P0018 Problems
 

So I got problems. I recently put on a JDL catless UEL header and JDL catless frontpipe. Got a moto-east tune for it. The day I put it all on I threw a P0018 code. I have put the stock ecutek tune on as well and still throw the code. I have datalogged countless times. I did have some times where one of my intake cams would go -29 when the target value was 0. I could rev the engine and the -29 would over shoot to +39 and settle back around 0. I thought this was the issue that was throwing the code. To me it looked like clogged OCV passage/filter. I got an engine flush and oil change. Data logging since then does not show the -29 anymore but I am still throwing P0018. Both intake cams seem to respond well and lag slightly behind the target angles. I assume this is normal since electricity is faster than oil. I have swapped OCV's from side to side and cam position sensors from side to side. Still get P0018 (not the other side).

Any idea what is going on guys? I am out of ideas.

Could it be the Crank position sensor. I assume I would have more problems if that were the case. It seems to drive fine once i reset the DTC's but then I get the code and it become "detuned" (slight limp mode).

Please help!!!

Oh. Mine is a 2015 BRZ.

take a look at the TSB s in the sticky at top of this thread, that cam code is likely covered under warantee on 2013 and prior cars if still covered its a known problem

unfortunalty might be related to the cam sensor the oil control valve,the cam sproket actuator or even the ecu itself, the TSB shows how to diagnose but its quite complex and could possibly require multiple parts to be replaced

p0018 is cam position correlation error, ie difference between what ecu asks for and what the sensor feedback reads.

I have done the first steps of the TSB, swapping the OCV's. Did not change sides. I also swapped cam position sensors. Looking at my logs the Actual Intake Angle and the Intake Angle Target are really not that much different. Which is why I am so perplexed. It does seem that the Bank 2 intake is slightly slower to react than the Bank 1.

From my understanding about the sprocket issue, a spring pops out and the phaser gets stuck in a position (advance I think). Mine is not stuck and seems to respond appropriately 99% of the time.

As for the ecu, the Intake angle targets is what the ecu is putting out. Those seem to be correct. Once I clear DTC is seems to run fine for about two runs (car on...run...car off....car on...run...DTC).

I will post some graphs with my intake angle reaction times.

http://www.datazap.me/u/osudoc11/p0018?log=0&data=5-6-9

Here is a data log once the error occurred. I have intake targets and actual angles on here. As well as the total number of DTC's. It goes from 0 to 1 which should correlate about when the error got triggered. I could use some more experienced eyes on this. Let me now if you need more data.

Im not an expert on this ,but you may not be able to go off where the DTC is triggered, as you don't know what parameters the ecu is using to determine when to set the DTC, it may average errors over time or some other weird algorithm to determine when to set DTC, like with DTC for cat efficiency or even IAM drop on knock.


Ive seen a few instances of these cam DTC and the dealers spend considerable time trying to fix and unfortunately often replace many items before problem is resolved, often they give car back after parts replaced to people only to have code re-occur a week later. It just appears to be a difficult one to diagnose right down to the exact cause.


and it possibly a cumulative error of several parts being on the limit of tolerance.

I would have a look at the cam sensor as per this TSB as well. It not same codes but it one to check and it costs stuff all to fix just shims.


S-SB-0033-13 : Hard Start, Rough Idle, Camshaft Sensor or Misfire DTCs


Im just guessing here but their is a couple of instances when both cams are targeting opposite to what the ecu is asking for (target angle) seems weird but you probably need to log a good car and check. Wether this indicates problem with sensor or ecu no idea just looks weird


see mark in log below



http://www.datazap.me/u/osudoc11/p00...rk=416-416-740

Thanks for the tips. Any idea why all of this would occur after catless header and loading the tune? I just seems to be a weird coincidence.

As for the spot you referenced on the log, I think that has to do with the delay between the signal and the time for the system to react. If you shift the target value to the right a bit everything seems to line up. I would imagine some delay is normal.

The tune likely asks the ecu to move the cams in a more aggressive (ie rate of change) manner than the stock tune, and may ask for more overall magnitude of movement.





This may have uncovered a latent defect in the cam actuator system .




I have heard of this on at least one other car, ie car was fine, guy loads tune (used on many other cars without problem), then gets cam DTC codes, goes back to stock tune, no DTC codes.




you probably need to log a car with no defect and compare the logs.

possible "lazy cam actuator" issue may be due to this


think post here is the one

http://www.ft86club.com/forums/showp...postcount=2970

its supposed to look like this

http://www.ft86club.com/forums/showp...postcount=2973

what is your rom calid if your pre B01C or D then you should update to B01C or D


ie if 700 900 A00 A01 B00 go to B01

I am 2015 so it is D00C. I have loaded the "stock" ecutek rom that comes with the program. I am somewhat concerned that maybe these ecutek rom's and the rom my tuner provided do not have the updated specs regarding the P0018 code. I know the manufacturer updated a flash which fixed the issue for a lot of cars. Do you know where I can get a manufacturer stock flash that I can load with ecutek?

D00C should be fine it should have all those updates as you say.


This cam issue is a bit unusual for a 2015 car they sorted it pretty well ib about 2013.


check you did not dislodge the cam actuator/sensor plugs at front of motor.


if you want a stock D00C rom I've posted up most of the stock roms here see end of post.


just download it , unzip it then you should be able to flash it with ecutek, without adding any racerom features.


Those roms were pulled with tactrix or OFT so their not altered in any way.






http://www.ft86club.com/forums/showthread.php?t=62332

I have checked all the connections I can see and everything seems to be fine. I am getting signal from the cam position sensors and to the OCV's when I data log. I would think if something were loose then I would see interruptions in signals. Is the P0018 pretty specific for the driver side intake cam? With these complex systems now-a-days is it possible one of the others is messed up and it is just throwing that code? I put the header on myself and the 02 sensors are down near the passenger exhaust cam. If I dislodged or damaged anything I would suspect it would be around that.

Thanks for all of your help by the way. This is an extremely frustrating issue for me. I spend a lot of money and time to make my car the way I want it and now it is turning against me.

the 02 sensor wiring or sensor faults would cause different codes. like P0420, but in tuned roms they are disabled especially if rear 02 has been re-purposed as flex fuel sensor input.


Im not sure if P0018 is specific to a specific side, judging by the number of TSB and the complex diagnosis of these cam issues its appears very difficult to isolate the exact cause.


As I said previously I am not an expert on this fault, however its reasnobly common and I have read of several instances of similar faults. Dealers have often had cars for several days and had several attempts for fix these cam issues, often replacing multiple expensive assemblies including the ECU to correct the problem.


It fairly unusual in a 2015 model.


My suggestion would be to return the car to stock including ecu rom and return it for repair under warrantee (you did say it now exhibits the fault with stock rom tune I believe). It likely the fault may be in manufacturing tolerance if several components and the combined effect is the fault your seeing. These assemblies OCV, cam actuators/sprockets ECU ect are all expensive , too expensive to trial and error swaps on.


Its extremely unlikely that the problems are a result of your mods, but I would swap everything to stock just to avoid a hassle at dealer.

So I am trying the stock tunes you provided first. They are in .hex format. I wanted to look at them before I flashed. I have ecutek which only seems to open .rom files. How do I flash the .hex file with ecutek?

Ok I think I'm making a little progress. I data logged some more today and threw the code. I had just reset the DTC's. I was just idling and not even moving when the code was thrown. All of the cam angles were essentially zero, as were my VVT intake angle targets. The only thing I noticed that could be off was the VVT Aim angle and the Exhaust hold learn value. These are both expressed as a % and were about 55% at idle. I may be mistaken but I thought the Aim angle was supposed to be 0% at idle. Any advice would be helpful. Also if anyone can data log these values on their car and post them for comparison that would be great. Thanks.

Update. So after logging some more the VVT aim seems to correlate with the OCV duty % which match up pretty close. However, I still don't have an explanation for the Exhaust hold learn value.

Steve, I talked to a local tuner and to him it sounds like the tune. I was going to give the stock tune you provided a try but having trouble converting .hex to .bin. Ecutek uses .bin.

Update again. Loaded a completely stock tune provided by steve. Still got the code within ten minutes. At this point I will be putting the stock header on and taking it to the dealer for them to figure it out. If software changes have not helped then this must be a hardware issue. I don't see how the modifications could have caused an internal hardware issue though. Steve, do you seen any need to put the stock overpipe/frontpipe back on. I was just going to put the stock header on. They shouldn't give me any trouble about the rest of the exhaust, right?

If you're taking off the header you might as well replace the front pipe at least, you can leave your over pipe but yes they will say something if it's not stock and your having an issues, that's their ace in the hole most of the time.

Well the front pipe and overpipe connect via vband. And it sucked getting it snaked through.

Correct it should not matter about the overpipe\front pipe. But some dealers will try and blame anything

Those cam faults are a pain to fix and require a lot of labour parts diagnosis, if it was me i send the car in completly stock engine wise just to be sure and also any other weird electrical stuff like aftermaket headlights tailights etc anthing they can blame on electrical or engine problems

So I understand that the cam position sensors are supposed to have certain waveforms. I logged the cam positions at idle. They all have regular waveforms except intake angle 2. Here is the log.

http://www.datazap.me/u/osudoc11/tim...0&data=0-1-2-3

Seems reasonable to suspect their is a problem with intake cam 2


suppose it could be


The Cam sensor itself
The clearance between cam sensor and pickup
Wiring issue
Ecu input issue

So I just got a multimeter to check electrical on the components. Cam position sensor and OCV valve seem to check out fine.

From my earlier data logs, one of the intake cams was going to -29 degrees at idle. This hasn't happened since my engine flush and oil change. Reviewing the TSB, it seems that anything less than -28 degrees can trigger the code. I am wondering if this is the dreaded ecu glitch that just can't seem to get rid of the code.

Update. More data logging. It seems that one of the cams oscillates more than the other. This is most noticeable at idle. The set point is 0. The "good cam" stays pretty close, within about 0.4 degrees. However, the "bad cam" oscillates between -0.6 and 0.6. Watching the OCV duty it seems that when it gets to -0.6 the duty will increase slightly to bring it back to 0. But then it overshoots a little and hits 0.6. Thus the dugy cycle drops to bring it back to 0. But it overshoots a little and hits -0.6...and so on. It seems just a little sluggish. As mentioned I have swapped ocv's from one side to the other, so I doubt it is that. Any ideas on other bottlenecks that could be causing this "slow" response and oscillation?

Here you can see that when VVT is not activated Intake angle 2 is more retarded. It evens out when VVT becomes active. Shortly after, the exhaust VVT becomes active. This causes a slight delay in intake VVT, as to be expected. However, intake angle 1 recovers nicely. Intake angle 2 oscillates quite a bit longer and slowly recovers. Either oil is being impeded to the cam phaser or the cam phaser is sticky.

http://www.datazap.me/u/osudoc11/tim...log=0&data=0-1

This really is a nice graph.

osudoc11, did you get any further? Im interested as to what the final fix is.

@osudoc11 is correct. I've had this code twice. The fist time it happened I the dealer replace the cam sensor. The second time I had a considerable oil leak from a banjo bolt on my oil cooler. Apparently I lost enough oil that the intake cam was delayed at one point. Just filled the car with oil, wiped the code and everything was working fine.

I had issues with the p0016 code and they ended up replacing my VVTI gears, and gave me a new ECU. Less then a 1000 miles later. I got rod knock. This is what I'm dealing with now.

Rod knock? That sucks. Dealer is going to replace cam sprocket and ecu. You sure you're getting rod knock? Can you log and see that the knock sensors are picking it up.

This sucks I have the same code P0018 and I'm hoping there's a solid fix for this. I've been limping around for over a month and half now :(


Sent from my iPhone using Tapatalk

How strange. I also got P0018 immediately after Moto East installed my JDL UEL/tune.


I started discussing this in post #6. It was a hellish time. I've driven my car 2,870 miles since October 11th 2015. That's something around 230 days.

http://www.ft86club.com/forums/showthread.php?t=97944

Hey Laika, I just read your issue with the p000d now. That sucks. Did you get it fixed? Did you put the tune back on after the initial cam sprocket fix? It really seems like some engines just have this failure recurrently and subaru really has no clue what is causing it. After this subaru it is going to be something without variable valve timing for me.

yes its weird, I suspect that a lot of aftermarket tunes "push" the vvt system harder ie ask for a greater rate of change of cam timing than the stock tune. Hence this may uncover latent defects in the vvt system in some cars

I did eventually reupload the tune. Probably about 400 miles (if not, much less) before I had the OCV failure (second issue). Allegedly the car is fixed now. Ever since I got the car back, my oil light will stay on a few seconds longer at initial cold start up. Kinda dismissed it. I had still fresh Subaru 0w20 in it from the initial failure because they had to drain and replace oil. After changing out to Eneos 0w20 (what the car has spent most of its life on), the light issue didn't change at all. So now I'm a bit worried because there's clearly a lag in building oil pressure here. Only a matter of time until till I find out what it is. In a bad way. Hoping to diagnose it before that.

I don't claim to know anything about tuning but I'd venture a solid bet that this is exactly right. Any thoughts on my oil light issue described above?

might be as per posts here
, the oil pressure switch in these cars is set to about 3psi, basicly total oil pressure loss to trigger


http://www.ft86club.com/forums/showthread.php?p=1789776

Well I can only speculate as to your oil issue. The simplest and hopeful explanation is a failing oil pressure switch. The most detrimental explanation would be that when you were having your issues earlier your rod journals weren't getting enough oil. This could cause wear (spun journal). Now it may be a little to big and allowing oil pressure to escape. I would think if it were this issue you would have a little bit of knock. I honestly feel like a lot of these OCV issues could be solved by checking the OCV filters. Unfortunately they put them in the freaking camshaft housing assembly. So essentially you have to take it down to the cams to get to them. You think they would have learned with the EJ series engines. Eventually Subaru was telling the dealers to just take the filters out on the EJ's.