Ecoboost torture and teardown

[bob urz]

http://news.pickuptrucks.com/2011/01/what-the-inside-of-a-torture-tested-ecoboost-v-6-looks-like.html

bob

[mike]

On 1/29/2012 8:45 AM, bob urz wrote:
> http://news.pickuptrucks.com/2011/01/what-the-inside-of-a-torture-tested-ecoboost-v-6-looks-like.html
>
>
> bob
Looking at the pictures made me realize that I don't know how cam
chain tensioners work.
I can understand why you'd want one on the slack side of the chain.
But why the other one on the taught side?

[jim beam]

potential for variable valve timing. whether frod bother or not is
another matter, but it's what some other manufacturers do.


--
nomina rutrum rutrum

[jim beam]

On 01/29/2012 08:45 AM, bob urz wrote:
> http://news.pickuptrucks.com/2011/01/what-the-inside-of-a-torture-tested-ecoboost-v-6-looks-like.html
>
>
> bob

"and run for 300 hours to replicate the equivalent of 150,000 customer
miles"

anybody here average 500mph?

and 3000 hours is not a lot - 50mph is a real high average speed - most
vehicles its closer to 30 with traffic, idling, parking, etc. add to
that the fact that at 50mph the engine has good lubrication with full
hydrodynamic separation, and at idle it doesn't, this so-called "torture
test" really is in reality a cake walk compared to what it's going to
encounter in the real world.


--
nomina rutrum rutrum

[jim beam]

On 01/30/2012 07:24 PM, Rebus wrote:

> jim beam wrote:
>
>> On 01/29/2012 08:45 AM, bob urz wrote:
>>> http://news.pickuptrucks.com/2011/01/what-the-inside-
> of-a-torture-tested-ecoboost-v-6-looks-like.html
>>

>> "and run for 300 hours to replicate the equivalent of
> 150,000 customer
>> miles"
>>
>> anybody here average 500mph?
>

> "@Darryl: As the story states, Ford said 300 hours on the
> dyno was the "equivalent" of 150,000 miles / 10 years of
> testing because of the hot/cold shock tests they did to
> the engine while it was on the dyno. The dyno test wasn't
> mile-for-mile."
>
> The engine was also run in Baja and did some other tasks
> in the real world.

yeah, irtfa, thanks for checking. there's no way that 300 hours
replicates 150k miles of real world - it has to be a typo.


--
nomina rutrum rutrum

[Brent]

I am not surprised that you are unfamiliar with accelerated life
testing.

It would be reasonable for 300 hours to be around 10 years of use in
such testing. 150K/10 = 15K/yr.

[jim beam]

but i am - i've done it. but of course, since your agenda is mere
mindless gainsay, you /have/ to manufacture an untruth since you've got
nothing else.

>
> It would be reasonable for 300 hours to be around 10 years of use in
> such testing. 150K/10 = 15K/yr.

using your "math", that's 30 hours of driving a year. completely
ridiculous.


--
nomina rutrum rutrum

[Brent]

Sure you have Jim... and yet you have already demonstrated your
ignorance of what it is. A total unfamiliarity with the theory
and operation. Although I suppose you could have 'done it', as
a know-nothing tech who just did what he was told with no knowledge of
the principles behind or meaning of his tasks.

>> It would be reasonable for 300 hours to be around 10 years of use in
>> such testing. 150K/10 = 15K/yr.

> using your "math", that's 30 hours of driving a year. completely
> ridiculous.

There you go demonstrating your ignorance of what ALT is again.
Accelerated means that each hour under test represents some much longer
time span in actual use.

[jim beam]

if you know the arrhenius equation, you can do accelerated life testing
for things like semiconductors, but it doesn't work that way for
mechanicals.

hydrodynamic separation, what happens when a lubricant "floats" the
elements of a bearing so that they no longer touch, only happens at high
engine speeds. thus you cannot replicate 1000 hours of "city traffic"
by 100 hrs of "freeway" because they're different tribological regimes.

it's like saying that accelerated fatigue testing of mild steel
represents real world, except that it doesn't because, by definition, it
can't give sufficient time for carbon diffusion, which is a known
fatigue crack arrestor.


--
nomina rutrum rutrum

[Brent]

You are the master of misdirection to spackle over your demonstrated
ignorance. Accelerated life testing does indeed work for mechanical
systems, wether or not an arrhenius equation is applicable for the
material or all the materials in question, as all accelerated life
testing is not material aging. Accelerated life tests must be designed
properly for what is being tested.

> hydrodynamic separation, what happens when a lubricant "floats" the
> elements of a bearing so that they no longer touch, only happens at high
> engine speeds. thus you cannot replicate 1000 hours of "city traffic"
> by 100 hrs of "freeway" because they're different tribological regimes.

More misdirection is that is not what is being done here and that
accelerated life testing of bearings does exist.

> it's like saying that accelerated fatigue testing of mild steel
> represents real world, except that it doesn't because, by definition, it
> can't give sufficient time for carbon diffusion, which is a known
> fatigue crack arrestor.

Your attempt to find exceptions that don't work to construct a rule is
nothing more than a futile attempt to misdirect and mask ingorance you
have already put on display. I would not expect an accelerated life test
designed to simulate years of use of the engine to be particularly good
at finding fatigue life of a a stamped carbon steel alternator bracket
anyway. But hey, if exceptions make you feel better have at it. But
ultimately it doesn't change your condition.

[gpsman]

Duh. Straw man. He obviously exhibits understanding of the premise
of ALT.

He finds fault with this specific ratio, which, of course, you avoid
addressing in favor of building upon a false premise of your own
creation.
-----

- gpsman

[mike]

I can't resist getting involved in a pissing contest. So here goes.

Accelerated life testing is great as long as you're accelerating the thing
you're testing.

I judge the reliability of a vehicle by how many times I have to take
it in for repair and how much it costs.

Running an engine on a dyno has zero to do with whether the starter can
start the engine over a period of 20 years sitting outside in the
snow and road de-icer.
Or whether some aerodynamic design sucks wet road grit into an area
with an exposed oil seal or electrical connection.
Or whether wear is increased in the first few seconds after starting
because the oil ain't flowing yet. Or whether the blower has some resonance
that eventually tears the blades apart at 55 MPH.
Or whether the coolant decomposes into something that eats some
gizmo made out of the wrong material. Or whether some seal can't
stand the flexing...you could go on forever....


There are also statistical issues. Test a hundred and you get some
statistics. Test one and you have the choice: Discard the bad result
or FEATURE a good result at an auto show.

Summary.
Poor correlation between things accelerated and failure modes in actual use.
Zero statistics...sample of ONE that wasn't a "lemon".

[Brent]

What makes you think accelerated life testing is limited to just the engine?

There are tests developed for many different aspects. For instance, your
starter would likely be tested by cycling it in temperature chambers and
exposing it to salt spray.

Statistics are typically done in the development program and over the samples
pulled from the line over time. This is just the story of one sample pulled
from the line. Such samples should be pulled out for testing on a regular
basis, as well as other subsystems and entire vehicles.

[mike]

You must have missed the second line of my post. I'll repeat it her so
you can't miss it:

Accelerated life testing is great as long as you're accelerating the
thing you're testing.

We have no details, but I'd bet that an engine running 300 hours on a dyno
wasn't started very many times. And they didn't spray road de-icer on
the engine.

The question wasn't whether the starter was tested. It probably had
its own test plan. (I'm on my fourth starter on my Honda Motorcycle.
Somebody missed the boat on reliability testing that one.)

The claim was that
the engine was reliable. And MAYBE the part they torture tested was
reliable...
In this case...sample of ONE.
You think they would have published the results if they'd been bad???
You think they might have avoided risky reliability scenarios?

You've probably seen the hot actress on TV claiming that she lost
100 lbs. on some fad diet. There's no disputing the results for
THAT case. But you'd have no trouble finding hundreds of others who
had different results. They advertise successes, not the many more
failures. There's always that YMMV disclaimer that absolves them
of any responsibility for misleading you.

I'd be careful drawing global conclusions from a carefully scripted
marketing extravaganza based on ONE sample.

Encouraging result, yes.
Proof of long-term reliability, not so much.

How about a side-by-side identical test on a Chevy and a Toyota and...and...

And how about we drive 'em over curbs and drag the pan on rocks
and haul sand for a while and get 'em stuck in the mud up to the
frame and all the other stuff that gums up the works?

Hauling a trailer across country ain't the same as hauling it
the same number of miles around town in traffic.

I once loaned my car to a roommate. He got it stuck in the mud
and got the clutch so hot that the steel starter gear fell off
the aluminum flywheel. Think that would have showed up in
an accelerated test?

I do loves me a good pissing contest....

[Brent]

I read it just fine. Then you went on about how it was only the engine
tested. So don't expect to save face by pulling that out of the context
you created around it.

> We have no details, but I'd bet that an engine running 300 hours on a dyno
> wasn't started very many times. And they didn't spray road de-icer on
> the engine.

What does that have to do with "jim beam" and yourself not understanding
how products and parts there of can be tested to simulate years of use
in weeks or less? This is the same context you used before, that somehow
because every single test wasn't demonstrated for you, you act as if
they do not exist.

> The question wasn't whether the starter was tested. It probably had
> its own test plan. (I'm on my fourth starter on my Honda Motorcycle.
> Somebody missed the boat on reliability testing that one.)

Could be a variety of factors from component manufacturing quality to
business decisions regarding known factors. You'll have to read Honda's
internal documents to find out.

> The claim was that the engine was reliable.

I neither made nor defended such a claim. You might want to take an
argument on that up with the voices in your head, but I have no opinion
either way regarding the reliability of that engine.

> And MAYBE the part they torture tested was
> reliable...
> In this case...sample of ONE.

Since I was pointing out that "Jim Beam" doesn't know what he was
babbling about and not drawing conclusions, what does sample size have
to do with anything here? If you are concerned with sample size you'll
need to get your hands on internal specifications regarding development
testing and line sampling and testing. However sample size is irrelevant
to pointing out that someone doesn't know what they are talking about
when they claim X amount of use can't be simulated in Y about of time.

> You think they would have published the results if they'd been bad???
> You think they might have avoided risky reliability scenarios?

A typical production line sample test is the same or lesser test than
what is done during development, so yes it was low risk because
samples of these engines were already tested and passed these things and
more in development.

> You've probably seen the hot actress on TV claiming that she lost
> 100 lbs. on some fad diet. There's no disputing the results for
> THAT case. But you'd have no trouble finding hundreds of others who
> had different results. They advertise successes, not the many more
> failures. There's always that YMMV disclaimer that absolves them
> of any responsibility for misleading you.
>
> I'd be careful drawing global conclusions from a carefully scripted
> marketing extravaganza based on ONE sample.

You would have noted, if you had decent reading comprehension, that I
drew no conclusions from the test. Only pointed out that as usual "Jim
Beam" didn't know what he was babbling about.

> Encouraging result, yes.
> Proof of long-term reliability, not so much.
>
> How about a side-by-side identical test on a Chevy and a Toyota and...and...
>
> And how about we drive 'em over curbs and drag the pan on rocks
> and haul sand for a while and get 'em stuck in the mud up to the
> frame and all the other stuff that gums up the works?
>
> Hauling a trailer across country ain't the same as hauling it
> the same number of miles around town in traffic.
>
> I once loaned my car to a roommate. He got it stuck in the mud
> and got the clutch so hot that the steel starter gear fell off
> the aluminum flywheel. Think that would have showed up in
> an accelerated test?
>
> I do loves me a good pissing contest....

I really don't give a shit about about your personal experiences and
silly tribal make arguments. Let me know when you learn how to read and
comprehend rather than make attempt lame misdirections like that above.

As to your touching story of you being foolish enough to loan your car
to someone who not only got it stuck but then proceeded to burn up the
clutch, it was probably a business decision assuming that all the other
customers didn't want to pay extra to keep the ring gear on the flywheel
when someone burned up clutch by slipping it exceesively.

[jim beam]

no shit sherlock. and you're blowing my mind with your ability to get
sidetracked. the arrhenius equation is seldom if ever used in
mechanical life testing analysis. the context in which it is relevant
however, just went WHOOOOOSH.

> Accelerated life tests must be designed
> properly for what is being tested.

and running an engine at an average of 50 mph isn't it - for the reasons
stated before.

>
>> hydrodynamic separation, what happens when a lubricant "floats" the
>> elements of a bearing so that they no longer touch, only happens at high
>> engine speeds. thus you cannot replicate 1000 hours of "city traffic"
>> by 100 hrs of "freeway" because they're different tribological regimes.
>
> More misdirection is that is not what is being done here and that
> accelerated life testing of bearings does exist.

<sarcasm> really? i never knew that!!! </sarcasm>

now go ahead and state what are the different tribological regimes that
would be experienced real world [hint - i've already outlined part of
this], and how you would do accelerated testing for them. and do so for
each of the bearing types used.

>
>> it's like saying that accelerated fatigue testing of mild steel
>> represents real world, except that it doesn't because, by definition, it
>> can't give sufficient time for carbon diffusion, which is a known
>> fatigue crack arrestor.
>
> Your attempt to find exceptions that don't work to construct a rule is
> nothing more than a futile attempt to misdirect and mask ingorance you
> have already put on display.

as opposed to your "contributions"? right.

> I would not expect an accelerated life test
> designed to simulate years of use of the engine to be particularly good
> at finding fatigue life of a a stamped carbon steel alternator bracket
> anyway.

red herring. you're wriggling and squirming because you obviously don't
understand the implications for more critical engine components.

> But hey, if exceptions make you feel better have at it. But
> ultimately it doesn't change your condition.

your condition is that of a rodent.


--
nomina rutrum rutrum

[Brent]

Once again Mr. Beam accuses the other person of irrelevant the side
track he created.

>> Accelerated life tests must be designed
>> properly for what is being tested.

> and running an engine at an average of 50 mph isn't it - for the reasons
> stated before.

Who stated that was done? Nobody. Once again you interject irrelevancy
to try to spackle over your ignorance.

>>> hydrodynamic separation, what happens when a lubricant "floats" the
>>> elements of a bearing so that they no longer touch, only happens at high
>>> engine speeds. thus you cannot replicate 1000 hours of "city traffic"
>>> by 100 hrs of "freeway" because they're different tribological regimes.

>> More misdirection is that is not what is being done here and that
>> accelerated life testing of bearings does exist.

><sarcasm> really? i never knew that!!! </sarcasm>
> now go ahead and state what are the different tribological regimes that
> would be experienced real world [hint - i've already outlined part of
> this], and how you would do accelerated testing for them. and do so for
> each of the bearing types used.

I'm not going to follow your attempt at saving face by derailing the
conversation into something else. Your further attempt to hide your
ignorance and save face is noted.

>>> it's like saying that accelerated fatigue testing of mild steel
>>> represents real world, except that it doesn't because, by definition, it
>>> can't give sufficient time for carbon diffusion, which is a known
>>> fatigue crack arrestor.
>
>> Your attempt to find exceptions that don't work to construct a rule is
>> nothing more than a futile attempt to misdirect and mask ingorance you
>> have already put on display.

> as opposed to your "contributions"? right.

Lame attempts at deflection doesn't change anything. And since you're
not here to contribute, but simply insult people to boost your ego and
troll to get your jollies... well...

>> I would not expect an accelerated life test
>> designed to simulate years of use of the engine to be particularly good
>> at finding fatigue life of a a stamped carbon steel alternator bracket
>> anyway.

> red herring. you're wriggling and squirming because you obviously don't
> understand the implications for more critical engine components.

HAHAHAHAHAHAHAHAHAHAAHA! Nice projection there Beamo.

>> But hey, if exceptions make you feel better have at it. But
>> ultimately it doesn't change your condition.

> your condition is that of a rodent.

*yawn* the true level of your discourse is reached again.

[jim beam]

piss and moan. rodent.


--
nomina rutrum rutrum

[Brent]

HAHAHAHHAHAHHAHAHAHA!

You need some new tricks.

[C. E. White]

"bob urz" <so...@inetnebr.com> wrote in message
news:jg3t39$9hm$1...@speranza.aioe.org...
> http://news.pickuptrucks.com/2011/01/what-the-inside-of-a-torture-tested-ecoboost-v-6-looks-like.html
>
> bob

There are errors in that artice. Instead of getting it second hand, why not
get it from Ford directly?

http://www.ford.com/trucks/f150/experiencef150/?section=torture:home/ -
there are movies here with a voice over to explain the testing - mostly pure
advertising copy.

http://media.ford.com/article_display.cfm?article_id=33286
http://media.ford.com/article_display.cfm?article_id=33603

In the links above Ford claims the dyno testing was equivalent to 150,000
miles of use, but they also clearly state it only took 300 hours. They claim
the dyno test is equivalent to this by doing all sorts of stresses (hot /
cold cycling, full throttle, etc.). But the video does clearly say they are
"simulating" 150,000 miles of use. This is advertising hyberbole. I am sure
their test is really, really tough but 150,000 miles of use would require
at least 3000 hours which is 125 days. Ford's dyno test was only 300 hours.
300 really tough hours, but I don't think the test really simulates 3000
hours of use in the "real world." In some respect the Dyno tests, as
described, are proably worse than typical real world usage, but in my
opinion, there is no substitute for time when you are trying to prove
something is really tough. On the otherhand, I know a couple of users who
would probably exceed any torture these tests subjected the engine to.

To me this whole "torture test" was an advertising stunt. I suppose it is a
good one and it is more "real world" than some of thos silly Tundra
advertsing stunts, but it is still just a stunt.

I suppose these sorts of stunts impress some people, but do they really
change peoples' minds? I am on a Ford specific mailing list and several
people on that list have F150's with the Ecoboost engine. All seem
enthusiastic, BUT, I'd like to see how happpy they are in three or four
years.

Ed

[Brent]

Such a test simulation could go either way. It usually depends on how
old the test is and how well feed back from the real world has shaped
the test over time. An old test with good feedback over the years will
be pretty good. A new test can be wildly too harsh or missing key
elements.

These tests aren't just for giggles either, if done properly and
using historical data of real world use they can be used to predict
warranty costs.

> To me this whole "torture test" was an advertising stunt. I suppose it is a
> good one and it is more "real world" than some of thos silly Tundra
> advertsing stunts, but it is still just a stunt.

I believe they took a standard test and turned it into an advertising
stunt. It's a fairly standard practice to a take a production sample and
subject it to a development test or an abbreviated version there of. The
advertising stunt was what they did with the engine after the dyno run
was completed.

> I suppose these sorts of stunts impress some people, but do they really
> change peoples' minds? I am on a Ford specific mailing list and several
> people on that list have F150's with the Ecoboost engine. All seem
> enthusiastic, BUT, I'd like to see how happpy they are in three or four
> years.

Nah... 10-20 years ;)