Carbon fibre makes Australian debut
csiro.au
csiro.au
Is that the current production rate for 737s?
https://en.wikipedia.org/wiki/Boeing_737#Orders_and_deliveri...
Or are some significant fraction of airframes scrapped due to defects before the next stage?
As an aside, here's an interesting article if you're interested in the process works (or worked at the time of publication): http://www.seattletimes.com/business/boeing-aerospace/boeing...
Carbon fibre also has many more degrees of freedom than aluminium. Not all of these degrees and their interactions, e.g. this arrangement of fibres with some other resin cured at that temperature, are well understood.
My opinion though, if the strength is matched if not exceeded of a metal airframe, but weight is I don't know effectively halved or more... less fuel, farther flights, should be cheaper air fare, possibly less maintenance if carbon fibre doesn't have fatigue problems... but the cost.
As a cyclist, I have a perception of carbon as kinda fragile; take a good hit, and you really need to replace your fork. This makes me think the carbon will mainly be used for internal structure, which would still greatly reduce the weight of the plane (and this the base amount of fuel used), and bring down those lifetime costs. But if it's too fragile, then the overall lifespan of the plane is reduced, which would counterbalance the gains.
Equally interesting to ponder what technologies are evolving because of our current situation
I'd certainly not bet any money on that happening.
I, for one, believe him.
To respond to GP, if you can make an aerospace component lighter, once it's made, the thing the component goes on will be better than if it were heavy. Planes have to carry enough fuel to sustain altitude and also to push the plane to the destination. If the plane is lighter, you use less fuel to sustain altitude, saving money every moment the plane's in the air. If you can cut off enough weight, you can tweak the design of the plane itself because it won't need as much lift at speed, making it even more efficient (or faster per unit fuel burned).
As this article points out, the new facility in Australia is state of the art. They claim it may produce better carbon fiber than anywhere else. So imagine Boeing made a massive investment in equipment to make their own, only to have someone come along and do better. They make airplanes and let the industry figure out how to make the fiber, that way they can always buy the best if they want to.
There is a big push for new production processes / material sources.
Here's an example of technology using cheaper and less energy intensive acrylic instead: https://phys.org/news/2016-03-ornl-low-cost-carbon-fiber.htm... Or a plasma method making it from PAN but faster: https://www.youtube.com/watch?v=SLmU0BmLmFE
So there is potential for large drops in price and increases in supply in the future. Probably the wind energy market is driving this.
I remember going to one of the shipyards (I think Vanguard... I can't recall) and watching them make carbon fibre parts. I was rather shocked how labor intensive it is. Maybe not labor intensive is the right word but it just takes a long time to make a single pole or mast.
If I recall you take long sheets of carbon fibre and wrap around a mold and wait a little bit and then cure it. You have to do this like 40 times.
I am probably missing a bunch of details as I was fairly young but I do remember it takes a long time to make carbon fibre parts.
So even if carbon fibre is produced at scale the molding and producing of actual parts probably has to be sped up as well?
i believe there are also thicker sheets that can be vacuum-bagged onto a surface, so you don't need to layer it.
Of course with better composites you got more stiffness and could also put the weight in the most advantageous position.
You also have to use expensive, stiff epoxy resin if you want to have the mechanical properties of carbon fiber. Fiberglass is typically laid up in cheaper polyester resin since glass fiber are stretchier than carbon fiber.
Would you mind expanding on that? I'm curious around the barriers to mass-produced carbon fiber.
A glass fiber/polyester panel will be no worse than a carbon fiber/polyester panel. So if you choose polyester you'd be a fool to use a reinforcement more expensive than glass fiber.
So as a practical matter, the decision to use carbon fiber reinforcement implies the use of epoxy resin.
To make a bike you need a special mold/jig for each part for each bike size. These cost money to make so to scale out the process is expensive (I.e. to have a lot of workers producing bikes concurrently). With metal tubing you can just make a lot if it and then cut, bend and weld it. Every worker doesn't need an expensive jig. And like you said, just laying up the fibre on a jig is time consuming and manual.
Anyway, they can make extremely light aluminium bikes now with really good ride characteristics and they won't break if you drop them the wrong way!
They were going to use it in the Nissan Delta Wing, but i think alot of that fell by the wayside when they elected to use the Monocoque from the Aston Marton AMR-1 rather than a bespoke tub. Never really heard much about it after that.
Bombardier spent a lot of money in trying to locate the Learjet 85 fuselage lay-up process in Mexico, instead of Belfast in the UK, only to find that the resulting parts were delaminating because the process had been developed for Belfast's climate and sea-level pressure.
Eventually the 85 project was 'suspended', as it remains today. Unfortunately most of the staff at the Querétaro site were laid-off as a result.
That said I just saw a talk yesterday about DNA folding and of course trying to replicate the experiment in a different lab caused a tenth of degree variation which voided the whole thing.
Ireland is a damp place; it doesn’t surprise me that carbon fibre layup didn’t work in the same way in Mexico!
It's also the where a fairly large printing company has it's central printing facility. Inside the main warehouse over tons of paper a fine mist if intermittently created, keeping the paper malleable. In a less damp environment this misting process might have to be intensified.
Shift in humidity = adjustment in drying of raw material & tweak to drying time as the product comes off the line = variability in total production time, for example.
Surprisingly, as referenced in the Bombardier example, adjusting to ambient environment is often manual/overlooked: it’s trial and error on the manufacturer’s part to make the adjustments necessary.
https://www.thisamericanlife.org/radio-archives/episode/241/...
It's a short tale about a sausage factory that moved from an old facility to a brand new state of the factory and discovering their sausages no longer tasted the same, despite using the same ingredients, etc. and how the eventually worked out what the difference was.
Depends on the industry. Some, like semiconductor chip manufacturing, go to extreme lengths. One plank of Intel's success is a strategy to measure everything they can, and calibrate every copy of a factory until they all line up. Don't try to figure out every possible causal connection, just control and copy exactly everything you can measure.
The 2017 Formula 1 season gets underway in Melbourne on March 26[1]. Formula 1 regularly uses some of the most advanced technology[2], and they make heavy use of carbon fibre. Since this was patented, the methods etc. should now be public. So, once the licensing stuff is worked out, F1's carbon fibre suppliers can rework their
The benefit to F1 is in teams being able to apply changes faster, taking lessons learned from Melbourne and applying them as quickly as possible, without having to wait as long as they do now for new parts to be constructed.
I'm looking forward to this!
[1]: https://www.formula1.com/en/latest/headlines/2016/11/f1-fia-... [2]: At least, what they can get away with! Some examples from the past: http://www.redbull.com/us/en/motorsports/f1/stories/13316206...
The comment implies that the time it takes for prototype teams to respin is largely driven by the availability of raw materials...I'm simply not following the logic.
That said, an improvement in carbon composite manufacturing could still help them if the amount of manual labor could be reduced.
That's the lingering thought in the back of my mind that kept me on the skeptical side. Thanks for chiming in.
I feel like you should've yelled "bingo!"
Even if I am willing to believe they are actually counting the strands (could be the case, especially if they are really long), I can't believe that this machine produces exactly 500 in each batch, so I suspect something is missing after "five hundred". "500 micron long strands", "500m of strands a minute" or something like it?
Also, regardless of whether a unit is missing, how does this compare to the competition?
2. Re comparison to competition: I have visited a carbon fibre production plant in France. They use a different process, and do not extrude the fibres on site. Instead they buy polyacrylonitrile (PAN) fibers - these are used in their own right to make "acrylic" cloth, but there they are the raw material. They convert that into carbon fibre with I think tens of thousands of strands at a time. So, 500 strand production line seems to be sub-commercial.
https://phys.org/news/2016-03-ornl-low-cost-carbon-fiber.htm...
"Researchers at the Department of Energy's Oak Ridge National Laboratory have demonstrated a production method they estimate will reduce the cost of carbon fiber as much as 50 percent and the energy used in its production by more than 60 percent."
The carbon fiber industry had overcapacity in 2013, but demand is now up a bit.[1] The biggest producer is Toray (Japan), but there are lots of others. Wind turbine blades are starting to use carbon fiber. Automotive isn't yet a volume user, but the fiber industry is struggling to get in the door.
Right now, the limits on using carbon fiber seem to come from the problems in making stuff out of it. The process described in the parent article ends with a spool of thread. As of 2014, that sold for about $20/Kg, which is cheaper than 3D printer filament. Making an auto part out of that cost $80/Kg. Some new technology was supposed to bring the manufacturing cost down.[3] But as of two years ago, the cost of the base material wasn't the limiting factor on use.
[1] http://www.compositesworld.com/articles/supply-and-demand-ad... [2] https://blog.carbonfibergear.com [3] http://www.reuters.com/article/sgl-fibres-idUSL5N0MP2RP20140...
Maybe the OP's article theorises why this may be the case, but I can't load it. Is it just that it's no longer purely a trade secret among facilities?
- an open patent rather than trade secret means that many manufactuers can make it with the comparitively lower barrier to entry of a patent licence rather than inventing a process internally
- what you said
- Their process is somehow revolutionary (pun not intended) and makes it cheaper to manufacture
I'd be interested if anyone knows more!
I presume we have other obstacles to overcome first ;)
Hilariously, most drafts I see/do tend to score at the postgrad level.
they didn't invent wifi, they invented modern wifi
from the horse's mouth, the lawyer representing the company in the relevant case: "CSIRO did not invent the concept of wireless LAN, it just invented the best way of doing it, the best way it's used now throughout the world,"
In addition, it does seem they had a strong desire to make use of their patent, approaching potential partners, and creating spun-off companies, centered around making use of the technology
CSIRO patented its use in WLAN setups (including WiFi).
It's controversial because they're only major first was to implement WLAN setups using OFDM with forward ECC. In any case they were merely taking something already used in cellular phone networks and applying to another kind of broadcast setup.
> The premier world engineering institution, the IEEE, created a working group for the evolving 802.11 wireless standard in 1990, a full three years before CSIRO filed for its key wireless patent. The group voted repeatedly on which way to go forward and produced heaps of records, but CSIRO didn't even participate in the 802.11 committee. The group published the first 802.11 standard in 1997 and CSIRO came forward years after the fact.
The working group was created in 1990, the patent was filed in 1993 and the 802.11 standard first published in 1997.
Long story short: CSIRO did participate in the process through a third party (the first company they licensed their tech to). They also agree to the IEEE's request that they would licence the IP on 'fair, reasonable and non-discriminatory terms' (note: not free). When they approached vendors and made an initial offer of $4 a pop, vendors refused to pay and ended up in court. Honestly I think wifi vendors simply misjudged the probability of CSIRO taking them to court if they didn't cough up.
The Commonwealth Scientific and Industrial Research Organisation is, as the name suggests, an honest-to-god pure and applied research body.
If CSIRO researchers invent a thing and the CSIRO patents it, they damn well deserve their fair royalties.
https://mobile.nytimes.com/2007/03/05/technology/05music.htm...
I must be confused. How are they a patent troll if they actually do the the R&D part of the patent? A patent troll would buy the filed patent and just spend money enforcing it to collect cash...
I am a bit biased because I'm Australian and fully support the CSIRO. They're a non-profit entity[0] - their mandate isn't to make money. The money they do collect from royalties is rolled back into their research.
[0] In that they don't produce a profit for stakeholders. They're a government agency though, not a registered charity.
They now operate as 'Data61' on a semi-commercial basis. Again, I don't think this is necessarily a bad thing: I'm actually working with some of the NICTA/Data61 guys on one of their projects, and they are doing a number of these across the APS. If anything, it's been great to unleash these guys on the APS, where IT expertise is sorely lacking. They seem to be very good at what they do (and have a heavy focus on applying machine-learning to real-world problems).
Their new CEO, Larry Marshall actually created a number of US startups centred on lasers/optics (sorry, don't know the proper term) and holds a number of patents in this area. I met the guy when he gave a presentation at my workplace, and this carbon-fibre work is very much in line with the strategy he outlined. He's a decent guy who, IMHO, has been given a pretty unfair run in the press (who have taken a number of his statements out of context and portrayed him as some kind of anti-science luddite). Incidentally, he reads PG's essays :)
I don't really have a point here, just providing some background information in case you find it interesting (and apologies if I sound like a bit of a cheerleader). This shift in CSIRO's focus might have been the one and only thing our idiot government has gotten right over two terms of government (or, at least, the one thing they didn't totally screw up).
EDIT: And just to concur, the assertion that CSIRO is a patent troll doesn't pass the laugh test.
Most of the people everyone outside Australia who've heard of CSIRO at all know of them because of what is viewed as a shakedown of WiFi manufacturers. It seems more sad than funny.
Oh well...
I'm not asserting the Arstechnica article is wrong and the Register article is right. But it's an alternate view from what is, as far as I can tell, an unbiased observer. I'm sorry it doesn't line up with the views that you quite obviously hold, but that's just how life is sometimes. You might also wish to consider whether your dismissal of The Register's assertions is logical, as it basically amounts to:
P: I don't like A.
P: A is asserting X.
C: Therefore X is wrong.
I haven't got the slightest idea why you view The Register's article as being of superior quality to Ars's, regardless of how you stand on the issues, since it's just a small, sarcastic ("Evil Aussie tax on Wi-Fi") attack on what the author views as the weakest parts of what Ars wrote. Even a blogger would put more effort in.
But it seems to me the Register adds a useful, but douchey ("Cisco was scammed!") elaboration on the role of Radiata, sidesteps the question of whether the amount CSIRO was asking was RAND (which is the whole point of everything), and makes some very pathetic attempts to appear balanced in the last two paragraphs. "The hyperbolic statements made by Australia’s science minister and various press should be challenged." But not in this article.
Sorry, but this is crappy writing not worthy of deeper consideration. At least Ars tried to write something.
edit: thanks for really putting the effort in with your reply!
Well it's certainly been a laugh and a half.
Having said that, I'd liked to have seen them handle the wifi case(s) better - I'm not sure I could say how, however. That said, I also believe that they're warranted in their attempts to get some compensation out of the contributions they made that were eventually adopted wholesale across the industry - you'd expect a private organisation to see financial benefit from it and I don't think that's much different if it's government-funded either.
In both that case and this, whether their specific contributions to a developing technology warrant being patentable or not will differ from person to person. I'm sure I don't agree with Stallman on this, for example. I'm not going to try to say it should be one way or another definitively but to paint the CSIRO as a patent troll is a tremendous leap of illogic that requires a fundamental ignorance to what they do. At best, you could claim that their aggressive use of patent litigation doesn't align with how you feel about patents - and I could certainly see why some would feel that way. That's more a question of the role of patents themselves and their scope than a question of the validity of the CSIRO as an organisation, though.
NB: I am Aussie, to the extent that may make me biased given my tax funds this org.
Edit: it's also worth noting that the guy you cited on Reddit is claiming they 'feel like' a patent troll because their website looks similar to Intellectual Ventures' one. He openly admits to being wholly ignorant to who they are and hsvinfno knowledge of their history or purpose.
The main difference is NPEs target smaller companies and then talk up their claims in the initial phases, then try to settle out of court as quickly as possible so they don't incur alot of expense and can avoid discovery, which is where their patents may get knocked out completely.
Plaintiffs like CSIRO may not be trolls, per se, but they use the same hired gun damages experts that the trolls do (like Ocean Tomo). The amount of money they seek is completely untenable in the context of standard essential patents, if only because our phones and computers would cost astronomically more money if everyone holding one of the thousands of allegedly standard-essential patents got the kind of money plaintiffs like CSIRO ask for in these cases.
As an aside, patents don't really get "knocked out" in discovery. They get knocked out by defendants filing IPRs, seeking re-exam by the PTO, and occasionally at trial. It's unbelievably rare for a judge to invalidate a patent on summary judgment (which happens post-discovery).