TOPIC: 2007 Subaru Forester 2.5 L, Canadian

This is my wife's main car, with 220,000 km on it. The groove is straight forward, with no change necessary to the IAC, as it is already routed to the side over the throttle plate axle.

This year of Subaru has no PCV valve, but has what is known as a PCV System with a electronically controlled solenoid going into the air box/silencer. There are 2 breathers coming from each valve cover and routing into the air intake box; another hose, split into 2 from the intake manifold that is attached to a hose going into the engine on the passenger side below the intake manifold, and one more hose coming again from the same area below the manifold to the electronic solenoid going into the air intake box. I t'd the hose that was originally coming from the split hose from the intake manifold into the hose going into the electric solenoid and capped the hose coming from the intake.

I was unable to get into the fuel trims: I need a different scanner, (dropped it on pavement), so I am unable to provide them. The initial mileage was 31.5 average, and it seemed to increase to 42 mpg initially, but is now back to 31.5. I noted no difference in the 0-60 run (12.4 secs) after the groove.

I plan on re-checking the vacuum lines, and also re-doing the groove. I did the new angle on this throttle body, but it ended becoming more elliptical in the centre,(ended up being further away from the plate in the centre when compared to the area closer to the axle).

Anybody with knowledge or experience on this type of PCV system, does it look like I T'd in the best place? I am not getting any blow-by or leaks, and the oil is still up so it seems to be doing its' job.

More info later!


Here is the TB with the opening for the MAP, blocked while cutting



Here is the finished groove; note placement of MAP route on side over throttle axle.


In this picture, the 2 parallel hoses below the Blue plug wires are T'd together, with the near hose feeding from the intake just to the left, and the next one feeding from a similar outlet, but on the rear of the intake on the right hand side of centre. Notice that this is attached to the hose that goes down into the engine somewhere below the intake on the left (passenger) side. This is the hose that is T'd and re-routed. Note the curved hose in the upper left corner of the picture: one end goes into the white electronic solenoid at the top left edge of picture, and the other goes down into the engine, below the manifold on the left side of the intake, very close to the hose mentioned earlier with the parallel lines. The curved hose is the one that will be tied into.



This picture is just a close up of the previous photo.



Here is the completed T. I used a nylon T, and there was no need to add any hose, as the two hoses in question passed within an inch of one another. Note the capped port on the left end of the upper hose of the 2 parallel ones (the ones running below the blue plug wires)

Work in progress!

Luke

Hi Luke! TracyG here. My first impression is that if the Subie got an increase in MPG at first but then dropped off--it's not the Groove itself. If the Forester got more MPG, did it ALSO get a power increase? By chance did you sniff the exhaust pipe before/after? I cup my hand next to exhaust pipe then bring my cupped hand up and sniff that. I can also feel how much moisture is in exhaust this way. My current belief is that water vapor is a good sign of complete combustion.

Personally if the car got more power and mileage increased at first, I would look to issues outside the Groove--vac. leaks, what the computer says about fuel trims, O2 sensor voltage values/switching rate, MAP or MAF values, etc. (need OBDII scanner, obviously.) With that many miles on vehicle, how old are the O2 Sensors? Remember they get "slow" when old, you could even check the plugs/connectors on O2 wires clean any corrosion w/ brakeclean or contact cleaner if present. My 85 Subaru had lots of OEM vacuum hoses on engine I'd left 'em alone, but when examined they were hard so I replaced lots of 'em. Might try another computer re-learn procedure, heck even try pulling/replacing fuses for Computer/engine function. I'd pull fuses while batt. is disconnected, give it time then re-insert 'em while batt. is still disconn. Sounds like computer had second thoughts about new engine operating conditions.

ON the PCV thing--Sounds like subaru got smarter about this system! As long as both valve covers are vented into the incoming air stream before the TB and manifold vacuum is capped you should be good.

In short might wish to leave TB alone for now look at most everything else first!

My 2 cents, Luke!

TracyG Gadgetman Reno

Hey Tracey,

Just finished pulling everything off, checked all vacuum lines (this Sube only has a few), and all were holding vacuum. This car has been service by the dealer, so I will check to see when they replaced 02 sensors last. Since everything was taken off, I filled in the old groove, and put another one in, a bit more carefully this time, following a scribed mark using a simple jig which gave me a 28 deg cut if the back edge of the burr was at the line. Just finished wiping the ECM, tomorrow I will start the relearning process. To answer your one question, there was no increase of power the first time round. 0 to 60 was still about 12.4 seconds.

I will see how this goes, then go further. Gotta get a scanner that works now. To see the trims in action and the 02 sensors flip sure tells alot.

Thanks for your input, Tracey!

Luke

I just took a few minutes and took a look at all you've said and posted on your Subie, Luke. I want you to have the best gains, and be able to DELIVER the best gains possible for everyone. So, troubleshooting an install is real important. This experience is going to be a great asset for you in the future.

Let's go back to the basics. You say there is no power increase, and that it never had one. This is an almost 100% indicator of an issue in the vacuum.

The only exception to this is electronics added to the system. This tends to confuse the computer, locking its memory. (Dan Merrick may be able to give more detail on this.)

So, given there are no electronics, and the Groove is cut correctly leaving no holes, it MUST be vacuum related.

Period.

To figure out where the problem is, you should do a very simple procedure. Cap each line, one at a time, starting with the largest moving to the smallest. I strongly suspect you are going to find one of those large lines are allowing air through the system. That, or your IAC is stuck, allowing air into the intake at all times.

No matter what, this is clearly a vac related issue, and a substantial one.

When you get the right line capped, you will know it!

Keep plugging away at this one, and remember that just because the MFGR designed something into the engine, it does NOT mean it was done with you benefit in mind!

Oh, and you will probably benefit from slowing down a tad. Make sure you have double-checked EVERYTHING as if someone ELSE had done it. I can't tell you how many times I have sped through a job that I was CERTAIN was right, only to have someone else walk up, take one look, and embarrass me completely!

Thanks Ron. Next chance I have I will check all vacuum lines again. I will check the pcv system as well as the iac valve. I did not see any leaks through the throttle body on the first or this second try. I am currently just in the re-learning mode on the ECM and so will post results later.

Luke

Well, help me out with the explanation, guys! I did a performance test with the Subie today with the second groove, and the 0-60 time was 10.8 secs, consistent across several runs, shifting at 5000 rpm like before. So, it is 1.4 seconds faster now. The mileage check is just starting this tank. So what happened? I did check the vacuum lines before and after the first groove: all good. I did not check the IAC as Ron suggested, so maybe, although it was idling steady at about 700 after the first groove. The second groove did change: I used my jig on the second cut and was able to be much more accurate than the angle I did the first time. (It looked uneven to me and I was not happy with it.)

So, once I get mileage figures, I will report back in.

Just a note: my mileage figures for the first cut may have been off, as it was figured out over only half a tank, as I wanted to know the results sooner than later.

Thanks for your help.

Luke

Hey Luke, well I guess never argue w/ results, eh?

The engine/car must like your second Groove more. Did you take pics of second Groove? I'm still curious 'bout the jig you mention, is it a device or simply a throttle angle setting? As Ron said the Groove needs to be as close to throttle plate as possible. Only exception to that for me so far is Drive-by-wire GM Vortec 5.3 L. That one needed some spacing away from throttle plate maybe 1/8' or so. It took me practice to get comfortable w/grooving by hand, the Panavice helped me a bunch.

Perhaps computer was locked? Re-learn(s) should help that. I suppose just keep monitoring the beast...Oh, spark plugs I forgot have you re-gapped +20%? I also check tire press. it makes some difference.

TracyG Gadgetman Reno

Hey Tracy,
I don't know if the performance issue computer related or not, as right after the first groove before the mileage check, I did the performance test and there was no improvement. The re-learn procedure was the same both times as far as I recall. Perhaps it is the groove, perhaps there was a hose that I put back on not all the way after I did the vacuum tests. I don't know. Big improvement this time with the performance.

I did not take pics of the second groove, as it was similar to the first, but smoother and straighter. The jig is actually a simple tool that is simply a marking tool that gives a reference point to cut line nearest the plate. I can modify it with spacers to get the appropriate angle correction depending what angle the throttle plate is at. It is not a throttle plate angle setting and to work properly, the top of the cutter must always start the cut by touching the throttle plate on the top and the scribed line at the bottom. Once the cut starts, you keep pulling down at the same angle till the shaft bottoms out.

I do have a number of questions: from what I gather, if the groove is too shallow for instance, you can get a power increase without a mileage increase. You can also get a power increase with a mileage increase. But can you get a mileage increase without a power increase with the groove? I would assume not. Can you get a power increase (along with a mileage increase), and then have the computer compensate for whatever, and delete the mileage gains but the car maintain the power gains? Or, are both hand in hand when related to the computer compensation, that if the mileage gains are deleted by the computer over-compensating, then power gains go as well?

Thanks for the inputs and knowledge!

Luke

As for some of your answers Luke,the computer usually will learn to get mpg increase some take longer than others.

And yes the ecu will get better mpg and not have a major power increase,although they usually work together.

If the customer acts like its to snappy on take off open throttle slightly,

then groove, this will still get the mpg results but not quite as aggressive.

The ecu is a learning computer it will adapt but the components must all be in good working condition.

Happy GROOVING DAN GADGETMAN MO. 669 229 4871 Call if you need any help....

Update: In fall time, the mileage showed in increase of ~10%, from 34mpg up from 31.5. Remember, all the hoses have been checked, and are holding vacuum and there are no pinhole leaks in the groove. The power increase is still there, as noted about a week after the groove by my wife as she noted that she was going particularly faster than intended going to and coming back from school. I am planning to put on the new MAPster and fine tune it. Will post when it is done.

Luke

I now have an Actron Elite Autoscanner Pro, so I hooked it up and did a "record data" at 60 MPH, and then re-set the ECM as I filled in the groove and re-cut another.

The first scan showed open-loop at idle, until I was moving and about 25 mph or so. When I recorded at 60 mph, the STFT and LTFT both were active, and the engine was in closed loop status. After the ECM wipe, I went through the re-learn procedures, but all the while, the engine remained in open-loop status with no, absolutely no, fuel trim movements. After coming home, the LTFT showed -3.2 at idle, with the STFT at 0, and still in open loop. I just checked it now after about 150 Kms, and the LTFT moves up to about 6.3 at 60 and down as the speed decreases, the short term doesn't move at all off of 0, and the engine still is in open loop status at all speeds. Strange. I have been doing a bit of research, and it looks like from 2007 and up, because of emission standards, there is a long cross-over time between open and closed loop. I don't know if the computer is learning anything, as there is no STFT movement, although the scanner shows the voltage movement from the upstream 02 sensor. Perhaps it takes X amount of distance, or startups before closed loop happens. There are no pending codes. I will be talking to the Subaru garage soon. Any ideas, anyone?

Just a few notes: This 07 Subaru Forester, besides having no PCV valve, but rather a PCV System, this vehicle also has no IAT (it's fly by wire), and the MAP is a 5VRef system, and should be able to use the MAPster. I am not going to put on the MAPster until the system is on closed loop status and I can do a comparison.

Luke


Luke

I don't know if you got the sitemail about the new YouTube, and wanted to make SURE you got to see the video on fuel management systems. That will answer many of your questions, to be sure.