Just Did Head Gasket...MISFIRES, LEAKS

Snapped a timing belt on my EJ253 w/ AVLS back in June. Acquired a JDM engine.

 

During the swap I did many things including:

 

• Six-Star head gaskets (following Six-Star's torquing instructions, not Subaru's)
  
• USDM spark plugs (NGK FR5AP-11 / 5463)
  
• *Half* of the studs (apparently one set only covers one head, so I staggered them out between the two)
  

Engine starts, runs and drives but it:

 

• idles poorly,
  
• throws misfire CEL codes (P0301 to 4) and
  
• leaks coolant from the (L) side.
  

No oily residue in coolant reservoir. No mayonnaise on dipstick. Based on the symptoms my friend thinks it's a head gasket issue. It's sitting at the dealer now awaiting a leakdown test; they'll call me in about 8 hours to give me the verdict.

 

I just dropped thousands of dollars replacing and servicing the engine as well as replacing a slew of other parts. I'm incensed that the very thing I tried to prevent happened not 300 miles later.

 

I fear I will get a phone call I don't want. Assuming I hear the worst, what would you guys recommend?

 

Thanks.

 

----------

 

Inspection of engine w/ my phone:

 

https://youtu.be/y1hVQkWoMuA

https://youtu.be/Ivp3mi9MJu0

https://youtu.be/ko803buRKqo

https://youtu.be/7Ig8adG_rv8

https://youtu.be/QAseT8kTEdo

I don't know.. I don't think it's the head gaskets..

 

I'm not familiar with the jdm engines, but at around the 20 second mark in the last video I see what could be a coolant hose missing a clamp, it looks like there are indentations where a clamp was.

 

Here's a screencap from your video, and then a random picture from my car, of around the same area.

Misfire could simple be a coilpack connector being interchanged during assembly.

 

Double check your timing marks as well.

 

Coolant leak could be from the coolant crossover pipe not being correctly torqued or a seal not replaced.

 

Might be a simple fix for both issues. Don't just jump to the worst conclusion .

[*]*Half* of the studs (apparently one set only covers one head, so I staggered them out between the two)

 

Uhhh...never heard of this. ARP stud kits come with enough head studs to do the entire engine-have used several sets of these and purchased from a couple of different sources. I have no idea what you received.

[*]*Half* of the studs (apparently one set only covers one head, so I staggered them out between the two)

 

Uhhh...never heard of this. ARP stud kits come with enough head studs to do the entire engine-have used several sets of these and purchased from a couple of different sources. I have no idea what you received.

 

When I did a rebuild I think I used felpro bolts and I only received enough for 1 head. Ended up reusing old ones for the passenger side and new ones for the drivers side. Going on 4 years and 40K at stage 2 with no leaks.

To clarify, did you swap out half the bolts for new bolts, or did you swap out half the bolts for studs? I would expect either of those situations to be a tragic fail, since it would be essentially impossible to torque evenly. Regardless of whose torquing procedure you use, torquing two different fasteners to the same torque setting will yield different preloads, creating uneven pressure along the block/head plane.

When I did a rebuild I think I used felpro bolts and I only received enough for 1 head. Ended up reusing old ones for the passenger side and new ones for the drivers side. Going on 4 years and 40K at stage 2 with no leaks.

 

JmP6889928, you're right. It doesn't make a whole lot of sense. In the past week I read the 24 Hours of LeMons forums and learned about ARP studs. Next time I'll use them.

 

Hirschman, you hit the nail on the head (not the cylinder head). I *did* order a FEL-PRO stud kit: ES72212 from RockAuto, to be exact.

 

To clarify, did you swap out half the bolts for new bolts, or did you swap out half the bolts for studs? I would expect either of those situations to be a tragic fail, since it would be essentially impossible to torque evenly. Regardless of whose torquing procedure you use, torquing two different fasteners to the same torque setting will yield different preloads, creating uneven pressure along the block/head plane.

 

*SIGH* :\ Yeah. If it weren't for the "get it done" factor I would have waited until morning and got another six head bolts.

 

I put three of the new bolts on each side of the head and staggered them with old bolts in their respective old positions across each head.

 

About to call the dealer and see the verdict. Will keep you all posted before I leave for work (employer blocks many things, including LGT ).

I don't know.. I don't think it's the head gaskets..

 

I'm not familiar with the jdm engines, but at around the 20 second mark in the last video I see what could be a coolant hose missing a clamp, it looks like there are indentations where a clamp was.

 

Here's a screencap from your video, and then a random picture from my car, of around the same area.

 

 

apexi, we have a winner. The dealer concurs.

 

 

They also commented on a lot of oil (clear & new) leaking all over the place. I've directed them to give that a look.

 

 

They stamped it with a rejection sticker for Virginia State Inspection for an exhaust leak and a taped-up hole in the driver floorpan. I'll just have to be ridin' dirty for a while.

Misfire could simple be a coilpack connector being interchanged during assembly.

 

Double check your timing marks as well.

 

Coolant leak could be from the coolant crossover pipe not being correctly torqued or a seal not replaced.

 

Might be a simple fix for both issues. Don't just jump to the worst conclusion .

Timing marks look good, yes?

 

Sent from my SM-G960U using Tapatalk

I know things weren't fully assembled in that picture, but do you remember if you put the guide back over the crank pulley?

 

https://www.gatestechzone.com/en/news/2016-06-subaru-spacer-tool

Coolant leak could be from the coolant crossover pipe not being correctly torqued...

 

Turns out that's exactly what it was. *PHEW*

 

Dealer service tech found a significant oil leak coming from the oil control valve ((L) cylinder head)...

 

 

...so I'm going to talk to the machine shop, who is attached to a full-service repair shop, to see what they can do.

 

The valves *may* have an adjustment issue, so I'll have the aforementioned shop check that as well.

 

A lean code (P0171) persists, and upon spraying brake cleaner on the intake manifold gaskets the RPMs dropped a little. I'll have to do them again. ~shrugz~ I'll also clean the MAF sensor. It's not the EGR valve, as the pipe didn't burn our hands while the engine was running.

 

A wiper (which they replaced), a hole in the driver's side floor and an exhaust leak at the mid pipe 'Y' (just aft of the rear axle shafts) render it 'REJECTED' for Virginia State Inspection:

 

• Wiper - replaced by dealer.
  
• Hole in floor - Virginia inspection manual (which he showed me) says if I patch it up w/ metal attached by rivets, screws or welding it'll pass (1). I'll get some sheetmetal, some body sealer, some screws or rivets and a B.F.H.
  
• Exhaust leak - Meineke is suggested.
  

I'll hold onto documentation, order parts and have my plan ready to explain to the constable on patrol.

 

Sincere thanks to everyone for their suggestions and moral support. It feels good to not be the only one out there who loves hooning cars.

 

Reference:

(1) "19VAC30-70-270. Floor pan." Virginia State Police Motor Vehicle Safety Inspection Manual (October 3, 2016). http://www.vsp.state.va.us/downloads/Official%20Motor%20Vehicle%20Safety%20Inspection%20Manual%20(November%2015%202016).pdf

I know things weren't fully assembled in that picture, but do you remember if you put the guide back over the crank pulley?

 

https://www.gatestechzone.com/en/news/2016-06-subaru-spacer-tool

 

I did, albeit without the Gates GAT0001 Spacer Tool, as Jeg's has them backordered. I broke out my tape measure and did my best, which wasn't much. This time I installed a milled Tomei part (https://www.amazon.com/gp/product/B00COYNEMQ), not the rubbish stamped OEM one that FAILED on my hard shift:

 

I did, albeit without the Gates GAT0001 Spacer Tool, as Jeg's has them backordered. I broke out my tape measure and did my best, which wasn't much.

 

The proper tool for the job is a set of feeler gauges, like $3 hanging by the cash register at your auto parts store.

Update:

 

*JDM* EJ253 w/ AVLS:

 

-USDM OEM air filter

-USDM intake manifold (except ignition & injectors)

-JDM ignition assembly

-JDM fuel injectors

-JDM-spec spark plugs (NGK 2300)

 

Still have stubborn P0301 to 4 (misfire) and a pending P0171 (lean bank 1) codes. No other ECM codes. Took longer than usual to start last night. Idles poorly (alternates between poor and higher, smoother idle). Drives acceptably.*

 

-Heads recently rebuilt

-Head gaskets recently replaced and torqued per gasket supplier instructions*

-MAF liberally sprayed w/ CRC MAF cleaner & reconnected securely

-Intake manifold gaskets replaced; surfaces scrubbed w/ 360-grit cloth and EGR pipe & bolts tight

-Spark plugs replaced & torqued to 20 ft-lbs (recently-installed USDM / 5463 ones were white after only a few hundred miles...concurs w/ P0171 lean code)

 

Will try a few more troubleshooting steps throughout the week:

 

• Will confirm all air duct clamps are secure

• Will confirm:

[emoji830] block-to-air chamber hose

[emoji830] valve cover-to-air chamber hoses

are connected securely

• Have yet to do a brake cleaner "spray" check (for RPM drop) on:

[emoji830]air duct

[emoji830]said hoses

[emoji830]throttle body gasket

[emoji830]new intake manifold gaskets

• Will confirm EGR pipe is tight and not overheating

• Will confirm intake manifold bolts are torqued to USDM spec

• Have yet to check or clean oxygen sensors

• Have yet to plug exhaust leak at 'Y' pipe

• Have yet to run P0171 diagnostic procedure in Mitchell1 DIY

• Have yet to run misfire diagnostic procedure in Mitchell1 DIY

 

Thanks for your help.

 

 

 

 

 

Sent from my SM-G960U using Tapatalk

Have you got a spare set of injectors to swap out?

Have you got a spare set of injectors to swap out?

Swapped out USDM injectors for JDM ones on Saturday. I no longer notice the idling issue though the codes came right back.

Can you also see what the sensors are reading via obd while its running ?

 

What does the ecu read as the coolant temp for example and is it a realistic value ?

 

What are the fuel trims when running?

If idle is surging up and down, then STFT are bouncing all over the place. Front O2 sensor causes this when it goes bad and doesn't necessarily set a code. I would go there first as I've fixed about 7-8 cars with this problem. Of course, these were all turbo cars, but the O2 sensor does the same thing on all of them in controlling the ECM reading the oxygen in the exhaust.

Can you also see what the sensors are reading via obd while its running ?

What does the ecu read as the coolant temp for example and is it a realistic value ?

What are the fuel trims when running?

 

I will try to get some numbers captured and posted OMW to the kennel and Hertz tomorrow.

 

I haven't even checked the intake clamps or even done a spray check yet. Work and sleep are passionately committed to stonewalling DIY work on my cars (hence my Legacy sitting for four months). I'm renting a Focus wagon and heading to MI this weekend, so I won't be able to work this problem nor deal with the VA inspection failure items.

 

Next week if I continue to get stonewalled I'll have to either seek out a shop or sacrifice sleep. It's been 500 miles with these lean and misfire codes, and I can't imagine blowing a piston is any fun.

 

If idle is surging up and down, then STFT are bouncing all over the place. Front O2 sensor causes this when it goes bad and doesn't necessarily set a code. I would go there first as I've fixed about 7-8 cars with this problem. Of course, these were all turbo cars, but the O2 sensor does the same thing on all of them in controlling the ECM reading the oxygen in the exhaust.

 

The idle surging up and down hasn't recurred since I swapped out the USDM injectors and plugs for JDM ones on Sunday morning (only got 4 hrs of sleep), but the P0171, 301, 302, 303 and 304 codes came right back. Next week once I beat life back for a few hours I'll put a multimeter on the sensor and check resistance / backprobe / etc. Thank you for this valuable insight.

Are you sure you have the correct wires to the correct spark plugs? If you're missing enough for codes in all four cylinders it is possible.

 

It also really sounds like it could be really bad valve clearances. That could cause poor enough compression to have misfires.

Stopped by NAPA after dropping off my cat. Clamps and hoses seemed tight but gave them all a little more. Did some spray checks. Spraying this one from the intake manifold to the hexagonal fitting on the block promptly stalled it.

 

Not sure what it's called but it doesn't appear to have ever had clamps on it. Could this be the culprit or should I keep looking?

 

Sent from my SM-G960U using Tapatalk

Can you also see what the sensors are reading via obd while its running ?

 

What does the ecu read as the coolant temp for example and is it a realistic value ?

 

What are the fuel trims when running?

Warmed up at idle.

 

Sent from my SM-G960U using Tapatalk

I circled the pcv valve in red below. On my car, there is no clamp at the bottom or top of that hose. I'm thinking about replacing my pcv hose sometime, it's rock hard at the bottom where it attaches to the pcv valve. Oddly enough I think it might be normal for that hose to be hard at the bottom though, it almost feels like it's reinforced, unless it really is just that hard from being old.

 

Also, 1.6g/s is slightly lower than what I think it your MAF reading should be, does it ever bounce up above 2, so that it's between 2-5g/s when idling? Might want to try cleaning your maf sensor, making sure to hit the two little resistors that are kind of hidden.

STFT1 is maxed out at 25%, the engine is defiantly lean or the ecu thinks it is.

 

Can you get a reading when the revs are higher also?

1.6g/s is slightly lower than what I think it your MAF reading should be, does it ever bounce up above 2, so that it's between 2-5g/s when idling? Might want to try cleaning your maf sensor, making sure to hit the two little resistors that are kind of hidden.

 

That's the thing! I hosed those resistors down with CRC MAF Cleaner Sat nite!

STFT1 is maxed out at 25%, the engine is defiantly lean or the ecu thinks it is.

 

Can you get a reading when the revs are higher also?

I'll be back on Monday after lunch. I'll grab some screenshots of higher revs as I'm leaving the airport.

 

Happy Thanksgiving, folks, and thank you for walking this issue w/ me.

 

Sent from my SM-G960U using Tapatalk