By on December 31, 2009

Showing some Legacy? (courtesy:subaruforsale.org)

TTAC Commentator NICKNICK writes:

Mr. Mehta, my 1999 Subaru Legacy 2.5 GT often loses power in the first hundred yards of driving. It has never failed to start, and it has never stalled during the period of power loss. It does not matter if the car is started cold, or if it’s been running for an hour on the highway and restarted after a 3 minute stop for gas. When it loses power, it doesn’t lose ALL power—it feels like it has about as much power as it would idling in gear. Flooring the accelerator does nothing…until it does, at which point it wants to break ya neck. This happens about 80% of the time. If I catch it starting to get fussy, I disengage the clutch and wait three seconds. The problem time passes while I coast, and I can be on my way without any more trouble. The other 20% of the time is trouble-free like a normal car. After the first hundred yards it runs great, sounds great, and gets great gas mileage.

I’ve got a data-recording and live-viewing OBDII reader—it has turned up nothing. The TPS seems to work fine, there’s nothing odd from the MAF, and I don’t see anything amiss in the fuel trim.

It needs air, spark, and fuel to run. It seems to get air just fine, as it runs well every other time. It has an old-school throttle body with a cable that runs to the pedal on the floor—no chance of a computer taking over and shutting me down. It never stalls, so I think I’m always getting spark. As for fuel, that’s my big question. If it has idle-level power and then goes nuclear, that sounds like not enough fuel going straight to lots of fuel.

One more thing: If I turn the ignition to ON but don’t start the car, and I wait for 60-90 seconds before starting the car, there’s never a problem–same thing if I actually start the car and wait a minute to drive.

My guess: bad fuel pressure regulator or vacuum line to the fuel pressure regulator. The fuel pump is electric and has time to pressurize the line when I wait for a minute with the ignition on. Fuel filter and PCV valve are new. What do you and the Best and Brightest think?

Sajeev Replies:

Nice job on the diagnosis, you’ve covered many of the branches on the diagnostic tree. Because your Subie is 10-ish years old, there’s a chance it has a significant amount of mileage (100k or more) on the clock. So if your Oxygen Sensors are original, replace them now.

But don’t take my word for it, even if it happened on a previous Piston Slap. And then again on another. Oxygen sensors measure air-fuel ratios several times a second, so if they are just a little off calibration, they’ll cause a bog or hesitation for a brief moment. Like the Civic in that previous Piston Slap, maybe the O2s are just bad enough to not trip an OBD-II code. Maybe.

But O2s are not all of it, nor are they the first issue to address: check for bad vacuum lines and replace them, odds are you have several at this age. Buy or rent a fuel pressure gauge, getting one with a hose long enough that you can drive with the gauge taped to your windshield. If the gauge ever dips below Subaru’s recommend operating range, you know for sure you have a fuel pressure problem. Then you can attack the question of what to replace: regulator, pump or just a filter?

Good luck with your diagnosis, you’ve saved yourself hundreds of dollars with your legwork and intelligent analysis.

(Send your queries to mehta@ttac.com)

Get the latest TTAC e-Newsletter!

14 Comments on “Piston Slap: A Legacy Of Hesitation?...”


  • avatar

    There is a TSB about hesitation and the fix is to move the knock sensor then disconnect battery for 20 mins….
    Take a look in case it applies to you. It’s a $0 fix.
    Thread here:
    http://www.imprezasport.net/forums/index.php?showtopic=785

  • avatar
    cdotson

    I was going to mention knock sensor as it sounds like timing may have been pulled way back.  I have a 99 Legacy SUS with the EJ25D that I had to replace the knock sensor on because it kept tripping the check engine light and fuel economy went to crap, but I never noticed the spotty/intermittent hesitation.
     
    You could confirm this if your diagnostic tool can display timing advance in real time.  You should notice a substantial spark retard during hesitation episodes.
     
    I’ll have to keep the knock sensor “TSB” in mind if I ever get around to replacing the head gaskets on the sucker (been parked for years so who knows).

  • avatar
    superbadd75

    Sounds like a fuel pump to me. Especially when mentioning the 30-90 sec wait.

  • avatar
    jet_silver

    This happened intermittently on my ’93 Legacy and I traced it to a sticking IAC valve.  On my car it would stick in cornering downhill and wasn’t so easy to find.  Sticking IAC valves are easy to fix – you spray in some fuel system cleaner.  (I’ve not been under the hood of a ’99 and it’s entirely possible they changed the design so this is no longer a failure mode.)

    • 0 avatar

      If it’s anything like a 1993 Ford IAC vs a 1999 Ford IAC, cleaning it with any light oil/cleaner won’t help. It might actually hurt.  So check the forums or a Subie service manual to see if they mentioning cleaning the bores of the IAC.

  • avatar
    johnny ro

    Also check: main air tube from TB to intake manifold. If it is molded rubber, like on my 80-90s VWs, and held on with hose clamps, and its slightly loose, it can do what you describe. Intermittently. The TSB sounds more likely though.

  • avatar
    golden2husky

    I believe fuel delivery to be the culprit, but I’m not sure the oxygen sensor is to blame.  Since you said this problem will occur when totally cold (as well as other times), that says to me the sensor should not be part of the problem because when stone cold, the oxygen sensor is ignored and you car operates in open loop mode.  The ECU substitutes real time data with pre-programmed values until closed loop operation.
    You are starving for fuel for the problem time period.  If it always acted this way off idle, a fuel filter would most likely be the culprit.  I, too suspect the fuel pump.  Upon turning the key to the run position, your fuel pump should run for a few seconds and then stop, assuming you don’t bump the key to “start.”  There is a check valve in the fuel pump that prevents leak down of fuel when the pump is off.  Perhaps the check valve is stuck and the pressure and all the fuel leaks down when you turn the car off.  There may be enough fuel in the injector rail to support idle, but not enough to accelerate the car.  After 90 seconds or so, the fuel pump “reprimes” all the lines and the car runs properly….my 2 cents…

  • avatar
    obbop

    Fuel line collapsing due to age.  Replace all rubber fuel lines.

  • avatar
    niky

    Definitely sounds like a fuel issue. My car does something similar.

  • avatar
    abaddsm

    Def not 02 sensor related. Cars will ignore the 02 sensor when engine coolant temps are below certain thresholds and even for a predetermined time after start like 60 sec. before going into open loop mode. iac also wouldnt effect this. me guess fuel pressure issue. put a fuel pressure gauge on the schrader test port on the fuel rail and see if you get over 40 psi at idle .  Since you have a mechanical throttle open the throttle by hand and see if your fuel pressure drops off significantly.

  • avatar
    MBella

    After you check all the vacuum hoses for leaks, perform the fuel pressure test. You can get a decent gauge at harbor freight for cheap.

    If the fuel pressure is good,  perform a power balance test. Just unplug injectors one by one, to see which one drops the rpm the least when you disconnect it. This will be your problem injector.

    If fuel pressure isn’t up to par, now you have to isolate the problem. With the fuel pressure gauge attached, disconnect the vacuum hose at the pressure regulator with the car running. The pressure should rise to its maximum, and the car should stumble. If you have a hand operated vacuum pump, attach it to the regulator. As you increase vacuum, the pressure should drop. If it doesn’t your problem is the regulator. If it works as it should, it will likely be the pump. The check valve theory makes some sense as well.

    Replacing the fuel filter at this point might not be a bad idea either. The odds aren’t very high that this will fix your issue, but it should probably be replaced anyway, and you have nothing too loose.

  • avatar

    @NickNick
    For our collective knowledge, please let us know what the fix was when this issue is resolved.


Back to TopLeave a Reply

You must be logged in to post a comment.

Subscribe without commenting

Recent Comments

New Car Research

Get a Free Dealer Quote

Staff

  • Authors

  • Brendan McAleer, Canada
  • Marcelo De Vasconcellos, Brazil
  • Matthias Gasnier, Australia
  • Tycho de Feyter, China
  • W. Christian 'Mental' Ward, Abu Dhabi
  • Mark Stevenson, Canada
  • Faisal Ali Khan, India