|
|
Yea Ernest,
I've read your dissertation. As you know, I'm a practical man, and as far
as what you have written goes, it all sounds eminently theoretically
feasible. However,
until you actually put it into practice, you won't know if it DOES work.
So, ..."Good Luck"!!
However, just as there are many different types of wheels, and many
different types
of cars, they all boil down to the same principle. Sometimes, if you try
to simplify something too far, you end up with a system that causes more
trouble than the slightly more complex system that is known to work.
As with most systems, there is a concept called "critical mass", and its
analogy "critical complexity". Mitsubishi, for example, know ALL about
this with their truck wheels that break and kill people. You'd think by
now, ANYBODY could build a wheel that wouldn't break, wouldn't you?? Some
little Japanese genius engineer thought he could literally "re-invent the
wheel". He did, but it wasn't fit for the purpose for which it was
intended.
When you look at the new "returnless" fuel systems, they are really no less
complex than the previous systems, only a bit CHEAPER to manufacture and
install!! (corz most of the stuff is in one place in the tank - but it's a
bugger for mechanics when it comes overhaul time!!) Cost is usually the
driving force in these things, not necessarily efficacy, functionality or
serviceability.
This is what I meant by "re-inventing the wheel". Something that neophyte
Experimental aircraft people always seem to want to be doing, and it's
usually accompanied by lots & lots of enthusiasm as well as false premises
and
pre-suppositions which are normally the hallmark of youth (but not
necessarily), ignorance, and lack of understanding.
PRESSURE REGULATORS
To wit, with the EFI pressure reg. Your assumption is wrong. They do NOT
maintain a SET pressure (at least in an EFI system). The fuel rail pressure
is supposed to be dynamic, and must be
able to vary with the MAP (manifold absolute pressure). At idle, because
the MAP is a pretty good vacuum, the rail pressure is just under a Bar/1Atm
LESS than at WOT
where the MAP is essentially atmospheric, so essentially, from Idle to
WOT, the rail pressure varies just under 1 Atm/Bar or 14 PSI at Sea Level.
*Unless you have the SAME pressure drop across the injectors at ALL times,
which is the purpose of the EFI regulator, (it has a port which connects to
manifold pressure - so why else would that be there?), then it becomes
difficult to accurately meter the fuel at
different MAPs.
Things get more complicated as you climb and descend. As the barometric
pressure drops, you need less and less pressure in the fuel rail to keep
the pressure DIFFERENTIAL across the injectors the same. The MAP at WOT @
10,000 ft is about 10 PSI.
If the pressure drop is NOT the same, you are going to be forever chasing
the correct A/F (Air Fuel) ratio
as you climb and descend, not to mention differences between one day and
the next. Which is fine if you want to do that, but that's NOT the idea of
a modern sophisticated EFI system. It IS possible to simplify any system
beyond the point of minimum functionality.
VAPOUR LOCK & HEAT SOAK & HOT START etc.
I'm not even going to go here in detail. I just don't have the time (nor
the inclination) to detail all the possible problems associated with your
proposed system. Except to say that the system must be
designed to be able to purge air and vapour from the system at all times. I
got caught with something like this a couple of years ago, and it took
three people and several hundred dollars of dyno time to figure out the
problem. Just bad fuel system design by the owner of the car, and all
three of us missed it until one bloke had an "Aha" moment. Wasn't at all
obvious. I also seem to remember that several people have had weird "fan
out" problems in their aircraft due to various fuel issues. We STILL don't
know what caused Paul's problem. So again "Good Luck"!!
Just on one issue - that as the temp changes of the fuel ,
which expands/contracts with changes in temp, and therefore changes fuel
density. So you then have to adjust the mixture to keep the same A/F ratio
(corz you need more fuel at the lower density and viccky verka). I can see
this being a REAL pain in the butt, sitting at idle on the ramp on a really
hot day waiting for taxi / take-off clerance. As the temp in the fuel rail
gets hotter, the mixture leans to the point where the engine cuts out. You
beauty!! Just another bit of un-necessary workload for the busy pilot!!
Anyway, again, "Good Luck"!!
FILTERS
One final word. Fuel fiters are there to protect mechanical fuel devices
from grit, among other
reasons. SO, assuming you are going to gravity feed your EFI pumps, you
still
need a low presure filter/fine screen BEFORE the high pressure pump, (to
protect the
pump from damage due to grit in the fuel). You also need a high pressure
filter AFTER the HP pump, as
close as possible to the firewall/fuel rail, to protect the injectors from
HP pump wear / disintergration detritus, as well as corrosion products that
can
occur in the supply line over time. The filters also need to be easily
accessible for service/inspection/replacement. Otherwise, sooner or later,
you WILL have injector/fuel supply problems!
Cheers,
Leon
----- Original Message ----- From: "Ernest Christley" <echristl@cisco.com>
To: "Rotary motors in aircraft" <flyrotary@lancaironline.net>
Sent: Tuesday, February 15, 2005 6:15 AM
Subject: [FlyRotary] Re: More on Header Tanks, Venting & Pressure Reg
Position
Leon wrote:
>Talk about being "on the case", ... the guy who wrote this is a genius!!!
I
>confess I couldn't do better myself. ALL the info, ALL the details, and
>ALL the diagrams:
>
>See:
>
>http://www.sdsefi.com/air27.htm
>http://www.sdsefi.com/tech.html
>
>Enjoy!
>
>
Leon,
There is a reason that there are so many types of wheels, not the least
of which is that there are so many types of vehicles. The SDS system
looks to be designed for multiple tanks and to keep high pressure fuel
out of the cockpit. But can't things get so much simpler when you have
only one tank that can gravity feed the high pressure pumps, and you
allow that 40psi is all that high of a pressure? I think I've done my
homework, and will now turn it in for grading.
I've attached a improved diagram of what I think would be a good fuel
system for a Dyke Delta. Two high pressure pumps under the tank feed
into seperate filters (which may need to be switched to before the
pumps...there seems to be some counterviews that I haven't sorted
through). The line then Ts into a regulator with a single line going
firewall forward. A gascolator/drain will have a seperated pickup in
the tank that is lower than the feul supply pickup. The beauty in this
system is its simplicity.
I've looked reviewed lots of post, taken notice of the cautions and
arguments on both sides, and listed my reasoning below.
Problem: The pressure regulator must go after the fuel rail.
Solution: I can not find anything in the way these regulators work to
suggest why this would be. The regulator will maintain the set pressure
at its intake side by dumping excess fuel out the other side. Like a
parallel zener diode, it matters not at which point it is placed on the
bus. Nor does it matter how many curves the power or ground circuit have
before or after the zener. The voltage will be maintained at the proper
level. Likewise, the injectors are T'ed off of the fuel rail, and I'm
just adding another T for the fuel rail. Logically, it is the same,
even though it's not physically.
Problem: High pressure fuel running under the cockpit.
Solution: I believe it was Tracy that said if there is a cracked fuel
line, your nose will tell you long before there is enough fuel to be
hazardous. I agree, and will call this acceptable risk in exchange for
a simplified system which will be more robust overall (It's guaranteed
not to break if it isn't there.)
Problem: Vapor lock
Solution: My understand of the term 'vapor lock' is that the pump is
full of air /fumes, and can't generate enough pressure to force anything
through. This vapor can be caused by sucking air from the tanks, or
heat boiling the gas in the lines. The gas pickup (in a recess in the
bottom of the tank), will be covered with fuel untill there is only a
cup of so left. The pumps in this case will be 6ft from the firewall,
with nothing more than a 1/4" aluminum line to conduct heat. No chance
of sucking air as long as I have something to pump, and heat is a
non-issue.
Problem: Fuel rail heat soak during flight.
Solution: Approach the problem from two ends. Don't let heat get to
the fuel, and don't leave much fuel sitting around under the hood.
1)shield the exhaust.
2)provide for blast air around the fuel rail
3)insulate the fuel rail (layer of micro and a couple layups)
4)fabricate a low volume fuel rail from 1/4" tube and a few T-fittings.
With these precautions, I don't see how the fuel could possibly pick up
enough heat in flight to be a problem. I would be dependant on how long
the fuel remained under the cowl before being used. At full throttle, I
estimate around 15gal/hr of fuel used. That equates to .004167gal/sec,
or .02672cu.ft/sec. With 1/4" supply lines, and 2ft of line firewall
forward, there will be .0006814cubic feet of fuel firewall forward at
any time. If I've done my math correctly, there will be 39.2sec for the
fuel to pick up heat. I can't get a cup of coffee warm in the microwave
that fast, and I'm here we're dealing with an insulated structure that
will mostly suffer from radiant heat.
Problem: Hot start. This would be the biggest problem. Park the plane
on a hot day, and residual heat in the engine will boil the fuel in the
hose that is firewall forward, and push fluid back up through the
regulator.
Solution: Hit the cold start button on the EC2. Mixture full rich. If
it doesn't crank in a few turns, hit the cold start button on the EC2.
(What a hack, using the cold start button for a hot start) The point it
to clear out the hot, pressurized fumes. If the engine will run on the
fumes, then when it does catch, expect it to run way lean and rough for
thirty seconds until the fumes get blown through. There shouldn't be
much fumes there. Only the line under the cowl will be heated. As the
fuel warms then boils, it will push the rest of the gas back toward the
fuel regulator, locking the fumes in to the high part of the system
which just also happens to be the hottest (the Delta sits about 9degree
nose high on the ground). As the fumes expand and move down the fuel
line, it'll eventually reach the firewall penetration. Notice the
dogleg in the fuel line on the passenger side of the firewall, from low
to high. The fumes will have to push the fuel down the dogleg and all
the way back to the fuel tank to cause the pumps to vapor lock, all the
while being cooled by the aluminum line. Since trapped gases are a
fairly decent insulator and any hot liquid gas would rise to the top,
I'd expect they'd only reach an inch or so behind the firewall. A hot
start will only be a problem till it uses up the fumes that can occupy
the 2ft of 1/4" tube in front of the firewall.
Ding! Ding!
Class is now in session. I have my pen and paper in hand, ready to take
notes.
----------------------------------------------------------------------------
----
----------------------------------------------------------------------------
----
>> Homepage: http://www.flyrotary.com/
>> Archive: http://lancaironline.net/lists/flyrotary/List.html
|
|