This weekend, I'm going to visit a ~12yr-old 914 and try to troubleshoot a recent development. We noticed that the (differemtial) fuel pressure drops from 0.15bar at 30"MAP to something close to 0bar at 40". This causes the engine to stutter, the wastegate to open, the power and RPM to drop, and then the cycle can repeat (if we allow it).

Since we knew we could get within spec (placard on instrument panel says 0.25bar +-0.1bar) at 30", we flew the plane to get some idea of what's going on. What we noticed is that as we climbed higher, the fuel pressure would drop. By throttling back bit by bit as we climbed, the fuel pressure would stay within the acceptable window. We got to 5k' feet or so and then decided to glider around for a while, so we didn't test if the fuel pressure would climb as we descended.

Looking at the Rotax documents, this screams "regulation problem". However, while I was gone an A&P tested for vacuum leaks and proper regulation and he said it all looks fine. 

If it were a fuel pump/fuel filter issue, then the fuel pressure would only drop at high volumetric flow rates, it certainly wouldn't have dropped  linearly with increasing altitude, only to return when the throttle was brought out a little bit more.

Thoughts? Any easy checks to do while I'm in front of it? Could the power drop be due to anything other than fuel starvation relating to overly low float bowl pressures?

P.S. The placard and manual says 0.25bar +-0.10bar, but is that really correct? Reading through the way the float bowl works, I feel like in order to run the engine at the correct fuel-air mixture that the regulated pressure should be very, very close to 0.25bar.

  • Re: 914 fuel pressure drops when going to 40" MAP on the ground, or at lower MAP when climbing

    by » 4 months ago

    The fuel pressure reading is (or should be) differential pressure with the reference pressure being the airbox pressure. You should see a pressure line running from the fuel pressure sender to the airbox to provide this reference. When measuring differential pressure, the reading you get from the sender (for any stable absolute pressure) will fall as the reference pressure (airbox pressure) increases.  In other words, if the fuel pressure was not being regulated, and was simply remaining constant, your pressure reading would fall as airbox pressure increases.  The differential pressure sender is meant to see the relative pressure that the carburetors see.  

    Assuming you have already tried using the aux pump, and that both pumps are working properly, then I would look very carefully at the pressure lines and fittings connecting the fuel pressure regulator to the airbox. These are not typical vacuum lines, they are pressure lines when under boost pressure.  So a vacuum test may not expose a leak that happens under pressure.  I would also want to use an analog differential pressure gauge to verify the fuel pressure sender is working properly, although the engine stumble you are experiencing leads me to believe the fuel pressure problem is real.  But, never assume anything!

  • Re: 914 fuel pressure drops when going to 40" MAP on the ground, or at lower MAP when climbing

    by » 4 months ago

    (This is a differential probe on a plane which has flown for a while before. I've updated the original post to clear that up.)

    Good point about the pressure lines leaking under positive but not under negative pressure. That makes me hopeful the A&P simply missed this nuance.



  • Re: 914 fuel pressure drops when going to 40" MAP on the ground, or at lower MAP when climbing

    by » one month ago

    Problem found: it was a busted airbox pressure line coming from the airbox to the fuel pressure regulator (yellow hose in the screenshot). So the regulator was indeed measuring ambient air pressure.

    My engine is plumbed slightly differently from what's shown in MMH_912-914 Series_ED1_R6.pdf, Fig 73-5. The yellow line has no tees, it goes only to the regulator. The blue line, on the other hand, has the tees, unlike in the image. Same air pressures throughout, but a cracked line affects things slightly differently in the two variants.


    35698_2_Screenshot 2023-04-09 at 6.17.38 AM.png (You do not have access to download this file.)

    Thank you said by: RotaxOwner Admin

You do not have permissions to reply to this topic.