08 Outback XT Engine Rebuild Crank No Start

[tpm55]

Looking for some help.  A few months ago I picked up an old 08 Outback XT 5 speed manual that had a bad cylinder 4 exhaust valve.  Since I picked it up, I bought a new shortblock, had the heads rebuilt, and have replaced nearly every vacuum line along with having the injectors serviced.  I finally was able to get the engine in the car and attempted the first start.  No such luck, cranking with an occasional sputter, smells of fuel out the exhaust.  I can hear the fuel pump prime, I don't think fuel is the issue but very intermittent spark.  I am getting the following codes P2008, P2431, P2433, P0340, and P0345.  The first 3 I am not so concerned about as I don't think they would cause a no start, although I could be wrong.  The camshaft sensor codes are the ones I have question marks over obviously.  The car did run prior to the teardown, just not well due to the burnt exhaust valve, but at that time the sensors appeared to work (never got those codes).  I have checked all grounds and I have added my own similar to the iWire setup, so I don't believe that is the issue.  I unhooked the passenger side cam sensor as I can get to it, cleaned it and checked all terminals and seemed fine.  Didn't do anything.  My battery died from all the cranking, so I got a new one and still nothing.  I also retimed it about a half a dozen times and I know its correct, I have pictures if needed.  Anyone with any suggestions or helpful hints?  Ill pry get the multimeter out and check some continuity when I get the chance and make sure the harness is still fine.  I also checked the crank sensor just to be sure and that appeared fine.  Everything has OEM parts as far as I know, was always serviced at a subaru dealer by the previous owner, AISIN timing kit, new spark plugs, new PCV valve, OCV valves, AVCS Cam gears were cleaned and serviced by Outfront Motorsports, new gaskets and seals pretty much everywhere.  I am at a loss at this point as to what is causing these issues. Can 2 cam sensors just go bad?  I know they are old but seems odd to me.  I got the OEM service manual recommended spark plugs and I assume the come pregapped correctly?  I know its never safe to assume. Maybe I am overlooking something simple?  I don't want to just buy parts to throw at the car, never a good idea, but having a known good cam sensor would make things a little easier I suppose.  Thanks in advance for any thoughts or suggestions.

[tpm55]

Also, I do have an accessport, I had tuned it to the stage 1 map prior to the rebuild, should I go back to the base map for the first start?  I can't imagine that will make a difference but it was just another thought I had.

[KZJonny]

Location? Maybe someone is near enough to loan you a set of cam position sensors? Or try a set from a scrapyard?

You've already listed all the likely problems and have plans to address them.

 

There is also a rare, but known problem where the cams themselves get magnetized and screw with the pickups.... Can be seen with a magnetometer (gauss reading)....
Demagnetizing them is possible, and seems to be the solution.... tho admittedly it is uncommon, it may be worth investigating...

[tpm55]

Sorry, I currently reside in Maine, just updated my profile. Ill see if I can find a set somewhere or at least one good one to see if it solves one of the codes.  And magnetized cams? That sounds less than ideal, ill keep that in mind if nothing else comes up/works.  I'd have to look up how to get them demagnetized, I am sure you have to take them out which would be mildly annoying.  Thanks for the response!

[tpm55]

And seems like a stupid thought, but I recently watched a youtube video from off the line performance about rebuilding an ej257, basically same car other than the exhaust AVCS as well.  They mentioned that cam sensors are sided? Didn't make sense to me seeing as the parts numbers are identical on the diagrams.  But if I switched them up could that cause an issue? 

[kzr750r1]

As I recall it's the same part for the cam sensors. Can't remember right now if I replaced mine on the last rebuild. Have had them go bad or pop up a fault.

Since you have an access port will clearing the codes return with the same again?

When the cam sensor went out on mine the ex wife and kids were on their way back from san diego. I had to talk her thru plugging in the access port and clear codes.

Ran fine after and two new cam sensors ordered before they arrived.

I may have some old spares in stock will look this weekend.

You mention timing it several times? I'm puzzled by this as it's not like we have a distributor to adjust.

[tpm55]

So I have reset the codes using the access port a number of times and they come right back. I have unhooked and checked both the tgv code as well and the secondary air injection and that comes back as well. None of that had an issue prior to rebuild which I found odd. Although I did disassemble everything. Including the intake manifold and cleaned everything thoroughly. I wonder if I messed up the motor/flap in the tgv.

 

As for the timing issue. Yes it should be simple and is straightforward. The retiming all started after I realized I had bought an 11mm oil pump instead of 10mm on accident. It probably wouldn’t have made a difference for me, but reading the findings of killerB I figured since it was out I should just do it right so I installed that and retimed it. At that point I tried three times to retime it, lining the marks all up and then turning it over by hand a full 2 revolutions and the cam lines all lined up but the crank seemed off a half of tooth, no matter how many times I retimed it. But I eventually figured it should be fine after reading some similar issues people have run into. So when I tried starting and had both cam sensor codes that was my first logical solution, to check timing, and it did seem off at that time. So I took it apart again and retimed it and at that point I know the drivers side exhaust cam was off one tooth, though that was the best it ran, just had no throttle response at all. Finally I did it one last time, actually counting teeth, so I know it’s correct and I left both timing covers off for now to continue to confirm. So next in line is to check with 2 good know cam sensors.

[KZJonny]

If it comes to this.

[KZJonny]

Also, I believe @jaylew is somewhere in your neck of the woods, tho I cannot speak to him having any spare parts. Given his propensity for extremes of violence as well, I would be cautious about not returning things to him in a timely manner.

I wonder if the video you watched was talking about the intake and exhaust sensors being 'sided'. I've never heard of the passenger v driver side cam sensors mattering, since as you say, it's the same part on both sides?

 

I have spare set sitting on the workbench in my basement, but I'm in Canada, and I would bet shipping + return shipping would get you most of the way to just buying a new one anyway.

[tpm55]

That was a very interesting read, I did read a lot of that thread when I first got the car and got a lot of misfire codes and didn’t know at first about the burnt valve. Didn’t see the most recent posts regarding the magnetism issue. I hope to god I am not having that issue, and simply hope that new sensors fix the issue, but with my luck I don’t know. 

[KZJonny]

Fingers crossed for you it’s something simple that was overlooked.

Welcome to the forum as well.

[kzr750r1]

Ok now on the same page. You have to turn in a bunch more till all the marks will line up again because of the uneven number of teeth.

http://www.youtube.com/watch?v=iCGTdaffoiA&t=291s

[xt2005bonbon]

The number of teeth between pulleys is what really matters to me, as per FSM.

Stupid check: did you make sure the white coil plugs are in the front and black in the back?

Also, it could be possible that one of the intake camshaft pulleys is not aligned exactly right with the camshaft dowel pin. I hope that's not the case. It is somewhat easy, upon install, to misalign that pulley with the camshaft. Some guys had that problem. It's rare but it can happen.

[tpm55]

Yup, black is back, I did double check that again. I never thought about the cam alignment on the dowel. The teeth counting worked out with all lines in alignment on the cams, so I believe timing is good to go. If the new cam sensors don’t fix it, I’ll check wiring and if that is all good, next step would be checking the cam sprocket alignment, from there who knows, maybe the magnetism problem as mentioned, and check some fueling and confirm the injectors are firing with proper fuel pressure as well. Might need help with checking all that if needed.

[tpm55]

SOLVED

So the other night I was over at my garage looking over some different things after having done a bunch of reading in forums about different codes I had.  Some people had indicated that a faulty alternator could cause hard starting and cam sensor codes due to a weird output signal it was giving, so I unhooked it and tried starting again and same issues popped up.  At that time a vaguely remember looking over at the bulkhead wire harness connector and thinking the latch wasn't fully down, but for whatever reason I didn't address it at that time and had to leave anyway.  But it kept nagging me over the next day or so and at work yesterday so I decided quickly to run over late last night after work and check the connection.  It was indeed not fully seated, I unhooked it, blew everything out, inspected everything, and fully connected it.  Went to start, pretty much as soon as I turned the key the car fired, ran beautiful, no CELs.  "Keep it simple stupid" I have heard it a million times when working on jet skis, motorcycles, or whatever else.  I would say this is the largest rebuild I have undertaken, have learned a lot and has thoroughly enjoyed it.  But it ended up still being something I completely overlooked and super simple, THANKFULLY.  Thanks to everyone who replied, I am sure things will still pop up over time and Ill be on here frequently.  Next up is break in time I suppose.  Lots of theories out there on that ahah

[xt2005bonbon]

Very happy for you. It was in fact a simple fix after all!

[kzr750r1]

Great news! Take it easy on the breakin and buy your oil and filters in bulk.

[xt2005bonbon]

yeah. lots of engine braking.

[tpm55]

Well the fix was short lived.  Was running the car trying to get it up to temp for the thermostat to open and burp the coolant, and the idle became a bit erratic, no abnormal noises or anything.  Eventually it just shut off.  I attempted to restart and eventually it did, rev'd fine (nothing crazy), but idle still remained rough.  Once the coolant temp was at 185 it shut off again and wouldn't restart, just sputtered.  Have general misfire codes along with pretty much every cylinder misfire code, and the LH Cam position sensor code.  My assumption is that the sensor may actually be bad causing the other issues (misfires).  Biggest issue is the LH (driver's side) sensor appears to be a bear to get at without tearing a bunch of stuff off, naturally the easy one to get at and change out in a few minutes appears fine for now.  Anyone ever remove the driver's side one without removing a bunch of stuff? I read someone had gotten one out on a WRX but had to dremel the TGV or injector protector a bit to pull it out, ideally I'd like to avoid that. I had ordered a new one anyway because of the issues I had originally had so ill change it out and see if that fixes the issue.  If that doesn't then Ill be most likely getting new coil packs as I don't know if these are the original ones or if some have been replaced.  After it shut off the first time I started data logging on my accessport to see if there were any other issues that could potentially be causing the erratic idle and abrupt shut off, everything looked good as far as i can tell at a quick glance (I am new to more advanced car diagnostics/tuning), but A/F mixtures seemed fine, biggest thing I saw was a few seconds before the car shut off the Fuel correction went +25 from much smaller values, I would assume that could mean vacuum leak somewhere?  But it came out of the blue so not sure.  I guess I will first assess the cam sensor issue and see where that gets me, I knew I was going to regret not putting new sensors on right away when the engine was out, I replaced pretty much everything other than the engine harness itself and all of the sensors, live and learn I suppose.  I know heat increases resistance, so I am going to assume cold start would be fine but once warm the sensor starts to malfunction, or the wiring has an issue at temp.

 

Another side note, didn't notice any leaks other than a pretty bad one at the rear oil banjo bolt in front of the turbo.  I switched to the IAG AVCS and turbo feed braided oil lines and I think I used the incorrect crush washer between the bolt and the cylinder head, so it isn't sealing properly.  In their kit it comes with a thicker 1.5mm one instead of the 1mm standard.  Easy fix other than trying to get at that banjo bolt.

 

And I need a new catback exhaust at some point.  Noticed at the y-section in the back there was exhaust and water dripping out, so there has to be a hole somewhere.  It appears to be the original exhaust and it has spent all of it's years in the northeast, even if it was garage kept, so its no surprise.  Hopefully I can pick up a cheap OEM 2015+ STI take off exhaust from someone.  They appear from the outback forum to bolt up with fairly little modifications other than some hanger bending, and the sound gets a little deeper.  Sounds like a win-win to me if I can get one cheap.

[kzr750r1]

Sorry you have to open up the third level of hell to change the sensor. For good measure you should change the other cam sensor and crank sensor. Coil packs are a good measure as well. But you can make that decision later to see if the other sensor changes help.

Finding leaks is good to find early.

My first turbo change was a mess. In clocking the turbo (BNG 16G) after install popped the drain hose out of position. Didn't notice until there was a lake of T6 on my driveway. 

 

[KZJonny]

If it behaves well for a little while on a cold start, it could be worth putting a hair dryer/heat gun on the cam sensor while the car is still running well.

If this causes the erratic behavior to happen faster, (warm sensor failing) than it did the previous time, it would certainly help to condemn the sensor and justify the work. I didn't remember the cam sensor being that hard to get at, but my engine in peices all over the garage atm, so everything seems easy to access.

[tpm55]

The passenger sensor is a breeze, the driver's doesn't appear to be easy to get to from briefly looking at it, but maybe once I dig into it more it won't be as bad as it appears.  I just couldn't really get my hand down in close to even unhook the wire connector, maybe it won't be that bad though once I figure out a plan of attack.  The hair dryer/heat gun is a great idea though.  I didn't get a chance to go back over to my garage after it had cooled down, but it definitely didn't want to fire back up easy once it was hot. Ill report back later in the week once parts come in and I have time to work on the car between work and school.

[xt2005bonbon]

Also. Some food for thought..

I have two OBXTs. Both suffered over time with some grounding issue between the MAF sensor and MAF wiring harness. AFR corrections were going wonky when warmed up (overly negative corrections). Found the problem by having the engine idling, while starring at the BtSsm app on my phone (or accessport), twisted the MAF wiring harness at the MAF. Corrections were going all over the place while I was doing that. You could also hear changes in idling. Had to replace the MAF connector with a pigtail from iWire. Did that on both cars. Not saying it is your problem, but it is good to know.

[tpm55]

I am open to any and all suggestions, I appreciate it! I knew going into this I would be battling issues such as those I am having so far. Old car with old original equipment. Hindsight is always 20/20, I seriously considered replacing pretty much all the electrical(engine harness) in the engine bay along with the sensors, and last minute convinced myself it would be okay. Here I am now wishing I’d done it, but it is what it is. Learning a lot as I go regardless. When I completely rebuilt my old Honda xl600r dual sport years ago, I made a lot of mistakes but learned a ton, and I know that bike inside out including all of its unique characteristics because of it. This car is already turning into that, and I am fine with it haha, as long as I can eventually get things sorted that is.

[birkhoff]

Your post about timing and re-timing seems like a loose end you should follow up.  When you say the mark is out, you mean the timing mark on the sprocket and not the paint mark on the belt, right?   kzr750r1 is correct that the belt marks only align at setup time and after rotation nothing aligns again: the number of belt teeth is not a multiple of the number on the gears or crank.   But, all the notches must align -- two full rotations and crank mark aligned you should see all the cam gears in alignment with themselves, with the cover notches etc.  That is the basic check (and no binding or crunching noises from within!) for the belt install.  

xt2005bonbon also mentioned the cam gear dowels.  Easy on the phenolic gears, but the alignment of the pins on the intakes is finicky.  There is a stamped dot on the actual toothed part that can help, but in the end, it is a bit of a 'by feel' thing and is it not unheard of to get the gear on but in a rotated position with respect to the cam. It will mess up the back of the cam gear, but it will sort of seat and torque down.  Needless to say this will be a problem for running.

That the engine ran and ran ok, for a while, is inconsistent with either of the above, but since you are two steps back after one step forward,  keeping these things on the table may be a good idea.