentum. That dictated the layout at the top, and then it was a matter of producing some curvy sides in the smoothfoam using the hot-wire cutter.
Once I was happy with the layout, it was time to glue things into place - I tried a few glues out on non-critical pieces, since some glues melt the expanded polystyrene that the smoothfoam is made from.
Eventually I settled on using Gorilla Glue, liberally applied using a caulk gun. Once the glue had had a little time to set, I applied waterproof silicone all around the edges of the foam where it joined to the board or to itself. I wasn't *sure* if the gorilla-glue was water-soluble, but I knew the silicone was waterproof.
I chose 4 of the Jebao DCP-15000 pumps to act as water-returns. One of the advantages of these pumps is that they are DC, and controllable in their output. That means I can tune the water volume to suit the final system, and not be worried about there being a mismatch in the flow requirements and production.
One annoying niggle was that the Amazon comments said the pump came with a 1" hose attachment. They don't, it's 1.25", so I had to rush to Home Depot and cobble together something to allow the 1.5" slip output be throttled down to 1" to suit the tubing I'd bought.
Then it was a matter of drilling holes for, and inserting bulkheads in each of the starting raceways, putting 90-degree elbows on the top, and 90-degree hose-barbs on the bottom of each bulkhead, and linking up the hose-barbs to the pumps. Sorted.
With all that done, it was time for a test. The pumps were all linked up, and switched on, the water started to flow, and fall over the end of the first module - rather messily in fact since it wasn't being plumbed back into the sump at that end. Still, I ended up with
At the end of the video, to the far left, you can see the second piece of the course waiting for its SmoothFoam sides. These pieces will be joined together by placing a thin piece of the foam between them, covered in blue painters tape for colouring, then clear plastic tape for waterproofing, thus:
Once the foam is in place in-between the boards, I intend to clamp the boards together using standard woodworking clamps, two per junction, one on each side. The pressure exerted by the clamps, together with the compressibility of the foam ought to provide a pretty good seal for the water to flow over
And so the next step was to figure out the obstacles to make life more difficult for the boats, which also meant making it not too difficult or we end up with just a log-jam of boats and, of course, no fun.
The chicken/egg problem was that to get the correct placements and sizes of the obstacles, I had to be running the water, but that also meant the obstacles had to be held in place or they'd just be pushed down the course by the water. The solution was … practical, if not elegant:
The below shows my very keen "helper" demonstrating the plusses and minuses of the current course layout. In the end we trimmed off a fair amount of the islands because they were causing the boats to get blocked more than they were helping the water flow. In some cases we actually had eddy currents going back up the course!
Up until now I'd just been draining the water directly into the sump that recirculated it back up to the jets at the start of the course but I needed something to let the kids pick up their boats without going near the pumps… I decided that 4 pumps would need 6 drains, and made a 4' square table for the recovery-pool at the end…
With the course sorted out, and some alterations made to the islands to make the flow better, I decided, Clarkson-style, we needed "more power!"… So another pump was installed (for a total of 5, now) with a directional head about a third of the way down the course. That meant I could add water where the natural slope of the driveway would have made it less flow-ful.
After the rapid realisation that 6 drains would in fact not be sufficient (the volume of water pumped under pressure through a 1" pipe is far greater than the volume of water that drains under gravity through a 1" bulkhead), and the plumbing-in of that extra pump, we had something that looks like…
Finally, I made one more 4' by 4' table, so the kids could have somewhere to build their lego boats without lego being spread all over the driveway…
We also used one side of this to hold 8 hooks, each of which had 2 artist-smocks hanging off it, so that the kids didn't get completely soaked while they were playing on the table. At the party, some of them even used those smocks :)
The project was a resounding success, judging from a technical viewpoint, from the comments of the various guests to the party, and from the kids reactions. I got a lot of questions asking where I'd hired it from (despite the, to my eyes, rudimentary construction). There was a lot of disbelief when I said I'd made it - I'm not sure if that was a compliment or not [grin]
Having said all that, there are a few things I'd do differently if I were starting over from scratch…
However, these were minor niggles on a day where pretty much everything went right. I can't say thank you enough to my wife for all her planning and hard work in organization and execution, and to all the guests who came over and enjoyed the day with us.
Summary: If you're going to throw a kids party, combining Lego with splashing water is a really good way to keep 30+ kids quiet. We also had a bounce-house, play-house and swings, so it wasn't all the boat-race, but that's what stole the show in terms of the outdoor things-to-do :)