The famous stumble...fixed?

Yes, as heiche said gives a smoother signal.

When I drive up the same long hill on rt84 a multi-lane interstate, in I drive, I can barely feel the stumble now. In S or S# its smooth.

This is on my 09 Spec B. The wagon has no stumble and hasn't had pretty much forever.

I do recall the first 2005 Legacy GT I test drove back in early June 2004, had it real bad. That car was so new, they had to remove all the plastic bags that cover the seats. I think it had just under a mile on it. I drove a few more that month and none of them had the issue.

Thsnks Max, I will try that mod.

My 2012 has the stumble issue and its very noticeable on my custom Stage 1 tune. On the OTS S1 tune I notice it ever so slightly if I really think about it.

Max, so you just ran the FPR vac line from the FPR to the BPV signal hose instead of the stock location on the manifold?

Correct, Cobb posted this thread explaining why to do this mod.

Here is how I did mine with the 05 FPR:

http://i160.photobucket.com/albums/t188/covertrussian/Cars/05%20LGT/Engine/Fueling/FPR/FPR%20Vacuum_003.jpg~original

Definitely going to do this mod. How are you guys capping off the cylinder #4 nipple?

Good place to attach a boost gauge to .

I had one of those parts store nipples on there, they don't last at all though, they tend to dryrot and crack after only a few weeks.

Jesus, 200+ posts in this thread and the fix is as simple as a t-connector and some hose?

Facepalm.

Eager to try it out, though.

Good place to attach a boost gauge to ..

This is what I was thinking, that way I can eliminate a T and some hose from my BPV reference line, and eliminates the need for a plug. And the variance of cylinder #4 probably isn't enough to notice on a 2" analog gauge anyway, correct?

This is what I was thinking, that way I can eliminate a T and some hose from my BPV reference line, and eliminates the need for a plug. And the variance of cylinder #4 probably isn't enough to notice on a 2" analog gauge anyway, correct?

So I had an interesting experience when I moved my feed to cylinder 4. The boost gauge would bounce around from 10-15psi. See:

http://i160.photobucket.com/albums/t188/covertrussian/Cars/05%20LGT/Interior/Gauges/Boost/th_Boost%20Gauge%204th%20Cyl.mp4

I figured it maybe this is why cyl 4 is a bad feed, so at that point I added a spare inline filter (same as the one right before MAP sensor)

http://i160.photobucket.com/albums/t188/covertrussian/Cars/05%20LGT/Interior/Gauges/Boost/CIMG1305.jpg~original

Still had the same issue, at this point I decided to throw away the small see through vacuum line that came with my Innovative gauge and used a thicker line (maybe 7/32's?) and that fixed that bounce!

http://i160.photobucket.com/albums/t188/covertrussian/Cars/05%20LGT/Interior/Gauges/Boost/20151110_201426.jpg~original

TL;DR This is a long way of saying that cyl 4 feed is perfectly fine for a mech boost gauge, and to throw away the cheap thin vacuum lines that come with most boost gauges

That video is amazing. So that rapid fluctuation in air pressure is just what the FPR would see when hooked up to the stock location. I remember reading that the FPR is actually responsive enough to then cause corresponding fluctuation in fuel pressure - not good.

Just another reason to re-route your signal line on your FPR to a more common plenum source like the BPV.

My Defi gauges come with pretty thick line.

Tell me what you guys think of this plan:

I have a Tial 50mm bpv, which is supposed to run off 1/4" vac line, not 1/8". Currently I'm just running a reducer off the 1/8" line that feeds the factory bpv. I'm thinking of running my boost gauge off the cylinder #4 nipple, run my FPR off the factory bpv nipple on the plenum, and then T my Tial BPV onto the much larger brake booster vac reference line so the Tial gets the flow it's designed for.

What say you?

One way to really tell is to do one of those hood mounted Fuel Pressure gauges, but I'm not that cool to drill a hole in my aluminum hood so meh!

I've actually had issues with this boost gauge and hose on my Nissan too, it would only read -15inHg vacuum tops, but boost read fine. When I moved to a rubber hose it started reading the correct -23inHg on deccel. I always figured the thin hose just didn't deliver enough vacuum to the gauge, then when I moved the gauge to the Legacy and teed into the MAP sensor it worked well enough that I didn't bother switching out lines. Only when I moved it to cylinder 4 did it start fluctuating like that.

In the video that Cobb linked (

Another thing that could have helped is the new hose is soft and pliable (rubber) while the old hose was a very hard see through rubber, but not quite PVC hard. Maybe the new house is soaking up the pulsations while the old one would deliver them straight to the gauge? If that's the cause a PVC hose would really make the gauge jump around.

I have had the stumble/stutter so badly at times that I wanted to sell the car and curse Subaru forever. At it's worst, I would severely lose power while in traffic, often at very bad times, such as when making a left turn across traffic. It got so bad that I left an envelope with family telling them probable cause in case I was ever in such an accident. (Seriously)

What is most aggravating is Subaru's apparent indifference to this persistent issue, not to mention others such as my car smelling like a fuel-air bomb most mornings. I guess their actuaries run the numbers and have decided that the lawsuits are cheaper than the fixes. That and that we will continue buying Subarus no matter how poorly the treat us.

That said, I did the fix shown above over the weekend, and I can say that I spent more time buying the parts than doing the fix itself. My car hasn't been stumbling of late, so I can't say if it has fixed everything or not, but the car isn't running any worse.

My 2-cents worth on the repair: I bought a barbed plastic 1/4" tee at the hardware store, and used a razor saw to shorten it up a bit and simply ran the hose to the BOV into the tee. No trimming required. The hose steps down from the 1/4" to something more like 1/8" going into the FPR. I found a smaller dia hose that would slide inside the 1/4" hose. Alternatively, I'd get a coupler connector and just stretch the smaller hose. Thanks to those who have posted all the good information about this issue.

I've been messing with the street portion of the timing map and AVCS map, and I was able to to induce some serious stumbling/hesitation.

For those that tried the new FPR vacuum location and have longer fuel hoses (or in bay fuel filters) and are still stumbling, what tunes are you running?

... my car smelling like a fuel-air bomb most mornings.

Well this shouldn't be. Your LGT should smell like any other car. Are you talking about a raw fuel smell? (if so, look for leaks, like the fuel hoses/clamps under the intake manifold). Or are you talking about exhaust?

That said, I did the fix shown above over the weekend, and I can say that I spent more time buying the parts than doing the fix itself. My car hasn't been stumbling of late, so I can't say if it has fixed everything or not, but the car isn't running any worse.

Nice work! Hope it ends up being the solution for you.

Definitely want to look into this, just because..

1) Would doing the "$0.16 bpv mod" and hooking the FPR line up after the BPV in any way mess with this mod? AFAIK, the "pressure" reference line still shows boost, just as before. My guess is no, but wanted to ask.

2) what would happen if our cap-off of cyl 4 happened to be bad and popped off? Would cyl4 (or maybe all of the intake?) loose a little boost, so it would run a little rich, and down on power a little? I.e. a non-catostrophic failure? Or could it make things go boom?

I have a gtboostdig20 already teed into the BPV line, so I COULD simply swap the two. I will probably try that first and see how bad the readings are. I have, IIRC, 5/32 plastic tubing.. Hopefully it doesn't cause all kinds of nasty things to happen with the boost gauge..

1. How would you hook the FPR line after the BPV? Since BPV's are end points you can't run something after it in a serial configuration.

2. You would be lean in vacuum (sucking in extra air) and rich in boost (pushing air out). Since stock the Open Loop fueling is based on closed loop values (mostly vacuum), there is a chance that your D learning will be so high that it will lean out in boost.

Going rich is safer then going leaner, but still has it's own issues, like washing the oil off the cylinder walls causing extra wear and contaminating the oil with fuel, also bad.

1. How would you hook the FPR line after the BPV? Since BPV's are end points you can't run something after it in a serial configuration.

Hmm. Either I am way off on this, or I messed up my terminology. Here's what I think I know: there's a 6-10 inch (roughly) air line that goes from the cyl #4 (shown capped off in one of the pictures) to the FPR, which is further towards the bumper (driver's side), and not too far from the MBC, etc. This gives a pressure reference to the FPR, telling it how to adjust fuel. This is the "not so great" source.

The blow-off valve, which attaches to one of the outputs of the TMIC, has an outlet line which sends a reference PSI signal to various other parts, as well as the "bleed off boost" line, which is thicker, and goes back to the turbo inlet.

I have teed into the blow-off valve's pressure reference line for my boost gauge for years, so I'm pretty sure it's correct and works. My thought was to plug my cyl-4 FPR feed into this tee instead, and move my boost gauge input to the cyl-4 feed. If the boost gauge didn't like the result (as yours didn't at first), then I might add a second tee to the blow-off valve pressure reference line to add the boost gauge back there.

Did I understand correctly? Which term did I use incorrectly? BPV? Could be..

2. You would be lean in vacuum (sucking in extra air) and rich in boost (pushing air out). Since stock the Open Loop fueling is based on closed loop values (mostly vacuum), there is a chance that your D learning will be so high that it will lean out in boost.

 

Going rich is safer then going leaner, but still has it's own issues, like washing the oil off the cylinder walls causing extra wear and contaminating the oil with fuel, also bad.

Right. I do understand long-term it's all bad. I was more worried about: "oh, something's wrong - let's check that cap in 2 minutes when I can pull over" time frame - if I blow the cap off during boost, how quickly would something break? I'm going to try my best to cap it well, but.. Stuff happens.

I think you got the order of air travel mixed up. The way it works is, the line starts at the intake manifold, then it runs to the BPV, intake manifold provides suction and pressure to keep the BPV closed under boost and open under vacuum. Thus when you tee into the BPV line, your actually stealing vacuum/boost right before it hits the bypass valve.

Now the BPV does have the bypass portion, which is the ~1" hose, when the intake manifold is under vacuum, it opens the BPV, which then lets the pressure in the TMIC bypass into the turbo inlet. You shouldn't touch this hose at all.

If cylinder 4 nipple comes off you should be fine for the most part, depending on how big of a boost leak it makes. I wouldn't beat on the car when that happens though. I would suggest carrying spare nipple covers in the trunk, they are cheap and small. Or better yet just use that source for the boost gauge and then you don't have to worry about it popping off.

I think you got the order of air travel mixed up. The way it works is, the line starts at the intake manifold, then it runs to the BPV, intake manifold provides suction and pressure to keep the BPV closed under boost and open under vacuum. Thus when you tee into the BPV line, your actually stealing vacuum/boost right before it hits the bypass valve.

Ah, interesting. I always thought it was sending the pressure from the TMIC on as a reference. So it's actually the control signal FOR the BOV, not an output? Then the condition of the BOV itself doesn't matter (for the purposes of this mod).

Any harm/loss of efficiency in teeing into it there, as I've done? And did I understand correctly that this IS where others teed-in for the FPR mod?

You got it on it being the control signal for the bypass valve(bpv)/ blow off valve (BOV). That signal tells the BPV/BOV what to do basically.

If everything is in perfect condition there is no issue in teeing into the line, since both the BPV/BOV and the FPR are enclosed diaphragms that don't leak or pass air through. This is why it's not advised to tee a boost gauge into the BPV/BOV or FPR line. If your boost gauge is crappy or springs a leak, you could be affecting two of the most crucial components to for keeping your engine and turbo alive.

...I was more worried about: "oh, something's wrong - let's check that cap in 2 minutes when I can pull over" time frame - if I blow the cap off during boost, how quickly would something break? I'm going to try my best to cap it well, but.. Stuff happens.

If it ever were to pop off I'm pretty sure you'd notice it pretty quickly, especially at idle. You may not notice on a long flat stretch of highway, but as soon as you tried to accelerate up an incline you'd notice something wasn't right. It probably wouldn't feel bad enough to make you want to pull over "RIGHT NOW!", but it would be noticeable. I highly doubt that it would cause any long term damage to your motor. It certainly isn't healthy, but if vacuum/boost leaks were catastrophic then people all over the world would be blowing up their motors every day. Just my $0.02

I got around to doing this mod and am seeing a pretty large reduction in my fuel trims, definitely worth while!

And no more stumble!