From: Sky2high@aol.com
Message-ID: <f6.33b48ae.2709fd58@aol.com>
Date: Mon, 2 Oct 2000 11:01:44 EDT
Subject: VMS Fuel Flow Transducer
To: lancair.list@olsusa.com
Reply-To: lancair.list@olsusa.com
Mime-Version: 1.0

          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>>
          <<  Lancair Builders' Mail List  >>
          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>> 
>>

Fred, 

I have tried a second transducer with identical results.  See the further 
discussion in the next section.

I would guess you are running a Continental engine while Brent (I/O 540?) and 
I are running Lycomings.  Continental's injection system returns excess fuel 
back to the fuel source. Some of the greater-than-actual usage may be due to 
this and perhaps Continental publishes the rate at which it returns fuel.  
Maybe Continental systems require two flow transducers, out-to-divider and 
return-to-tank, where the difference would result in a more accurate reading. 
 In any event, read on--

Brent, thanks for breaking down the possible sources of inaccuracy -- Maybe 
those with accurate readings could respond information about their 
installation.  Let's look at the items you mentioned:

1) Turbulence - Although you mentioned vibration, my horizontal, padded, 
engine mount frame installation should not experience much vibration. I am 
using Teflon, SS braid, fire sleeved lines and 6" of the input line is 
straight while the exit line uses a 90 degree elbow to get the fuel up to the 
spider.  However, because the spider is fed with smaller fuel lines, there 
are fitting reducers used on the transducer and these could introduce 
turbulence.  I would have thought that turbulence would reduce the flow 
indication.

2) Temperature - This might be significant for me!  My front mounted throttle 
body sends the fuel back under the engine  to the transducer which is mounted 
in the hot, dead air behind the engine and then up thru the back baffle so 
that only the last foot is in cool incoming air.  Heat indeed!  I am cooking 
the fuel from the time it enters the cowling until it reaches that last foot 
before the spider.  I will try a blast tube on the transducer (for what it's 
worth) and start to think about rerouting some fuel lines.

A friend with a pusher Cozy MkIV and VM1000, reverse cools the engine (air 
exits the top) so that all the fuel components (including the pumps, lines 
and transducer) are in the cool incoming air.  I will ask him if his 
flow/totalizer is accurate.

3) True Density - Where do you buy your light weight fuel?  I calibrated mine 
by volume and find the probes to be very accurate, even the short one in the 
header tank.  I would have thought the VMS computer would account for any 
flow/density question using the third-party sensor.

4) Tolerance - I agree there might be a tolerance band, but my limited 
experience with two would indicate that the band is very narrow, as it should 
be.

Philosophical or Metaphysical view, sometimes described by the letters IMHO:

In my now ex-Skymaster (Continental powered), fuel flow was indicated in 
pounds per hour (pph) except at full power where the indication in red was in 
pounds per square inch (psi) and I believe this analog gauge was measuring 
pressure all the time, displayed in weight (works if you know the orifice 
size), and then mentally converted to gallons per hour (using the inaccurate 
6#/gal).  Thank God I wasn't in Canada using Imperial Gallons!  The rear 
engine always indicated 3-4 pph more than the front and, sure enough, after a 
long flight, the rear tank (right wing) always took a couple of gallons more 
as measured by the "very accurate" FBO fuel truck flow meter.  Since the fuel 
quantity gauges were notoriously unreliable, I always relied on flight time 
-- my mental totalizer, backed up by the fuel truck and checked by the "flow 
gauge."  The flow gauge was pretty accurate for an analog needle gauge and it 
was mainly used to logically lean the engine or adjust the fuel flow during 
climb.  Good enough for its' implicit accuracy band.  

I have different expectations about digital gauges and computers.  I expect 
to see a very accurate display of sensor collected data.  I expect engine 
parameters (temperatures and pressures) to be within 2 or 3 percent, why not 
flow?

I will not regress and fly primitive equipment (172) in order to fully 
appreciate modern, digitally displayed misinformation.  

I will take your suggestion to determine the offset (after some above 
mentioned experimentation) and have VMS cut a new EPROM.  In a discussion 
with a VMS rep at Oshkosh, he suggested the same for the warning on fuel 
levels -- I want the warning indication switched from 4 gals to 0 gals for 
the wing tanks.  Too bad I don't have more control over the parameters or I 
would fix them myself.  

Modern systems should provide that parameter driven controls are changeable 
thru the device itself.  For example, the Garmin 430 was not producing enough 
sidetone and I thought I had to return to the installer for some internal 
adjustment -- not so, a special power-on sequence gets you to a series of 
pages which control internal functions, one of which is the output level for 
the sidetone.  Voila!!

Scott Krueger
N92EX
The 'E' is for Experimental.
The 'X' is for the slashes from Occam's Razor.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
LML website:   http://www.olsusa.com/Users/Mkaye/maillist.html
LML Builders' Bookstore:   http://www.buildersbooks.com/lancair

Please send your photos and drawings to marvkaye@olsusa.com.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>