Bob,

 

Thanks for the informational response.

 

Of course the water would either have to be mixed into the fuel or sitting in the sump.  The common way to get water mixed into 100 LL is during fluid introduction into the tank such as when filling up.  I monitor all fill ups but the tanks could be filled up with a water/fuel mix - at least its filled up. It would take a great deal of still water in the wing tanks before any part of the probe was submerged enough for it to register.  This is possible in the always-full header tank, but the float provides backup information about actual liquid making the tank full or not full.  The logic of the always-full header means that it was full of good fuel the last time it was used, thus avoiding sudden surprises on takeoff.  Preflight sump checking should take care of the still water problem whether or not capacitance probes are used. 

 

Fiberglass tanks, unlike metal ones, resist the formation of condensation as a source of water.  In my case (the one I keep defending) my aircraft is always hangared, thus it is unlikely that condensation could affect me at my home airport.  The Lancair is so small that traveling often results in finding hangar space.  However, it certainly has spent overnights parked under the stars, rain clouds and, uh, out in the cold.  That certainly includes the brutal long-term parking conditions at Redmond, OSH or SunNFun. When I travel, the tanks are filled with whatever the FBO is using - the request is for 100 LL.

 

Yes, the LML is important so that all may benefit from experience, both good and bad.  The system I use, taken in the context it is used, is safe enough plus it eliminates other more common failure modes such as those we see across GA from the use of selector valves.  The few fuel related Lancair crashes I have reviewed are most often the result of mistakes made by the pilot and in some cases the result of known low levels of fuel aboard.

 

This discussion has been most illuminating but, unfortunately, has been focused on fuel monitoring systems and not the more frequent and deadly accident cause - flying too slow, too low.

 

Scott Krueger

Flying my slow build since last century (1996).

 

PS Apparently, JP4 is unreliable as a fuel since it leaks out so readily ;<)    

 

> I have no idea how the capacitance probes you talk about work.
> Here is how the VM Fuel system probes work:

Scott:
The probes that you describe are just the same as the ones that Gary described.
Good for you in figuring out how to lower the output frequency to
match between a smaller sensor
and your VM system. In that system, capacitance is translated into a
frequency so that it
can be transmitted with little effect of electrical noise.

The capacitance that is being measured is a tube dipped in the fuel
with a concentric wire.
That forms a cylindrical capacitor with a dielectric that is (ideally)
either gasoline or air.

The dielectric constants are:

air 1.0
gasoline 2.0

So the capacitance of the probe doubles with it filled with gasoline
instead of air.

But some other liquids have much higher dielectric constants:
ethanol 30
methanol 33
water  80

A probe that is 10% full of fuel and 1% full of water would read just
about full.


If the concentric tube sensor is fully immersed, then area is constant
and the observed capacitance
is a measure of the dielectric constant of the fluid. This is the
basis for detecting additives in racing
fuels, such as alcohol: http://www.foxvalleykart.com/fuel2.html
That also means that a few percent of alcohol (ethanol or methanol)
would cause the tank to read more full
than it is --- while also reducing the fuel value of the liquid in the tank.
In other words, there is a way to measure the composition of the fuel,
as well as its level
with a simple redesign of the sensor and the controlling electronics.

> Oh well, you use your experience and I'll use mine.

I think the point of having a mailing list like this is so that we can both
benefit from both of our experiences. I'd like to think that I could
learn from someone
who landed an F-4 with 6 gallons of JP4 on board without having to try
it myself.
He turned back to base with maybe 20 seconds to spare when the tanks were still
reading in the upper half, but the gauge readings were a little lower
than expected
and were "bouncing around more than normal".

Think about the decision process, and how much room was actually left
for thinking
about whether there was really a problem. If Bill had waited another
minute before
turning around, his story would be different, and we might not have
had the opportunity
to learn from it.

-bob

--
For archives and unsub http://mail.lancaironline.net:81/lists/lml/List.html