Jump to content

1/144 Hindenburg Scratchbuild: I think I've made a terrible mistake.


Recommended Posts

I bought an FDM printer last year, specifically with an eye to doing some more airships in the future.  Well, I found some sweet-fool drawings of the Hindenburg a while back, and spent some time on it over the weekend.

 

aR5eGzg.jpg

 

The problem is, apart from punching the basic dimensions into Solidworks and having a general feel for the size, it never really sunk in just how big it would *actually* be in 1/144. I knew it would be about double the size of the Q-Class Zepp I'm building.  I knew it would be noticeably longer.  And I knew that as a result, it would be a more massive object.  But I’ve been too focused on modelling it in CAD to really step back and understand the full scope of the beast.

 

Just now, I plopped a model of my existing Zeppelin next to it to render a comparison and was like, ha ha ha, it really IS bi.... oh, crap... maybe I should get a tape measure.

 

hPa1RoP.jpg

 

It’s 5’ 7” long and 12” in diameter.  Gulp.  Having some serious second thoughts right now.  The thought of sanding and painting something that big is terrifying.  I have no idea where I'd even put it.  And I've got half a dozen once-in-a-lifetime-type builds in the queue as it is.  But dang... it would be pretty awesome.

 

Still, I'm pretty pleased with the way the CAD is progressing:

 

EAAqxVk.jpg

 

Edited by MoFo
Link to post
Share on other sites

I’m running into the same thing with my project. It looks cool on the screen but when I scale the thing out it’s huge. Like keep it in the garage huge.

 

I’ve heard good things about Solidworks. Have you ever used Blender?

 

Edited by Major Walt
Link to post
Share on other sites
On 4/3/2018 at 8:36 PM, Major Walt said:

I’m running into the same thing with my project. It looks cool on the screen but when I scale the thing out it’s huge. Like keep it in the garage huge.

 

I’ve heard good things about Solidworks. Have you ever used Blender?

 

 

It's interesting, because I usually have the opposite problem - when I'm designing parts, it's easy to forget how tiny they are.  I constantly have to remind myself that details which feel crude and clunky and terrible when you're zoomed in on the screen are actually only .1 mm thick and will be fine on the finished piece.  

 

Never used Blender.  I'd like to get into some of the sculptural software like Blender or ZBrush, but it's an entirely different skillset and approach to modelling.

Link to post
Share on other sites

Some updates:

 

I've revised the tailplanes and have split both them and the gondola off of the main envelope.  They'll be printed as separate parts and should(!) just tack on later.  The tailplanes were actually pretty much completely re-done - I wanted to tweak the geometry of the upper and side planes (which are all the same), and then created the lower tailplane (which is different).

 

1IVE76V.jpg

 

I'm still debating whether I should add surface details to the parts.  The skin is basically smooth, with just a subtle colour difference from the structure visible, and smooth parts will be easier to print and finish, but because they're such large pieces (each tailplane is almost a foot long) it needs some detail to break up the large expanse.

 

674de58171605c51589c4ef6c82c27f5.jpg

 

(...and looking at this pic, I see a few more things I need to tweak!)

 

 

I've also started roughing out the engine gondolas.  These have clamshell shutters at the front, to regulate airflow to the radiators.  I'm *really* tempted to engineer some kind of functional solution for that...

 

3iEjXVB.jpg

 

And lastly, now that the gondola is split off, I can begin detailing the interior.  Complete with easter egg.

 

4LXNq4A.jpg

Link to post
Share on other sites

Started adding windows; revised the tail planes yet again:

 

7huGuJj.jpg

 

Full view along the bottom, showing both updates.  It's quite busy underneath, actually.  There are also a number of cargo hatches, but I'm probably not going to add those to the CAD - FDM printers don't do fine recessed details well, so it will be easier to either scribe the doors, or add some sections of decal film as 'raised' panels.

 

wE9bX8s.jpg

 

And a view with the engine pods in place:

 

7POM6qA.jpg

 

Next it needs reinforcing strips along the envelope, and some subtle scalloping to the facets.  And locating points for the vents up top.

Link to post
Share on other sites
On 4/14/2018 at 11:31 PM, janman said:

Something very different

 

You mean you're not sick of seeing 6' long airships at all the model shows you go to?  :woot.gif:

 

To give a sense of scale, here are a trio of test prints I ran, all at 1/1000.

 

AuCkHUj.jpg

 

On the bottom is a WWI-era Q-class Zeppelin - the one that I scratchbuilt previously - which was already massive in 1/144.

Above that is an R-class WWI Zeppelin.  These were the first of the 'Super Zeppelins', larger and more aerodynamic so they could carry a heavier payload (nearly as much as a B-52), higher, in an attempt to avoid British air defences.

On top is the Hindenburg.  It almost looks like a different scale.

 

Extrapolating that to the 1/144 version, I've also run test prints of the gondola and engine pod.  They look tiny in the CAD renders, but will be quite substantial on the actual model:

 

aqlirxR.jpg

 

Again, these are being produced on an FDM printer - the simplest, cheapest, but crudest of all 3D printing technologies.  When you read about Makerbot, or the Dremel Idea Builder, or just general 'hobbyist' 3D printing, it is an FDM printer.  They work sort of like a cross between a computer-controlled Etch-a-Sketch and a glorified hot glue gun, using motors to precisely extrude layers of plastic on top of each other, to create a finished product.  The up side is that they're relatively inexpensive to build and run, relatively simple to operate, and relatively fast to print, but the down side is that they're poor at reproducing really fine details, so the're ill-suited for most modelling purposes.  An FDM printer will work well for a Zeppelin because it's basically a big, smooth shape; if it was full of panel lines and fine surface details, they would look terrible.  I'll be using this printer to produce the Zeppelin's envelope - the balloon itself - but detail parts like the gondola and engine pods will be run on a different kind of printer than can render finer detail.

 

Vto5MIR.jpg

These parts were run at .04mm layer height, which is *extremely* fine for an FDM printer (it's about half the thickness of a sheet of paper), and has generally resulted in nice, smooth layers.  However, you can still clearly see the stepped-look on shallow angled places like the bottom of the gondola.  

 

xdf0jpF.jpg

However steeper, more vertical angles look pretty good.  Fortunately, most of the final prints will be steep angles, like this.  A quick sanding and a heavy coat of primer and the parts should be smooth.  This is ultimately why I bought the printer in the first place - it was a ton of work filling and sanding my previous Zeppelin, and the parts weren't as crisp as I would have liked.  If I can get a sharper, more precise balloon with less surface prep, I'll be thrilled.

 

rl6Q7Zz.jpg

I was actually pleasantly surprised with the fidelity on the engine pod - you can see the slightly recessed clamshell doors on the front.  Not perfect, and not as detailed as I'd like, but if I was going for a simpler display piece rather than tons of internal detail, it would actually be usable.

 

EN5h6x7.jpg

Lastly, something that interests a lot of people...  FDM prints are rarely *solid*.  There are a bunch of reasons for this - it's much faster since you're printing far fewer lines with each layer, and it saves an enormous amount of material; plus the finished parts are still more than strong enough.  So the software used to generate the code that drives the actual printer - it converts your 3D object into individual layers and plots the movements and settings for each layer - can also 'scoop out' the inside of each file and generate a honeycomb-like 'infill' to maintain rigidity.  Hence, this grid-like pattern visible on the bottom (or, top) of the gondola.  The printer itself isn't actually all that complex or sophisticated, it's the software that runs it that's smart!

 

 

 

 

Link to post
Share on other sites
21 hours ago, MoFo said:

 

 

AuCkHUj.jpg

 

 

 

 

 

The bottom test print seems to be average for that kind of stuff at 7", the second one is big at 8". And the last one is monstrous, just too big in girth and length. Oh wait, you telling me this wasn't some NSFW forum I entered by mistake? ;)

Link to post
Share on other sites

Side note about the fact that most FDM prints aren't solid...  I just imported one of the envelope sections into the print software to see how long it would print if it was solid.

 

209 days, 10 hours and 55 minutes.  Yikes!  If I did the whole thing as 100% solid prints, it'd take 748 days and weigh 90kg (about 200lbs).  Yet another reason why 3D printers aren't the magic machines most people think they are.

Link to post
Share on other sites
  • 1 month later...

EY9lYYg.jpg

 

I made a thing!

 

Started printing with the (top) rudder/tailplane.  It's... pretty massive.

 

3o7DuxU.jpg

 

It's actually a 12-piece assembly.  Because the rudder hinges hang way out into thin air, they would require support material to print, which would have given a poor finish. 

 

cijQmTY.jpg

 

To get around this, I split them off to print separately, and included locating holes for hypo-tubing pins which would let me accurately and securely join the parts.  Strength being important, because the rudder is poseable - it's printed with a tube inside, so I can feed a length of hypo tubing through the rudder, capturing the hinges.  Which illustrates yet again that, while many people think of 3D printing as a magic bullet, click-print-and-get-perfect-parts, the reality is far away from that ideal...

 

 

So just how big is it?  Well it's hard to get a feeling for how *massive* it is in photos - it's more than an inch thick, but here are a few comparisons.

 

1/48 F-15 tailplane:

 iIkUXFZ.jpg

 

1/48 -109 rudder:

EvEwf7W.jpg

 

(the Hindenburg's rudder would be more than a foot tall in 1/48!  kinda tempted to print one just for giggles :woot.gif:)

 

and for something in-scale, a 1/144 P-51:

89iHOpu.jpg

 

The Mustang fuselage would *easily* fit inside the rudder, let alone the main tailplane.

 

 

Link to post
Share on other sites

I put together paper kits of the Hindenburg and Graf Zeppelin about 40 years ago. They was 1:200 and they were pretty damn big. My brother was an airship fan so I built them for him. I built a 747-400 in 1:200 so he'd have something for scale. Also, the old Hawk vac-formed Zep; if you built it as the Los Angeles was 1:200.

paper kits 1c.jpg

paper kits 2c.jpg

Edited by Grey Ghost 531
Link to post
Share on other sites
  • 2 years later...

Spurred in part by the LED thread elsewhere, I figured I should post an update (though it feels like screaming into the void...).

 

I've been picking away at this, slowly but surely.  Last summer was spent on the electronics - apparently it's not enough of a challenge to design an build a 6' long model kit, I also had to add lights, sound and motion.  I believe they call that 'mission creep'.  😞 

 

r2o7Kbr.jpg 

 

So...  a basic schematic of the electronics inside. 

 

tcYWKL9.jpg 

 

And focusing on the lights and motors.  It's been simplified to not show every single LED and switch - I'm running each section on its own switch: corridoors, spotlight, nav lights/beacons, passenger deck, crew deck, etc.    Each individual engine will also function, working off its own switch, because... why not.  😄  Lights are a mix of sources: LED strip lights for the corridoor (which will be roofed in and get a few exra details), an array of 3mm LED bulbs for the spotlight, discrete SMD LEDs for the position lights and passenger areas, etc.  All are warm white, to be closer to incandescent.

 

The motors are 3mm diameter pager motors - tiny, but they'll slide in to the 1/144 Maybachs in each gondola - run through a potentiometer so the throttle can be 'tweaked'.  Mostly so I can adjust it to something suitable rather than just rely on the motors OOB, but I'm sure some viewers will also want to play with them. The motors and nav lights on the gondolas posed a big challenge: how to come up with a secure but fine electrical connection so I can plug the pre-wired gondolas into the pre-wired balloon during final assembly, after everything is painted and ready?  My solution was fine brass rod and tubing - .2mm rod (which will simulate plumbing in the gondola) was soldered to the LEDs and motors, and will slot into .4mm brass tubing, which was soldered to my wiring harness and will serve as the struts supporting the gondola.  The rod has a slight ( shaped bend to ensure a snug connection inside the tubing.  Lastly, a 3D printed block inside the balloon has holes bored at the various correct angles to ensure everything slots correctly into place (I hope...)

 

The audio side was a little more straightforward in terms of planning the layout, I just had to figure out the implementation.

 

Ovj7DAY.jpg  

 

The heart of the system is that little PCB in the middle - the DFPlayer Mini.  It's basically a tiny (1" square), cheap ($2) MP3 player that you can program to function in a bunch of different ways.  I went for the simplest option: each button runs through a resistor, and each resistor value is mapped to a specific function in the player; when it detects the signal from a button press, the it checks the voltage (determined by the resistor value) and does that particular function.  In this case, it's to play a specific audio track (yes, including that one...).  Or, TL;DR, you press a button and it plays a sound.  🙂  And again, the speaker's output runs through a potentiometer for volume control - which actually WILL be useful.

 

...and it's all been set up to run on a 5V power supply, so either a simple phone charger + USB cable, or a 5V wall transformer, or even a powerbank, if I need the portability (if it ever goes to a contest)

 

Next up: the actual wiring (it's ugly, but it works!)

Link to post
Share on other sites

So, the 'for real' electronics, now, starting with the spotlight, 'cause it looks cool.  🙂  I've seen diagrams that show a simple, single spot style, and some that are more like an 'array'.  Not sure which one I'll end up using in the end, but I'm thoroughly pleased with how the array style turned out - the LEDs are all a nice, snug fit inside, all angled and aimed at the same point:

 

FdKewCh.jpg


So the heart of the electronics is a chunk of protoboard PCB, which I've built the circuits off of for strength and stability - it's basically your standard fibreglass circuit board material, with standard spaced holes for attaching components.  The DFPlayer is to the left (well, centre in the pic, really); power connects on the left side of the board; individual components/circuits plug in via pin headers for tidyness, and the aforementioned row of resistors is to the right. 

vsPDDJ2.jpg

 

...and with the buttons in place.  Each of these triggers a different audio file - a couple of tracks with ambient noise, a couple of period news reports, of the Hindenburg crash, a couple of pre-crash recordings, and a couple of interviews with survivors.

 

K2B5Pjs.jpg 

 

The switches for the individual lights/motors get installed via pin-headers on the obverse of this board, but that's harder to photograph because they're also attached to:

 

89YG07L.jpg 

 

The support posts!  Okay, difficult to get excited about, but this was one of the weirdly challenging problems to solve.  I wanted the supports to look relatively fine and delicate (no brass plinths or pedestals), be fairly tall, and to house all the wires for a clean finished appearance.  So, okay, some sort of rigid tube... like carbon fibre.  And the more I worked on it, the more I realized that it would be a really, really good idea if I made them removeable, so I could troubleshoot, and didn't have to have the model permanently attached to the base (kinda obvious, really...)  So the challenge became, how do I get a strong and secure, yet easy to remove attachment between the posts and the guts, and how do I make it small enough to not have a gaping hole in the bottom of the balloon?  I thought of a number of possibilities, from micro HDMI connectors (too big and too sticky) to pogo pins (too finnicky and delicate), and ended up deciding on a custom shaped block of pin headers (this is what actually started me on the pin headers as a connection).  I glued 3 lengths of 3 pins to a section of PCB, epoxied it into a length of CF tubing, then ground it to shape so that it slots into a larger tube.

 

ZLf78mS.jpg 

 

Something like this.  The 9 pins give a nice, secure fit that slots in fairly easily and comes out with a slight tug, but won't pop out accidentally.  And there's zero chance of crossing connections and shorting things out.  The other half of the equation was just as important, both providing a secure, precise locating hole (the one in my other Zepp is a little imprecise) AND being able to support the weight of the finished model on a handful of electrical connections.

 

NdHvIop.jpg

 

My solution here was to print out a support brace (black) to hold a length of CF tubing in place on my aluminum spine.  Another printed part (white) spans the gap between this locating tube and the envelope, which helps bulk it up and add stiffness.  My CF support tube slots through this locating tube for a nice, precise, 90` fit.  I then drilled and filed through the aluminum spine so that the support post can connect with the other half of the header block, which was soldered onto a strip of proto board (for rigidity), and then bolted down with some large diameter washers, so that it's held firmly in place with no way to bow, flex or shift.  Lastly, I glopped a bunch of hot melt glue on the wiring around this internal connection to act as strain relief, and some additional insurance against things popping apart in the future - I really don't want to have to hack this open to repair a simple solder joint!

 

CaFFCkb.jpg 

 

Net result being, a simple connection between the support posts and internals, giving me exactly the kind of fit I'd wanted.

 

And here's another view of the assembly.  Funny how much engineering it takes to make 'simple':

 

RbuVcB2.jpg 

 

Apart from that, it was a relatively simple matter of measuring out lengths and distances, then wiring up the various components.  It looks rather messy inside: 

 

OzK1EMA.jpg 

 

but hopefully this gives a feeling for how it all works.  And might help someone else planning on wiring up their own model.

 

 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...