View Full Version : DARPA Flying Missile Rail
unmanned_droid
7th Sep 2017, 18:22
Well that is pretty damn interesting.
I'd be up for this as a professional venture, but not being american or a known name it would likely be a waste of time.
It is an interesting problem though. I've been mulling over how you build and operate military aircraft in a modern World War situation.
In my opinion, We're not able to lose many aircraft over the battle field because of the numbers available for replacements.
When your replacements run out, there is significant time taken to actually manufacture a modern combat aircraft, and when all your time intensive long lead time items run out, the time taken to get those replaced - what are we doing in that time? A lot of those long lead time items are things like very high quality steel and titanium forgings which are not made locally and can require significant 'seasoning' times before being ready for use.
Having an incredibly cheap way of boosting the effectiveness of an asset whilst at the same time reducing that assets exposure to loss has to be the way forward. And I'm not talking posh LO UAVs - it's got to be firebee type technology with even cheaper very freely available materials that do not require expensive manufacturing methods or storage regimes. Yep, it would be cool to be a part of this. I wonder just how ghetto they would go?!
Lonewolf_50
7th Sep 2017, 20:38
Love the idea, interesting challenge on the manufacturing and capability side.
Mechta
7th Sep 2017, 20:49
The V1 flying bomb has got to be the closest to this concept so far. Ground launched and air launched, with a very simple engine, and airframe parts made from either wood or metal.
From Wikipedia:
Almost 30,000 V-1s were made; by March 1944, they were each produced in 350 hours (including 120 for the autopilot), at a cost of just 4 per cent of a V-2, which delivered a comparable payload. Approximately 10,000 were fired at England; 2,419 reached London, killing about 6,184 people and injuring 17,981. The greatest density of hits were received by Croydon, on the south-east fringe of London. Antwerp, Belgium was hit by 2,448 V-1s from October 1944 to March 1945.
Fonsini
10th Sep 2017, 15:05
That would be a project I would love to be a part of. Keeping the cost down would be the major challenge as it would be easy to fall into the trap of designing a mini UCAV.
Nice find ORAC.
Fonsini, you remind of a discussion I had with a contractor back in the 1970s who was looking to interest the RAF in wartime disposal drop tanks for the Jaguar force. He had a design based on those of WWII (http://www.warbirdsnews.com/warbird-articles/necessity-mother-invention-paper-drop-tanks-wwii.html) which he reckoned he could produce at short notice for about £200 a piece. The RAF wasn't interested as they wouldn't believe the price and thought anything aircraft qualified would have to be at £6K each - at which point it wasn't economic in peacetime against much more expensive reusable drop tanks.
What is the price point in peacetime where you could persuade an accountant to start buying FMRs which have have to be thrown every every trip against buying a reusable platform you can bring home but which costs 200+ times as much?
It's the same as the story that the engines for a Mig-21 had a life of about 100 hours because the expected service life of a Mig-21 in war was about 90 hours - why spend money on better materials building an engine which would get shot down anyway? Better to spend the money building cheap engines and airplanes continuously in peacetime so you had the in place capacity to do the same in wartime. But no peacetime accountant will pay for such extravagance.
Maybe computerised lean production with rapidly rerolled production lines, coupled with 3D printing of both plastics and metals, have changed the equation.
unmanned_droid
10th Sep 2017, 18:05
As I said....just how ghetto are you willing to go (obviously in an aerospace engineering context...)?
3D printing isn't as rapid as say punching out shapes with a press or vac forming plastics and some aluminiums are far more available in useful shapes than just about all aerospace grades.
When I was designing a research test I wanted to purchase some L shaped extrusion material in a common aerospace grade. Not possible. Had to buy blocks and machine away the vast majority.
riff_raff
11th Sep 2017, 03:13
This idea sounds wonderful on a Powerpoint presentation. But I can't see how it would be possible to quickly achieve a production rate of 500 units/month "on-demand". The proposal to produce the rail using rapid prototype processes like DLMS is probably not realistic. CNC machining from bar or extrusion would be far faster and cheaper. If they want to minimize lead time, they should focus on optimizing the CNC machining process, minimizing any dedicated fixture/tooling required, and designing the rail so that it can be produced on a wide variety of common CNC machines.
The most important thing they can do to reduce lead time for getting these rails into service would be to pre-qualify several established, reliable vendors to supply them on short notice.
It's really trying to copy the SpaceX model isn't it?
When SpaceX started they got sticker shock when the companies with virtual monopolies quoted them what they charged NASA/DoD. For example for an actuator they were quoted over $120K - so they went away and built in-house to a higher standard for $3,700. They ended up building over 80% in-house at fractions of the going market prices (and use a lot if 3D printing etc).
Of interest SpaceX don't patent any of their in-house designs. According to Musk he sees patents as simply a way to provide China with the technology without them having to go to the bother of having to steal it....
unmanned_droid
11th Sep 2017, 10:39
I'm fully in agreement with Musk on the patent point. Besides getting anything patented (whether it be a piece of specific tooling or a whole aircraft configuration) is an expensive pain in the arse just to support an industry that doesn't need to exist. I imagine we all know our work when we see it.
I disagree on the point that this is trying to copy the SpaceX model.
I see this as having the capability without having the assembled hardware until right before you need the hardware. From what I see, It's about having a couple of containers worth of kit in a storage facility that could be rapidly deployed and set up to build the harware when needed, and it's about having access to materials that can be sought in quantity with no time delay. There is also a certain amount of pre-bought hardware - electronics and engines for example that need to be carted around with the build kit.
A rate of 500 units a month for essentially a target drone from zero isn't to be sniffed at. It would take compromises that I am not sure those without knowledge of RC aircraft production and having the thought process to consider that quality on a long term aerospace product isn't what is needed, would consider to be acceptable.
riff_raff
13th Sep 2017, 02:30
It's really trying to copy the SpaceX model isn't it?
It took SpaceX almost 6 years (2002 to 2008) to achieve the first successful launch of their Falcon rocket. The company model DARPA should be trying to emulate is North American Aviation during the early 1940's when they built and test flew the Mustang prototype in around 8 months time.
fltlt
13th Sep 2017, 03:14
The future is here folks, I know the items are favorite WWII aircraft, but watch the video.
All designed to be printed in sections, glued together and hey presto you can have an airframe ready for power plant/controls, the only limit is how many 3D printers you have access to.
Upgrade the design on the fly so to speak, simply send out an updated printing program.
Want it printed in colors, no problem.
All the design, build, test, certify if need be, done at one central location.
If you really want to produce en masse, this is the way to go:
https://3dlabprint.com/
There are other companies out there just getting into this field, just Goggle 3D printed planes.
PPRuNeUser0211
13th Sep 2017, 05:19
It took SpaceX almost 6 years (2002 to 2008) to achieve the first successful launch of their Falcon rocket. The company model DARPA should be trying to emulate is North American Aviation during the early 1940's when they built and test flew the Mustang prototype in around 8 months time.
Riff,
No offence to NA, I love the Mustang as much as the next guy, but I'd suggest that building a frikkin space rocket is *somewhat* more complex than something controlled by cranks and pulleys. Don't get me wrong, both companies did a great job, but I'd suggest taking ten times as long to develop is a reasonable penalty for getting to space....
unmanned_droid
13th Sep 2017, 12:20
The future is here folks, I know the items are favorite WWII aircraft, but watch the video.
All designed to be printed in sections, glued together and hey presto you can have an airframe ready for power plant/controls, the only limit is how many 3D printers you have access to.
Upgrade the design on the fly so to speak, simply send out an updated printing program.
Want it printed in colors, no problem.
All the design, build, test, certify if need be, done at one central location.
If you really want to produce en masse, this is the way to go:
https://3dlabprint.com/
There are other companies out there just getting into this field, just Goggle 3D printed planes.
Having worked alongside a major aerospace Additive Manufacturing development team, I know that this solution isn't quick enough, deployable or in any way affordable for a task like this, besides other more sensitive issues.
fltlt
13th Sep 2017, 13:51
Having worked alongside a major aerospace Additive Manufacturing development team, I know that this solution isn't quick enough, deployable or in any way affordable for a task like this, besides other more sensitive issues.
I beg to differ, one has to make the leap from aerospace to modeling techniques, each with its own prerequisites, strengths and weaknesses.
Straddling both worlds opens up relative (user dependent definition) low cost/high volume production of somewhat niche products.
Not for manned, high tech fighting machines.
unmanned_droid
13th Sep 2017, 13:57
I beg to differ, one has to make the leap from aerospace to modeling techniques, each with its own prerequisites, strengths and weaknesses.
Straddling both worlds opens up relative (user dependent definition) low cost/high volume production of somewhat niche products.
Not for manned, high tech fighting machines.
You'll need to explain this a bit more for me:
"I beg to differ, one has to make the leap from aerospace to modeling techniques"
Making a leap from a sector to a technique?
Additive manufacture can and does work in aerospace (although right now it's the current poster boy - look what we can do!!) But, for this specific application?
Desktop Metal (http://newatlas.com/desktop-metal-3d-printing/50654/)
fltlt
13th Sep 2017, 15:24
You'll need to explain this a bit more for me:
"I beg to differ, one has to make the leap from aerospace to modeling techniques"
Making a leap from a sector to a technique?
Additive manufacture can and does work in aerospace (although right now it's the current poster boy - look what we can do!!) But, for this specific application?
Take the Additive mindset out for a moment. This has to do more with design and manufacture for purpose and cost, airframe only.
That could, and I do stress could change, although as you point out, additive and in fact the whole promise of 3D printing is still a somewhat look at me.
Scale of production is not manhour nor size of facilities dependent.
You could have 1000 printers under one roof or 4 under 250 roofs or any combination thereof. If one needs to manufacture on the other side of the country/world, relatively (there's that word again) cost effective.
Just my tuppence.
There are other issues on the logistics side, but nothing that is a show stopper.
unmanned_droid
13th Sep 2017, 15:38
Desktop Metal (http://newatlas.com/desktop-metal-3d-printing/50654/)
Thanks for the article. No build times stated (including cooling times!) As well as some other bits they appear to have left out (a lot of sintering is done in a vacuum).
Sintering isn't really a technology that would be looked at for parts that get load cycled in comparison to other ways of forming metals. I'm not hot on sintering (hohoho) because it's just not used in aerospace structures, but it wouldn't be my go to choice. Then again, this application is very limited in terms of life so it may well be fine if it actually can be scaled.
unmanned_droid
13th Sep 2017, 15:40
Take the Additive mindset out for a moment. This has to do more with design and manufacture for purpose and cost, airframe only.
That could, and I do stress could change, although as you point out, additive and in fact the whole promise of 3D printing is still a somewhat look at me.
Scale of production is not manhour nor size of facilities dependent.
You could have 1000 printers under one roof or 4 under 250 roofs or any combination thereof. If one needs to manufacture on the other side of the country/world, relatively (there's that word again) cost effective.
Just my tuppence.
There are other issues on the logistics side, but nothing that is a show stopper.
Right ok, with you.
My argument isn't could you do it necessarily...because, as you say, you can set up however many you like and can buy or get access to. It's more, why would you do it?
Thanks for the article. No build times stated (including cooling times!) no vacuum and 100 times faster than current sintering.
https://www.desktopmetal.com/products/production/
"Breakthrough Single Pass Jetting (SPJ) process delivers speeds up to 8200 cm3/hr–100x faster than laser-based systems. With zero-tooling needed.. The Production system is based on a new approach to metal 3D printing—Single Pass Jetting (SPJ). Created by the inventors of the binder jetting and the single pass inkjet processes, Single Pass Jetting builds metal parts in a matter of minutes instead of hours."
https://techcrunch.com/2017/04/25/desktop-metal-reveals-how-its-3d-printers-rapidly-churn-out-metal-objects/
"Desktop Metal calls its core technology “microwave enhanced sintering.” The company’s printers put down layers of metal and ceramic powders that are mixed in a soft polymer. The cartridges and alloys that work with the printers are made by Desktop Metal and other major providers in additive manufacturing. Once a mixed-media item is printed, it goes into a furnace where it is rapidly cooked. Heat burns off the polymer. Gases are filtered by charcoal. Meanwhile, the metal is fused together but at a temperature that won’t make it melt and lose its shape. Wherever ceramic was laid down in a printed design, metal remains separated and doesn’t fuse. The pieces created by Desktop Metal machines can be separated by hand...."
unmanned_droid
13th Sep 2017, 16:36
no vacuum and 100 times faster than current sintering.
https://www.desktopmetal.com/products/production/
"Breakthrough Single Pass Jetting (SPJ) process delivers speeds up to 8200 cm3/hr–100x faster than laser-based systems. With zero-tooling needed.. The Production system is based on a new approach to metal 3D printing—Single Pass Jetting (SPJ). Created by the inventors of the binder jetting and the single pass inkjet processes, Single Pass Jetting builds metal parts in a matter of minutes instead of hours."
https://techcrunch.com/2017/04/25/desktop-metal-reveals-how-its-3d-printers-rapidly-churn-out-metal-objects/
"Desktop Metal calls its core technology “microwave enhanced sintering.” The company’s printers put down layers of metal and ceramic powders that are mixed in a soft polymer. The cartridges and alloys that work with the printers are made by Desktop Metal and other major providers in additive manufacturing. Once a mixed-media item is printed, it goes into a furnace where it is rapidly cooked. Heat burns off the polymer. Gases are filtered by charcoal. Meanwhile, the metal is fused together but at a temperature that won’t make it melt and lose its shape. Wherever ceramic was laid down in a printed design, metal remains separated and doesn’t fuse. The pieces created by Desktop Metal machines can be separated by hand...."
Yeah, but you need a vacuum or another gas to prevent impurities during the sintering process.
Ans as for the stated build times and what they do and do not include, well throwing around things like factors of a hundred gets my spidey senses tingling.
I look forward to seeing where this goes.
riff_raff
15th Sep 2017, 04:35
That Desktop Metal machine is limited to a part size of around 13". And if the finished part requires close tolerance interfaces, it will probably need to be finish machined using conventional CNC processes after being sintered.
Less than that, 10" x 6.7" x 6.7" post sintering. But remember this is their first model and is designed for studio use and to fit through an office door. It's the technique, not the machine that is of interest. They claim it doesn't leave support material to be removed and parts can be finished by bead blasting.