Paycheck

Ok, good to know. I guess I'll start by swapping the parts from passenger side to driver side and see if the problem goes with either the OCV or the sensor. At least they're easy to get at. More soon. $$$

But you replaced both sensors and both OCVs at the same time, right? So it's not clear what the actual root cause was, correct?

Hey, just wanted to reach to see if you had any luck fixing your issue. Also, just to confirm....the code you get is P0340, right? Thanks, $$$

Now why do you guys gotta go and give me hope!? I had planned on selling the car in the spring but now I have to try swapping my LH/RH valves and see if my problem follows the valve. Seriously though, at this point, I'll try just about anything. But this could absolutely be something that manifests in the cold. And I could see it causing the P0340 code in my case as well. Once I get over covid I'll try pulling the valves and see where it leads. Thanks for the info! $$$

Thanks, CR. I'll take a look at those when I can and give an update. If it's leaky injector o-rings, can just the o-rings be replaced (if it's even worth it)? Or am I looking at new injectors? $$$

The MAF sensor is on the cold air intake. It's a MAF and intake air temp sensor combined. That got swapped out in the summer when I was modding the car. I got the part from car quest so it's not a Subaru OEM part. The AFR sensor is on the passenger side exhaust header right before the up pipe. Some people call it the "front O2 sensor" but I think that's incorrect. The other sensor on your downpipe is an O2 sensor and it's mostly to verify that your cats are working correctly. The front one is more for fuel ratio feedback. Anyway, I tried removing it last summer while replacing my up pipe. I was trying to get at the exhaust bolts behind it, but I couldn't get it out. It looks like the original one...seized solid. I ended up being able to get at the bolts behind it by using an open ended wrench and coming at it from the passenger side wheel well with the steering turned all the way to the right. So I never did remove the AFR sensor to get a good look at it. I saw an another thread that (if it's bad) unplugging it will fix the idle issue. You'll get other codes obviously, but it will at least indicate the issue. I tried unplugging the AFR sensor yesterday after the engine was cold to see if it made a difference. It didn't. The cold weather issue is kind of an idling one, yeah. All the details can be found here. See my super long post near the end of the thread. https://legacygt.com/forums/showthread.php/removing-05-gt-camshaft-position-sensor-p0340-263392.html $$$

Well, last night when got home I took a chance and downloaded the latest LGT map / firmware. Upon doing so, the car started right up with no issues. It was already warmed up at that point, so I left the default map in there and waited until this morning to try again. This morning, although it wasn't SUPER cold, it still chugged and sputtered away as it normally does. So there's still some other issue at play here that I've been struggling for years now to find. I'll keep looking as long as the weather stays cold. Tuned map is back in so i don't blow something up. Thanks for all the insight. $$$

Ok. Huge WTF here! Just opened my starting ROM in ECU flash to look at the version. It seems the previous owner was doing some screwing around. Original ECU backup shows an '04 Baja Automatic?!?!? Just a tad different than what the correct one shows. I expect the ECU itself isn't even the correct one for my vehicle. Not even sure if I can flash a Legacy GT ROM in to it. Either way, I don't know if flashing the ECU to the latest factory will solve my cold weather problem or not. But I feel like I should regardless. But that means Phil will basically have start over with the tune process. Don't think he's gonna do that for free! So I guess the question now is....if I download the CORRECT factory ECU file just to see if it starts in the cold, can I then put the last modded tune back in and I'm back where I started? Assuming of course that it doesn't brick my ECU altogether. $$$

MAF is brand new. Replaced it in the summer when I was doing the other mods. Now THAT is exactly what I needed to know! Thanks, CR! I've heard the term "pro-tune" a few times and I'm not really sure what you mean by that. I'm not tuning it myself. I'm getting it E-Tuned by Dynotech in New Zealand. They are one of the recommended e-tuners here on the forums and as far as I can tell, Phil really knows his stuff. I went that route because I have a catless exhaust and the guys in North America that I reached out to didn't want to do it due to EPA regulations. Fair enough. Anyway, this problem predates the mods and tune I did this summer, but Phil told me that the car had already been tuned once before (which I didn't know), so that's what prompted my original question. I'll do some comparisons as you suggest and update when I can. Thanks again! $$$

Yeah, I'm just clutching at straws at this point. I'll try logging data and seeing if something seems out of whack, but it's hard to know what to look for. What it "seems" like is that when really cold, the computer wants to flood the engine. So maybe my AFR sensor is bad and not throwing a code. I believe it's the original one (16 years old at this point). But if it was going bad, why would it perform just fine when warmed up and during summer. Also, I would expect my tuner would have seen some anomaly and raised a flag. Who knows. When I get a chance maybe I'll throw a heat gun on my header and get it nice and toasty before starting.

I guess I was thinking of it more like a firmware vs. software change. You can upgrade the firmware on your desktop computer without changing the software. You can even update the OS without changing any installed programs. But if I understand you correctly, in the case of my ECU, it's all one and the same. A brand new ECU will be completely overwritten by my current tune. In which case....square one.

Hey folks, I have what is most likely a stupid question. But as a chronic ignoramus, I'm gonna throw it out there anyway. When getting an e-tune, am I downloading the entire ECU software to the ECU? Or only tune related data (ie. fuel trims, target boost, etc). The reason I ask is...I've been having problems in cold weather and want to get my ECU flashed with the latest and greatest firmware from the dealership (assuming there's an update available - it's an '05 LGT). However, since my car is modded and tuned, I'll also need to download my latest tune afterwards. So will the tune download overwrite the firmware update and basically put me back to square one? Thanks, $$$

Well, I would consider trying that if I had one handy. Plus, it runs great when it's warm and yet still gave me trouble when the ECU itself was warm (brought it inside). So I don't think it's the ECU itself. I've asked my tuner for help in diagnosing this, but I don't expect a reply right away. He's many time zones and an equator away. He told me that it had been tuned before and I would like to know if there's something in there that the previous owner might have done that's screwing it up in cold weather. Not sure if it would still be there after a second tune or not. Plus, I don't know if he started with my base map that I sent him and then he changed it, or if he just grabbed a default MAP and added the appropriate changes to that. We'll see when he gets back to me. I think he's running out of patience with me, and I can't blame him. This tuning process has dragged on for months for various reasons.

Hi there... I figured I would chime in with my own similar issue. Hoping to find some glimmer of hope of coming up with a solution. Last post in the thread below is a summary of everything I've seen/tried over the last several years. OP, let me know if this sounds like what you've seen on your own car. https://legacygt.com/forums/showthread.php/removing-05-gt-camshaft-position-sensor-p0340-263392p3.html Thanks, $$$

Well, I might as well do what I can to figure it out while the weather is cold. Who knows, maybe someone can offer some insight. I no longer think this has anything to do with the original thread, but since this topic has all my diagnostic history in it, I'll continue on. I'll be brief. (haha! no not really) Symptoms: 1 - Cold weather ONLY (below 10 deg. C) - no code 2 - Try to start the car. Struggles and won't idle. It'll "catch" briefly but won't run. Keep trying and trying.... 3 - Eventually, the CEL comes on and the engine starts right up and will idle. 4 - CODE : P0340 - Camshaft Position Sensor Circuit A. 5 - At this point, if I clear the engine code it will immediately sputter and stall and the process begins again. If I DON'T clear the code it will idle until warm 6 - as the car warms up, it is drivable and runs fairly well up to a point. It also seems to be running very rich in this state. Hard throttle will result in bogging and sputtering 7 - JUST before it gets nice and warm (optimal temperature) the car will start to sputter and want to stall again. If I'm driving at the time (ie. not stopped at a light) it will backfire a LOT. Flames out the back, the whole 9 yards. More indications of running super rich. 8 - Pull off to the side of the road and let it idle for a little more to get nice and warm (still not clearing the code). During this time, if I feather the throttle, I can keep it running. But if I hit the gas hard, it sputters. Usually this occurs at 2000 RPM and above. I will also get misfire codes on all 4 cylinders. Lots of sputtering and misfiring at this point. The neighbours probably hate me. 9 - after 20-30 minutes of idling, I can carry on it runs AWESOME. No issues at all. I can even clear the code at this point and it will NOT come back as long as the engine remains warm. 10 - On a really cold day, if I stop for say, 20-30 minutes, the engine temperature will end up just below the "sweet spot" and I'll be at about step 7 above. And the process repeats. What I've tried: 1. New MAF sensor 2. New Temp sensor 3. New Cam Position Sensor (pasenger side - North America) 4. New Crank Position Sensor 5. New TGV Sensor (passenger side- North America) 6. New spark plugs 7. Cleaned throttle body 8. Replaced Intake gaskets with thicker ones (orange o-rings) 9. Tore apart my wiring harness and checked for anything obvious on the three wires going to my cam position sensor 10. Replaced the cam position sensor connector (repinned) 11. Replaced B20 bulkhead connector under my TMIC. This connector carries the signal wire from the cam sensor (connector housing only - did not repin) 12. Replaced TGV sensor connector (connector housing only - did not repin) 13. Went through the diagnostic procedure for the cam sensor. Stopped just short of the last step which is "Replace ECU". 14. Removed the ECU and brought it in the house to keep "warm". Then plugged it back in to the car and tried starting. No change. 15. Took ECU apart and looked for bent pins, bad solder, etc. All appears fine. I don't think the mods are important because this issue predates the work I did on the car this summer. But here they are... 1. Catless Invidia up pipe 2. Catless Invidia down pipe 3. Invidia 3" catback 4. VF52 turbo 5. AVO TMIC 6. Grimmspeed 3-port boost controller 7. Grimmspeed cold air intake 8. Perrin turbo inlet 9. Cobb 4 bar MAP sensor 10. Tune (still in process) Some web searches indicate that this isn't a unique problem with Subarus, but the causes seem to be varied. So my gut tells me that the code is completely misleading me and that the root cause is actually a fuel / throttle / mixture issue that is causing it to run rich / flood when cold and that throws off the cam sensor in terms of when it expects to see the signal pulse. Just a guess. Finally...I'll add that the previous owner had put on a Viper remote start system and had apparently had the car tuned once already (which I didn't know until my tuner told me). Not sure if either of those things could have an effect. Now, for anyone who still cares, any suggestions would be appreciated. <3 to you all and thanks for the help so far. This is my first Subaru ever and I absolutely LOVE the car...when it's warm. My wife hates it of course, but that's only because she can't see past this issue. $$$

Well, I appreciate your encouragement. Sadly, this wasn't the issue. I was so optimistic, but the car still hates the cold. I'm going to see if my tuner (Phil at Dynotech - NZ) is willing to let me pick his brain a little about the ECU. But he's already gone over and above and has been very patient as this process has dragged on WAY too long. If he can't offer any insight, (which I doubt) then I'm done. Very discouraged. Selling it in the spring. $$$ P.S. On a positive note, once I sell it my wife will be a lot happier.

Ok....This is what I hope to be my penultimate update on this thread. The reason for this is that, first, I just wanted to use the word "penultimate" in a sentence. Secondly, I think I've finally found the root cause of my cold weather starting issue. I won't know for sure until I finish putting my wiring harness back together and let the car sit out in the cold overnight, but I'll know for sure this weekend if I've fixed it. Now for the details. First of all, as a reminder, the P0340 code is "Camshaft Position Sensor - Circuit A" fault. The "circuit A" part of that code is the tricky part. Dave over at Cryotune told me that this code is "pretty reliable". And believe he's correct. The bottom line is that the ECU sees something on the cam sensor circuit that it doesn't like and sets the code. However, I went through the entire diagnostic procedure in the manual and everything seemed ok. So after replacing multiple connectors and still no success, I decided to unwrap my entire harness so I can look for potential trouble spots on any of the three wires going to that sensor. Everything looked good. But while doing this, I found a loose connector on my passenger side TGV sensor. Now, the 0 Volt wire (black/yellow) that comes from the cam sensor is bundled up with several other 0V wires from other sensors and goes to the ECU on a single wire. One of the sensors that share that bundle is the TGV sensor. (see picture 1). So upon closer inspection, I saw that the TGV sensor housing was damaged. So, the connector doesn't click into place and the housing itself is gouged out (see picture 2) which could easily result in the ingress of moisture, poor connection, etc. And since this connector shares a 0V wire to the ECU, I think it may have triggered the cam sensor fault even though it's being driven by a totally different sensor. The sucky part of course is that I can't get at it without removing the turbo, which means the down pipe has to come off too. So yeah...that's awesome. But I will know in a couple of days if this is in fact my problem. More to come very very soon. Thanks for reading. $$$

Well, it looks like I spoke too soon. Sunday was a little cooler and damp in the morning and unfortunately the car was up to it's old tricks. Still sputtering and stuttering. I'm pretty confident I'm on the right track though, based on the readings I was getting on my meter. Next I'll replace the connectors on other end of the cable in question. It's directly underneath my TMIC and it looks a little long in the tooth. Plus, any moisture that gets through my TMIC will basically spray right on top of that connector. Not the best design in the world. Anyway, I'll replace that connector and see what I get. $$$

Ok, so if anyone still cares, I have finally found and fixed the pesky cold weather start issue that I've been having. First I'll say that if you live in the Detroit area, NEVER go to Hodges Subaru. They ripped me off for $1300 for changing my wiring harness which they never did. I only discovered this while doing all the other mods to the car, but there it is. Had they done what they charged me for, this problem would have been fixed long ago. But since they lied to me, I was chasing my tail on other things. On even wrote them an actual snail mail letter explaining that I discovered their deception. No reply. Bottom line, it was the connector. In cold weather, and more recently in very humid weather as well, it would stumble and sputter on startup, wouldn't want to run and give me the P0340 code. Once I discovered that I was lied to, I started from scratch and lo and behold, it was the connector. Repinned it and put a new housing on it and it's fixed. Thanks to iwireservices for that. 3 years later and tons of money wasted and the problem was fixed in 30 minutes with a $16 part. https://www.iwireservices.com/ Now to finish the tune and the sleeper shall awaken. $$$ P.S. Lots of negative reviews coming your way, Hodges. Every social media weapon I have will be aimed at your garbage dealership.

Yes, I'm in the process of getting an e-tune. I went with DynoTech in New Zealand because none of the North American ones I looked up would do it because I don't have a cat on the car. Oh well. That's ok. Phil (dynotech) is the MAN! $$$

Thanks, Rhino, I actually have already gone ahead and ordered the Cobb 4bar kit for the LGT. I also ported the waste gate which helped a little bit with the boost creep. I figure, if my tuner is telling me I need it, I had better get it done. He's the expert, not me. The only things I know how to do on a car I learned from YouTube or this forum. LOL. $$$

Awesome. Post pics when it's all done. Mine is getting tuned now. Just waiting on new MAP sensor and boost gauge. Then it'll be DONE and I'll post pics. $$$

Hi there, I just replaced my exhaust a couple of months ago on my 2005 LGT (non wagon) but it should be pretty similar. For me, the downpipe was one purchased item and the catback was a second purchased "kit". That got me all the way from the turbo exhaust port to the tailpipes. I bought all Invidia. Fitment was great and it looks (and sounds) amazing. You'll likely also need a new turbo to downpipe gasket and downpipe to cat-back gasket. The Invidia cat-back kit is basically 4 individual parts (2 mufflers, mid pipe, y pipe) and it comes with the necessary nuts, bolts and gaskets for those parts. You should probably also get new hangers. On the non-wagon version I needed 5. Lastly, I suggest you get 5 new fasteners to connect the downpipe to the turbo exhaust flange. PB Blaster is your best friend when removing old nuts/bolts. But mine were all hacked to bits and seized and I needed extractor sockets on some and a blowtorch on others. Replaced them with all new ones. I believe 2 of them are a nut/bolt combination (where the heat shield bracket mounts) and the other 3 are just nuts that mount on to the studs in the turbo flange. I also replaced the up pipe because I was doing the turbo at the same time. If you're up pipe is 16 years old, I recommend replacing that too. My catalytic converter in my UP was actually rattling around once I removed the pipe. Probably not a good thing. I also went catless, which is causing me some overboosting issues, but it sounds like a monster now. Love it! If you're considering doing the up pipe, post again here and I'll give you a few tips. The seized bolts weren't fun to deal with, but I managed. Good luck. $$$ P.S. Make sure you use anti seize on all the turbo-downpipe connections. Well worth it in case you have to get back in there later.

Hi gang, I'm in the process of tuning my 2005 LGT and I'm being told I need to upgrade my map sensor because it's maxing out. My tuner is saying to get a 3bar sensor, but I'm also being told by a vendor that the stock one IS 3 bar. However, my data logs suggest that my sensor maxes out at about 18.4 psi or just over 1bar. I tend to trust my tuner/data more than the vendor, but can anyone tell me for sure what the stock sensor is? Thanks, $$$

So, it's been a while since I posted anything about the camshaft code I've been getting (P0340 - Camshaft Position Sensor Circuit A) so I figured it's worth a quick update. It's summer here in southwestern Ontario, but we actually had a cool morning a week or so ago. Not crazy cold....about 10 deg C. Just cold enough to reassure me that my cold weather problem is in fact STILL an issue. I thought I was on to something when I saw the sensor ground was commong to the ECT sensor ground. I replaced the engine coolant temp sensor in the event that it was at least part of the problem. No dice. I've been modding the car for a couple of months now (very slow process with Covid / work travel / quarantine / etc.) but I've finally got all the parts on, got all the air leaks fixed and I'm now getting it e-tuned. (Shout out to Phil at Dynotech in New Zealand. Doing an awesome job.) During my mods, I found a broken ground strap under the engine on the passenger side. From the looks of it, it's been missing for a while. The engine bolt was nowhere to be found and all that was left of the ground strap was the metal crimp end bolted to the chassis. So I replaced both front engine ground straps. Now for a little background on my diagnostics - see my above post with the manual pages. At step 5 of the diagnostic procedure, it wants to see a value greater than 1 megohm. I was measuring about 950Kohm, so I'm just underneath the "healthy" threshold. So, from what I know about car electronics (which isn't much) a bad ground can cause all kinds of weird and seemingly unrelated issues. So who knows....maybe all this time it was just an electrical issue that only happened when cold (ie. before the metal heats up, expands and makes a better connection somewhere so the condition goes away). We shall see. I won't know for sure until the weather gets cold again. If this isn't it, my next step will be a complete tear down and repair of the engine wiring harness. That sounds fun! But other than that, the car is running great. Can't wait for the tune to be finished and it's zoom zoom time. Peace, $$$