[rescue] LCD monitor diagnosis

[Don Y]

Geoffrey S. Mendelson wrote:
> On Thu, Apr 27, 2006 at 03:12:51PM -0400, Mike F wrote:
>> I wonder if something like this is even feasible; the big problem to
>> overcome in a weight-based system is g-force. 
> 
> It's really a case of it works, don't fix it. People are happy with what
> they have, the guages work to the standard of accuracy people expect
> (no kidding, they standard is based upon them) and everyone knows how
> to fix them.

Yup.  It's not like tracking fuel consumption in an aircraft, etc.

> Let's play a game. We design a fuel guage system that uses ultrasonics to
> map the inside of the tank. It's a bunch of sensor consisting of a microphone
> and a loudspeaker (to use easily understood terms) run by a sequencer and
> a microprocessor. 

Pressurize the tank.  Note the difference in pressure over time...

> If you scan the tank using five sensors in ten seconds, you could use a
> real expensive processor like a Z8 which costs about $2 with the ram,
> rom and analog to digital converter.
> 
> If you went with high performance, like a scan every second or less, you
> would have to go to an ARM processor, which would cost about $25 for the
> processor and the needed extra chips. 

You'd be surprised at how cheap things get when you are talking
BIG volume.  We used to just say we were buying *plastic*
(regardless of the functionality of the Si inside!)

> So now we have a 100% accurate guage of how empty the tank is, and assuming
> no major deformities that occur after manufacturing and calibration,
> the amount of fuel in the tank.
> 
> For legal purposes, I have not investigated the patent history of such
> a device, but if it can be patented hereby I claim patent rights.
> 
> Going on, look at the ramifications of the change. 
> 
> 1. It costs more because it is new.
> 
> 2. It requires certification of some sort or insurance companies won't
>    insure the car.
> 
> 3. It requires training and documentation.
> 
> 4. If a unit costs $10, it will add $100 to the cost of a car.

That may be an arguable point... but, it will obviously add
more than $10 to the car's "selling price" (cost is a funny word)

> 5. If a car with such a unit has a gas tank fire, it will become necessary
>    to defend the device in the court of public opinion as the press will
>    create a story that it caused the fire and run with it
>    (c.f. how Ralph Nader got started.)
> 
> 6. If a housewife runs out of gas, and decides to walk to the gas station
>    and gets raped, murdered, or even mud spattered, a hot shot lawyer
>    will sue.

Yup.  Many "design choices" are based on non-technical issues.
Medical devices being a prime example (in which litigation
and exposure to it are bigger issues than the technological ones).

It's just a matter of doing the cost/risk/benefit analysis and
deciding how *much* you'll gain in the wash.  Any *claim*
(i.e. assertion) you make can just as easily become a liability
as an asset.

> I'm sure there are other issues, but I have to be a daddy (my real job).