electronic throttle

News, thx for link. According to this the dodginess of my memory is
certainly confirmed.

Let's see what clare@snyder says. In my interpretation gelling would imply
the precipitation of wax.

DAS

To reply directly replace 'nospam' with 'schmetterling'
--

 

Good point.

I like to read the whole technical report as opposed to what was
published in a paper.

JT

(Who remembers years ago a VW beetle icing up - A little "dry" gas took
care of the problem quickly)

 

DAS wrote:

Link to the full AAIIB report:

http://www.aaib.gov.uk/publications/formal_reports/1_2010_g_ymmm.cfm

No, that was suspected pitot probe icing, but no conclusive report has
been issued and the search for the bulk of the wreckage continues.

 

Have at it:

http://www.aaib.gov.uk/publications/formal_reports/1_2010_g_ymmm.cfm

Advise when able.

 

Hah! Yeah - like LED assembly replacements for incandescents that you
buy off of ebay that use LED's designed for maximum current of X, and in
the assembly they each dissipate 1.3X so they can advertise brightness
and compete on an even footing with their competition that is doing the
same thing. So what if the LED's fail in 6 months - they have your
money and you already gave them rave reviews.

 

So there was water in the fuel? There was an AD put out on those
engines requireing the replacement of the fuel pre-heater heat
exchanger is all I know "for sure".
I also know diesel and Jet A can gel in the cold, which is why
fuel-heaters are used on diesels, and I assume why they are used on
jets.

 

Idle Speed Control

They fail from overheating, generally - and then USUALY from being
under-spec'ed.
Capacitors have a finite life - but using a 135C rated cap instead of
an 85C rated cap makes it last WAY longer. Something like double for
every 5 degrees below spec, and half for every 5 above. Not 100% sure
of the number - but very close to that from what I remember. And that
is for aluminum electrolytics.
Solid tantalum and dry Mylar caps last virtually forever if not run
way over the rated voltage.

Any discrete component run too close to the limit in a vehicle control
system constitutes BAD engineering. Virtually all automotive
electronics spec 135C rated parts, ar at the VERY minimum, 105C.

Commercial grade is 85C, and they fail in personal computers a lot
faster than the higher grade 105s.

 

Quite pssibly how they do it - I was just showing a "simple" example.

 

Both Jelling and Gelling are correct english usage.
Jell-
to become clear, substantial, or definite; crystallize: The plan began
to jell once we all met to discuss it. –verb (used with object). 3. to
cause to jell. ...

 

Correct. Most common cause of fuel gelling (or jelling)

Truckers often add a "ant-gel" (not french for anti-freeze) additive -
Cheapskates just throw in a bit of gasoline.
Most over-the-road diesel trucks have a fuel heater and circulate
heated fuel back into the tanks to keep them from setting up in cold
weather.

 

I had my 49 beetle vapour lock and ice up on the same one-day trip
with weather in the high 80's F ( and roughly 100% humidity)

 

Looks like they suspect water in the fuel, but it does not fully rule
out Gelling (the fuel WAS at -10C) because they only identified
"probable" causual factors.

 

You Do know that pulsed LEDs CAN run at significantly over rated
current almost indefinitely, providing much brighter output than
normal, with no visible flicker??
As long as the LED is not caused or allowed to OVERHEAT, it will last
virtually forever.

Jim Weir of RST Engineering has demonstrated this and provides some
neat "overdriver" circuits.

 

There is ALWAYS water in fuel.

 

Good luck with your accident investigation career.

 

Oh - no problem. I was just having fun thinking it thru with you. I
have no idea how *they* do it. But that's how I would do it.

 

Different situation, but I had a car overheat in what is called a winter
hurricane in CO in sub-freezing whether one time - the snow blew so hard
that it clogged the radiator fins, and the engine overheated. Reminded
me of "Oh Suzanna" - "The sun so hot I froze to death", except the opposite.

 

Yes - you pulse it at higher *instantaneous* current, but you duty cycle
it such that the *average* power is within what it can tolerate. BUT -
the human eye registers the pulse peaks more than the average, so it is
a trick to run it within the rated average power but the eye *perceives*
it as more power than it really is.

 

A general rule of thumb for practically *all* areas of physics: Rates
double for every 10°C rise. That is true of chemical reactions, as well
as thermal characteristics of electronics - and is true for life and
failure rates of aluminum capacitors. So in general - you cut the life
of an aluminum capacitor in half with a 10°C rise in operating temperature.

Glad you finally qualified that. If you hadn't, I was going to.

I will say that tantalums have *very* little tolerance for much
over-voltage at all (and, like aluminums, are polarized - i.e., can't be
reverse biased by more than a volt). But, like you say, if kept below
their rated voltage will last virtually forever. I design in tantalums
a lot - except that the supply of tantalum in the world is finite, and
prices constantly creep up and up over the years (relative to other things).

Don't forget ceramics - they have their place, but there are many
different types of them, each with their own pros/cons - to many
variations to discuss.

There are other plastic types too - each with it's own special properties.

Also good idea to protect components from over-voltages/transients with
some series resistance/impedance plus Transorbs™ (zener diodes
especially designed for clamping over-voltages/spikes). Not unusual to
have spikes over 100 volts in automotive systems.

Absolutely - several well-known computer "manufacturers" (Dell for one)
got burned about 3 (IIRC) years ago when cheap aluminums got put into
their MoBo's. Cost them a pretty penny to do the warranties on all
those failures - they lost some customers over that too.

I avoid using aluminum electrolytics in my designs if at all possible.
But in cases where they are otherwise the best solution (tantalums get
impractical at larger capacitance values), I **only** use
super-high-reliability brands and types within the brand (they are rated
for so many hours within the temp. rating) - and spec. the voltage
rating for well over what they will actually see. I would not use an
aluminum cap. from Radio Shack in *anything* except for a temporary
breadboard experiment.

 

i follow it, but i don't see the point. i've had a tps sensor go out,
and i don't see how this would get around the worn wiper connection that
caused that failure. and i don't see how having two code loops look at
the same sensor helps at the computer either since the wiper is common.
to get "redundancy" you need two [or more] sensors. two [or more]
computers too.