Distortion ?

[enginesrus]

In years past with some engine designs there was some consideration for distortion of various components by various causes. I'm just wondering in this new world of design that is more dependent on computers, how much time is really spent on real physical parts or assembly's in determining distortion? And some examples.

 

[CWB1]

Required testing time certainly has decreased over the years but at the OEM level every component is still analyzed, tested, and understood to the n'th degree, often more as a matter of process driving quality than anything. Not sure if it meets your definition of distortion but the example that comes to mind is pistons. With FEA today we can fairly quickly and easily play all manner of games with piston profiles and see the effects on cylindricity at various temperature and loading conditions to a pretty high degree of accuracy. The standard validation process still typically drives piston scuffing and seizure testing to at least a minor extent however.

 

[BrianPetersen]

The standard validation process only needs to be the last "sanity check". It's a whole lot easier to do all the initial design, and where appropriate the initial trial-and-error and optimization, using finite element methods on computer simulation, than to do cut-and-try the old way.

Finite element methods can tell you the distortion of the piston (let's say) in-cylinder when exposed to whatever the thermal and mechanical stresses are imposed in a running engine. It isn't realistic to measure that physically in real time inside the engine. The old way involved cut-and-try until the engine stopped grenading or burning oil. Whole lot easier to get it right by finite-element methods on the computer.

 

[dgallup]

While I agree that FEA simulation can get you much closer to a nearly optimal solution you still need the final validation testing. Most FEA simulations can only go so far in modeling all the simultaneous real world physics. Taking the piston example, you would need both thermal and structural solvers coupled with fluid dynamics and combustion to really know the exact piston shape in a running engine. I'm sure the F1 teams do that but I'll bet a run of the mill engine design just gets some ball park boundary conditions applied. After all, you car engine spends most of it's time running down the road at 2000 rpm producing 15 horsepower, not at redline producing 400.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[enginesrus]

That is what I thought was going on now, though I would guess due to costs and such that there is very little actual real time testing of engine parts, other than start it and go for it on a dyno and monitoring all the normal parameters that are easy to obtain. I just can't see OEM's spending a bunch of time on coatings and strain gauges for all minor details that they feel past the FEA with flying colors. The only exceptions to this would be maybe manufactures of very large industrial engines, where a small problem can cost huge dollars to resolve once the product is sold and buried deep in ships or generating plants. And explains some of the problems I read about various OEM's are having with their engines.

 

[SwinnyGG]

If anything, more time (a lot more...) is spent on physical testing in 2018 than was spent in 1980/1970/1960/whatever year you want to use as a reference.

The difference is not the amount of time spent on physical testing- it's the goals of that testing.

In 1960 they used long-term dyno tests to determine things like 'will this camshaft break' or 'how long will this oil pump last'.

Now those things are handled in simulation, and the physical testing is used to answer much harder questions, like 'simulation said this additional 4.8% squish volume on the new cylinder head should increase combustion efficiency by 3.2% and reduce NOx formulation by .35% at MBT, is that true?

 

[enginesrus]

It would be so interesting to know what tests if any where performed on engines like the Ford Focus RS, that seems to have a problem area.

 

[SwinnyGG]

"If any"??

Thousands and thousands of hours of physical testing are a guarantee.

The Focus RS head gasket problem has been identified- a batch of engines were assembled with the wrong gasket. The gasket from the 4-cylinder Mustang/Explorer engine is very nearly exactly the same (but with different part numbers). Ford is replacing all affected head gaskets and cylinder heads at no cost regardless of mileage.

 

[enginesrus]

And they are still failing any guesses why?

 

[BrianPetersen]

Head gaskets are failing after the recall work is performed?

Incompetent dealer mechanics? Short-cutting assembly sequences? Not cleaning parts adequately before assembly? There's a ton of stuff that can go wrong with a dealer-performed service procedure as opposed to the comparable procedure being done on the assembly line using new parts and calibrated tooling. And this is a pretty highly-stressed powertrain, which leads to being more sensitive to mistakes.

 

[SwinnyGG]

If people are experiencing head gasket failures after the proper service is performed this would be the first I've heard of it.

I'm sure there's people who didn't immediately drive to the dealer and drop their car off, and have had failures after the bulletin was issued.

 

[RossABQ]

First thing I would ask an RS owner who had a HG failure after replacement of the faulty HG is, "How much did you crank the boost controller?" Second would be "Which aftermarket tune are you running?"

 

[enginesrus]

I've read about some instances. And strikes me funny why such a design for the supposed power levels they are extracting from those engines. Trying to cut costs doesn't always save money in some instances.
Maybe someone at the engineering dept should have looked at this.

http://www.enginelabs.com/news/supporting-your-cylinders-open-semi-closed-or-closed-deck/

 

[SwinnyGG]

'Supposed' power levels?

As in... you don't believe them?