Subaru WRX Forum banner
1 - 20 of 20 Posts

·
Registered
Joined
·
24 Posts
Discussion Starter · #1 ·
Tried NASIOC, and after years of the issue, I’m trying here as well. Grab a coffee and sit... This might take a while...

The car is a 2008 WRX with VF52, TP stg 2.52 and supporting mods.

As the title says, the engine stalls with no CEL and no engine codes. The issue started in 2017 and I still haven’t been able to figure out what is wrong.

It started as in this video:

then evolved to this:
.

Before any of this started, the only work done on the car was a new clutch and a/c evap was replaced. A year after the a/c was repaired, I found the relay box under the glove box hanging from its wires, but putting it in place or jerking the wiring around didn’t recreate or fix the issue. When the issue from the first video happens, you can hear all the relays from that box clicking on and off. When any of the issues happen, I don’t loose anything electrical, meaning no lights go off, no dimming or anything alike, speedo keeps reading speed, fuel level keeps reading, radio stays on, blower motor on, accesories like gaugues, homelink mirror and radar stay on, etc. I believe that if it was a short circuit or a loose ground, I’d loose something electrical or a fuse would blow, but none of this happens. The car has no trouble starting, and as seen in the video, when any of the issues happen, the engine starts back up in a matter of seconds and by itself, without the need of me turning the key. I’ve wiggled the key around when the car is on based on a theory that the ignition cylinder could be bad, and messed around with the plug as well, but this doesn’t recreate or fix the issue.

During that time, I’ve replaced MAF, MAF plug (one pin was corroded and loose, causing too rich and too lean condition codes, but the issue was already existent), crank sensor, all ignition coils with new oem ones, new spark plugs and new connectors, main battery harness, main wire from engine fuse box to positive battery terminal, ebcs (tested with newer oem and grimmspeed), different brake switches, new battery, new fuel pump, different radios, no radio, clockspring and angle sensor (to add bluetooth buttons, but issue was already existent), different keys and key fobs, disabled alarm, pedals had to be replaced due to broken welds (issue was already existent), Cobb V2 and V3 and their respective OBD cables, and probably some other things I can’t remember, but all done trying to find the cause of the issue. I’ve checked and rechecked all my grounds and connectors and nothing loose was found. I found a harness hanging over the trans because of a plastic clip holder that broke off its base, but I tied it to prevent movement or contact with the trans and made no difference. I once removed the dash completely to work on the broken welds, issue existent, and found no loose wiring or torn cables. I tried disconnecting the front A/F sensor with the engine running and it doesn’t recreate the issue. Both issues happen with the a/c on or off.

I had the Cobb V2 and upgraded to the V3. I’ve tried different maps and uninstalling it, but the issues remain.

I recently got a full system scanner, iCarsoft HNM II, with which I was able to find one code that keeps popping up every time the issue from the second video happens, which is C0047. When the issue from the second video happens, ABS, VDC and Hill Assist lights all turn on, and go off if I shut the engine off and on again, but the code stays until it’s cleared. When I clear the codes, all lights go off except the ABS, which stays on for about a minute before turning off. Based on the FSM, this code relates to the ABS control module having a CAN comm malfunction. One code that popped up once was B0221, and from Google I found that the same issue has been going on since 2005, but only on models with Auto Trans. There’s even a service bulletin about it showing a flowchart to diagnose the issue. My symptoms are the same as described in this bulletin, with the exception of Auto Trans codes and that my odometer doesn’t show “Er Hc” as described, which I believe is because I don't have Auto Trans. Here’s the service bulletin: http://ww2.justanswer.com/uploads/n9320h/2010-07-16_222648_sub.pdf. I tried replacing the ABS module, but the one I got, even though it had the same exact part numbers as mine, came off an Impreza (non WRX) without VDC, so a new code pops up, C0045, which is “incorrect vdc specification”. Can’t use it because all its functions are disabled, meaning no vdc, hill assist or abs.

I went to the dealer to have it scanned with the Subaru Select Monitor. The dealer stated that a bunch of CAN codes were activated and could not be reset. They claimed that the ecu was bad and had to be replaced. I kept asking why they said the ecu was bad and everything turned to “could be bad” and wanted me to pay $1600 for a new ecu so they could see if this was actually the issue. There was no guarantee that this would solve the problem. I took my iCarsoft scanner and found all the CAN codes the dealer said were activated. I then asked a friend who owns a ‘14 WRX to let me scan his car, and all the same CAN codes were activated in his as well, with the difference that his car has no stalling issues as mine does. I cleared all CAN codes in both cars. My friend’s WRX hasn’t had any codes pop back up, and mine only gets the C0047 when the issue from the second video happens.

Apart from this scanner, I also used FreeSSM software with VAG OBD cable. With this software I’m only able to read sensors the same way the Cobb AP does, showing max/min values while reading live data, but this software keeps reading when the issues happen. The Cobb AP crashes and restarts, so logs just cut off without showing any weird behavior prior to the issue, and it did the same with both V2 and V3. The only thing I could notice with this software is that the rear O2 sensor sometimes drops to zero and stays there, but after a while of this happening, I’ve seen it happening without causing the issue, so I don’t think it’s related to or causing the issue.

I tried connecting the green test connectors and scanning it, and got 4 codes:
  • P0140: rear O2 sensor no activity
  • P1152: front O2 sensor circuit low
  • P1153: front O2 sensor circuit high
  • P0852: neutral switch input high

That last code, P0852, will always pop up when scanned with green test connectors plugged in, but I found it to be normal. It was on in both my car and my friend’s WRX, and it goes off if you move the shifter. It happens either way, meaning that if it was in neutral and then put in gear, the code clears, and the same happens if it was in gear and then set to neutral. The code clears by itself without the need of clearing codes or resetting the ecu. I tried clearing the other codes, but they don’t clear. My friend’s WRX had no rear O2 sensor code, and the front O2 sensor showed 2 codes as well: stuck lean and stuck rich, but same as mine, they could not be cleared.

I have not found a single person other than me with this exact same issue. It’s been really hard to find an ABS module that came off a car with my same specs, seeing as the FSM states that if it’s not, it’ll have to be reprogrammed with the Subaru Select Monitor, but the dealer refuses to reprogram a used part. This issue has never prevented the car to keep running. Every time either issue happens, the car just starts back up as if nothing’s wrong. The issue from the first video leaves no trace, but the second video gives C0047.

I haven’t tried new O2 sensors, cam sensors, different fuel pump module, replacing abs sensors... From what I’ve read, any of these sensors would trigger a fault code to identify them, but no P codes show up with any of the issues.

Apart from what I’ve mentioned here, anything else you guys might have an insight on? I’m leaning towards selling the car and let someone else deal with it, but $$ is tight to be getting into a new(ish, er) car.

Thanks in advance.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #3 ·
Ignition switch has been tested for wiring, and using different keys. The FreeSSM lets me see this switch live and it doesn’t change from ON when the issue happens. Like I said, all the wiring has been tested, wiggled around, checked for grounds and even continuity, which wasn’t mentioned. The ecu can be cloned for a lot less than getting a new one, but that would mean about 2 weeks without being able to use the car due to the process. I saw the same behavior and codes on my friend’s ‘14 WRX when we scanned it, but he has no issues.

Is there anything else worth checking that wasn’t mentioned in the original post?
 

·
Super Moderator
Joined
·
11,775 Posts
I'm thinking on it. I'm not an electrical guru, but I'll try to bounce some ideas around and my thought process for diagnosis of this.

In the second video, when it dies out is the clutch still engaged? It looks like you loose the tach and the temp gauge. When you mentioned the AP crashes it almost makes me think something in the ecu is at fault and its not a mechanical issue. However I would expect with an ecu problem the speedo would go bonkers as well.

Because the issue is happening randomly and under load and running unless you had a fluke testing everything when it happens it may be hard to spot just testing grounds and such while stationary.

I'm not sure what all is part of the TP 2.52 but I'm assuming something along the lines of intake, turbo back, TGV delete. I don't suspect any of those would cause this issue or be related. TGV deletes can cause weirdness when idling cold, but nothing like this.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #5 ·
I'm thinking on it. I'm not an electrical guru, but I'll try to bounce some ideas around and my thought process for diagnosis of this.

In the second video, when it dies out is the clutch still engaged? It looks like you loose the tach and the temp gauge. When you mentioned the AP crashes it almost makes me think something in the ecu is at fault and its not a mechanical issue. However I would expect with an ecu problem the speedo would go bonkers as well.

Because the issue is happening randomly and under load and running unless you had a fluke testing everything when it happens it may be hard to spot just testing grounds and such while stationary.

I'm not sure what all is part of the TP 2.52 but I'm assuming something along the lines of intake, turbo back, TGV delete. I don't suspect any of those would cause this issue or be related. TGV deletes can cause weirdness when idling cold, but nothing like this.
In the second video I’m at wot with speed in 3rd gear. I don’t press on the clutch on any of the videos. The temp gauge jumps to high, same as it does when you’re loading a new map with the AP. I’ve tried different maps and even uninstalling the AP and the issue still happens. As for grounds, what I’ve checked is that there’s good contact and continuity between them and the chassis.

The TP map is the same as a stg 2, but for a VF52, which the 08 doesn’t have. Fuel pump, dp, KN Typhoon, ebcs, one step colder spark plugs, and that’s about it. As you said, if an ecu issue was the cause, I’d be loosing more than just the rpm, and getting more codes apart from the C0047 that the ABS throws. I say that the AP crashes, but it’s more like it resets itself. It stops logging, the screen flashes and readings on the display reset their max/min values.

Another theory I read about was the alternator causing random electric issues. Mine has never been replaced, so it’s over 10 years old. It’s one of the To-Do list I have. While scanning live, I also noticed that the fuel pump duty never goes to zero, even when the revs drop to zero. I already replaced it to discard it from the probable causes. Also, I had a loud exhaust for a couple of months, and when the issue happened, you could still hear the exhaust as if you just let go of the gas and were decreasing speed in gear. Where I’m going with this is that the engine is still turning when I loose rpm signal, but because it’s in gear. So, it’s like I loose rpm signal, but the engine isn’t really off. Hope I’m explaining myself correctly.

Like I said, haven’t found a single person with the same issue, or even remotely similar, but the service bulletin I linked proves that I can’t be the only one. The problem with that bulletin is that their conclusion is that this is a “temporary unstable communication phenomena”, and the solution is to just clear the codes. Obviously, clearing the codes has done nothing for me.

I keep checking for new stuff every day, wiggling cables, checking plugs, etc., with the hope of someday finding the actual issue and solving it.
 

·
Super Moderator (Actually a SuperSpy)
Joined
·
19,820 Posts
I was thinking about this and am not an electrical guru either. However the problem may be similar to that of the Phrygians. When Midas tied Gordias' ox cart with a knot so complex that nobody could see where it started or ended, nobody could untie it. The oracle had predicted that whoever untied the knot would rule Asia. Alexander couldn't untie the Gordian knot any better than anyone else, plus he really needed to get on with the whole conquer the world thing so he drew his sword and cut it. Worked great.

Let's say for the sake of argument you have a sensor with a faulty contact but the fault occurs so sporadically that you may be unlikely to ever solve the issue. Consider either buying a grounding kit or buying a bunch of wire and running a series of ground lines, from the passenger McPherson tower to the intake manifold on one side, to the throttle body, to the body of the alternator, to the intercooler to the other side of the intake manifold and to the negative pole of the battery. Just make sure you never, ever connect anything to a + source; everything goes to earth only.

This may accomplish absolutely nothing but you'll be adding money to the economy and it'll keep you occupied.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #8 ·
I was thinking about this and am not an electrical guru either. However the problem may be similar to that of the Phrygians. When Midas tied Gordias' ox cart with a knot so complex that nobody could see where it started or ended, nobody could untie it. The oracle had predicted that whoever untied the knot would rule Asia. Alexander couldn't untie the Gordian knot any better than anyone else, plus he really needed to get on with the whole conquer the world thing so he drew his sword and cut it. Worked great.

Let's say for the sake of argument you have a sensor with a faulty contact but the fault occurs so sporadically that you may be unlikely to ever solve the issue. Consider either buying a grounding kit or buying a bunch of wire and running a series of ground lines, from the passenger McPherson tower to the intake manifold on one side, to the throttle body, to the body of the alternator, to the intercooler to the other side of the intake manifold and to the negative pole of the battery. Just make sure you never, ever connect anything to a + source; everything goes to earth only.

This may accomplish absolutely nothing but you'll be adding money to the economy and it'll keep you occupied.
I thought about the grounding kit for a while, but never got around to doing it. Not a bad idea. Some people have mentioned several grounds that came on newer models and mine didn’t have. I added an oil pressure gauge and disabled the oem sensor. One of the things I tested was moving that sensor’s plug around and even making contact to ground to see if I could recreate the issue, but all it did was turn the oil light on in the cluster.

My power steering was leaking at one time and you could see oil puddled on the head, but the issue started way after this was fixed. The leak left oil around the crank sensor, but I tried cleaning and replacing it after the issue started and it didn’t change the issue. The cam sensor on the passenger side still has some oil around it, but I don’t see how to remove it without taking the inlet manifold out. I’ve wiggled its wiring around and caused no issues. Could a cam sensor cause rpm signal issues?
 

·
Super Moderator (Actually a SuperSpy)
Joined
·
19,820 Posts
I suppose it could? So could a crank sensor. However I’d expect a CEL from either.

Remind me, have you cleaned the IACV? Have a new gasket on hand.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #10 ·
My thoughts exactly... either sensor with an issue should cause a CEL. If you watch the second video again, you’ll notice that the CEL flashes for a fraction of a second, but it stays off. If it was an actual fault, the light should stay on even if the fault clears, but it doesn’t, and neither pending or history codes show up. The only time I get a code is when the issue from the second video happens and the abs, vdc and hill assist lights stay on. If I shut off and turn back on, the lights go off, but the code remains. I’d expect the same behavior if an engine sensor failed.

One odd thing is that, when I’m at the garage with the engine on, wiggling wires and testing stuff to try to recreate the issue, none of the issues happen. I can leave the car on wasting gas for up to a half hour and nothing happens. Sometimes it’ll be days, weeks and even months before the issues arise, and sometimes they’ll happen several times during a drive.

I’m saving up to have the ecu cloned, but that also means planning ahead because I have to ship mine out to Portland (I’m from PR), along with the spare ecu I get, so I’ll probably be about 2 weeks without the car, which is the only one I have. In the meantime, I’ll have to keep trying to find an exact match to the abs module, which I’d like to do before I swap ecu’s.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #11 ·
I’m keeping a log of the things I do and check, so that when any of the issues happen, I can discard them. Today I removed the rear O2 sensor and disabled the corresponding codes. I had a theory that this sensor could be bad based on its voltage. It only cycles when I’m on the gas pedal, but the cycle to me is erratic. I can be maintaining speed and you’ll see it jump all around, from 0.00 to 0.92, but at idle and no movement or gas applied, it reads a steady 0.03. Sometimes it’ll drop to 0.01 and stay there as well.

A few days ago, while reading voltage on it, the issue from the second video happened. It happened so quick that when I glanced at the reading, I don’t know if I saw 2.03 or 20.3ish, but I don’t think it’s 20 because the battery only reaches 14 volts. Now, a short on it could give a false reading, right?

I’ll drive it around for a week or so without the sensor to see how my fuel trims behave in comparison to having it connected, but also to see if any of the issues arise.

On a side note, found an ABS module with matching part numbers and matching car model, meaning 08’ WRX with VDC, so I’ll be ordering it by the end of the week. The first one I tried had matching part numbers, but came off an Impreza (non WRX) without VDC, which I didn’t know were available. The second one had matching part numbers and did come off a WRX, but the plug was damaged, so I wasn’t even able to test it. I’m waiting on the refund on that one to order the new one I found.

I’ll update as soon as I’ve checked on the fuel trims and the new ABS module has been tested.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #12 ·
So I’ve driven a few days, not many miles, without the rear O2 sensor. So far, the fuel trims are a bit different from what they were with the rear O2 sensor on. At idle, af learning 1 reaches between 7 and 10 (positive values) before they start to come back down, af correction 1 maxes out at 25 for some time, then starts to come back down as well. During normal driving, meaning no wot, some traffic, going through stop lights, etc., these numbers drop to zero or go to negative values. As for af correction 3, which is determined by readings from the rear O2 sensor, it’ll be between 0 and -0.5 while driving and maxes out at -5.00 at idle. I think this happens because there’s no O2 sensor, but I’d assume this shouldn’t happen BECAUSE there’s no rear O2 sensor. At least voltage reading hasn’t moved from 0.00.

As for the issues, they haven’t happened, not even hints of going to happen, since I removed this sensor. Still, it’s still too early for me to say that this solved it because, as I’ve said before, it’s been up to months without any of the issues happening and they do resurface.

I’ll keep driving it like this for a while. I have a long trip coming next week, which I’ll probably use to put some miles on the car and see if the issues arise.

I’ll keep updating as I keep performing tests.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #13 ·
On a side note, some bad news...

The day I removed the sensor, I also worked on the power steering pump suction hose. It was leaking right under the clamp on the pump side, so i cut about 1/2 inch from and clamped it back on. Due to this procedure, the system has to be vented/bled to remove any air remaining. The car was on lifts, so I turned it on without the canister cap on and turned the steering wheel to both sides until I stopped seeing air bubbles coming out through the canister.

Apparently, either I did this too hard or too fast and ended up damaging my clockspring. No dash lights are on, but my horn stopped working. After some google searches I found some people with the same issue saying that turning the steering wheel to one of the sides made the horn work. I performed the test and the horn worked, which is why I believe my clockspring is damaged.

Now, as I’ve mentioned before, I replaced the clockspring with one from a 2013 so I could add the bluetooth buttons. I’m not actually sure if the clockspring got damaged when I worked on the power steering hose, or if was damaged before because I hardly ever use the horn. I do remember it working when I first installed it, but I can’t tell if the horn stopped working before or after I worked on that hose. Either way, I’ll need a new clockspring. I still have my oem one, but since it’s not for bluetooth buttons, I’d also have to change the steering wheel controls and reprogram the adapter to make it work with my aftermarket headunit, which is too much work to do twice.

I found some people who have successfully repaired them, but I’ll need to tread lightly as I don’t want to do more damage than it’s already done.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #14 ·
Updating...

Rear O2 not the cause. A day after my last pos, got the issue from the second video, and the day after both issues happened intermittently. This time 3 codes were activated: U1221, B0221 and C0047. Looking again at the FSM, it all points to the ABS module.

Decided to go through google again combining this 3 codes and found a Legacy forum with someone else with these codes, but different symptoms. In their case, dash went crazy, radio flickered on and off, and a few more electrical issues that didn’t allow the car to start.

In my case, nothing electrical goes off, flickers or dims during any of my issues. One of the guys said that his issues went away after replacing the alternator. It was working properly and charging correctly, but apparently the voltage regulator was causing volt spikes that created the issues he was having.

Another guys talks about the neutral safety switch. I thought about this once, but seeing as the Cobb AP showed the signal correctly as I moved the shifter through gears, I didn’t pay that much attention to it. After reading the Legacy thread I decided to give it another look.

I went through the FSM and found a testing procedure to check resistance in the switch. I disconnected mine today to do so, but I don’t get any readings from the switch whatsoever. My multimeter doesn’t change when I’m on the pins and there’s someone inside the car going through the gears.

I reconnected the rear O2 sensor today and decided to leave the neutral safety switch plug hanging so I can drive around for a few days and see how it behaves.

I’ll update with my findings.
 

·
Super Moderator (Actually a SuperSpy)
Joined
·
19,820 Posts
What effect has pulling the fuse on the ABS module and can that help you troubleshoot the module?
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #16 ·
What effect has pulling the fuse on the ABS module and can that help you troubleshoot the module?
Haven’t tried that, so I’ll give it a try after a few days of driving without the neutral switch.

I did try a different module, but got a code for “incorrect specification”, which disabled the system completely, and another code for “pressure sensor”, which was due to not reading the master cylinder pressure when the brake pedal was applied. I didn’t have vdc, abs or hill assist, so due to safety issues, I only used it for 2 days. I could feel the difference of not having hill assist, and also did some hard braking to verify abs, at which times the front left tire always locked up when the brakes were applied hard. I didn’t get any of issues shown in the video during that time, but as I explained earlier, it can be days, weeks or even months without the issues happening after I change something.

Having the neutral switch disconnected hasn’t thrown any codes so far. I have a long trip tomorrow, but still don’t know if it’s my car I’ll be using. I’ll just keep driving it like this until something happens. I’d have to give it at least 4 months of no issues happening to be sure this is the reason, or that I’m probably on the right track.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #17 ·
Updating...

Took a 100 mile drive without the neutral switch. I didn’t remove the DTCs corresponding to this switch, so at around the first 15 miles, the check engine light came on and gave P0852, neutral switch input high. None of the issues happened when the code popped up. I was doing some slight acceleration at half throttle, reaching around 12 to 15 psi of boost before it happened. I explained in my first post that this code was always active when scanning with the green test connectors connected, but cleared if I moved the shifter from neutral to gear and vice versa.

I kept driving with the check engine light on and later on, the issue from the first video happened. It happened while driving at cruising speed (cruise control disabled due to check engine), but the odd thing is that when it did, the check engine light (and limp mode) went off. Everything went back to normal, no warning lights on the cluster, no codes when scanned, no limp mode, cruise control could be activated, etc. I didn’t stop the car, didn’t turn it off and didn’t reset the ecu. The code, check engine, limp mode, etc., went off right after the issue happened, meaning the issue caused it to reset. I tried to force the code to pop up again doing the same slight acceleration and 1 or 2 wot’s, but it didn’t come on again.

All of this happened on my way to the destination, which was half the trip. On the way back I was doing a whole lot more wot’s, at different speeds, different gears, from different rpm starting points and got the issue from the first video a few times.

What never happened was the issue from the second video. I had cleared the C0047 code from the brake system, which was the only code active before the trip. Today I disabled the codes related to the neutral switch and will be driving it some more like this to see if the C0047 code pops up again, or if any of the issues arise.

I’ll keep updating as I keep testing.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #18 ·
Updating...

As mentioned in my first post, one of the things that was worked on before the issue started was the a/c. Last week I decided to remove the dash to have a look at all the grounds hidden behind it based on the FSM. The ones that were visible and accessible were fine. They were not loose, dangling or broken. What I did find was that the whole blower motor housing, the duct from that housing to the evaporator and the plastic that covers the evaporator lines were loose and some screws were missing.

The grommet that seals the evaporator lines going through the firewall was also loose, and I noticed because I could see the engine bay through that opening. The bolts that hold the blower motor housing to the firewall were there, but they weren’t fully tightened. And apart from the a/c system, the metal plate that covers the ecu was missing the top nut that holds it to the floor.

When I reinstalled everything, I properly tightened the blower motor housing and added the nut missing from the ecu cover plate. Battery was disconnected during the whole process, so the ecu reset.

Since I did that work, the car is now behaving very differently in terms of AFR. Before, my AFR would be around the 13’s at idle, and now it sits around 15, popping up DTC P0171 System Too Lean. AFR Learning 1 goes to pos 25 at idle and stays there unless I start driving.

My rear O2 has been giving low voltage for a while now, between 0.03 and 0.04. It goes up while driving, but erratically, giving readings between 0.02 and 0.98, not staying at any given value steadily, even while cruising. My guess is that, whatever I did while reinstalling the dash and a/c components, is providing better ground to that whole area, causing the O2 sensors to read properly and provide the correct info to the ecu. The relay block that clips on the side of the blower motor housing holds the relay for the O2 sensors. I mentioned that those relays start clicking very fast when the issues happen. Maybe the work I did gave the ground it needed to work properly and not loose power.

It’s only been a week, so I still don’t trust the car, but none of the issues have happened since I worked on the dash. I’ve been doing some pulls at wot and half throttle to try and recreate any of the issues, but nothing yet. I still have to reconnect the neutral switch and activate the codes accordingly, but I’ll also go back to my previous setup and tune as I’ve found that tune or no tune has made no difference in any of the issues.

I’ll update if any of the issues arise again.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #19 · (Edited)
Updating...

Had the issues happen a few times during the time I haven’t posted. Since then I added a ground cable from the trans bracket that holds the oem ground from the battery, to the block beneath the tmic, and the issue from the first video hasn’t happened, at least at idle. I’ve had the revs drop to 0 while doing wot’s, sometimes throwing all the lights and codes, and other times not, so I can’t say for sure that the issue from the first video is fixed, but I’m guessing the ground I added might have something to do with it.

Today I made a discovery. The relay block under the glove compartment has relays for O2, throttle, seat heaters, ignition, and in STi’s only (mine’s a WRX), there’s a relay called “Main” in the FSM, which I don’t have. I decided to pull out the one that the FSM denotes as “Ignition”. When I turn the key to the ON position with this relay removed, the issue from the second video happens (Temp needle pegged to Hot, ABS, VDC and Hill Assist lights on, check engine light off, no start).

323524



Car won’t crank or attempt to start. When I read the codes, I can’t access common P codes, but I can access all other systems, i.e. ABS, airbag, TPMS, etc.

Now I have to figure out whether the relay is bad, or if it’s loosing power or ground.

I’ll see if I can find a new relay to replace and update when I perform more tests.
 

·
Registered
Joined
·
24 Posts
Discussion Starter · #20 ·
Updating...

Performed some tests from the FSM. Didn’t get anywhere, but I figured out that codes B0221 and U1221 are the same codes based on the description given by the scanner and the FSM. The difference is that the B code shows up in the integrated unit and the U code shows up in the CAN system, but they are the same code.

Based on what I’ve found in terms of wiring, I’m inclined to think that the ECU is going bad since it appears that I’m loosing communication in the wire that goes from the ECU to the main relay that I tested and recreated the issue.

I’ll start the process of getting my ecu cloned, but it’ll take a while to gather the money for the spare ECU and payment for the service, so I’ll update when I can.
 
1 - 20 of 20 Posts
Top