Propane Motor Fuel Injection

[Johnny]

>> I have been experimenting with trying to get the stuff to stay in a liquid
>> state all the way to the injector, off and on, for quite a while now. I
>> could never quite get the line pressure to be stable enough under varying
>> throttle conditions. Part of the trouble was I was not able to get the right
>> injectors. I am thinking now that I can just pump up the tank with nitrogen
>> to boost the tank pressure to about 300-400psi or so, regulate it
through >>a
>> liquid type high pressure regulator to get the line pressure stablized at
>> about 150psi or so, go with the high pressure injectors you guy have
>> mentioned here, sized as tiny as possible so that I can get fairly good
>> metering at idle.

>I  don't  think  pressurizing  the  tank  with  nitrogen  will get you
>anywhere, the pressure will still decrease rapidly as the  propane  is
>used up and the fuel volume in the tank decreases.  Also, the standard
>filling  equipment  probably  won't  be  able to push fuel in at  that
>pressure, (you wouldn't be able to fill your, uhh,  tractor  off  the
>tank anymore)  and  the propane system tank, lines, valves, etc.   are
>only rated at 250PSI test...

I meant use a tiny little nitrogen bottle with a regulator that feed into
the propane tank while in use. When refilling the propane tank, the nitrogen
is disconnected. Of course you couldn't use a standard propane tank for
this, but I didn't think that was that big of a deal.

>I  been  doin some thinkin on this over the past couple of days, and I
>think we may be approaching the problem bass  ackwards...    With  the
>standard EFI system, the fuel pressure is regulated in order  to  make
>injector on-time  calculations  easier.   Since the injector is really
>just an orifice, the  pressure  needs  to  be  constant  in  order  to
>maintain a predictable flow across the orifice.  Most EFI systems have
>the  fuel  pressure regulator tied into manifold pressure, so that the
>fuel  pressure  remains  at  a  set  amount  above  manifold  pressure
>regardless of throttle opening.  All this is done so that the computer
>doesn't  need to dink with flow metering or additional calculations to
>come up with injector pulse width.
>
>Now,  if  we  were to plumb the injector directly onto the liquid line
>from the tank, and put a pressure transducer on the fuel line so's  we
>would  know  the  fuel  pressure  at all times, we could theoretically
>compensate the injector pulse width to give the right  fuel  flow  for
>the instantaneous  line  pressure  that we have... voila - no need for
>messy regulators, nitrogen, or any complicated mechanical stuff. Since
>(last time I checked, anyway) propane follows the laws of physics wrt
>flow  across  orifices,  it  should  be  relatively simple to have the
>computer compensate for existing pressure.  Whaddya think? Am I out >to
lunch, or did I stumble onto something here...

I like it. By far the simplest way to go, and it would seem to be the most
accurate in that you aren't required to meet a certain pressure all of the
time but instead just compensate for what it may be.

>Still have the problem of where to find the proper  injector  for  the
>system, though.    Stock injectors are all made to run at 20-50 PSI, I
>think liquid propane would blow it up real good...  Might have to have
>2 injectors, a tiny one for idle/cruise, and a big nasty one for  WOT.
>The system I have in mind would be a TBI type system, I'm not sure you
>could do this with a port injection system unless you had a really big
>engine...

We could go to staged injectors if we had to, but as long as the line
pressure stayed within a fairly narrow range I don't see why we would have
to do staging anymore than you would have to with gas. The key is the
injector itself. Getting one that will handle the line pressure and also
getting one that is small enough to supply just enough fuel at full power.

Granted, I am sure we will see a drop in line pressure at WOT, but If the
lines were sized properly, and the fuel rails (if we used fuel rails) were
large, acting as a fuel manifold, I would think we could get pretty close on
getting the pressure at the nozzel to equal the pressure at the bottle. Then
as long as our transducer was at the fuel rail, we should be able to
compensate accurately enough.

Great idea!

I have been digging all of the hardware I have out from the last time I
kicked this idea around. I still have to follow up on trying to find the
correct nozzels though.

-j-