1999 Legacy Terrible Mileage Sluggish, Smooth Running Engine

[Deeron]

I've seen a few similar threads and have followed the advice on them without much improvement. My legacy (about 150,000 miles), with a brand new head gaskets and timing belts was running sluggishly with very bad fuel economy. I discovered a leak in the lowest point of the filler neck and replaced it without much change in efficiency. I have since put in a new knock sensor and front O2 sensor, which maybe made some small improvements, but the car is still sluggish and getting like 13 mpg (though I can't be sure because the tach/odometer work intermittently. The engine sounds fine with no misfires, vibrations or erratic idling and no CEL. I've sprayed the engine compartment thoroughly with carb cleaner and found no vacuum leak.

 

Could the catalytic converter be bad and would that cause such terrible fuel economy? Is there any definitive way of testing a cat? 

 

If the timing belt were off one cog wouldn't that trigger a CEL? I'm certain it was right when I put it on, though.

 

I don't want to throw expensive parts like an MAF sensor or catalytic converter at it without being certain that they are the problem. 

 

Any ideas?   

[1 Lucky Texan]

easy to check the belt's timing, pull the covers, set the marks - find the tooth count for your engine for a triple check.

 

If you reset the ECU (with a scanner or with a battery disconnect) does the car run better at first?

 

check for a stuck caliper, maybe a brake is dragging?

[mikec03]

What's missing from the OP's post is information about how was the car running and milage before the HG change.

 

If your timing belt is off, you would not get a code other then misfire.   If you have any doubt about your timing belt settings, then you have to take off the cover and check it.  This is a big job, but what else can you do?   By the way, I hope that you know that the hash marks are the required to be used, not the arrows.

[Deeron]

The car ran fine before the head gasket change and it definitely got more normal fuel efficiency. It also ran fine after the head gasket change until I put about 500 miles on it which is when the problem started. 

 

A little update. I had the car idling in the driveway while I wrote the original post and when I came back the CEL was on - code P0325, even though I just installed a new denso knock sensor. I decided to test the voltage to the knock sensor and it read 3.18V which is below the expected 5V reading. Could this low voltage be the issue?

[1 Lucky Texan]

If the ECU responds to that knock sensor by drastically pulling the timing, it will make the car very sluggish and likely be detrimental to mileage.

 

On some engines, care must be taken to route the cable/connector in the same ways as the original knock sensor. There is a lump in the casting that can lead to cracking a knock sensor if it isn't oriented correctly. maybe it took 500 miles to break?

[wtdash]

Your Speedo's not working? Or just the Odo? And/or the Tach?  And the CEL light does work (key On/Engine off), but there are no codes, correct?

 

I'd suspect maybe a clogged CC, too...if you can Remove the exhaust  - yeah Loud AS Heck - you could test it...probably a better way, though.

 

And this is the Phase 2 2.2, not the EJ25D, correct?

[Deeron]

Yes, it's the 2.2 and you're correct, it's the speedometer and the odometer that aren't working. Yes the CEL does work and it actually came again yesterday with a p0325 code, even though that knock sensor is brand new and definitely oriented correctly. 

 

Lucky, I didn't change the knock sensor during the head gasket job, but it did end up being cracked and threw a faulty knock sensor code 500 miles after the HG replacement so I put a new one in. Now, the new knock sensor is giving a faulty code which has me worried that it's the knock sensor circuit. Also, I reset the CEL yesterday, took the car on a 2 mile drive, and yes, the engine had more power. So, the CEL reset does lead to improvement. 

[Bushwick]

A clogged cat can/will reduce power, and it CAN physically (substrate) change over time, depending on how damaged it might be i.e. typically part of a cat substrate melts first, other part still allows flow, so it reacts fairly normal under lighter throttle, but WOT will choke. From my own own experience, a CEL typically isn't tripped until some of the substrate actually breaks free. When it breaks free, the exhaust bypasses the restrictive ceramic substrate section and flows through the empty cavity, which causes the "raw" exhaust fumes like a cat-less car regardless of how long the engine has been running. They can fail from something as simple as a constant misfire that's ignored too long, incorrect gas additives or additives not mixed correctly, poor maintenance with plugs/wires, contaminants entering the exhaust, etc.

 

Easiest way to check is get a vacuum gauge, connect it to a vacuum port on the manifold (stay away from any fuel-related vacuum ports if any) or "T" it into one, and observe the readings. Gauge can found for roughly $15-20 at a parts store. Might need to buy a separate packet of plastic "T" connectors.

https://www.hemmings.com/magazine/hsx/2012/02/Testing-Engine-Vacuum/3709071.html

https://www.motor.com/magazine-summary/mastering-the-basics-reading-a-vacuum-gauge/

 

Google "engine vacuum" if you want to look for other sites. Some have handy illustrations to follow. Also, be aware that IF the ceramic substrate breaks free, it has to go somewhere. Unless you are running a glass pack or straight pipe, that usually means it'll end up in the muffler(s). If the muffler is internally baffled, it'll probably get lodged. If you "kick" your muffler and hear a hollow rattle inside, guess what? Also, probably more of a PITA, is IF the substrate ends up jammed in a section of pipe, like say at a bend, a size reduction section, etc. But a vacuum gauge should be helpful there if it's enough to affect engine vacuum or severely restrict flow. Options will be to try and ram something sharp into the chunk and break it up, or remove the section to get it free.

 

FWIW, I just had to replace the cats on my Lincoln Cont. after a misfire was ignored by the previous owner (it had an electrical issue at the time and was driven enough to damage the cats (probably 50-100 miles). When I got the car, I had that fixed, and e-checked it. It fast passed. About a week or 2 later, I was getting wiffs of "raw" exhaust out the mufflers, but it was mild. After the factory cat pipe was dropped, you could see a chunk of substrate caught on the piping where a "Y" was, which was part of the actual cat housing. Other cat on opposite side of engine was still intact. Also, it took AWHILE to pop a P0420 code. Like 20-50 miles w/o popping it, then while light cruising at low speed, it'd pop. I could reset it with hand scanner while driving, and it'd go another 50+ miles. In this situation, since the other cat was OK, the readings were still apparently "good enough" at the O2 sensors (2 up, 2 down) that it couldn't detect it immediately.

Edited May 12, 2018 by Bushwick