Firefox MIG-31 movie aircraft

Hahahahahaha! I'm in Perth. No, the other one. I wish I was as close to you guys as you are to each other! Now you're gonna find out you're related. LOL. Pretty sure I played the Firefox arcade game too but my memory sucks. I do recall seeing the machine but that's all.

“it's so subtle I thought it was camera lens bend.”

Yep definitely real. It’s one of the problems with the design, like the Jawa crawler, the Nostromo and various other 70s/80s movie models - there are a ton of unexpected odd shapes designed to befuddle the eye, haha.

“Been reading the thread about the studio scale Narcissus you are involved with if I have that right, haven't got to the end yet!”

Haha, neither have we. There’s big news coming up fairly soon though, I believe.

“Printer is RepRapPro Ormerod 2, seemed to be the case if you didn't mind having to build it yourself from a kit and the fettling/swearing involved setting it up, and a few cheap mods afterwards, you can print as well as something 5 times the price.”

I had a brief look at the site yesterday (didn’t have much time, we’ve got an extra floor going on the house starting today – builders erecting scaffold right now – eek!) It does look interesting for the price. I’ve built one printer from a kit (Makergear M2) and it’s for sure the way to go IMO; these are not machines you can treat as if they were fridges or TVs. Building a kit gives you a great deal more familiarity with the stuff you’re gonna need to get familiar with sooner or later anyway, extended swears vocabulary included. :lol

“Also I know if anything goes wrong it's easy and inexpensive to fix myself. Plus the build size is pretty big, in fact I am going to mod it so I can print up to around 40cm high, currently about 19cm max.”

What’s the standard build volume? I could be interested in something larger too, looking down the track. The M2 is a good machine but only has 20x20x25cm, decidedly not the biggest on the market. Forces me to be smart about parts breakdowns though.

“custom firmware which has added some really useful print controls so you can change the print speed and extrusion while printing. I think it also runs smoother and tracks corners more accurately now.”

Now that I would love. You can do it via G-code of course but that presumes you know in advance that you’re going to need to. A proper (not G-code) pause function would also be pure gold.

“I have only been printing about 2 months so am still learning, things like silver filament looks cool, but is a bitch to get to stick to the bed, probably whatever colouring makes it silver also makes it set slower and behave more elastic than other colours.”

Yeah, they all have their own quirks. I’m still working my way through different colours; the consensus seems to be that clear PLA is easiest, with various other colours getting trickier and ABS a good deal tougher. (Are you using ABS btw? You’ll want to be unless you plan to mould and cast this? PLA has a lifespan, it’s biodegradable in the medium-term timeframe. Much easier to print though.)

Do you have a heated bed, and are you using blue tape or kapton? Hair spray? I stuck it out with nothing but the heated glass for months, which was stupid. I get much better results on kaptop, with hair spray – even for PLA. The other big mistake I made was ignoring everyone’s advice about measuring your filament. Reel to reel it really, really, really varies in thickness. I thought I could just eyeball it and of course I was dead wrong.

Definitely get a smaller nozzle. Mine’s .35mm so I can print details down to .7mm. I’m mostly using Cura at present, it does 90% of what I need and the visual interface is golden. I am really interested in your remarks about Slic3r as I gave up on it last year during their “we forgot how to generate support material that isn’t like rock” period. Might have to take another look.
 
“The seam is around halfway through the canards, with the seamline being the base of the print for each piece.”

Ah right. Makes sense, precludes being able to sweep them, though.

“Beefing them up a bit will work, even it they require a bit of smoothing by hand to the edges. Could possibly leave hollows to push something like 1mm round steel though to give more stiffness?”

Exactly, there’s no hardship sanding an edge. Steel or brass rod for sure - you’d want to leave a bit of a tolerance – make the holes 1.2mm or something like that – but it would work. Personally I would print them as separate pieces with a matching hole in the fuselage, so you cut the rod with 5 or 10mm excess sticking out of the wing and it takes care of the alignment for you. Or a slot in the fuse, so that you can do the sweep-back thing if you want to, haha. Treat the fins and wingtips the same way IMO.

“Haven't mentioned this yet but planning on having an aluminum subframe for the 1/24”

Good plan. :thumbsup

“.... aren't we all “

LOL!

“I wish the WB pics were mine (I would've shot hundreds more!)

RIGHT there with you.

“but apparently its part of a VIP tour where photography is not permitted so those are the only proof I've seen.”

Arrrgh, noooo, why is it always the way?!? I know Will B had some good pics of one of the minis; he only ever published a handful though, they’re the ones with the 2001 timestamp. Is it the same model?

“As far as the miniature is concerned I'm amazed it's still in such good condition although Jon you have a very keen eye for all the idiosyncrasies which is really encouraging-“

Seconded! :thumbsup :thumbsup
 
The 3D printer did nearly go out the window a couple of times in the first week with many uses of the F-word in conjunction with the C-word and the W-word, last one rhymes with banker.

Most of them were down to the steep learning curve, two were down to production problems, the cable to the heated bed failed literally inside the crimp joint after initial testing, handy to have an electronics background with things like that and a test meter to hand.
Then I thought I had a extruder motor or mainboard failure, turns out they had failed to mention that with a firmware update (as a safeguard, AAARRGGHHH!) the extruder motor will not run unless the hot end is at at least 180 degrees. It seem obvious to me now but people will think that something has failed and waste hours like I did, even the tech support people didn't know! Sorry just needed to vent that ;)

Build size is quoted at 200x200x200mm you can squeeze 210 on the y-axis, but the z is actually more like 190 max, realised I could gain 4mm by adjusting the bed lower but as I got it pretty much dead on level by hand/eye it's not worth it. As the construction is quite simple and elegant all you need to be able to print higher is a piece of 20x40 ali extrusion and think it's 12mm chromed stainless steel round bar, possibly a better ali reinforcement bracket that hold the Z to the rest of the printer, but that's easy to make with bit of 5mm ali and a 3 holes with a drill press.

Bed is heated and they recommend by default you kapton tape it, which I agree with as I tried printing straight to the glass top and it stuck but probably half the force required to get it off, ok for small bits, but wouldn't be reliable for large parts. I can use ABS, but the stink it gives off is nasty and where I have the printer at the moment not really workable, but I can move it to a bigger workshop where the fumes wouldn't be that big an issue, but would then maybe need an overall cover to keep the temp up over the whole part to avoid warping.

However I have done some research into PLA as I thought biodegradable meant it would literally rot away with time, but I think they overplay the biodegradability of the material. I can't find the blog page now, but a guy who started early in 3D printing left some in a compost pile for 4 years and nothing happened to it. Seems to be unless you get it near to a constant long term 60 degrees C nothing actually happens. I know it gets hot in Oz, but I think you would die before the model! Also if you seal the surface with paint it further stabilises it.

There is also another good alternative from makesolid.com called PET+, does cost quite a bit with shipping as it is only sold by them in the US as far as I know. There is a testing vid here https://www.youtube.com/watch?v=qpFZEl-cXv4 no affiliation btw, just found out about it from a tech mailshot.

You can pause mid print no problem, I have had random odd glitches printing simple things where it seems to add the next x,y change to the current move, but seems ok with anything complex, or maybe the glitch is so small you they don't show.

Yes on the filament variation, I thought the greater unevenness on the test print was the filament, possibly is some of it. Then realised my X-axis belt was a bit slack due to a minor accident caused by the endstop sensor wiring coming loose which resulted in a fairly horrible grinding sound as it tried to move the head into the Z axis column, and forgot to check it afterwards.

I had some medium priced black PLA that printed really well, there did seem to be some variation in the thickness, but the printed results look good even though the black shows any surface unevenness more.

**EDIT**

Found another similar material produced by Colorfabb http://colorfabb.com/xt-copolyester might get a sample to try as they have an EU store and pound to euro is a good buy at the moment.
Yes the supports were not great in Slic3r but I think they have improved them, having said that a number of people seem to say build the support into the model so you know what you are going to end up with. I did this with part of a figure print as neither Slic3r or Cura would add support material to places that turned out needed it.
 
Last edited:
Regarding the canards, I think it must have been another model they used since I can't think of a way to have them in the normal position without an ugly gap where the back edge would need to go when they sweep back, and conversely infront when they are swept back, in fact you can see the gap infront in this one

https://drive.google.com/drive/folders/0Bz-XCi-Q52kXeE1ONlFLUHhjUTQ?usp=sharing > Nose Canards detail

Using the steel rod method for mounting you could have both normal and swept interchangeable, having the rods meet and interlock inside the model so they stay level.

I will scan the whole Cinefex mag pages when I have time as they cover some details of the models construction. I think the reason the model has survived is that it is mostly resin so pretty damn tough, there is a pic of the guys working on the moulds.

Also added a couple more pics to the Studio Model folder here:

https://drive.google.com/drive/folders/0Bz-XCi-Q52kXeE1ONlFLUHhjUTQ?usp=sharing
 
Last edited:
WOW. Just noticed this thread. Im working on a 1/18 scale Firefox. Its ready to print as money allows. Great to see another getting done. Ive been dying to do this project since i finished my X jet. Cant wait to see yours.


Al
 
A friend of mine whos into Firefox and has been doing research for many years had it drawn up. And we are going to print it out and mold and cast it. Its scaled to 1/18 for now. We may make a larger version after, we both like large scale models
 
“The 3D printer did nearly go out the window a couple of times in the first week”

I know that feeling. I still get weird mystery failures that are sods to track down and can’t print adequately with ABS; the stuff just will not stay down. I’ve not yet tried blue tape – it should work on Kapton, but even if I get adhesion, the print head itself clearly isn’t happy. The stuff is just that bit too viscous at any temperature; it jams all the time and I end up having to strip the head. I’ve got an E3D aftermarket head but haven’t installed it yet. Got to try that.

“extruder motor will not run unless the hot end is at at least 180 degrees.”

Yeah, that’s fairly universal I think. You don’t want to be trying to drive a 1.75mm solid through a 0.35mm nozzle. Re the bed, you’ll need to re-level it from time to time. Automatic bed levelling is a nice feature to have that has been coming in with some newer machines, but I wouldn’t pay a large premium for it. The huge number of quite small build-volume machines is something I don’t get; you have to ask why they don’t provide for enlarging Reprap types at least, as part of the design. I’d get frustrated with anything less than a 250mm long bed. (300mm would be nice, print a helmet with that. Been eyeing off Taz machines lately.)

“However I have done some research into PLA as I thought biodegradable meant it would literally rot away with time”

Well that’s a little reassuring, but my understanding was that it’s something like ten years ish, and it does weaken/rot/something or other. Ten years would have seemed like forever once, but decades seem to fly by these days…
We get up to about 45 deg C here in summer but there have been days over 50 in some places, fortunately Baghdad weather is rare and yeah, don’t think I’d survive it. :lol And yeah there’s quite a few new filaments coming out, bit by bit. (Laybrick is a lot of fun for printing buildings, for example.)

“You can pause mid print no problem”

Huh? In Cura, or some other host? Belts are a problem if you don’t keep them good and tight. Mine have started to leak fibreglass so I suppose I’d better find some spares. Re PLA colours I like blue and green, for some reason they’ve worked best for me out of all the coloured stuff. I’m using some red PLA right now that’s way undersized at 1.3mm. Upping the feed rate to a value which worked took a crazy amount of experimenting; in the end I’ve found a feed rate thickness setting of 1.72mm only is what works. Any more than that is far too much. Surprising. The red stuff also doesn’t produce as nice a finish, for some reason.

“Yes the supports were not great in Slic3r but I think they have improved them, having said that a number of people seem to say build the support into the model so you know what you are going to end up with.”

It’s not a bad way to go but I go with Cura ‘line’ support 90% of the time; it’s effective and easy to remove. I think about print orientation a lot and design overhangs on an angle wherever possible, avoiding the need for support.

Re canards, it’s been an age since I’ve watched the film, but don’t we see them move? Sliding doors would be my guess. But yeah - steel rods for both versions would keep it nice and simple.

IIRC one of the flying models was converted into a stage model. It might have been epoxy-glass construction for light weight. Not sure about the others, they might be heavier polyester-glass.

OMG, 1/18. It’s getting like buses around here. You wait for hours, then three come past together. :lol
 
"I know that feeling. I still get weird mystery failures that are sods to track down and can’t print adequately with ABS; the stuff just will not stay down."

ABS printing does seem a bit of a nightmare that's why I was looking at alternatives, the one thing people talk about are Elmers glue stick, basically the glue stick you used as a kid to stick paper together. Also something I saw yesterday mentioned PEI sheet which I had never heard of seems to to work very well http://reprap.org/wiki/PEI_build_surface

"You can pause mid print no problem”

"Huh? In Cura, or some other host?"

You can pause with the standard web control interface, the custom firmware fork someone called dc42 did adds the speed and extrusion control

"The huge number of quite small build-volume machines is something I don’t get; you have to ask why they don’t provide for enlarging Reprap types at least, as part of the design. I’d get frustrated with anything less than a 250mm long bed. (300mm would be nice, print a helmet with that. Been eyeing off Taz machines lately.)"

Yeah, I went to an expo about 2 years ago and all the print machines had tiny build volumes, and were about £1000, crazy. I mentioned increasing the Z axis on mine, thinking about it it wouldn't be that hard to increase the Y also. I was tempted by a brand new unused Taz machine on ebay UK a little while ago, it sold for about 1/2 retail, but that was still about twice what I paid for mine.

"Re canards, it’s been an age since I’ve watched the film, but don’t we see them move?"

One place but it's a pretty much a top down shot which that frame is part of, obviously they tried to hide it.

"IIRC one of the flying models was converted into a stage model. It might have been epoxy-glass construction for light weight. Not sure about the others, they might be heavier polyester-glass."

They aren't specific in the Cinefex article, it does say they were trying to do carbon fibre for the RC ones but was too tricky in a short time so used straight epoxy is the wording. The RC ones flew well in testing but they didn't end up using them as where they were shooting the ground was to rough and the wind too severe.
"OMG, 1/18. It’s getting like buses around here. You wait for hours, then three come past together. :lol"

I know but things seem to happen that way, all or nothin', least we are working on different sizes, lol
 
Last edited:
First want to say thanks to Will who started the original Firefox project on here as I found out the 2D plans I started with were actually drawn up by him. Wouldn't have got this far without his help.

Been working on the rear section for the past few evenings, needs some smoothing and detailing but think I have got the complex end of fuselage section right.

Firefox rear 19-6-2015.jpg

Going to add some panel lines and do a test print of the engine section over the weekend to see how it works out.

Bit ahead of myself but started looking at colours. The original paint job is going to be near impossible to replicate, take the often asked what colour is the Millennium Falcon (hope saying that doesn't bring anyone out in boils) and multiply it by about 10. There was a basecoat upon which many layers of shading and lacquers were added. So I started looking for a base colour which covered the majority of the paint scheme. Spent a few hours looking at car colours and for some reason Toyota/Lexus seem to have a wide range of shades similar to what I was looking for. Found a handy website with multiple pictures of the same car colour in different lighting conditions, so armed with a few possibilities visited my local car paint supplier.
Took a number of pics in daylight of about 20 different colours in the same colour palette, noted my phone showed them a bit over blue to real life and came home. On uploading the photos was pleased to see they appeared less blue on screen so it was going to be ballpark right. Quickly realised that what I thought was my starting point colour was way too blue and light, but luckily in a random last 4 selection hit one that is very close.

Firefox colour.jpg

Second from left highlighted in black. If you base the colour on the full scale one or the ice flow model you will probably think I am loosing it, but the 1/12 hero is pretty close to that overall. It will need darker almost slate grey shades and silver/blue, in fact the colour to the right of the highlighted colour in the pic would probably work well for the lighter shades in misted over the main colour.

Jon
 
Last edited:
Superb Jon - Keep it coming :)
I also wasn't aware Will was responsible for the renders - he really did his homework
The colour choice is fascinating - as it appears a different shade in most every shot I guess its subjective but its great to have a starting point.

Did you see these? found them while searching for thinkinrussian.org
(which does seem to have disappeared :(

Mig31Four_3.jpgMig31Eight_3.jpg

Looking forward to seeing these test prints..!!
 
Last edited:
Jon I'm right there with you on colour. For me it's the hero miniatures every time; that's what gets the bulk of screen time and that enigmatic bluish silver-grey has always been fascinating to me. I think you've done a great job finding a good colour match there and can't wait to see it on a model. It's rare to see a fan model built in anything other than black and that just doesn't work, for mine! Others will disagree of course but I'm delighted you're going this way. :)
 
Held off the test printing as I realised that there were a few things off, and decided it wasn't worth doing before all the structural parts were done. So many small tweaks, you realise how accurate you can be when you are moving things only around an inch in real scale terms on a 63 feet long plane!

Anyway, the inset sections around the exhausts I am calling done after many hours tweaking, mainly because there was never any real detail there as the engines are always lit on the hero so there didn't need to be as it would get obscured.
Had a hard time getting it right as I had made a couple of assumptions about angles of certain pieces which I realised were wrong, once those were corrected it was a eureka moment as it started matching the reference pics, had a celebratory cup of java over that one ;)

Took a bit of artistic license with the exhaust manifolds as they are kind of plain. So found reference pics of the SR-71 Blackbird engine internals and detailed them in a similar fashion, after all it's the only plane that's got anywhere near the Firefoxs Mach 6...

Tail planes are mostly there, they were slightly undersized and the tilt angle was wrong, now they match the hero shots. Built the defence pods, undecided if the ones by the engine should be blended in or left as is. That section comes out with the engine internals on the hero and I think there may have been two sets that were a bit different - one lit and unlit as some of the shots just don't seem to show the same thing.

Have done a bit of test printing of the exhaust manifold sections which were a pain as the edges get so thin, needs a bit more adjustment on print settings for that.

So current state
Firefox rear 27-6-2015.jpg

and a close up with lighting/shadows
Firefox rear 27-6-2015 engine detail.jpg

More tomorrow.
 
Last edited:
It's great isn't it, when your project starts matching the ref in multiple self-reinforcing ways and you know you've got it right, haha.

Looks good! :thumbsup
 
Had some 3D printing problems, kept getting issues, partly down to cheap filament, and partly down to learning optimal settings for what are complex shapes. Kept getting problems, and after starting the same print 3 times, realised I was going through - insanity is doing the same thing but expecting a different outcome situation!

Realised I was trying to print probably too fast, and with not enough infill, trying to save time on prints, but probably wasted more time in the process. I was trying to get the print time down as most of the complete engine section was quoted at about 12 hours, which sounds a long time but not a big deal if you leave it reliably printing overnight, and it is quite a big portion of the model.

I have pinned down what the main problems are including a banding effect I was getting on the engine exhaust, which seemed to be down to certain layers getting infill rather than just a perimeter when the thickness went from somewhere between 2-3mm, turning off infill and adding more perimeters fixed it, but have realised it's down to a default infill overlap to perimeter setting of 15% causing a bulge.

I had to stop the rear section print shown in the pics about 3/4 through, again know why it happened and how to stop it, but though what the hell and kicked of the rest before going to bed last night, sure enough it screwed up in the same way, see the insanity quote above!

Anyway nearly threw it in the bin this morning, but coming back to it this evening quickly glued it together and realised it didn't look that bad, al least it looks right, which was why I was doing it. There are shots of the exhaust version 1 and 2, version 2 I split it into two parts and I realised I was going to have printing problems otherwise. To give a sense of scale the entire engine section in this one is about 200mm wide.

Firefox rear test 03.jpg

Have also tested producing panel lines, the test slices shown are about 8mm high and the panel "cut" lines about 0.5mm wide, you can see the 0.2mm layer resolution in the second one. I don't think this will reliably work at anything less than 1/24th scale, maybe 1/32.

All pics in this folder https://drive.google.com/drive/folders/0Bz-XCi-Q52kXR2QwR2h6SEQ1cjA

Have some new material quality filament on it's was which from what I have read will help with some of the issues I am having, and I need to mod the printer to stop the show stopping issues. What happens is you get a small curl or lift on a perimeter edge where there isn't complete infill, this seems to be a trait of using PLA filament as it stays soft for some time, good for layer bonding, but bad for stuff like this. When that happens high enough the Z axis sensor by the head runs into it and either a) knocks the part off the bed, or b) if it's stuck well enough, as in this case case it causes the drive belt to skip so everything after that is offset or in other words screwed up.
Might not happen with the new type of material as from what I have read although you print at a higher temp is sets (glasses) faster, really hope so as I have 2.2kg coming to make to price reasonable.
 
Last edited:
Your 3D print experience sounds a LOT like mine, haha. Did any of your problems also stem from filament diameter consistency issues? I had a lot of trouble with that before I realised the experts really meant it when they said it makes a huge difference, aheh. I've started paying a lot more attention on that front, and slowed my prints down. I go with about 22% infill for most jobs. I find you can usually leave a printer going overnight without supervision and expect an OK result provided the reel isn't badly wound (ugh, that's cost me a few print jobs) and you have good initial bed adhesion (which is why I almost always use rafts). And, obviously, that all your other settings are good for the reel you're using.

Try playing with your settings for lifting the head - there should be some Z axis settings that could help with that curl problem. Also, your bed fan - does it have a shroud forcing the air around the nozzle/on to the part? If not there's bound to be a mod part on Thingiverse. One of the first mods I made and is great for *everything* to do with PLA. Oh yeah - also, check your "time per layer" setting. If it is too fast, especially on a small layer (i.e. especially if you don't have infill - sounds like the case here) - the print is retaining too much heat and you get warps etc. Um, and playing with bed fan speed too.

And nozzle temperature haha. Oh god these things are a nightmare of interrelated settings factors ain't they? :D

I use 10% infill overlap most of the time. You shouldn't get bulges even with 15% so I'd check that filament thickness if you haven't already - if the system thinks it is eating less plastic than it really is, that's when I get bulges.

- - - Updated - - -

PS you pretty much have to adjust your settings for each new roll of PLA. When you find something that works well stop experimenting and just get on with it, is my advice. Don't be me, I keep trying different colours and makes, can't help myself.
 
Small update, worked on the tricky geometry of the wing to fuselage shaping and the intake inlet cut outs which flare outwards so the narrow tips nearest the front of the plane are as wide as the front fuselage widest point, something people seem to miss.

The wing body join looks a bit obvious at the moment as they are two separate pieces of geometry so there is no visual smoothing being added by the software.

Done a few other things, found some subtle shaping and angles on the front of the underside intake/gun port, added the rear underside fins and defence pods. Keep checking by using the match photo feature in SketchUp where you can adjust the perspective to get an almost perfect match to any of my reference pics.
 

Attachments

  • Firefox 15-07-15.jpg
    Firefox 15-07-15.jpg
    107 KB · Views: 1,122
You are really, really nailing this thing down! :thumbsup

It's never really occurred to me before but this thing is actually area-ruled. I wonder if that was intentional or a complete accident? Pretty amusing and probably a one-off for fictional aircraft!

You should post some comparos against the refs.
 
I have had to use a combination of Photoshop and Coreldraw to create this as inside SketchUp unless you can zoom in and out it it's really hard to see what's going on!

Realised I saved it after I was trying to line up what I have now realised as slightly sagging wing-tips, so my model is tilted towards the camera a tiny bit more than the hero orientation, like 1 degree or something, but it doesn't affect the horizontal dimensions. I will do some more, but they take time.

This is one of the best pics to match to as it looks like it was taken at a distance with zoom, which I have found reduces the distortion artefacts somewhat.
Firefox hero comparision.jpg

Forgot to post this, using this pic from Cinefex, I scaled and overlaid dimensions in Coreldraw, gives a good scale reference as the mount pipe must be 1.5 inches diameter, as far as I can tell that's what they used, certainly from that sizing it scales the hero correctly.


Firefox hero mount scale.jpg

Realised this is probably the shot where they are moving the canards back, sorry about the male nudity, must have been hot in there!
 
Niiiiiice! Thoughts:

* that nacelle section really isn't as tall as I've always thought
* on the nacelle outer side walls, is there a facet demarcation at the level of the wing, which carries aft?
* or is that only a panel line and change in the line of the sweep of the trailing edge of the nacelle?
* do you need to move the leading edge of your lower main intakes forward a little, possibly?
* I think you need more of a 'step' back to the main nacelle at the lower edge of the nozzle shrouds.

Overall, mindblowingly close to perfect.
 

Your message may be considered spam for the following reasons:

If you wish to reply despite these issues, check the box below before replying.
Be aware that malicious compliance may result in more severe penalties.
Back
Top