[Eth207]

After a little digging… P1309
Note the lhd/rhd differences.
If you search the FSM pdf in eg preview or acrobat reader, for a relevant plug name, eg B136, you’ll eventually come to the plug maps eg engine/cabin bulkhead. View attachment 70325

Wow okay I just saw this I’ll check it out. Thank you so much for your help on this I really appreciate it

 

[Eth207]

After a little digging… P1309
Note the lhd/rhd differences.
If you search the FSM pdf in eg preview or acrobat reader, for a relevant plug name, eg B136, you’ll eventually come to the plug maps eg engine/cabin bulkhead. View attachment 70325

Okay so before I saw this I ended up re-wiring my OCV's and putting them straight to the ECU, still no change. I think this means my ECU is definatly bad. Also with the OCV pins fully removed from the ECU, I still get the CEL. I think its safe to say the ECU has an internal short what do you think?

If that is the case, would it be possible to fix it?

 

[bigBADbenny]

The two options for the ecu are replacement or repair.

Replacement can mean dealership level coding for security: if the ecu is coded to the security handshake set, eg key chip, cluster, bcm etc. Some generations of JDM ecu do not have this limitation, potentially.

Repair can be attempted by any business that does circuit board repairs, including some tuners (eg sick chips in Australia). Providing the AVCS circuit relevant ecu plug pinouts and the FSM ecu block diagram will help a non-automotive tech to isolate the fault.

It’s interesting that the intake manifold grounding experiments almost fixed the issue.
I have no idea why, other than reverting to diagnosis of the energised harness wires with a low amp clamp, to eg check the current flows are heading in the correct direction.

This outlier fault was discussed on the HPA site, where a ground wire that passed a continuity test, but was frayed to a single strand, caused reversed current flow.

Staying Grounded - Troubleshooting Automotive Grounding Issues

Grounding problems are without a doubt the most common cause of electrical gremlins in the automotive performance world. Here's how to fix them.

www.hpacademy.com

Perhaps the auto tech you find with an oscilloscope can check that and the ecu too?

It’s potentially related to the clean and dirty aspects of ecu and chassis grounds.
It’s also important as some electrical control systems in Subaru are ground switched:

ECU Grounding at battery or cylinder head? | EFI Wiring Fundamentals

Having read and watched dozens of webinars, your course, and installation guides, I'm a bit confused by the advice Shannon Davis (Davis Technologies) gives h...

www.hpacademy.com

You might also perform a voltage drop test on the relevant harness wires: it’s a plug to plug, single wire test closing the loop with a multimeter set to volts, a 12v bulb and battery inline. But as you say, no change with redundant wires in place of the harness…
Unless you have say pushed or bent pins at the ecu & solenoid harness plugs.
Also always back probe any related plugs to the harness wire via the insulation, and check the ecu pins have continuity to the board. I also check all plugs and pins for integrity, discolouration/burntness, and use a little dielectric grease to promote continuity via insulation and lubrication and or to open circuit marginal connections.

There’s also checking for the spacer in the main harness plug/s at the firewall including press and shake tests, revisiting the big three and your aforementioned iwire harness kit and instructions.

Did P1307 come back?
If not, swap solenoids side to side and see what happens.

 

[Eth207]

The two options for the ecu are replacement or repair.

Replacement can mean dealership level coding for security: if the ecu is coded to the security handshake set, eg key chip, cluster, bcm etc. Some generations of JDM ecu do not have this limitation, potentially.

Repair can be attempted by any business that does circuit board repairs, including some tuners (eg sick chips in Australia). Providing the AVCS circuit relevant ecu plug pinouts and the FSM ecu block diagram will help a non-automotive tech to isolate the fault.

It’s interesting that the intake manifold grounding experiments almost fixed the issue.
I have no idea why, other than reverting to diagnosis of the energised harness wires with a low amp clamp, to eg check the current flows are heading in the correct direction.

This outlier fault was discussed on the HPA site, where a ground wire that passed a continuity test, but was frayed to a single strand, caused reversed current flow.

Staying Grounded - Troubleshooting Automotive Grounding Issues

Grounding problems are without a doubt the most common cause of electrical gremlins in the automotive performance world. Here's how to fix them.

www.hpacademy.com

Perhaps the auto tech you find with an oscilloscope can check that and the ecu too?

It’s potentially related to the clean and dirty aspects of ecu and chassis grounds.
It’s also important as some electrical control systems in Subaru are ground switched:

ECU Grounding at battery or cylinder head? | EFI Wiring Fundamentals

Having read and watched dozens of webinars, your course, and installation guides, I'm a bit confused by the advice Shannon Davis (Davis Technologies) gives h...

www.hpacademy.com

You might also perform a voltage drop test on the relevant harness wires: it’s a plug to plug, single wire test closing the loop with a multimeter set to volts, a 12v bulb and battery inline. But as you say, no change with redundant wires in place of the harness…
Unless you have say pushed or bent pins at the ecu & solenoid harness plugs.
Also always back probe any related plugs to the harness wire via the insulation, and check the ecu pins have continuity to the board.

Did P1307 come back?
If not, swap solenoids side to side and see what happens.

okay cool I’ll see if a shop can check it out. I live In Hawaii so I doubt I’ll be able to find a super sophisticated shop but we’ll see.

I think the OCV circuits internal ground in the ECU is interrupted (short to power, being that the current changes when power is applied to other circuits). because when I put a the ground wire directly into the back probed OCV connector ground wire, it almost fixed the problem. Whereas wiring straight to the ECU with brand new wires didn’t result in the same outcome. I don’t know if this makes sense but maybe that test made the ground a little better, but the fact that it was still shorting to power kept the problem present.

P1307 has not come back ever since I grounded the intake manifold better. My theory is that the 2 codes were unrelated to eachother, and the right bank was just lacking a good ground for whatever reason.


I had another one idea, being that both OCV’s should behave exactly the same, and my right circuit appears to be working. what if I just jumped wires from bank 1 to bank 2? I’m pretty sure a fault in either circuit will result in the ECU disabling AVCS entirely, but maybe I can bypass that somehow? I don’t even know why there would be 2 independent circuits if they should both function the exact same

 

[bigBADbenny]

Ok so at this point, swapping ocvs is a really basic blind test. When you ohmed the solenoids alone, did they have the same resistance?

Your idea to run both ocv’s from a single banks harness, that can work for eg tgv angle sense repurposed for external sensor use, eg wbo2, but for avcs, it’s unlikely to properly work as the system self calibrates after an ecu reset.

I guess you could try that but try eg just the solenoid power first then ecu ground, it will be immediately apparent if the control strategy is via power or ground: but will it fry the ecu further?

So just side swapping ocv’s is harmless and least intrusive.

So one aspect covered in the fsm around diag-47 is the test mode (temporarily join plugs under dash) and related system reset as mentioned in the p1307/9 etc diagnostics (top of page). These ecu level resets can be done with a dealership tool or potentially FreeSSM app on PC with tactrix or $15 vagcom kkl 409.1.

There’s a chance you can find a friendly dealer, Subaru specialist or local enthusiast to help with that.

 

[bigBADbenny]

Check this out, spiider dccd throwing those DTC’s on full lock:

SUHBARU - 99RSTi V8 EJ207 - now W/FlexFuel

The Story... So I have been registered to this forum for quite some time. I originally had a black 2.5RS as my first car. I loved the car dearly, it is what got me hooked on these AWD horizontally composed animals we call Subarus. With my RS I did minor stuff and stupid stuff. Eventually I...

www.rs25.com

So that’s an indication that this dtc is related to marginal system volts and keeping those clean and dirty grounds seperate (DCCD drawing lots of current on lock).

So pull your battery, charge it and measure volts, leave it for 24h and check the charge after the surface charge has dissipated, is it still +12.4V?

Flashing dash on key on/run means a dropped cel.

Or get your battery professionally load tested.

My car threw CPS DTC’s with a stuffed battery and very tired starter.

You’d hate to replace that ecu if you’d overlooked some diagnostic considerations 101.1!

I’ll keep googling regardless, for more clues…

 

[Eth207]

Ok so at this point, swapping ocvs is a really basic blind test. When you ohmed the solenoids alone, did they have the same resistance?

Your idea to run both ocv’s from a single banks harness, that can work for eg tgv angle sense repurposed for external sensor use, eg wbo2, but for avcs, it’s unlikely to properly work as the system self calibrates after an ecu reset.

I guess you could try that but try eg just the solenoid power first then ecu ground, it will be immediately apparent if the control strategy is via power or ground: but will it fry the ecu further?

So just side swapping ocv’s is harmless and least intrusive.

So one aspect covered in the fsm around diag-47 is the test mode (temporarily join plugs under dash) and related system reset as mentioned in the p1307/9 etc diagnostics (top of page). These ecu level resets can be done with a dealership tool or potentially FreeSSM app on PC with tactrix or $15 vagcom kkl 409.1.

There’s a chance you can find a friendly dealer, Subaru specialist or local enthusiast to help with that.

Yes I agree that it would be kind of redundant to swap the solenoids over, as they have the same ohms and I have actuated and logged both already, I thought that's what you meant when you said to "side to side swap" what do you mean by that? do you mean connecting right side wiring to left side vice versa?


I have been using the "Reset ECU" tool in rom-raider, would that have the same effect? I have been in the process of E-tuning so everytime I put my car into test mode and flash the ECU it kinda resets in that way too.

 

[Eth207]

Al

Check this out, spiider dccd throwing those DTC’s on full lock:

SUHBARU - 99RSTi V8 EJ207 - now W/FlexFuel

The Story... So I have been registered to this forum for quite some time. I originally had a black 2.5RS as my first car. I loved the car dearly, it is what got me hooked on these AWD horizontally composed animals we call Subarus. With my RS I did minor stuff and stupid stuff. Eventually I...

www.rs25.com

So that’s an indication that this dtc is related to marginal system volts and keeping those clean and dirty grounds seperate (DCCD drawing lots of current on lock).

So pull your battery, charge it and measure volts, leave it for 24h and check the charge after the surface charge has dissipated, is it still +12.4V?

Flashing dash on key on/run means a dropped cel.

Or get your battery professionally load tested.

My car threw CPS DTC’s with a stuffed battery and very tired starter.

You’d hate to replace that ecu if you’d overlooked some diagnostic considerations 101.1!

I’ll keep googling regardless, for more clues…

Very interesting. That would be pretty funny if this entire issue was just due to a bad battery. The battery is less than a year old but I will go ahead and get it load tested.

What do you mean by "a flashing dash on key on/run means a dropped cel"

 

[bigBADbenny]

Yes: literally swap the OCV’s left to right right to left.
Remember I cannot see your logs and I’m not doing the tests, so take my advice with a grain of salt, this is like a fun puzzle for me and a potentially expensive and time consuming exercise for you.
All I know is there’s no assumptions to be made in diagnostics and one needs to address the absolute basics first.

So we do know the DTC’s can be triggered by a low current situation when the car is running (eg DCCD draw), potentially on startup too.

DTC descriptions generally also refer to the affected system not necessarily the named parts.

I wanna say the battery issue is more battery current related than volts, eg if your cranking amps are low, it could cause your DTC/s.

So maybe try yet another ecu reset, a known good battery charge booster pack, or a known good battery with at least 600 cold cranking amps, and fire it up. An idle reset is probably not required on a DBC throttle, but you’ll need to let AVCS recalibrate, which might happen around that 60’c point or 10-15 minutes after cold start, around 5 minutes idle on hot start.

 

[bigBADbenny]

Some year models, the entire cluster lighting flashes, during cranking, which can indicate a bad cell in the bank of cells (lead plates submerged in battery acid) that make up the internals of a car battery.

iirc a battery with a dead cell needs replacement, especially the sealed type.

But this is more related to crank no start conditions?

 

[Eth207]

Some year models, the entire cluster lighting flashes, during cranking, which can indicate a bad cell in the bank of cells (lead plates submerged in battery acid) that make up the internals of a car battery.

iirc a battery with a dead cell needs replacement, especially the sealed type.

But this is more related to crank no start conditions?

Okay this keeps getting more and more confusing.

I got my battery tested at AutoZone and it checked out (I don't know how fancy their testing equipment is but he said it applies load to the battery) I also put a different battery in the car that I know is good. no change.

I took out the OCV's, I noticed the right side was drenched in oil, and the left was somewhat dry

I reset the ECU, swapped them to opposite sides and let the engine warm up:
-CEL remained the same (only p1309 for right side)
-VVT duty cycle remained at 9.41% for both sides
-OCV current stayed the same as before (however the left side no longer fluctuated when power was applied to other circuits, it just remained at 440mA while the right side was at 96)
-VVT angle was strange. Usually they are both at 0 and the right side will jump to 1 or 2 randomly, now the left side is the one jumping up like that.

Very weird how the CEL didnt change, and the oil thing is confusing me Im not sure what is going on.

again, Both OCV's have the same resistance and they both fully advanced my cam gears when I applied 9v to them during idle.

I tried thinking back to when the CEL first came up. And I am pretty sure i got the p1309 after changing out my turbo. I then pulled over and disconnected my OCV and plugged it back in, the code went away. A few weeks later the code came back, and then p1307 came too, and now here we are. I dont know if this is relavent info or not.

 

[bigBADbenny]

All info is relevant.
But wow, what a weird outcome so far.
If the rhs OCV wasn’t getting oil supply, the union screw/banjo bolt on the rear of the rhs head potentially having a blocked filter, one might expect the turbo to be also oil starved, meaning it should have blown up long ago.

If the banjo bolt on the rhs OCV manifold has a raised nipple on top, then that’s a filtered banjo bolt. They are interchangeable with non-filtered bolts, an aspect worth checking.

You could try a flood test on idle, pack the areas below with rags, crack the OCV manifold bolt slightly and see if oil issues forth.

Be conservative doing that bolt back up, they’re just over nipped up on the copper washer, preferably replace the washer, it’s a soft bolt, can break with too much torque.

It’s a super long shot, but I have heard of clogged oil supply tubes, also mentioned in the fsm in reference to P0011.

Your model was the (correction) second implementation of AVCS in any Subaru maybe, meaning early days, those DTC’s are apparently unique to (correction) V7& v8 STi, this is potentially part of the issue (vague diagnostics?) where they’re sometimes fixed with ground or current related fixes.

Googling continues:
Replaces V7 ecu with V8 ecu: either the shop took care of any security coding or it’s not a thing for this year Jdm ecu:

v7 ej207 throwing p1309 code - NASIOC

v7 ej207 throwing p1309 code Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo)

forums.nasioc.com

This is why a basic data log is a good first look: here the codes were in conjunction with a faulty (front?) exhaust sensor: the clue being afr stuck on stoich 14.7 afr and the engine being limited to an open loop fuel map?

P0031 and AVCS Fault codes - ScoobyNet.com - Subaru Enthusiast Forum

General Technical - P0031 and AVCS Fault codes - Hello, I'm currently chasing faults on my 06 Blobeye which are fault codes P0031, P1307 and P1309. I've checked the resistance of the heater elements on the front lambda sensor and everything seems fine. The car still runs normally (not in limp...

www.scoobynet.com

This page refers to the generic obdii origins of the code (non SSM), another good reason to make a datalog:

2005 Subaru Impreza OBD2 Code P1309 Misfire Detection Monitor Not Enabled - MyCarSpecs United States / USA

2005 Subaru Impreza OBD2 Code P1309 Misfire Detection Monitor Not Enabled

www.mycarspecs.com

Did you log or test your neutral safety switch?
Also your engine coolant sensor?
Brake lights all working properly?

I’m leaning towards the originating fault being an rhs bank 1 issue.

So I’d also ohm out the cam position sensor harness & plugs, inspect the cam sensor for sludge or swarf on the sensor face, check for oil in its plug etc. I have had one cmps harness resistance fail a resistance check…

Here’s one example indeed, leaking cam sensors being a known fault that can have drastic consequences: oil inundated harness…

Плавающие обороты после активной езды. Ошибки P0328,P1307,P1308

Добрый день.Ездим на impreza 00 г.в. МКПП 2 литра атмо GD9-002368.Проблема следующая - после активной езды (крутим двиг. до 5-6 тыс. оборотов) начинают плавать обороты как на ходу так и на холостом когда стоишь. На холостом плавают от 750 до 1500. И так продолжается пока не заглушишь двиг. и не...

sti-club.su

Here’s my logging how to in case you missed it: open source and AP:

Make and share datalogs! • club.liberty.asn.au

Do you have AP? Any idea who tuned it, or what tune you have?

 

[Eth207]

All info is relevant.
But wow, what a weird outcome so far.
If the rhs OCV wasn’t getting oil supply, the union screw/banjo bolt on the rear of the rhs head potentially having a blocked filter, one might expect the turbo to be also oil starved, meaning it should have blown up long ago.

If the banjo bolt on the rhs OCV manifold has a raised nipple on top, then that’s a filtered banjo bolt. They are interchangeable with non-filtered bolts, an aspect worth checking.

You could try a flood test on idle, pack the areas below with rags, crack the OCV manifold bolt slightly and see if oil issues forth.

Be conservative doing that bolt back up, they’re just over nipped up on the copper washer, preferably replace the washer, it’s a soft bolt, can break with too much torque.

It’s a super long shot, but I have heard of clogged oil supply tubes, also mentioned in the fsm in reference to P0011.

Your model was the (correction) second implementation of AVCS in any Subaru maybe, meaning early days, those DTC’s are apparently unique to (correction) V7& v8 STi, this is potentially part of the issue (vague diagnostics?) where they’re sometimes fixed with ground or current related fixes.

Googling continues:
Replaces V7 ecu with V8 ecu: either the shop took care of any security coding or it’s not a thing for this year Jdm ecu:

v7 ej207 throwing p1309 code - NASIOC

v7 ej207 throwing p1309 code Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo)

forums.nasioc.com

This is why a basic data log is a good first look: here the codes were in conjunction with a faulty (front?) exhaust sensor: the clue being afr stuck on stoich 14.7 afr and the engine being limited to an open loop fuel map?

P0031 and AVCS Fault codes - ScoobyNet.com - Subaru Enthusiast Forum

General Technical - P0031 and AVCS Fault codes - Hello, I'm currently chasing faults on my 06 Blobeye which are fault codes P0031, P1307 and P1309. I've checked the resistance of the heater elements on the front lambda sensor and everything seems fine. The car still runs normally (not in limp...

www.scoobynet.com

This page refers to the generic obdii origins of the code (non SSM), another good reason to make a datalog:

2005 Subaru Impreza OBD2 Code P1309 Misfire Detection Monitor Not Enabled - MyCarSpecs United States / USA

2005 Subaru Impreza OBD2 Code P1309 Misfire Detection Monitor Not Enabled

www.mycarspecs.com

Did you log or test your neutral safety switch?
Also your engine coolant sensor?
Brake lights all working properly?

I’m leaning towards the originating fault being an rhs bank 1 issue.

So I’d also ohm out the cam position sensor harness & plugs, inspect the cam sensor for sludge or swarf on the sensor face, check for oil in its plug etc. I have had one cmps harness resistance fail a resistance check…

Here’s one example indeed, leaking cam sensors being a known fault that can have drastic consequences: oil inundated harness…

Плавающие обороты после активной езды. Ошибки P0328,P1307,P1308

Добрый день.Ездим на impreza 00 г.в. МКПП 2 литра атмо GD9-002368.Проблема следующая - после активной езды (крутим двиг. до 5-6 тыс. оборотов) начинают плавать обороты как на ходу так и на холостом когда стоишь. На холостом плавают от 750 до 1500. И так продолжается пока не заглушишь двиг. и не...

sti-club.su

Here’s my logging how to in case you missed it: open source and AP:

Make and share datalogs! • club.liberty.asn.au

Do you have AP? Any idea who tuned it, or what tune you have?

Okay so all of my oil lines/banjo bolts do not have filters, they’re all flat bolts. I’ve checked all of them and I assume the previous owner removed all of them.

I worded it In a confusing way, but the left side was the dry side, opposite of turbo. But I think a flood test is a good idea I’ll see about that.

As for the Neutral safety switch, do you mean the neutral position switch? I’ve tried leaving it unplugged and also jumping it (I know certain years are default closed or default open so I tried both.) however in romraider, it appears the switch is working correctly.

I have not messed with the coolant temp sensor, and according to my logs it seems to be working

yes all of my lights work properly, and I do an open source e-tune with Eric from torqued performance via romraider and ECU flash


As for cam sensors, they all appear to have proper resistance, but they do have oil in the connectors no matter how much I clean them. I’ve been worried about that although I can’t help but overlook all of this info and think back to my OCV connectors. The left side did not have ground. is that not a clear indication that the ECU is bad?

we can narrow this circuit down to 3 things. OCV’s, wiring, and the ECU.

we know both OCV’s work (from applied power while Idling and seeing the advance on the log

we know wiring is good (i re-wired the entire system and it had no change.)

Doesn’t this leave the ECU left?

it could be a hundred different things. But we know that the connector does not get a proper ground from the ECU. are we beating around the bush here?

 

[bigBADbenny]

Thanks for that recap, I was about to do the same.
Time is also money, so it’s up to you if you want to try another ecu, potentially cutting to the chase.

Otherwise, and this is free to you other than time, make a datalog.

Vvt angle is in essence a cam sensor vs crank sensor calculated parameter.
So there’s minimal vvt angle to be logged, but we will see crank position as rpm, and from memory there’s maybe cam indication in the switches panel in RomRaider. Maybe Eric can advise on the subject?

If you have oil in the cam sensor plugs, this is a known fault, with oil seeping through the sensor itself via cracks in the plastic housing, also the o-rings that seat the sensor in the head: creating unmetered air vacuum and oil leaks. I upgraded to viton o-rings…

Because of the oil leaks it’s really important to check the main harness plug/s that lead to the firewall for oil contamination, as eventually the oil will get to the ecu. With new cam sensors and thorough cleaning you might get the issue under control, without having to replace or repair the engine harness: Subaru engine sensor oil leaks and loom oil contamination. • club.liberty.asn.au

It can’t hurt to follow the P0365/7 cam sensor diagnostics for the harness: the resistance checks and potentially a voltage drop test. On my car I had verified the main plug to ecu, so then I did the main plug to sensor plug, the lhs wires didn’t pass the resistance test so I back probed the affected wire and found the weak point at the first branch in the harness. I added a same gauge wire across the weak section in parallel. For sure one might expect your car to actually throw those cam sensor DTC’s, is it possible that the VVT codes override them?

As regards your banjo bolt filters, check behind the lhd drivers side (lhs) timing cover for where the OCV oil supply exits the head: this is the usual spot for a filtered banjo bolt: right where it’s hard to see and get to. If it has a nipple on the head, it might still have a filter.

But wait, we know we can discount an oil blockage because it’s been blind tested by using remote power to the solenoid during idle: full angle achieved. That’s if you’d logged VVT angle during the test and had evidence that the rising rate of the VVT angles matched each other. Say there was a partially blocked filter, you’d see a much lower rate of angle compared to a bank with good oil supply. In an extreme case of that lag, VVT calibration might struggle. It appears your car cannot make it to the calibration stage, it’s failing at the self check stage potentially.



Other considerations are things that were disconnected and reconnected during recent works, and since it’s a 20yo car maybe clean the ecu ground as well?

 

[Eth207]

Thanks for that recap, I was about to do the same.
Time is also money, so it’s up to you if you want to try another ecu, potentially cutting to the chase.

Otherwise, and this is free to you other than time, make a datalog.

Vvt angle is in essence a cam sensor vs crank sensor calculated parameter.
So there’s minimal vvt angle to be logged, but we will see crank position as rpm, and from memory there’s maybe cam indication in the switches panel in RomRaider. Maybe Eric can advise on the subject?

If you have oil in the cam sensor plugs, this is a known fault, with oil seeping through the sensor itself via cracks in the plastic housing, also the o-rings that seat the sensor in the head: creating unmetered air vacuum and oil leaks. I upgraded to viton o-rings…

Because of the oil leaks it’s really important to check the main harness plug/s that lead to the firewall for oil contamination, as eventually the oil will get to the ecu. With new cam sensors and thorough cleaning you might get the issue under control, without having to replace or repair the engine harness: Subaru engine sensor oil leaks and loom oil contamination. • club.liberty.asn.au

It can’t hurt to follow the P0365/7 cam sensor diagnostics for the harness: the resistance checks and potentially a voltage drop test. On my car I had verified the main plug to ecu, so then I did the main plug to sensor plug, the lhs wires didn’t pass the resistance test so I back probed the affected wire and found the weak point at the first branch in the harness. I added a same gauge wire across the weak section in parallel. For sure one might expect your car to actually throw those cam sensor DTC’s, is it possible that the VVT codes override them?

As regards your banjo bolt filters, check behind the lhd drivers side (lhs) timing cover for where the OCV oil supply exits the head: this is the usual spot for a filtered banjo bolt: right where it’s hard to see and get to. If it has a nipple on the head, it might still have a filter.

But wait, we know we can discount an oil blockage because it’s been blind tested by using remote power to the solenoid during idle: full angle achieved. That’s if you’d logged VVT angle during the test and had evidence that the rising rate of the VVT angles matched each other. Say there was a partially blocked filter, you’d see a much lower rate of angle compared to a bank with good oil supply. In an extreme case of that lag, VVT calibration might struggle. It appears your car cannot make it to the calibration stage, it’s failing at the self check stage potentially.



Other considerations are things that were disconnected and reconnected during recent works, and since it’s a 20yo car maybe clean the ecu ground as well?

Many good points, ill see about finding cam position parameters via romraider, that could help out a ton.

Ill get a data log going tomorrow too

Good point about the oil flow, I also confirmed it today by doing a flood test everything looked good.

That's very interesting about oil contaminating a wiring harness I have never head of that, great write up on it too! I think it is worth looking into. I mentioned yesterday how the problem started after installing a new turbo and initially I thought there was no connection between the two, but maybe that cam sensor near the turbo was on its way out and by messing in that area it just screwed it up?

I have gotten p0365 briefly in the past, this could all be connected

Today I was messing with that same side of the engine bay (fixing a vac leak) and after I finished I took it on a test drive. I didn't see it actually happen but I noticed on romraider the "max" indication on my left VVT read 40 degrees (indicating that it reached that value) by the time I was looking at it the value was back to 0, and the right side never moved, maybe it was just a glitch. But it is interesting that working near the cam sensor seems to be effecting things

Weird how the rhs cam sensor seems to be effecting the left OCV circuit?


The sensors are a bit expensive I don't know how to test if they are functioning correctly besides checking resistance, it would be nice to see if I could log them like you were saying.


Where is the ECU ground by the way? Are you referring to the big metal block inside the ECU?

 

[bigBADbenny]

Did you find the correct cam sensor part number? Be sure to get the correct type eg 3 wire vs 2 wire iirc, try jp-carparts com maybe?

Ok it’s promising you saw a change in VVT angle, patient has a pulse, if briefly.

If you have data logs maybe upload them to datazap online and share the link, I don’t have my laptop atm.

The cam sensors are hall effect iirc, so hook them up to a multimeter and wave a metallic object over the sensor face, google will have the test specifics.

jp-carparts lists two sensors, different prices so potentially the more expensive is an update part more resistant to leakage?
You’d hope so, but always fact check any of my info here

22056AA062 &
22056AA091

22056AA091 / SENSOR ASSEMBLY, CAM SHAFT / IMPREZA 4D SEDAN / subaru | part detail｜JP-CarParts.com

subaru・IMPREZA 4D SEDAN・SPARK PLUG & HIGHT TENSION CORD detail of the parts catalogs with pictures.You can find the parts you want.

jp-carparts.com

 

[Eth207]

Did you find the correct cam sensor part number? Be sure to get the correct type eg 3 wire vs 2 wire iirc, try jp-carparts com maybe?

Ok it’s promising you saw a change in VVT angle, patient has a pulse, if briefly.

If you have data logs maybe upload them to datazap online and share the link, I don’t have my laptop atm.

The cam sensors are hall effect iirc, so hook them up to a multimeter and wave a metallic object over the sensor face, google will have the test specifics.

jp-carparts lists two sensors, different prices so potentially the more expensive is an update part more resistant to leakage?
You’d hope so, but always fact check any of my info here

22056AA062 &
22056AA091

22056AA091 / SENSOR ASSEMBLY, CAM SHAFT / IMPREZA 4D SEDAN / subaru | part detail｜JP-CarParts.com

subaru・IMPREZA 4D SEDAN・SPARK PLUG & HIGHT TENSION CORD detail of the parts catalogs with pictures.You can find the parts you want.

jp-carparts.com

Awesome thank you for those part numbers!

I've been busy I haven't been able to do a log but I realized, when I actuated the OCV's during Idle and I was able to see the VVT parameters move to max value, doesn't that confirm that the cam sensors are working properly?

 

[bigBADbenny]

To make related value judgments you’d note, preferably log the maximum vvt angles achieved, the rate of change with external power off then on, and off again (attack and decay), and sustain: the powered-on test for a few minutes or more to catch any intermittent signals eg angle, duty cycle or volts, and include the harness press/shake test.

Then repeat the test also in neutral with very gentle throttle variations between 1500 & 2500 rpm, since oil pressure will be greater with rpm. Or as you mentioned previously, try to vary throttle to prevent the engine stalling as you perform the logged tests, eg get an assistant to help with either aspect of the proposed tests.

I’d then inspect the logs in MLVHD, being sure that the related parameters: closed loop, rpm, NPS, VVT L&R, AVCS duty cycle & volts, coolant temp, crank and cam switch, etc are all graphed with smoothing switched off/inactive in the app as smoothing hides dropouts.

I’d also log FBKC, FLKC, per or all cylinder roughness/misfires, knock sum, as per the logging guide.

Logging is free, apart from your time, make a gentle off boost driving log as well if safe to do so.

Eg my cars calibrated VVT angle lag was half a second or so, the lag was attack not decay, I wanna say the VVT dropouts were misfire related, eg my overdue really ash encrusted plugs, and the integral windup from partially blocked banjo filters (soot and varnish flakes) would cause the affected lhs bank to top out at 53’ as the calibration tried to catch up the lagging bank with the good (actually just better) bank.

So eg, and I hope not, a log might reveal maxed fuel trims, a persistent misfire or misfires, leading to the current hard code, or something preventing closed loop, in turn preventing AVCS operation, the DTC being a side effect. So logging casts a wider net as regards diagnostics, over just going by the DTC tests.

Oscilloscope tests are even more informative, a real time,universal to all vehicles diagnostic tool as I might have mentioned previously.

At this point replacing your ecu is still not to me 100% certain.

Also if you just went ahead and simply replaced it, did you get to the failure mode that potentially cooked it after the recent major works? Was the failure mode indeed a compromised ground that you’ve now fixed, or grounds you’ll upgrade with a big three and or iwire kit? Or is the issue something obvious we haven’t yet seen, or might see in logs?

I forgot to mention, I’ve seen the ecu ground being stated as “adjacent the ecu”: you’ll find it in the FSM cabin harness diagram/map or the grounds maps, both in the FSM wiring diagrams section.

 

[Eth207]

ments you’d note, preferably l

To make related value judgments you’d note, preferably log the maximum vvt angles achieved, the rate of change with external power off then on, and off again (attack and decay), and sustain: the powered-on test for a few minutes or more to catch any intermittent signals eg angle, duty cycle or volts, and include the harness press/shake test.

Then repeat the test also in neutral with very gentle throttle variations between 1500 & 2500 rpm, since oil pressure will be greater with rpm. Or as you mentioned previously, try to vary throttle to prevent the engine stalling as you perform the logged tests, eg get an assistant to help with either aspect of the proposed tests.

I’d then inspect the logs in MLVHD, being sure that the related parameters: closed loop, rpm, NPS, VVT L&R, AVCS duty cycle & volts, coolant temp, crank and cam switch, etc are all graphed with smoothing switched off/inactive in the app as smoothing hides dropouts.

I’d also log FBKC, FLKC, per or all cylinder roughness/misfires, knock sum, as per the logging guide.

Logging is free, apart from your time, make a gentle off boost driving log as well if safe to do so.

Eg my cars calibrated VVT angle lag was half a second or so, the lag was attack not decay, I wanna say the VVT dropouts were misfire related, eg my overdue really ash encrusted plugs, and the integral windup from partially blocked banjo filters (soot and varnish flakes) would cause the affected lhs bank to top out at 53’ as the calibration tried to catch up the lagging bank with the good (actually just better) bank.

So eg, and I hope not, a log might reveal maxed fuel trims, a persistent misfire or misfires, leading to the current hard code, or something preventing closed loop, in turn preventing AVCS operation, the DTC being a side effect. So logging casts a wider net as regards diagnostics, over just going by the DTC tests.

Oscilloscope tests are even more informative, a real time,universal to all vehicles diagnostic tool as I might have mentioned previously.

At this point replacing your ecu is still not to me 100% certain.

Also if you just went ahead and simply replaced it, did you get to the failure mode that potentially cooked it after the recent major works? Was the failure mode indeed a compromised ground that you’ve now fixed, or grounds you’ll upgrade with a big three and or iwire kit? Or is the issue something obvious we haven’t yet seen, or might see in logs?

I forgot to mention, I’ve seen the ecu ground being stated as “adjacent the ecu”: you’ll find it in the FSM cabin harness diagram/map or the grounds maps, both in the FSM wiring diagrams section.

Okay so ill do a log of just cruising around, and then one of me messing with the OCV's and ill try to figure out how to post it to that site like you said.

Should I try to wire both up to one battery that way the engine isn't unevenly adjusting timing? It might be able to maintain idle that way

I have not replaced the ECU yet but I am tempted because I found one for a "good" price hopefully it doesn't sell before I decide I need it

 

[bigBADbenny]

If it’s a good price, it’s money in the parts bank… Easy to cash out if not required.

That’s the only way parts swapping is viable, if cheap parts are on hand and the fault mode won’t blow up swapped parts

Yes, I did mean to wire up both ocv’s at the same time, could get fancy and extend the wires back to the cabin so you can test it solo?