F35 - Money well spent.

bob231

War Hero
I’ll just say this:
You actually mentioned the Gloster Meteor in print for anyone to see!!

Would you like some pics of apples and oranges to compare while you are on a roll? ;)

The fact you chose something like the very very very numerous Meteor and not something far more ground breaking in every sense like the F117 speaks volumes. The F117 didn’t do too bad for losses, especially when it’s considered that it did far more combat sorties in far far more contested airspace than the Typhoon has.
It would be nice if the RAF could still have a few thousand fighters of one type on the books like back in the Meteor days, but those days are long gone :)
My smileys were lost on you weren’t they? ;)

As for your last point, it’s falling on deaf ears I’m afraid (especially as your point on Watchkeever has zero to do with a production method....and you know that already).
I’d simply ask you: if you needed new F16s quickly, would you prefer them more quickly, or would you keep to old production methods........
After all, you already know how to fly them, maintain them and use them :)
Double quote as my point applies to both. The Meteor is more applicable as you're talking about novel materials, not novel design. Once you've got past the first jets (and the first discoveries of how wings and engines perform when they run hotter, higher, through more pressure differentials, etc), the chances of failure through material failure go down. Helped of course by better materials science, but empiricism is a good starter.

3D printing is fantastic for making something quickly. It's less fantastic if you care about the crystal structure of whatever you're making: to consider another of your examples, a wing root of fairly amorphous crystalline structure is significantly weaker than one that's been cast and cooled just so. It means you can replace the wing root much faster but it may need to be significantly larger or a different shape, which imposes design costs (in time and money) and possibly certification costs.

The same is true of composites. Composites would be fantastic in jet engines (relatively easy to work, cheap, very light) were it not for the list of things composites don't like corresponding very closely to the list of things found in aero engines (extreme heat, high vibration, hot oil, pressure differentials, sharp changes in pressure, requirements to withstand shock loading...). Marine gas turbines don't care as steel is (effectively) free and weight is irrelevant.

These technologies have fantastic possibilities, don't get me wrong. But they have some quite significant limitations, and until these can be overcome, you can't easily incorporate them into certain key parts of the aeroplane.
 

Slime

LE
Double quote as my point applies to both. The Meteor is more applicable as you're talking about novel materials, not novel design. Once you've got past the first jets (and the first discoveries of how wings and engines perform when they run hotter, higher, through more pressure differentials, etc), the chances of failure through material failure go down. Helped of course by better materials science, but empiricism is a good starter.

3D printing is fantastic for making something quickly. It's less fantastic if you care about the crystal structure of whatever you're making: to consider another of your examples, a wing root of fairly amorphous crystalline structure is significantly weaker than one that's been cast and cooled just so. It means you can replace the wing root much faster but it may need to be significantly larger or a different shape, which imposes design costs (in time and money) and possibly certification costs.

The same is true of composites. Composites would be fantastic in jet engines (relatively easy to work, cheap, very light) were it not for the list of things composites don't like corresponding very closely to the list of things found in aero engines (extreme heat, high vibration, hot oil, pressure differentials, sharp changes in pressure, requirements to withstand shock loading...). Marine gas turbines don't care as steel is (effectively) free and weight is irrelevant.

These technologies have fantastic possibilities, don't get me wrong. But they have some quite significant limitations, and until these can be overcome, you can't easily incorporate them into certain key parts of the aeroplane.

I was talking about a novel design concept.
Well, novel for military aircraft, but widespread elsewhere.

You should note I never mentioned printing wing parts, so it’s not an example of me advocating 3D printing.
I talked of using a laser to measure the aircraft, and thus being able to make bespoke wings to match the fuselage. I mentioned the MR4 as each aircraft needed a different fit for the wing roots.

You might be missing my point about how the NGAD programmes sees speeding up design, or changes in aircraft in production, and getting bogged down with plastic/metal printing.

The concept can (as advertised) affect all components, as well as spares supply and maintenance pricedures.


As for Meteor, comparing stats for 200 Typhoons, while ignoring there were a few thousand Meteors isn’t a good comparison imho.

I thoroughly get that the wisdom of ARRSE thinks those designing, building and actually testing in the NGAD programme for the last few years have all got things wrong :)

Edited to add:
Apart from my typo missing the ‘d’ from machined, you can see I mentioned the parts being machined to tight tolerances.......not printed :)

With ref to the above, just imagine if some bright spark had laser scanned each of the potential MR4 Nimrod wing roots.................

It would have saved loads of hassle, and each bespoke wing could been modelled and machine to tight tolerances instead of the complete horlicks of how things were done.
 

bob231

War Hero
I was talking about a novel design concept.
Well, novel for military aircraft, but widespread elsewhere.

You should note I never mentioned printing wing parts, so it’s not an example of me advocating 3D printing.
I talked of using a laser to measure the aircraft, and thus being able to make bespoke wings to match the fuselage. I mentioned the MR4 as each aircraft needed a different fit for the wing roots.

You might be missing my point about how the NGAD programmes sees speeding up design, or changes in aircraft in production, and getting bogged down with plastic/metal printing.

The concept can (as advertised) affect all components, as well as spares supply and maintenance pricedures.


As for Meteor, comparing stats for 200 Typhoons, while ignoring there were a few thousand Meteors isn’t a good comparison imho.

I thoroughly get that the wisdom of ARRSE thinks those designing, building and actually testing in the NGAD programme for the last few years have all got things wrong :)

Edited to add:
Apart from my typo missing the ‘d’ from machined, you can see I mentioned the parts being machined to tight tolerances.......not printed :)

With ref to the above, just imagine if some bright spark had laser scanned each of the potential MR4 Nimrod wing roots.................

It would have saved loads of hassle, and each bespoke wing could been modelled and machine to tight tolerances instead of the complete horlicks of how things were done.
My error, I read yours quickly having seized on a particular point to reply. You are, at least as far as my knowledge runs, absolutely correct.

It may also amuse to hear that there are a stack of Rolls-Royce gas turbines sitting somewhere in Dahlewitz, each of which has a coach-built set of amounts as the variation from Nimrod to Nimrod is so great.

I haven't looked into the NGAD; would I be wrong in assuming the concept is pushing standardisation very hard, so bolt-in-bolt-out is practical and bespoke manufacturing is required as little as possible?
 
I’d simply ask you: if you needed new F16s quickly, would you prefer them more quickly, or would you keep to old production methods........
After all, you already know how to fly them, maintain them and use them :)

...as the Canadians recently demonstrated when they bought the RAAF's old F-18s, rather than a bunch of (same name, similar shape, different aircraft) Super Hornets.

But do you really "already know how to fly, maintain, and use" a new-build Block 72 F-16, if you're currently operating Block 25 like the Air National Guard? Different radar, ESM suite, mission computer, nav system, cockpit layout, engine model? I'd rather hope that unlike the 737-Max, Boeing won't insist that "an F-16 is an F-16" and that the pilot conversion training is minimal...
 
I thoroughly get that the wisdom of ARRSE thinks those designing, building and actually testing in the NGAD programme for the last few years have all got things wrong :)

I don't think that at all - I think it has the potential to be very exciting indeed; and properly run, could deliver impressive results. However, I'm also skeptical of those who are trying to sell these excellent, very fine, clothes - stunning , so delicate you can hardly see them, Oh Emperor...

Take "Agile" development processes. Done at the right time, with the right people, in the right way, the results are impressive. Waved as a banner by muppets and slimy sales types who only think they understand it, and the results are rather different - even in something as comparatively simple as a few hundred thousand lines of software, for use in a benign business environment. Been there, done that - beware the slick PowerPoint demo.

Except this isn't "just a bit of software", or "oh, we're fitting together a few 3D parts". These are hideously complex items of safety-critical cross-discipline engineering, operating at the raggedy edge of materials science, electronics, software, physics, and aerodynamics. Lots of stuff affects lots of other stuff, often in unpredictable ways.

I've used the example of SpaceX Crew Dragon and Boeing Starliner before; only two years ago, NASA (and many informed observers) reckoned that Boeing was more likely to deliver a working, certified, safe, crewed capsule on schedule. It turned out that Boeing were running a bit of a cowboy outfit, while SpaceX were the professionals. How can you tell?
 

Flight

LE
Book Reviewer
The F-35 was also going to be a revolution in how things were made with a robotised production line much like cars....

Still having multiple maintenance depots worldwide is decidedly spiffing...
 
The F-35 was also going to be a revolution in how things were made with a robotised production line much like cars....

Still having multiple maintenance depots worldwide is decidedly spiffing...
Yes, well back when the F-35 project was started the trendy new thing in management consulting circles was the Toyota Production System. So, that's what the consultancies sold the US defence ministry. "Buy our consulting services and training packages and we'll teach your project managers how to apply this magical system so you too can increase quality while reducing cost."

Now the "Toyota production system" is passée and the consultancies are selling "agile". "Buy our consulting services and training packages and we'll teach your project managers how to apply this magical system so you too can bring projects to completion faster while reducing cost."

The people who created "agile" have said that what management consultancies are selling today is a travesty of the original concept, but none the less, let's just follow through with the idea. Let's see, the Americans have a problem with rampant feature creep extending timelines and driving up costs. What can we do about that? I know! Let's adopt a methodology based around the idea of the customer not really knowing what he wants so we just keep adding features until the customer runs out of money! What could possibly go wrong?
 
@Slime - In case you may think that I just like to criticise without offering any constructive suggestions, have a look here at what I said on this thread more than 7 years ago. That's at post 628, and we are well over 12,000 posts by now.

I talk about the need for shorter development timelines, more incremental developments, more standard interfaces, and less "big bang theory" of project management (doing everything all at once).

However, that isn't the same thing as "agile", or even what present day management consultants are marketing as "agile". It is though more about evolutionary change rather than being obsessed with revolution.

It doesn't require 3D printing or any sort of revolution in technology. It does require careful planning and the setting of industry standards so that companies can work together instead of creating proprietary silos of technology.

I don't have any objections to the pentagon changing the way they do things. I'm just sceptical that they have genuinely changed however, rather than just once again grabbed on to the latest passing fashion in an effort to avoid real institutional change.

The problem in the US procurement system is their institutional obsession with wanting to deliver the big project that changes everything. The terminology they use when they talk about "generations" is telling, as each new project must be a distinct break with the past, even if it takes a human generation to deliver on it.

Have a look at the link above, and tell me what you think of it.
 

bob231

War Hero
While I agree with the skill set and mind sets you mention, the 20 years gap is where huge differences can be made imho.

Right now, all over the world things get changed very rapidly.
If we considered a ‘thing’ being made, it will have a design spec and a parts list.
With a few clicks of a mouse the design can be changed. That would have been the same with previous things being made since CAD came into existence.

For me the game changer is when those clicks send the info direct to the CNC machines or printers making the parts, send the info to the automated parts warehouse and send the drawings or assembly instructions direct to the shop floor.

With the above system a worker on the shop floor can be adding a new component to the assembly line within a few hours of the design being completed, and using different parts that already fit and meet the new spec.

At the same time servicing schedules and spare parts storage/ordering will also have been altered.
What CAN be achieved in less than one shift would have taken a few month two decades ago.

The advent of plastic and metal printed parts can cut months from production of individual parts due to not needing to design and order jigs.

As per an earlier comment, the big question is whether the USA can actually get their industrial complex to use the new methods NGAD has promised.
At the risk of sounding continuously critical: this can be made to work, but it's very inefficient, and selling that inefficiency to the buyers (typically not very educated in manufacturing engineering) is hard.

Changing the design is easy; virtual testing of the design is an exercise in computing power only. Expensive but practical and allows for lots of design iterations without actually cutting metal.

As noted above, 3D printed parts are only sometimes applicable, though the production methods can be used to speed up making the jig too! They also have the appeal - for industry - of being multipurpose, though the state may find itself wearing some more capital costs to encourage takeup.

More broadly, though, you're missing how industry actually works. The automated parts warehouse and maintenance schedule may exist, but the industry supplying them probably doesn't: unless what you want is very similar to what you had and doesn't involve buying in anything not in your warehouse*, you might be able to rapidly reconfigure your production line to sit idle for two weeks while you wait for your suppliers to catch up.

Now, working on the assumption that every other aspect of the production line can be similarly reconfigured, you can reduce this time rapidly. Unless your suppliers are currently using their plant to make something else, because you've switched supplier to avoid vendor lock**. NGAD had corcumvents this - as described - by regularly ordering tranches of aircraft, though you hit the same problems as noted above with cost inefficiency through not committing fully.

What has also changed over the past twenty years is the complexity of aircraft, which now combine scarily cutting edge technology with pushing ever closer to the limits of current technology paradigms. That makes every aspect of the design less flexible, more prone to failure and more complex to change (second, third and n-th order effects are more likely to be significant). What you're describing is emphatically true for most consumer products, where getting the design right is typically unimportant compared to being able to rapidly change features or shift to respond to the market. This just isn't true for the military world***.

The reality is that substituting rapid changes of technology for good strategy and capable commanders is a red herring. It cannot be achieved at the level at which we want to play, barring a genuine change in the "combat aircraft" paradigm.

*Typically your warehouse doesn't exist, because Lean Manufacturing and nobody recognising the military demands make Lean impossible.

**This carries significant additional risk, which what I've read of the NGAD programme does not consider.

***The maritime equivalent is the scourge of modularity, which involves adding significant topweight and additional complexity to a Ship to allow it to supposedly be slightly better at one of a number of secondary roles, on the occasions you bring the right module to the last place to change it, or building a Ship which is worse at everything to make it more modular (see HDMS Absalon). Land is less vulnerable, although you can see a different-but-related problem in the lag between introducing a new rifle and it being effective.
 
Another example of new tech potentially making life easier are laser scanners that can measure items in 3D.
Reverse engineering a part so it can be modelled by computer now takes hours instead of days/weeks.
Good call, when we first tried in probably late 90’s early 2000’s the biggest issue was you had to get a paper drawing to measure it and sign it off as to spec. Couldn’t do it with the technology then so we couldn’t certify a flight worthy laser printed part.

The laser sintered a wax and what emerged you put through the casting process and melted the wax out of the ceramic and poured in the aluminum.
 
Yes, well back when the F-35 project was started the trendy new thing in management consulting circles was the Toyota Production System. So, that's what the consultancies sold the US defence ministry. "Buy our consulting services and training packages and we'll teach your project managers how to apply this magical system so you too can increase quality while reducing cost."

Now the "Toyota production system" is passée and the consultancies are selling "agile". "Buy our consulting services and training packages and we'll teach your project managers how to apply this magical system so you too can bring projects to completion faster while reducing cost."

The people who created "agile" have said that what management consultancies are selling today is a travesty of the original concept, but none the less, let's just follow through with the idea. Let's see, the Americans have a problem with rampant feature creep extending timelines and driving up costs. What can we do about that? I know! Let's adopt a methodology based around the idea of the customer not really knowing what he wants so we just keep adding features until the customer runs out of money! What could possibly go wrong?

Well, it was kind of on the grown-ups to say "hang on a second, we aren't making cars or websites, these methodologies are simply irrelevant, go away and try again". I mean you can't blame sleazy snake-oil merchants for trying to sell snake oil, that's in their very nature, a leopard can't change its spots, that's why there should be someone in charge instead of letting consultants run hog wild all over your project.

Agile can work if (and only if) you can "mock" all your dependencies e.g. you say "thing A is not ready yet but it's OK, we can simulate enough of it to proceed with working on thing B". Sometimes it is applicable to hardware e.g. the "Blue Circle radar" but 99.9% of the time it is not and everyone knows it is not and only a fool (or a government) would even consider Agile for a project like that.

 

Slime

LE
Or Nimrod air-to-air refuelling. Which didn't cause any problems at all, and we didn't lose an aircraft and fourteen men aboard because a "quick fix" that became permanent turned out to have lethal side effects even if we were fully familiar with Nimrod and knew all about it and had no worries about airworthiness or safety...



If the aircraft catches fire and explodes in mid-air, you can't use it.

I couldnt agree more, and you are just reinforcing the point I was trying to make.

While I had already mentioned the MR4, and it’s wing root saga, the loss of XV230 is a prefect example of an issue being known about yet no fix being introduced..........Even though it could have been sorted.

While the other Nimrod losses (for some bizarre reason a lot of people think only one was lost) included engine issues after servicing etc, XV230 suffered from a well known issue, and due to a non flexible (and non accountable) hierarchy the issue wasn’t dealt with with a proper fix, and was ‘fixed’ by having crew look at the pipes in question during refuelling.

The same system that could have been employed to sort the MR4 wing issue could have been employed on the internal piping.

Conversely, the cheapskate descision to buy J model Hercules without fire suppressant foam in the wings was reversed after the loss of an aircraft, and was yet another decision that cost more in the long run ‘to save money’.
 
Agile software development was invented largely to avoid developing software in the way the F-35 programme did it.

It will be interesting to see how the Aeralis project here pans out, as it's very much a pilot* project for doing things this way ahead of Tempest.

*Try the veal
 
Agile = the jump from frying pan into the fire.
Or, the leap from consent to confusion.
As I've heard it described.
 
Consent sounds like doing it the way it has always been done (except during wartime).

Would you prefer that?
That’s an interesting way of putting it.
One could also present it as a choice between a tried and tested process refined over many iterations to optimise on safety, and an uncontrolled and untested free-for-all.

Which would you prefer?
 
Consent sounds like doing it the way it has always been done (except during wartime).

Would you prefer that?

The problem with a lot of these project management ideas is that the theory is great, but the practitioners thereof are not as studious a disciple as they should be. What's worse, they often peddle the ideas like snake oil ( as has been pointed out ) or the new fashion. People often race and cut more corner than Schumacher , which results in the blokes in the pits having to double up to rescue the project.
 
That’s an interesting way of putting it.
One could also present it as a choice between a tried and tested process refined over many iterations to optimise on safety, and an uncontrolled and untested free-for-all.

Which would you prefer?

Tried and tested, doing it like it always been done, sounds to me like a fear of change.

As I mentioned before, do I think we are ready to jump straight to it? No, my comments were that we should invest in the F-35 (and perhaps Typhoon) until it’s a tried and tested process.

You sound like staying with the status quo of aircraft design and not embracing enhanced technology, which is why I also said, probably a good time to take a number of this generations university leavers to build this concept without the burden of ‘tried and tested’.
 

Slime

LE
That’s an interesting way of putting it.
One could also present it as a choice between a tried and tested process refined over many iterations to optimise on safety, and an uncontrolled and untested free-for-all.

Which would you prefer?

I’ll offer another way to look at it.

The tried and tested process you initially described.
Or
A tried and tested process that has been used successfully for years, but by people who need to count the pennies, or have a quick turn around.

The procurement model NGAD ascribes to is so common that even the average man on the street modifying a car etc can go to a myriad of companies and have parts modelled and fabricated or made at a fraction of the time and cost.

As per an earlier post, the initial methods you describe that optimised safety led to the loss of XV230. We aren’t talking about the loss days, weeks or even months after the piping issue was discovered, we are talking years.

Just to cite the absolute most ‘bog standard’ and down to earth example, and one using a ‘British’ product have a look at this vid.

 
The problem with a lot of these project management ideas is that the theory is great, but the practitioners thereof are not as studious a disciple as they should be. What's worse, they often peddle the ideas like snake oil ( as has been pointed out ) or the new fashion. People often race and cut more corner than Schumacher , which results in the blokes in the pits having to double up to rescue the project.

I am unsure where this ‘project management’ and ‘Consultant Ideas’ are coming from.

From what I have read, it is being pushed onto industry not taken from. Every consultant I have been involved with have only ever peddled someone else’s ideas, eg, Kanban, JIT, Six Sigma all taken from mass production of cars, and in some cases it worked great, in other, not at all.

So which manufacturing industry (not software) are these consultants getting this from
because I have only met ‘experts’ in consultancy with experience and never a leading edge think tank providing next level ideas.
 

Latest Threads

Top