Sudden loss of power

[89Ru]

2007 Outback 2.5i wagon, base.  110k miles.  5 speed manual.  Recent hard drive through the mountains late for a wedding avg 80-90 mph about a month ago without any problems.  Wife's daily driver.  On a morning drive all of a sudden it could not hill climb.  Any incline it loses power.  Weak acceleration from a stop.  I can feel AVLS kick in around 3500 rpm but it hardly helps power.  On flats or down hills it can eventually get up to 65-70 but it takes minutes.  On steep upgrades, it struggles to maintain 40 mph.  There is a rattle from the front with engine load that sounds like preignition.  No CEL.  MPG has dropped from 27 to 18.  Coolant temps running a bit hotter than usual, hot heat in cabin.  Not overheating.  

Recent work in the past year:  Has new radiator and water pump.  Timing components replaced last year with Gates kit by previous owner.  I think aftermarket wires per PO.  Recently changed to new NGK FR5AP-11 plugs.  Gates tb tensioner was leaking so it was replaced with OEM, $176, ouch.  Head gasket job last year.  Original exhaust.  During engine work I noticed wet oil around exhaust port 2nd cylinder, so I'm thinking a ring problem, it consumes a lot of oil but hasn't been starved.  Car is new to me one year ago.  After last year's engine work, we renamed this car to "damaged" and it is living up to this quite well.  About four months ago got a P0420 code (catalytic converter efficiency) but it went away and hasn't returned.  Last summer it flagged CEL p0026 and that forced it into limp home mode, had loss of power (would not go above 4k rpm) but this is different.  That CEL went away after replacing both Variable Valve Lift Diagnosis Oil Pressure Switches, and power was restored. 

So the possibilities are:  Cat or muffler obstruction.  Slipped a tooth on timing.  There is no engine fuel filter anymore compared to my 2002 and older.  Fuel injectors?  Plug wires?  Haven't tried anything like pulling plug wires, trying to finish an engine swap in a 2002 H6 that is taking all my garage time.  How are fuel problems diagnosed?  Pull injector connectors one at a time while running?  I don't have a fuel pressure gauge.  Any help is appreciated.  Going to start with plug wires, then fuel, then timing, then exhaust.  

[forester2002s]

Knock-sensor failing? (although this would usually set a P0328 CEL)

[89Ru]

If it's pre-ignition then I'd say the knock sensor wasn't sensing it.  CEL isn't burned out either.  ECU wouldn't limit power without flagging CEL right?

[1 Lucky Texan]

if it misbehaves while idling, squirt some starting fluid in and see if it smooths-out.

there's some vacuum gauge test that might find exhaust blockage.

[89Ru]

Idle is unchanged.  I like the vacuum gauge test idea. 

Edited May 5, 2019 by 89Ru

[78 brat]

Had this problem with an 08, upstream o2 was literally warped from terrible plugs..so much unburnt fuel. New plugs and assetone plus blow gun solved it. High compression motor problems lol

[montana tom]

If you can get it in a dark place , when its acting up . I think you will see the catalytic glowing .  Too much raw fuel being dumped.

[89Ru]

7 hours ago, 78 brat said:

Had this problem with an 08, upstream o2 was literally warped from terrible plugs..so much unburnt fuel. New plugs and assetone plus blow gun solved it. High compression motor problems lol

I like this thought!  PO had all wrong AND mismatched plugs.  I was like what were you thinking?  What's a blow gun?  Torch?

[89Ru]

29 minutes ago, montana tom said:

If you can get it in a dark place , when its acting up . I think you will see the catalytic glowing .  Too much raw fuel being dumped.

Eek!  So that's where my 9 lost mpg's are glowing to haha.

[78 brat]

9 hours ago, 89Ru said:

I like this thought!  PO had all wrong AND mismatched plugs.  I was like what were you thinking?  What's a blow gun?  Torch?

I ran assetone through the head pipe or whatever they call it, not our usual y pipe. But ran it back and forth through the honey combs because the fuel had partially plugged it..and the blow gun I threw a 2 foot piece of brake line on the end so I could run in the head pipe a bit. I'd do the same with the mid pipe as well.

But if I can find the o2 I'll take a picture. It literally warped the end. 

Hope this helps

[78 brat]

Oh yeah, and when I say blow gun I mean the valved head you put on an air hose. 

[89Ru]

On 5/4/2019 at 6:11 PM, 1 Lucky Texan said:

if it misbehaves while idling, squirt some starting fluid in and see if it smooths-out.

there's some vacuum gauge test that might find exhaust blockage.

Vacuum gauge shows more vacuum at 2k rpm vs idle.  If restricted, vacuum would drop.  Don't think there's a restriction.  New O2 sensors on the way.  

[89Ru]

Finally got some garage time.  Replaced both upstream and downstream O2 sensors with Denso parts.  Looks like an exact match to OEM.  Took a short test drive on flats, power is still dismal at heavier loads.  Much less than my 2.2L that has 290k miles.  So pulled out the plugs.  Gapped ok, just some brown ash, doesn't look burnt or oil fouled. 

Dry compression test results:

cylinder 1 205 psi

cylinder 3 165 psi

cylinder 2 180 psi

cylinder 4 185 psi

What's up with 1 and 3?  Just did valve adjustment in the last 5k miles when head gaskets were done and I admit I didn't check compression afterwards.  If timing slipped a tooth wouldn't all cylinders be affected?  Still baffled about this loss of power.  Took apart the air intake.  Not obviously blocked. 

Going to post the video of the vacuum test in a bit.  

[89Ru]

Intake vacuum test

Does this mean there is no restricted exhaust?  

[89Ru]

EZ30 intake manifold vacuum for comparison

Normally running vehicle.  From idle up to 3500 rpm.  Vacuum increases in EZ30, where the EJ253 rises, then drops after 2000 rpm.  Going to drop the exhaust.

[brus brother]

clogged cat?

saw this on a friend's van. started out gradual loss of power then wouldn't even move!

disconnected cat and ran that loud sucker home.

removed cat and dumped out pieces of honeycomb. Reinstalled as far as I know, it is running to this day.

[89Ru]

On 5/12/2019 at 7:53 PM, brus brother said:

clogged cat?

 

Front cat pic.  Looks like it's degrading.  Can't see other side without bore scope.  Some loose material but no big chunks.

[89Ru]

After checking the plugs and compression I reinstalled the air intake snorkel.  Didn't leave anything disconnected, afaik, but got P0102 and P0113 codes when I started it. 

P0102 Mass Air Flow (MAF) circuit is low

P0113 Intake Air Temperature Circuit High Input

Cleared them and they didn't come back on short drive.  Still low power as before.  Then took the cat off for pics above.  Just getting weirder before it gets better.  I think on this car the MAF and intake air temp are one sensor on the snorkel hose by the air filter.  Looks like I need a new front cat and maybe MAF sensor.  Other than P0420 many months ago these are the first codes I have gotten in a while.  

[78 brat]

2 hours ago, 89Ru said:

After checking the plugs and compression I reinstalled the air intake snorkel.  Didn't leave anything disconnected, afaik, but got P0102 and P0113 codes when I started it. 

P0102 Mass Air Flow (MAF) circuit is low

P0113 Intake Air Temperature Circuit High Input

Cleared them and they didn't come back on short drive.  Still low power as before.  Then took the cat off for pics above.  Just getting weirder before it gets better.  I think on this car the MAF and intake air temp are one sensor on the snorkel hose by the air filter.  Looks like I need a new front cat and maybe MAF sensor.  Other than P0420 many months ago these are the first codes I have gotten in a while.  

I am truly surprised, not a mouse nest in the air filter? Lol

[brus brother]

did you run with the cat disconnected? was the loss of power resolved or unchanged?

[89Ru]

8 hours ago, brus brother said:

did you run with the cat disconnected? was the loss of power resolved or unchanged?

Ran it with open exhaust manifold.  Sprayed the MAF with contact cleaner.  Power restored!  

Video of the loudness

[1 Lucky Texan]

some MAFs have been known to have poor solder connections inside.....

[brus brother]

yup a real varoom machine!

So after you reconnected the exhaust, was the power still restored or did it once again show signs of constriction?

As I said previously, my buddy took his cat off and cleared out the broken bits and then reattached. No codes or anything.

I read here in the past that people experiencing loss of power when going uphill resulted from gravity moving the bits of cat comb back into the air flow.

seems like you are narrowing down the causes.

[89Ru]

On 5/16/2019 at 4:36 PM, brus brother said:

So after you reconnected the exhaust, was the power still restored or did it once again show signs of constriction?

I read here in the past that people experiencing loss of power when going uphill resulted from gravity moving the bits of cat comb back into the air flow.

I haven't reconnected. My thought is if the cat is crumbling then even if I clear the obstructions with compressed air and solvent, it might leave me stranded in Iowa in the near future.

Below is a picture of some ferrous, i.e. magnetic, crud that I found at the upstream oxygen sensor.  If this stuff is coming from outside the exhaust system, my concern is a new cat will eventually become plugged. 

[89Ru]

After my brief noisy run with restored power after basically running without any exhaust system at all the other day, we're back to more mysteries.  Injected air into mufflers in an attempt to blow out particles, if any, trapped in the center pipe cat.  Didn't see any residue at the outlet triangle flange far upstream.  Installed new aftermarket exhaust manifold/front cat, no heat shields cuz they won't transfer over, with the new oxygen sensors.  Rest of exhaust is original.  Ran the new setup.  Thought it had more power initially but things went downhill, back to crappy power but maybe just a smidge better.  But wait, there's more.  Now some new symptoms.  During acceleration the CEL flashed rapidly on and off, on two separate occasions and then went out.  Stalled twice during idle or coasting into a turn.  On one occasion after it stalled, there was some weird shaking while cranking, a bit rough to start, but then it calmed down.  Idle seemed intermittently low/rough.  Could be due to resetting the ECU to clear a P0031 code that I got while running with both O2 sensors disconnected.  Maybe the ECU was confused after initializing with an obstructed exhaust.  Still running hot with pre-ignition pinging noises.  Knock sensor is original.  No codes.  CEL only flashed, didn't stay solid.  In the past I have had misfire codes on another subaru that were preceded by flashing CEL but they always flagged a code and had a solid CEL on afterwards.  

Going to drop the mid pipe cat next and run it.   No O2 sensors that far back to mess with.  I think I need to upgrade my code reader.