Ducati goings on. Part 2.

[Oscar]

The "tuneability" advantage of c/f was never exploited, it was scapegoated by the pudits as the "obvious" reason; logical fallacy: alien material = poor results...correlation = causation. As this ongoing trainwreck continues, whatever their unresolved undelying issue(s) may be, perhaps the ONLY thing Ducati has definitively proven is.... their problem is/was not frame material(s).

Mea culpa, I did not express myself thoroughly. I'm not pointing at the c/f frame as the culprit. What I was trying to say was - and in agreement with your point that the tuneability of the c/f frame appears to never have been properly exploited - Preziosi implied that by using that tuneability they could produce bikes with greatly varying characteristics.

The fact that the 'great leap forward' in '10 was from Stoner changing the forks for more flexible ones points to the bike as a whole being an aggregation of parts that did not work well together - forks, frame and swing-arm. The 'bend like a tree' philosophy sees the entire mechanism operating in harmony from one contact patch through to the other, with the influences of all of the elements that contribute to the reaction of the mechanism taken into account. Yamaha are the masters of that; Ducati seem to be at the other end of the spectrum - parts of the mechanism (mostly)work very well part of the time but the incredible number of crashes for Ducs overall in 2010 says that for some of the time they did not work at all well - they would combine to produce a bike that was highly unpredictable ('exactly wrong'...)

As to the comment that the '12 bike was 'cobbled together': (from the Manzania article quoted above)

[Tourn46]

As to the comment that the '12 bike was 'cobbled together'

Duct Tape is the material of Champions.

Especially when in a rush because Mama's Bolognese is getting cold.

[JoeKing]

The "tuneability" advantage of c/f was never exploited, it was scapegoated by the pudits as the "obvious" reason; logical fallacy: alien material = poor results...correlation = causation. As this ongoing trainwreck continues, whatever their unresolved undelying issue(s) may be, perhaps the ONLY thing Ducati has definitively proven is.... their problem is/was not frame material(s).

Mea culpa, I did not express myself thoroughly. I'm not pointing at the c/f frame as the culprit. What I was trying to say was - and in agreement with your point that the tuneability of the c/f frame appears to never have been properly exploited - Preziosi implied that by using that tuneability they could produce bikes with greatly varying characteristics.

The fact that the 'great leap forward' in '10 was from Stoner changing the forks for more flexible ones points to the bike as a whole being an aggregation of parts that did not work well together - forks, frame and swing-arm. The 'bend like a tree' philosophy sees the entire mechanism operating in harmony from one contact patch through to the other, with the influences of all of the elements that contribute to the reaction of the mechanism taken into account. Yamaha are the masters of that; Ducati seem to be at the other end of the spectrum - parts of the mechanism (mostly)work very well part of the time but the incredible number of crashes for Ducs overall in 2010 says that for some of the time they did not work at all well - they would combine to produce a bike that was highly unpredictable ('exactly wrong'...)

As to the comment that the '12 bike was 'cobbled together': (from the Manzania article quoted above)

I should be the one giving the mea culpa for deliberately distorting your point to recrudesce one of my pet peaves....my apology sir..

As a cyclist I have seen that sport literally transformed with the almost complete implementation of c/f bikes at all levels of that sport & expected a similar revolution in MotoGP. Ducati's failure to achieve any advantage is both disappointing & confusing considering your point re:Yamaha's "bend like a tree philosophy"; a c/f structure is better suited to do so than an aluminum one. I believe that absent their as yet undiagnosed "fundamental" problem(s) c/f could have potentially paid huge dividends...we'll probably never know.

I am saddened that Ducati's failure is being unjustly attributed in large part to the c/f (convenient cop-out) & more disturbingly, will forever banish structural c/f from MotoGP, closing an avenue of innovation & creativity.

[Oscar]

I am saddened that Ducati's failure is being unjustly attributed in large part to the c/f (convenient cop-out) & more disturbingly, will forever banish c/f from MotoGP, closing an avenue of innovation & creativity.

c/f is an amazing material and the lay-up, resin and cure-cycle possibilities give it by far the best range of variation of anything I know - but it does have problems with crash-worthiness when also used to maximise its strength/lightness/stiffness possibilities. I've seen the end result of an aircraft crash that would have been quite survivable in a 'glass-composite / aluminium aircraft but was a horrible double-fatality - the damn thing literally exploded, you couldn't have gotten more disintegration by dropping several hand-grenades into the cockpit.

I believe the problem was with the integral airbox idea - thus producing a box-beam of high stiffness in the lateral plane because it was a short structure attached to a wide and inflexible base: the ends of the heads. The fact that the alloy version made sod-all difference suggests that it was not material but design that was at fault. Pretty obviously, the Panigale is a narrower box-beam because it attaches (I assume) to only a single-width head, not a twin, roughly halving the size of the lateral web for much the same overall length.

C/F bicycles are still comprised of long, thin beams (AFAIK) that allow bending and thus high-stress points to be taken out over a considerable distance (relative to the cross-section of the beams). I suspect that a c/f perimeter frame concept could work well for motorcycles too - possibly quite brilliantly, though the problem of validating its structural integrity after a crash remains one that requires almost laboratory-grade facilities to achieve.

[Domino]

C/F bicycles are still comprised of long, thin beams (AFAIK) that allow bending and thus high-stress points to be taken out over a considerable distance (relative to the cross-section of the beams). I suspect that a c/f perimeter frame concept could work well for motorcycles too - possibly quite brilliantly, though the problem of validating its structural integrity after a crash remains one that requires almost laboratory-grade facilities to achieve.

CFRP bicycles look the way they do for one reason only, sporting regulations require bicycles to "look" like classic bikes (they required a seat downtube). This was a reaction to the trek y-foil which was a huge leap forward in performance but was killed off with regulations.

[JoeKing]

I am saddened that Ducati's failure is being unjustly attributed in large part to the c/f (convenient cop-out) & more disturbingly, will forever banish c/f from MotoGP, closing an avenue of innovation & creativity.

c/f is an amazing material and the lay-up, resin and cure-cycle possibilities give it by far the best range of variation of anything I know - but it does have problems with crash-worthiness when also used to maximise its strength/lightness/stiffness possibilities. I've seen the end result of an aircraft crash that would have been quite survivable in a 'glass-composite / aluminium aircraft but was a horrible double-fatality - the damn thing literally exploded, you couldn't have gotten more disintegration by dropping several hand-grenades into the cockpit.


Sad story..but it sounds like the designer forgot what "composites" mean & focused too much on strength & lightness at the expense of crash worthiness. F1 cars have c/f crumple zones which work quite effectively. As long as min. weights are in place, MotoGP frame designers would have no reason to lighten frames potentially compromising strength or crash worthiness.


I believe the problem was with the integral airbox idea - thus producing a box-beam of high stiffness in the lateral plane because it was a short structure attached to a wide and inflexible base: the ends of the heads. The fact that the alloy version made sod-all difference suggests that it was not material but design that was at fault. Pretty obviously, the Panigale is a narrower box-beam because it attaches (I assume) to only a single-width head, not a twin, roughly halving the size of the lateral web for much the same overall length.



If memory serves this whole topic was explored in depth on this thread. If both the alloy & c/f integral airbox frames & the current conventional alloy frame ALL exhibit similar handling anomolies, logic would suggest the problem lies elsewhere.




though the problem of validating its structural integrity after a crash remains one that requires almost laboratory-grade facilities to achieve.

I would think that frames could be sent out for inspection or an in-house facility could be built. I'm curious as to what plans Boeing has made for inspection after a crash or even for routine inspectionss after x-number of landing/take-off cycles.

[Zaphod]

I thought the issue with the CF frame on the waddling Duck was that it was too short, therefore too stiff where it needed to be flexible (laterally), and lacked any sort of adjustment options.

.............the lack of length, and adjustment options being due to the engine layout/configuration.......a layout/configuration that for some benine reason, they are stubbonly oposed to changing in any way, shape or form.

Hence the same issues getting carried from one bike to the next.

I have interperated JB and Rossi's request for the TSA frame as an attempt to build some adjustment into a bike that, by nature of it's "stick-in -the-mud" engine design, has none. I belive, from reading everything here and elsewhere, that they have achieved this aim, and have made the bike slightly better.........but it's still no good, and suffers from the same issues to a large degree.........because of the obstinant engine configuration.


Pretty clear, I thought.........and also explains Rossi's "Oh !,..who can be effing botherd !! " attitude.

........which, although much more well controlled, appeared to be the same attitude Stoner had at around the same time he must have been talking to HRC.

There is politics and personalities/ego's at play, which are things that can be drageed around and opined on for eons, but to me, the basic nuts and bolts of the problem is what I have written above.

Maybe, because of their passion and patriotisim, both to country and design, "Too many Italians spoil the bike".

They will get there in the end................look at Cagiva, it only took twenty years, but even they started to get somewhere..................in the end.

[Oscar]

I would think that frames could be sent out for inspection or an in-house facility could be built. I'm curious as to what plans Boeing has made for inspection after a crash or even for routine inspectionss after x-number of landing/take-off cycles.

Yep, both of those options are entirely possible, but you have to strip out the frame to do it - not an impossible ask for a motoGp team, even between warm-up and race start. Once you have built up a body of data, strain gauges at strategic locations might alleviate much of that necessity, also - but it's an iterative process. As for Boeing, commercial aircraft are required to have very frequent inspections (such as found the cracks in the Airbus 380 rib/skin joints) and Boeing builds up an incredibly comprehensive data base that relates structural problems reported with loading profiles (they can even tell their customers which pilots land hard / pull more G's in turns etc.). Every Boeing heavy has in-flight recorders for which the data is sent to Boeing, and they can then inform their customers what to replace on which aircraft according to the use it has had - and that has been going on since at least the 747 introduction, as far as I know, so is not a reaction to a specific material.

[dave_m]

In hindsight it doesn't look like the CF itself was the problem, as much as Ducati's general design concept. It may take a couple more race weekends to find out for sure, but the reports indicate the random lowside problem from the previous years has been fixed, so if they stick to an aluminum twin spar frame, it's hard to see how they'll bring the CF back in. I'm assuming you can't do a twin spar frame in CF...?

[TwoStroke Institute]

The theory falls flat on it's arse about taking months to build a new engine, regular main page contributor Chris Cosentino CNC machined a set of cases in the same time, in between work and family.
Ducati should be able to knock out a set of cases overnight. Budget I would believe.

Machining isn't the problem. It's the CAD that precedes it that causes the delay. Usual lead time for a fully-fledged race engine is 2 years - that is, a race engine that could compete at MotoGP level. More to it than just machining.

I find that hard to believe, clean sheet designs maybe but rarely do clean sheet designs occur most will follow on using components that are known to work.
Having only the benefit of a beginners belt in Soldworks and MasterCAM it pretty much does it all for you, FEA, components in motion, sections, casting cores etc etc, married with a decent engine simulator a single cylinder test mule would be a week away, not months. The time is taken finding out what is going to break and refining the power output/delivery.

http://www.youtube.com/watch?v=IL2HxmHV9gI

http://www.youtube.com/watch?v=Uq7A8pSC ... re=related

[Oscar]

I find that hard to believe, clean sheet designs maybe but rarely do clean sheet designs occur most will follow on using components that are known to work.
Having only the benefit of a beginners belt in Soldworks and MasterCAM it pretty much does it all for you, FEA, components in motion, sections, casting cores etc etc, married with a decent engine simulator a single cylinder test mule would be a week away, not months. The time is taken finding out what is going to break and refining the power output/delivery.

If it's only a new crankcase to narrow the vee-angle (or still to use the 90-degree V, but with the bottom of the case level and a proper oil pick-up mechanism, as Manzania suggests) then 3D printing can produce a casting master ( Duc use cast cases) very rapidly. Sufficiently large 3D printers are available, and using vacuum casting tecniques, the quality of the resultant casting is extremely good and subsequent machining can be done with confidence that voids won't occur to stuff up the production rate (don't ask how I know this, suffice it that it is being done here in Oz for major engine components by a specialist engine maker.) The cost of that process is viable for low-production volumes.

[Sideswipe]

The theory falls flat on it's arse about taking months to build a new engine, regular main page contributor Chris Cosentino CNC machined a set of cases in the same time, in between work and family.
Ducati should be able to knock out a set of cases overnight. Budget I would believe.

Machining isn't the problem. It's the CAD that precedes it that causes the delay. Usual lead time for a fully-fledged race engine is 2 years - that is, a race engine that could compete at MotoGP level. More to it than just machining.

I find that hard to believe, clean sheet designs maybe but rarely do clean sheet designs occur most will follow on using components that are known to work.
Having only the benefit of a beginners belt in Soldworks and MasterCAM it pretty much does it all for you, FEA, components in motion, sections, casting cores etc etc, married with a decent engine simulator a single cylinder test mule would be a week away, not months. The time is taken finding out what is going to break and refining the power output/delivery.

http://www.youtube.com/watch?v=IL2HxmHV9gI

http://www.youtube.com/watch?v=Uq7A8pSC ... re=related

I'm a Mechanical Design Draftsman by trade (although somehow ended up in electronics atm). I would consider myself very competent in almost all CAD packages. You'd be surprised at how long stuff can take to design. Its so much faster if its a simple modification to an existing item (assuming the design intent was captured properly in its construction).

Ducati had ZERO to go off before hand, they couldn't just grab a steel trellis frame and go "Now make this bit aluminium". As to modifications to the engine, it probably would have been easier but still would have required going over with a fine toothed comb look for errors that up with those modifications.

Then there's the "paper" side of things. People often often overlook how important documentation administration is, and it its a fine line between slowing up a job and ensuring its done properly. I'd assume as lead project engineer, Preziosi would have to sign off on every single component design coming from the engineering office. He'd need to check and recheck every calculation and FEA and check how it affects everything else (eg mounting points, component clashes etc.). Prezi has Ultimate responsibility for ALL design choices.

THEN, after the design is finished you have to get on with manufacturing details. Designing Casting Patterns is an entire career in itself. Theres also jig constructions, and machining, welding, assembly drawings and instructions are all extremely important documents to be created.

THEN you have to "Make it real" That means liaising with either internal or external workshops to get your stuff made, and anybody who has dealt with this knows that the turn around is more often measured in weeks than days. And often what you get back ain't what you asked for. Or someone does a simple stuff up. And you need to fix it and re-order it. More time.

Its never a case of "Just rotate the cylinders back.

[Kropotkin]

I find that hard to believe, clean sheet designs maybe but rarely do clean sheet designs occur most will follow on using components that are known to work.
Having only the benefit of a beginners belt in Soldworks and MasterCAM it pretty much does it all for you, FEA, components in motion, sections, casting cores etc etc, married with a decent engine simulator a single cylinder test mule would be a week away, not months. The time is taken finding out what is going to break and refining the power output/delivery.

http://www.youtube.com/watch?v=IL2HxmHV9gI

http://www.youtube.com/watch?v=Uq7A8pSC ... re=related

The top ends would require very little modification, everything north of the piston would be fine, probably - though it would be worth taking into consideration that we are going to have a rev limit for 2013. There may be one or two small packaging considerations, but that's it.

But then there's the bottom end. It's not just a matter of narrowing the V, it's also fining the right place for the balance shafts and weights. FEA needs to be done on the loads the engine + balance shaft produces, and weight pared down to a minimum where possible. These are all fairly obvious jobs, but they are also massively time-consuming. You could knock up something that would work on a test bench in a couple of weeks. But the chances of that engine lasting beyond FP1 are fairly slim. To make it last would take much, much longer.

[Cam D]

I find that hard to believe, clean sheet designs maybe but rarely do clean sheet designs occur most will follow on using components that are known to work.
Having only the benefit of a beginners belt in Soldworks and MasterCAM it pretty much does it all for you, FEA, components in motion, sections, casting cores etc etc, married with a decent engine simulator a single cylinder test mule would be a week away, not months. The time is taken finding out what is going to break and refining the power output/delivery.

http://www.youtube.com/watch?v=IL2HxmHV9gI

http://www.youtube.com/watch?v=Uq7A8pSC ... re=related

The top ends would require very little modification, everything north of the piston would be fine, probably - though it would be worth taking into consideration that we are going to have a rev limit for 2013. There may be one or two small packaging considerations, but that's it.

But then there's the bottom end. It's not just a matter of narrowing the V, it's also fining the right place for the balance shafts and weights. FEA needs to be done on the loads the engine + balance shaft produces, and weight pared down to a minimum where possible. These are all fairly obvious jobs, but they are also massively time-consuming. You could knock up something that would work on a test bench in a couple of weeks. But the chances of that engine lasting beyond FP1 are fairly slim. To make it last would take much, much longer.

The challenges of engine construction (steps involved/time etc) would make for a great interview with Preziosi or one of the engine builders

[Kropotkin]

The challenges of engine construction (steps involved/time etc) would make for a great interview with Preziosi or one of the engine builders

For sure. I'll see what I can conjure up...

[TwoStroke Institute]

Must just be me As a fellow I know made his own balnce shaft that was added to a CR500 engine cases with a minimum of fuss and he works in a coal mine.

There is no die casting in a race shop it is all sand cast with complex cores built up by laser sintering, done overnight straight off the CAD file, no problems at all. Remember Ducati would have some very skilled tradesmen in their in house casting plant. Most times a skilled tradesman can see things a roomfull of engineers fail to see. This core goes together with the water jacket core and the outer to make a RSA cylinder

If I was starting from scratch on a frame I would be carving up a CBR 1000 frame in the bandsaw, which would short cut your FEA by a factor of 10. FTR did the fabrication on the frame, something Ducati just had to make a drawing for.

Finding what is going to go bang on a engine is pretty easy, you hook every sensor you can think of to it, hold it flat out on the dyno and gradualy apply the load. However a 1000cc MotoGP engine would rarely be on WOT for more than 30 sec and even less in the lower gears. The engines rarely go bang these days because of all the CAD work especialy in the area of thermal efficiency, which is what extends engine life.

[Kropotkin]

Must just be me As a fellow I know made his own balnce shaft that was added to a CR500 engine cases with a minimum of fuss and he works in a coal mine.

Was it a 260hp CR500?

[TwoStroke Institute]

Must just be me As a fellow I know made his own balnce shaft that was added to a CR500 engine cases with a minimum of fuss and he works in a coal mine.

Was it a 260hp CR500?

I'll organise you a ride, it will feel like it when your on it Only 65HP per pot as well which even a mild tune CR 500 would be close to.

[Kropotkin]

I'll organise you a ride, it will feel like it when your on it Only 65HP per pot as well which even a mild tune CR 500 would be close to.

Must admit, a CR500 supermoto is close to the very top of my lottery must-buy list. I have a feeling I would not last very long on it, but it would be fun while it lasted!

[Hanuman]

Finding what is going to go bang on a engine is pretty easy, you hook every sensor you can think of to it, hold it flat out on the dyno and gradualy apply the load. However a 1000cc MotoGP engine would rarely be on WOT for more than 30 sec and even less in the lower gears. The engines rarely go bang these days because of all the CAD work especialy in the area of thermal efficiency, which is what extends engine life.

Engine failures are easy to root cause? So easy that HRC grounded their bikes due to a winking red light?
From some of what I read in here, engine design is a matter of a couple of CAD workstations, the student release of some CAD and CFD software and an NC milling machine...
Can you explain the correlation between thermal efficiency and engine life? I'm missing the connection?

[yzr750]

Finding what is going to go bang on a engine is pretty easy, you hook every sensor you can think of to it, hold it flat out on the dyno and gradualy apply the load. However a 1000cc MotoGP engine would rarely be on WOT for more than 30 sec and even less in the lower gears. The engines rarely go bang these days because of all the CAD work especialy in the area of thermal efficiency, which is what extends engine life.

Engine failures are easy to root cause? So easy that HRC grounded their bikes due to a winking red light?
From some of what I read in here, engine design is a matter of a couple of CAD workstations, the student release of some CAD and CFD software and an NC milling machine...
Can you explain the correlation between thermal efficiency and engine life? I'm missing the connection?

If you can keep each component at optimal temperature, whether it is too cool or too hot, it will last longer. Non optimal temperatures are the cause of many component failures that result in catastrophic engine failures.

[Hanuman]

If you can keep each component at optimal temperature, whether it is too cool or too hot, it will last longer. Non optimal temperatures are the cause of many component failures that result in catastrophic engine failures.

Cheers. Undoubtedly. I was interpreting it in the narrow confines of pure thermodynamic efficiency...not heat loss/metallurgy/poor design. But examining failure modes, is loss of mechanical properties due to heat really the root cause of issues...(over, for example, a nice too-early downshift tangling your valves)..or is it a secondary effect? Burning out an exhaust valve comes to mind, but it would be rare to blame the valve material over an over-lean mix.

[Oscar]

Cheers. Undoubtedly. I was interpreting it in the narrow confines of pure thermodynamic efficiency...not heat loss/metallurgy/poor design. But examining failure modes, is loss of mechanical properties due to heat really the root cause of issues...(over, for example, a nice too-early downshift tangling your valves)..or is it a secondary effect? Burning out an exhaust valve comes to mind, but it would be rare to blame the valve material over an over-lean mix.

Can be a complex set of circumstances. For example, I know of one brand of (air cooled) aero-engine that uses a head material optimised for CNC machining, that is JUST ok when operated strictly to temp. limits, but even a slight leaning of the mixture on a hot climb-out can take it over the material limit and cause warpage. It also has problems with cooling the exhaust valve port that then manifests in an overly-tight valve guide that causes valve stem breakage due to excessive friction against a slight side-loading from the lifter which can also result (too often does) in the valve not closing fully, hitting the piston and breaking the head of the valve off... a chain of circumstances that is triggered by a slightly too lean mixture but is a combination of material and design problems all ganging up together. Provided it's operated to limits it's a sweet engine - but totally intolerant of even slight abuse.

[Hanuman]

[quote="Oscar]

Can be a complex set of circumstances. For example, I know of one brand of (air cooled) aero-engine that uses a head material optimised for CNC machining, that is JUST ok when operated strictly to temp. limits, but even a slight leaning of the mixture on a hot climb-out can take it over the material limit and cause warpage. It also has problems with cooling the exhaust valve port that then manifests in an overly-tight valve guide that causes valve stem breakage due to excessive friction against a slight side-loading from the lifter which can also result (too often does) in the valve not closing fully, hitting the piston and breaking the head of the valve off... a chain of circumstances that is triggered by a slightly too lean mixture but is a combination of material and design problems all ganging up together. Provided it's operated to limits it's a sweet engine - but totally intolerant of even slight abuse.[/quote]

Which is (kinda) my point in the original post. Thermal efficiency doesn't necessarily mean durability/robustness. That is, in the narrow terms I held in my excuse-for-a-brain. A iron V-8 running 6:1 compression and baggy clearances vs TSI's (deservedly)beloved RSA engine...Which one would pop into you mind as the robust one, and which would be the one with the thermal efficiency...
[I would classify your aero-engine head as a metallurgy issue. But getting off topic with my semantics]

[Oscar]

[I would classify your aero-engine head as a metallurgy issue. But getting off topic with my semantics]

It is, you are right - but if it were water-cooled it could tolerate wider limits of lean mixture and the exhaust port problems would be minimised, so it's sort of a 'ganging-up' of factors that actually produces the problem: if the sums of airflow plus lean mixture plus material limits goes out of range, whammy. Engines operated always within limits will usually run their 1500-hour TBO (time between overhauls) - yet in real life, they often don't go more than about 350 hours, and can fail with less than 100 hours. My point here relates to TSI's comment that actual design and production can be very quick, but it's finding out what breaks and what doesn't in real life that takes time, even when the numbers all stack up on the computer.. Design genius can sometimes be the difference between seeing the numbers and having the 'feel' for where the weak link in the chain will occur and making the right decision to strengthen the weak link just enough against the competing demands for light weight and performance.