Subject: P-port maifold pressure tap
From: Rotary Engine
Date: 8/15/2005, 9:05 PM
To: AA-me


    >
    > > I had a long talk with Tracy on this subject.
    > > The question is what is the true manifold pressure. We need to
   know
    > > that number to set the pulse width of the fuel injectors. Since
    > > Tracy samples it for only a millisecond or so it could be
   affected by
    > > a lot of factors. We know there are pressure pulses in intake
    > > and exhaust systems. The question is; what part of the pressure
    > > curve is occurring when Tracy samples the pressure?
    > >
    > > Tracy samples the pressure for firing the injectors on rotor two
    > > right after (within a few microseconds) he fires rotor
    > > one's ignition plugs (about 20 degrees BTDC). So far I have not
   been
    > > able to talk him into taking many samples and averaging the data.
    > >
    > > We discussed how to measure the pressure wave in the p-port
   intake manifold
    > > as a function of e-shaft angle during one cycle per revolution
   of the
    > > e-shaft. The max pressure will occur at different times depending
    > > on the RPM and the length and diameter of the runners. So far
   suggestions
    > > have been; feeding the output of the high speed pressure sensor
   into a PC
    > > audio input card or the microphone jack of a note book
   computer. Tracy
    > > pointed out the problem with that are those are AC inputs and
   we need
    > > absolute numbers from  the pressure sensors. We also need to relate
    > > them to the ignition pulse. That led to  using a storage
   oscilloscope
    > > instead.
    > >
    > > The price of battery powered hand held storage oscilloscope
   have dropped
    > > recently and they are fast enough to do this job. At 7500 RPM
   it takes
    > > the e-shaft 8 milliseconds to make one revolution. In other words
    > > that is only 125 revolutions per second. Since the response times
    > > of the pressure sensors is only 1 Kilohertz most hand low cost held
    > > oscilloscopes are fast enough to do this. They are also fast
   enough
    > > to measure the pulse width of a fuel injector if you want to
    > > use them for that. This device will help greatly when designing and
    > > developing intake manifolds.
    > >
    > > Here is a intake manifold pressure pulse from a Mazda tech paper
    > > SAE 841017 on P-port intake manifolds.
    > >
    > > Here also is a pdf file for some hand held storage oscilloscopes.
    > >
    > > When Mark gets his p-port engine in his Mustang II airplane I
   will pop
    > > over to Denver and help tune it with what ever test equipment
    > > I can beg, borrow, bring, or steal :)
    > >
    > > Paul Lamar



    > Hi Paul,
    >
    > It seems like there are two system design choices available. The
   first
    > is to leave Tracy's computer alone and provide some mechanical
   low pass
    > filtering for the sensor.  A long, thin tube feeding a small air
   tank,
    > feeding (through a small orifice) another thin tube to the sensor
   would
    > probably work - resistive to resonant pulses.  This will get a stable
    > signal for the computer to look at - average out the ups and downs of
    > the manifold pressure during the engine's cycle, regardless of RPM.
    > That way the computer can have a realistic value for the amount
   of air
    > available to be injested by the engine.
    >
    > A few nice things about approach are that it's dirt-simple, and
   it works
    > throughout the RPM range.
    >
    > As I think you have suggested, taking a short snapshot of the MP at a
    > fixed point in time each cycle may mean that the pressure is low
   at some
    > RPM values and erroneously high at others.  That can probably be
   tuned
    > out via MAP vs RPM vs fuel curve, but it's clunky.
    >
    > The other approach is as you suggested - sample the MP sensor often
    > enough to get a realtime waveform of MP.  I can't see a real
   benefit to
    > this except that it might allow you squirt the fuel into the
   intakes at
    > a more optimized time during the cycle - allowing better charge
    > distribution, or something like that.  That's going to take a fair
    > amount of time to dial in, I think.
    >
    >  From the standpoint of fuel delivery requirements, I can't see
   the need
    > to get a high resolution picture of MP.  Even if there were a 10
   cycle
    > delay in the response time of the sensor, it wouldn't have an
   impact on
    > engine performance.  I'm probably missing something.
    >
    > One other concern I might have is that of mechanically damping
   the input
    > signal to the MP sensor.  I could imagine generating a resonance
   which
    > could really wack things out.  That might be a bit of a concern
   if you
    > try to get a high resolution picture of MP.
    >
    >
    > Regards,
    >
    > Matt-
    > VE N34RD, C150 N714BK


Matt,
Al, Rusty and I all discovered.  Making a the manifold [vacuum] tubing long
enough seemed to fix the problem.  Making it too long and adding
restrictions will then slow the response time.  The tubing to the fuel
pressure regulator also seems to help.

-- 
Dave Leonard
Turbo Rotary RV-6 N4VY
http://members.aol.com/_ht_a/rotaryroster/index.html
http://members.aol.com/_ht_a/vp4skydoc/index.html

-- 
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Copyright 1998-2005 All world wide rights reserved.