Sudden Increase in Power

[SchwarzeEwigkt]

So, I have a kind of interesting thing going on with my '97 Outback with a 4EAT. I'll have a sudden increase in power about a third of the way through my admittedly short trip to and from work. It's not much, maybe 10%. I had initially thought it was the engine switching from open to closed loop or something that has a hard transition on it, but that didn't makes much sense considering that it always does it climbing the same couple of hills. It occurred to me that it can't be warmup since ambient temperature has no effect on it. It's the same when it's 40 out as it is when it's 10. I've also noted that once the "boost" kicks in, it stays that way until the car sits overnight. It's like a valve opens or something.

 

I always notice it when accelerating up the same hill at about 40-45mph with the engine at 2700-3200RPM, so in third gear. It feels like a little VTEC kicking in, even. After that first time, the "VTEC stays kicked in, yo."

 

There's a short list of things that could cause that kind of behavior as far as I know. It did it in a more pronounced manner before I did plugs, wires, air filter and timing stuff, kind of a "wimpy" to "there it is" kind of transition more like a 30% increase in power. The O2 sensor was changed about 2k ago to correct a rich running/crappy milage problem. It didn't affect the power increase behavior. I did the fuel filter since I had one and it was overdue. That hasn't changed it either, though it seems to idle slightly better. PCV was done a while back, not that it should have anything to do with it.

 

It idles and runs as well as it has since I've owned it and gets acceptable milage (17-19mpg on snows and winter gas on a 8mi round trip commute).

 

I'm thinking that IAC isn't the problem since that closes at anything above idle (right?) and would probably throw a code if it was stuck. It's not open loop vs. closed loop since it would depend on ambient temperature as opposed to spatial position and I expect the new O2 sensor would have caused a change.

 

There's no vacuum leaks I can find and I'd think they'd cause other issues.

 

That leaves one other thing: the EGR. I'm theorizing that it sticks until the engine warms a bit and I get a low vacuum situation like half throttle up a hill at 3k*rpm, where it pops loose and starts injecting stabilizing gases, allowing the timing to advance a few degrees or something. Maybe it's a stretch, but that's why it's a theory.

 

As far as I know, it's the original, so over 207k on it. I'm pretty sure it works, though I honestly haven't the slightest idea how to test it. I would expect that the thing would be pretty well carbonized at that milage. This begs the question "How can I test it?" It also begs "Can I clean it out by taking it off and hosing it down with carb cleaner or Seafoam or something?"

 

What do you guys think?

Edited March 13, 2011 by SchwarzeEwigkt
Low vacuum happens when the throttle is open, not the other way around.

[SubaruFred]

You mention "boost", are you turbo-charged? If so check your wastegate and blow-off valve.

[SchwarzeEwigkt]

Nope. It's just a run of the mill EJ25D. I used the word "boost" to describe the little increase in power. I had hoped the quotes would clarify what I meant. Sorry.

[Olnick]

I always notice it when accelerating up the same hill

 

Wonder if it might have something to do with the pickup tube or sock before the fuel pump? Fuel would shift in the tank as the car changes attitude when you start up the hill. Just a thought.

[SchwarzeEwigkt]

It doesn't really feel like a fuel delivery problem, though. I've also noticed it when I make trips to other places in flat ground if I'm accelerating harder than normal, say I need to outrun someone slow to a lane merge or I'm feeling like driving in a slightly more spirited manner.

 

I've also considered that it might something odd about the coolant temperature sensor. No codes and the gauge operates normally as far as I can tell.

 

Dunno. I'm considering taking out the EGR and checking it out. I did notice that the Haynes manual wants you to remove the pipe from the exhaust manifold when you pull the EGR. Any reason I can't just disconnect that inlet pipe right at the EGR when I remove the valve itself?

Edited March 13, 2011 by SchwarzeEwigkt
Smartphone spelling autocorrect FTW.

[Olnick]

I've also considered that it might something odd about the coolant temperature sensor. No codes and the gauge operates normally as far as I can tell.

 

 

Might be worth looking into. The ECTS will not affect the gauge--they are not linked at all. And, IIRC, the ECTS doesn't always throw a code.

 

Good luck.

[SchwarzeEwigkt]

No kidding. The Haynes book said you can check it by unplugging the sensor or something and watching the gauge. Are there two sensors, one for the gauge and the one for the ECU?

[Olnick]

Exactly right! Two different sensors side by side on the crossover pipe under the intake manifold.

[Fairtax4me]

The sensor you want to investigate has 2 wires.

 

Have you checked for vacuum leaks? Anything with a couple few years of age on it is going to have some rock hard vacuum hoses that can crack or split at the ends. With EGR there are quite a few more than without, so chances of developing a leak are more than double.

 

Be sure to check the hoses for the PCV system, as well as the ones that attach to the bottom of the intake. Those can cause some issues if they get loose.

 

 

You can clean the IAC fairly easily and it may help. Subaru's IAC system constantly varies airflow though the valve even when off idle. SLOWLY pour half a can of Seafoam down the IAC tube while running. You will have to hold the throttle open slightly to keep the engine from stalling. I find a flat blade screwdriver jammed in the throttle lever works great to keep engine speed up around 1500 - 2000 rpm while keeping your hands free to hold the can.

 

Have you ever had to change the knock sensor on the car?