MOC: Apollo Launch Tower

[NathanR]

Ever since I got my 21309 Saturn V, I thought it would be cool to have a model of the launch tower to go with it.  Well, it may be a long time before I get a chance at building this model for real, but I have completed the digital design!

If anyone is interested, an LDD file is available from my bricksafe page: https://www.bricksafe.com/pages/NathanR/real-space/apollo-lut

 

 

 

Original first post:

Ever since I got my 21309 Saturn V, I thought it would be cool to have a model of the launch tower to go with it.  I'm clearly not the only one because a similar model recently reached 10k votes on Lego Ideas - however, I'm impatient and so I decided to start building my own.  I've been working in LDD due to a lack of physical bricks, and the first (very rough) sketch model is complete:

 

The Saturn V rocket was launched from the Mobile Launch Platform (MLP, nothing to do with ponies...), a two story structure on which the rocket was constructed and fuelled. It measured 169x135 feet, or 56x46 studs at 1:110 scale, and also carried a 380 ft tall launch umbilical tower (LUT).  The tower featured nine arms that provided electricity and other services to various parts of the rocket, and swung away just a fraction of a second before it blasted off.

The rocket did not stand on the engine nozzles, but was instead suspended over a single exhaust vent measuring 45x45 feet. This should be 15x15 studs but I've gone with 14x14 studs for ease of building:

 

I'm particularly proud of the design of the hold down arms:

 

These four clamps held the rocket down until all five engines were confirmed to be running at full power, at which point they retracted behind heat-proof shields and allowed the Saturn V to soar skywards. Three additional holddown arms were located around the pad and have yet to be modelled.  I'll add details once I have the core structure sorted.

The base of the MLP uses 1x4x6 door frames like the version that was on Lego Ideas - an inspired choice which I've temporarily copied here (I'm actively trying to avoid looking at the Ideas model so I don't copy it), but I'm toying with 1x1x5 bricks to represent the support columns instead.

I could really use some help and advice with this model though.  How do I make it structurally sound?

The MLP is a glorified box with a hole in it, but I've never built anything this big before and I'm not sure how to link together the baseplates. I had a few ideas for some internal supports: 

  

Do you think these would be strong enough? Should the outer wall be thicker? Should I criss-cross the entire hollow space with brick beams? How do I strike a good balance between strength and low parts count? 

You can view larger pictures and find the LDD file on bricksafe.  If there's interest I'll make instructions available when the model is complete.  Any advice or constructive feedback would be much appreciated!!

Edited July 6, 2018 by NathanR

[rollermonkey]

It depends on how strong you want it. I recommend that you look at the instructions for the Sydney Opera House and the UCS Millennium Falcon for some ideas how to create a rigid internal frame to a large model using Technic beams. I've built both, and the internal frames are key for each model. 

[NathanR]

On 11/09/2017 at 5:10 AM, rollermonkey said:

It depends on how strong you want it. I recommend that you look at the instructions for the Sydney Opera House and the UCS Millennium Falcon for some ideas how to create a rigid internal frame to a large model using Technic beams. I've built both, and the internal frames are key for each model. 

Thanks for the suggestions!  The base needs to be very strong, since it has to support the entire weight of the rocket and the launch tower.  I'm very familiar with the UCS falcon, but the Sydney Opera House was incredible.  I'm really regretting not buying that one while it was in the shops, it was an amazing build.

Trying to figure out the interior structure will have to wait a little bit though.  After testing different arrangements of plates for the very base of the MLP, it seems I can really simplify the internal wall structure by nudging the exhaust vent one stud away from the tower. However, this breaks the accuracy of the model.  I found a French website with loads of technical drawings, and at the moment the ten stud gap between exhaust vent and the wall on the right (in the photo above) puts the centre of the rocket exactly where the blueprints say it should go.  Of course, I may have to move everything over to accommodate the Launch Tower foundations.  Decisions, decisions...

 

Over the last week I've been getting in a mess with the base of the launch tower, since no individual side is flat:

The four supports all taper inwards.  I did consider trying to build some "Pythagorean Quadruples".  These are like the triples (e.g. the 3,4,5 triangle has sides which are all whole numbers), but work in 3D with a^2 + b^2 + c^2 = d^2.  An example is a 1,2,2,5... diagonal, I guess? But I think these would be too fragile in Lego, at least In don't see a way to anchor them securely and provide enough support for the rest of the tower.

So, I am back to the pythagorean triples, trying to come up with suitable triangles that will work in Lego bricks:

 

There's a lot of scaffolding, a lot of trial and error, a lot of angles that almost but don't quite work.  It's possible I've miscalculated the height at which each platform must be placed (which complicates things) but I'm also trying to look at triangles that are possible when (say) a stacked wall of technic bricks have one or two plates in there as well.  I don't suppose anyone knows of any guides detailing possible lego triangles?

[Dragunov2]

Very nice and then make the tracks and such so it can move

[NathanR]

Well, it's been ages since I posted an update but I'm still tinkering away with this.  I've done a lot of work on the lower levels of the tower, trying to figure out how to build the triangular framework "in system".  There are loads of constructions that are almost right, but put a bit of a strain on technic pins when you try and force them together.

I wa surprised how strong the technic framework at the base of the tower actually was, although I've swapped technic beams for a combination of axle pins and connectors because it looks better (I hope this won't weaken the structure).

One major problem I'm having is with the launch platform itself (the MLP). It's a giant grey box... and it's far harder to design than I would have ever imagined.  The outer walls are a combination of 1x4x6 door frames, and 1x6x5 wall elements, with top and bottom decks made from an assortment of plates.  I've tested a 16x16 corner with real bricks, it's strong but the slightest flex or twist causes the top and bottom plates to pop off, and the walls immediately fall apart:

I've studied the instructions for various large scale Lego builds like Sydney Opera House, Tower Bridge, the Taj Mahal, even the Maersk container ships, but I don't see where I'm going wrong.  How do I make this internal frame more rigid? Do I need thicker walls? Do I need to change how I overlay the plates top and bottom? Would technic bricks inside help lock it together better? 

I have an LDD file with some sketches here, and also some more photos in the gallery. Any ideas?

 

[zux]

Not really which part of your creation is shown on the picture with bricks going apart. But I'm surprised you've build bricks on top of each other with an overlay, which would interlock them and would allow vertical displacement so easily.

Technic bricks are very good for making sturdy construction if you do it the right way. Sadly I cannot suggest you better construction, as I don't really understand the point which is being stressed.

[NathanR]

13 hours ago, zux said:

Not really which part of your creation is shown on the picture with bricks going apart. But I'm surprised you've build bricks on top of each other with an overlay, which would interlock them and would allow vertical displacement so easily.

Sorry, I'm not too good at explaining things.  The bit shown in real bricks is a rough build of the base platform, the corner away from the tower and near where the rocket will stand.  I don't have many parts to hand, so I built the 1x4x6 door frames as a stack of 1x4 bricks, the 1x6x5 wall elements as a stack of 1x6 bricks.  

This is an exploded view, with a lot of bricks removed to show what I'm trying to achieve:

Around the outside, I have 1x4x6 frames for detailing, and 1x6x5 wall elements.  On the inside, I have a 2-stud thick wall around the flame pit. The 16x16 plates are so large, I had to support them in the middle with a 6x6 cross-shaped structure.  Originally it was standalone, but I've tried merging it into the flame pit wall, it doesn't make much difference. I've tried adding 2x4 columns to the empty edge of the 16x16 plates for more clutch, and adding extra bricks along the edge where the tan and light grey plates meet.

If you hold this assembled unit with the tan part in one hand, and the 16x16 plates in the other hand, and then give the slightest twist or flex, the 16x16 plates pop off.

The entire launch platform is a patchwork of plates measuring 46x56 studs, it's sheer size and weight means that the base will twist and distort a little bit whenever it is picked up. I would like to get it a little more robust if I can, I just don't see how.  Should I use smaller plates for the top decks, would they have better clutch? Should I fill the interior with criss-crossed 2x8 beams or something?

 

[NathanR]

Well, I'm still tinkering away with this. Starting to regret ever starting it to be honest.

I've gone through several drafts of the MLP, I can only build small sections at a time due to a lack of pieces, and pretty much everything I do falls apart far too easily.  My latest and greatest version (mostly because I can't test it!) features a base made of a three plate thick sandwich.  Unfortunately, the bricklink best price (not counting the walls) is ~70GBP, I'm beginning to think I'll never be able to afford the bricks to build this in real life.  

Any ideas how I could reduce the number of plates in the bottom three layers? The interior will be sealed up, it doesn't need to look nice, it just has to be fairly rigid.

LDD file is here if anyone is interested: https://www.bricksafe.com/files/NathanR/real-space/apollo-lut/V2_MLP_Draft.lxf

[NathanR]

I may have found a solution! I was looking at the 40174 Iconic Chess Set, which builds the board from four 16x16 plates.  The plates had a border of 1x12 bricks, and the seam was closed by a 2x10 central spar.  I've duplicated that technique here:

Can't help feeling that this is overkill and I could hollow out some of the supports and get the cost down. Bricklink says this bit alone will set me back ~175GBP or 200EUR, which is double what I was hoping, I may end up stuck with this as a digital MOC.

Also can't help feeling that it still won't be rigid enough to stop the top and bottom sets of plates popping off.

LDD file is available here in the unlikely event anyone want to take a look at it.  The next version will include the launch tower itself, I just need to figure out how best to build the 14x14 floors given the limited selection of large red plates available. 

 

[Shiva]

NathanR, the latest picture of the model, it looks strong.

Could more technic bricks be something that could make it strong too?

Regarding the cost. Big builds will cost much, nothing we can do about that.

40174 ? I got it delivered few hours ago :)

[NathanR]

5 hours ago, Shiva said:

NathanR, the latest picture of the model, it looks strong.

Could more technic bricks be something that could make it strong too?

Thanks! I did consider more technic, but those bricks are really expensive and I wasn't sure how to lock them into the outer walls.  Curiously, the most expensive part is the 1x12 grey bricks making up the outer wall. Large panels would be a lot cheaper, but then the they wouldn't lock the base plates together.

A new update - the tower is rising...

First three floors are done.  The second floor sits about one brick too high, but I had to do this - the sides slope inwards at 22.6 degrees, so the 14th hole of a technic beam will exactly line up with a stack 11 bricks high.  What I'm really proud of, is the way all the triangle support structures on the sides are legitimate connections.  One or two are pushing the tolerance of Lego bricks, but it all fits together without straining the parts.  On the MLP, you can also see the tail service masts and (I think) a generator , which I've temporarily borrowed from the whatsuptoday LUT design.  I'll be trying ti replace these at some point (some hard to get parts). Estimated cost now stands at 270GBP.

Next up will be the rest of the tower - 15 identical decks, topped off with a crane.  Just to give an idea of scale:

The grey plates sticking out of the tower are reference points - the lower one lines up with the 1x2 grille pieces on the SII-C, and the upper one lines up with the top of the Command Module.  So I've kind of recovered from that extra brick height in the base.

[Shiva]

8 hours ago, NathanR said:

Estimated cost now stands at 270GBP.

This is not the final cost, I guess. My guess would be that this will cost more than 500£. But it can be worth it. If 1 can afford it.

Good Luck :)

[NathanR]

7 hours ago, Shiva said:

My guess would be that this will cost more than 500£. But it can be worth it. If 1 can afford it.

But one can't afford it... My budget was only £300, any more I'll have to split the cost with next year or skip some upcoming Lego sets. I will have to see what the final design comes out like...

[Shiva]

Maybe have it as a long time build? Finish it in LDD and buy parts when you can afford?

[NathanR]

19 hours ago, Shiva said:

Maybe have it as a long time build? Finish it in LDD and buy parts when you can afford?

Possibly... that's how I got the parts for 42009.  Cost is definitely spiralling out of control though, I added in the railings for each deck of the tower - 480 1x1 clips, 240 1x2 grille tiles, something like 60GBP for these parts alone. 

I've started work on the swing arms now, and a very old design decision has come back to haunt me.  I moved the Apollo rocket one stud closer to the tower, so that the base could use a a 10-wide baseplate instead of a 9-wide plate.  Unfortunately, the rocket is now too close for me to the able to use the 2x2x10 girder (95347) as the main bulk of the arm - the arm hits the rocket every time you try and move it.  So I've had to come up with a brick built version:

It's a little bit crude but I'm not sure how I could build it any better 

The other problem is the two vertical beams behind the hinge.  There is only one pin holding each beam in place at the bottom, and another at the top (out of shot).  I'd like a couple more connections for strength, but I don't see any way to build them in. The pivot point sits two beams away from the main structure, cannot be slid left or right or it will collide with the main superstructure, 

Ideally I'd need a 3x3 plus-sign shaped technic beam, but they don't exist and I don't see how to make one out of existing pieces.

Also, if I extend the vertical beams up the tower to allow the other swing arms to be attached, the pin holes stop lining up with the higher decks.  Ultimately, the entire swing arm assembly (large bulky, and critical for the model) could be held on to the tower by just 5 pins over a height of about 3 feet.This just isn't going to work... Any technic experts got some suggestions?

[BrickWild]

Nice work! I hope it's finished in time before 2019!

[NathanR]

On 01/03/2018 at 9:04 AM, BrickWild said:

Nice work! I hope it's finished in time before 2019!

Thanks! You know it took me a few minutes to realise why 2019 was special...

I figured out the connection for the S1C inter-tank swing-arm (the bottom one), I had to build a 1x4 wall inside the tower to give enough support, and a couple of 3L pins with a 1L beam (pin connector?) narrowly missing the angled support strut.  It looks a little clunky for the back, but I'll have to use this same pattern all the way up to give a strong enough connection for the remaining arms.  Also, a pretty neat solution to transition two horizontal pin connections into a vertical axle - combining 48989 Pin Connector Perpendicular 3L with 4 Pins with two 15100 technic pin with pin hole, and locked together with a pair of 11214 Technic axle pin with 1L axle:

 

I've added in the rocket to try and gauge where to put the swing arms:

Firstly, this was not a good idea for my poor computer, which is struggling to handle such a complex scene - the entire model is currently on 3500 pieces, with a further ~1900 for the rocket.  Another 700 pieces are expected for the railing details on various decks. Swing arms, detailing fuel lines and the crane on the top should add at least another 1000 parts.

Unfortunately, after working out where all the connection points should go to attach the arms up the tower, I've discovered a horrible mistake.  The top 17 decks are formed from nearly identical subassemblies. The total tower is five bricks short of what it should be.  I could just plonk an extra deck on top, but that would be cheating, so now I have to go rebuild the tower and add an extra plate to each level. And then redo all the connection points, and re-assess where the arms have to clip in.

[NathanR]

And now I'm double posting, but right now I'm basically screaming in frustration.  It's proving impossible to put the swing-arms on the right tower levels and get them lining up with the right bits of the rocket. 

The bottom arm is in the correct place - the base of it should line up with deck 5, and also line up with the bottom of the black stipe on the rocket.  The top of the topmost arm should line up with the top of the black stripe on the rocket.

It turns out that my fancy sloping sides linking Layer 1 and Layer 2 are perfectly in system, but shift the entire tower two bricks too high (not just one brick as I'd originally thought).  So I need to redesign that section.  The red sloped beam shown here needs to shrink from 13M to 11M (10M would be better but there is no such beam). Only snag is, there is no connection that is strictly "in system". The best I can see is this:

But if you look closely, the pin holes don't quite line up.  Is this close enough? Would it work if I used an axle instead of a pin?

It would be a disaster if the red beam pulled the blue brick downwards, it would causes the deck to pop upwards slightly and weaken the attachment of the next 17 levels. Are there any better ways of designing this?

[Johnny1360]

Don't really have any insight how to overcome any of the issues you now face. 

However having followed this thread since day one, I am very impressed with what you have accomplished. It looks much better than I had originally thought it would and look forward to seeing it upon completion. 

Ever since I got the Saturn V Rocket, I have wanted to build a launch tower as well, however now I think maybe not, after seeing what you have dealt with and all those parts, incredible. 

Anyway keep up the good work, I may convince myself to do this after all. A great rocket deserves a great launch tower. 

[NathanR]

On 04/03/2018 at 7:57 AM, Johnny1360 said:

However having followed this thread since day one, I am very impressed with what you have accomplished. It looks much better than I had originally thought it would and look forward to seeing it upon completion. 

Thanks for the kind words! It's nice to know other people are interested in the build.

Thanks to @DugaldIC, @JonathanM, @Saberwing40k and especially @Didumos69 over in the technic forum, I think I have figured out the new triangular struts.  Took a lot of work with meccabricks, a spreadsheet and even a short python script to give me suitable triangles.  Here you can see the old (left) and the new (right) designs for the tower's first three layers:

The new design has lost some of the simple elegance of the originalism but it's the right height! (well... just about...). The second deck is 0.3mm too high to fully connect with the central support column, but the separation between bricks is typically 0.2mm so I'm not too concerned.  In any case, this will translate to a slight strain on the angled support beam.  The third deck is even closer, but I need to rebuild the model to remove rotational distortions and then redo all my calculations to make sure.

Updating the truss-structure on each side has also been a bit of a pain.  For the sides facing the rocket, the struts between decks 1 and 2 reach a little too low while the struts between decks 2 and 3 reach a little too high.  Unfortunately those are the only connections really available.  For the other sides, it's been a challenge disguising the fact that 3-4-5 triangles are too short, or don't quite fit:

The junctions at corners are really tight, the rotation of the beams means that some key pin holes just can't be used. Still, it all looks like it's going to work.  Now it's just swing arms, then more triangles to get the fuel pipes running up the side of the tower... 

 

 

[Shiva]

Nice progress :)

[NathanR]

Well, she's starting to look like a launch tower:

Unfortunately... Houston we have a problem.  The swing arms are correctly placed on the tower, but they don't touch the right bits of the rocket.  The bottom four arms are basically fine (this is Lego after all...) but the topmost arm, which is the one the crew used to get into the Command Module, is placed just over three bricks too high.  This is like having the astronauts unable to simply step into the spacecraft, but jump down 3-4 metres into the open hatch.  Some giant leap for mankind!

Worse, the tower is already a bit shorter than it should be. The column of yellow bricks shows the height of a 380ft tall tower, at 1:110 scale.  So where have I gone wrong?  After much head scratching and number crunching, it turns out the rocket itself is the problem!

 

Firstly, I checked the scale of the rocket. The width of the first stage (ignoring fins) is 10m in real life, 28 plates in Lego, so is to a scale of 1:110.6. The height of the rocket was 110.64m, the Lego model is 313 plates tall - a scale of 1:110.5.  So, it's not actually 1:110 as everyone says, but this just means the tower should be two plates short than it currently is.  I still have another 7-8 plates to explain.  So I decided to check each stage. 

The first stage was 42m, excluding the interstage ring, and is about 40 bricks tall in Lego, which works out at 1:110 or 1:111 scale. So far so good.  I can't find heights for the S-II or the S-IVb on the assembled rocket.  Official statistics always include the height of the engines, which don't count here because they are hidden inside the rocket (and I know have been changed because of limitations in the choice of Lego parts available).  The Luinar Module housing is the right height.  Then I looked at the Launch Escape Tower (LES). Including the white cover over the Command Module, the entires LES was 12.02m tall. At 1:110 scale, this is 11.3 bricks high.  The Lego model has it as 13.3 bricks tall.

So the rocket has the right overall height... but somewhere on the S-II and S-IVb, the rocket is two bricks too short, and they recover this height by making the LES a bit longer.  I suspect it's the black and white cone linking the S-II to the S-IVb, but as I can't get any proper height data I can't be sure.

So now I'm faced with a really fun problem... over 12 identical levels (ignoring the first three, I don't want to mess with the triangles at the bottom of the tower any more), how can I lose 9 plates of height?

 

Edit: Just confirmed, it is indeed the adaptor linking the S-II to the S-IVb.  I thought the S-II had been made taller to compensate, but I was wrong.  So, I either compress one layer in the middle to be half the height (which will look odd), shrink the whole tower one plate per layer (which leaves enough space to fit an extra deck on top to make top the correct height), or just place the crew access arm at an incorrect height (which will mess up the detailing later.  What should I do?

Edited March 12, 2018 by NathanR

[JGW3000]

5 hours ago, NathanR said:

What should I do

Obviously, get a beer!  Then shrink each level one plate, to maintain realistic looks. Great work, by the way 

Edited March 13, 2018 by JGW3000

[NathanR]

On 13/03/2018 at 2:08 AM, JGW3000 said:

Obviously, get a beer

There was a guy whose doctor said he should quit drinking and get a hobby.  He took half the doc's advice... He started putting ships in bottles...

After a break and coming to the model fresh, I looked again at the rocket.  The adaptor from S-II to S-IVb is actually correctly - there are two studs between the grille tiles and the slope bricks on the S-II, this is part of the conical adaptor.  The only error I can find is that the LES is two bricks two tall, probably due to including the height of the BPC (the cover over the command module) n the height of the escape rocket tower.  Shrinking the LES by two bricks puts the rocket back to 1:112 scale almost perfectly. 

Firstly, this means that the entire MLP is about two studs too big in each direction... For the sake of my sanity, I'm leaving it as it currently is.

Secondly, I can accept a tower height that is 1:112 scale, so I've adjusted the heights of a few levels to get the swing arms at the right heights. Everything is now looking pretty good:

I'm expecting a few problems from the position of the swing-arm axle, I'm not sure how easy it will be to link it up to a gear box in the top deck, but that's a problem for another day.

Over the weekend while eurobricks has been offline, I've focused on adding in the fuel pipes that ran up the side of the tower (see above). The transition from light grey to white is done using universal joints - not an ideal hinge, but the angle is about 10 degrees and click hinges won't fit.  I did consider using the "toggle joint", part 32126, but this isn't available in light grey (after appearing in the 10179 UCS Millennium Falcon, the part costs ~2 EUR each).  The final hinge in white is done using 32126 in white, which is mercifully available in two sets at the moment, but still expensive.

I guess the next problem is working out which level each pipe finishes on (and where it runs to after that), In haven't seen too many reference images and it's made worse by the fact that the real LUT had 11 pipes running up the side of the tower and I have only got room for 7.  But onwards and upwards... literally...

[Shiva]

Nice progress :)