Welcome to the North American Subaru Impreza Owners Club Saturday August 24, 2019
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Go Back   NASIOC > NASIOC Technical > Electrical & Lighting

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads. 
Reply
 
Thread Tools Display Modes
Old 08-08-2019, 08:20 AM   #1
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Unhappy Strange o2 sensor issue... Scottish Scoob owner needing some advice!

Hi all,



Figured i'd post on here to reach a wider audience across the pond. Hopefully someone can help!


Car is a 2000 JDM GTB Twin Turbo. Virtually identical to a v5/6 JDM impreza... but with an additional turbo and about 300kg of weight lol.

I've been wrestling with an issue for some time now. The car has been running poorly. Poor fuel economy, lumpy acceleration, lumpy on idle when started from hot, struggles to stay alive due to the revs dropping to 3/400rpm.


I Plugged in a SSM cable and the front o2 sensor (only o2 sensor on this model) voltage was reading 0.000v. I cleaned the plug... still 0v. Replaced the sensor with a bosch universal 3 wire and its still showing 0.000v. Voltage jumps to about 0.2v for a split second when the throttle is pressed but instantly drops back to 0.

I replaced the sensor with a brand new bosch universal 3 wire sensor. wired it in correctly.

When I test the black signal wire at the plug where the sensor joins the loom, there is 0.45v at idle (this is about right as this would suggest a normal AFR). This suggests the sensor is working and producing a reading/signal. At the ecu however, the voltage is different.


Service manual says that at the ecu plug, the o2 should see between 2.3-2.8v (iirc). At the ecu pin, mine showed 0.80v with the car idling and o2 sensor unplugged. When plugged in, it jumped to 1.1v but fluctuated massively (down to about 0.6v when revved in neutral and back up to 1+... never as high as 2v+.



I tested the pins on the loom-side plug on the car to see if they were getting power. The heater wire pin was easy to identify as it was getting a solid 12v. The other white wire was the ground wire for the heating element but was still getting 0.5v. The signal wire itself was also getting about 0.38v which I found a bit strange but I dont understand enough about auto electrics to make sense of it.



I tried bypassing the entire wiring to see if it was a wiring fault between the o2 sensor loom plug and ECU but didn't make much headway. I unplugged the sensor and ran individual wires from all 3 pins on the o2 sensor plug.

The heating element wires went straight to the battery positive and negative terminals to simulate the 12v and the ground.
The signal wire was then ran directly from the 02 plug and alligator clipped to a safety-pin/needle jammed in the back of the 02 signal wire at the ecu. On ssm it showed a constant .330v (usually what a broken o2 will show). It would not move despite revving the car and stank of fuel due to it telling the car its was running a lean mix.

Basically, i'm being told different things at 3 points in the system. The black signal wire from o2 is reading 0.45v (what it should be), the SSM software via obd plug is reading 0.000v and the actual ECU pin for o2 signal fluctuates wildy between 0.6-1.1v

Took it to a garage today and the mechanic was adamant that I should buy an oem sensor as universal ones are bad for subarus apparently. They 'don't agree with universal sensors' etc. This wouldn't be an issue but the oem sensors are like 180/$250+ and i'm a student lol. Is there anything else I can try before buying one?



This is the second universal bosch o2 inside 8 months, this is the reason Im worried it isnt the actual fault of the sensor. Im fairly confident there is something else weird going on with the ecu/wiring.



Any help is much appreciated, thank you guys
* Registered users of the site do not see these ads.
rorygtb is offline   Reply With Quote
Sponsored Links
* Registered users of the site do not see these ads.
Old 08-08-2019, 05:18 PM   #2
Cougar4
Scooby Guru
 
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:
2001 LL Bean Outback
Winestone

Default

The normal operating voltage range of the front O2 sensor should be between .1v to .9v. From what you stated about the issue it seems to me you might have a problem with the ECU input circuit for the sensor. Check the resistance between the wire for the sensor output and the input to the ECU with things disconnected at both ends. The resistance should be less than 5 ohms if things are good with that wire. Let us know what you get for a reading. Also check the resistance between one end of the sensor wire and ground. Do the same test with the power lead to the heater of the sensor while the ignition is OFF. Both of those test should show an OPEN or an 'OL' reading on the meter, meaning there is no connection to those areas, the lead is isolated.

Here is some reading for you to check out:
http://www.autodiagnosticsandpublish...r-testing.html
Cougar4 is offline   Reply With Quote
Old 08-09-2019, 06:06 AM   #3
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

Thanks very much for the reply mate- very informative, helpful advice!

Will try this when I get home from work today and update you.

Also... (unsure if this may be contributing to my problem or not), I did test the amperage between the -VE battery terminal (with the positive strap still on) and the disconnected -VE strap. It showed 1.5a (I understand it should be 0.1-0.5amps) so I think there is certainly electrical current leaking somewhere. I plan to pull all the fuses systematically to isolate which circuit the leak is on.

I did pull the SBF-5 30amp fuse and it didn't change so It may be unrelated, however.

thanks again,

Rory
rorygtb is offline   Reply With Quote
Old 08-09-2019, 06:17 AM   #4
Cougar4
Scooby Guru
 
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:
2001 LL Bean Outback
Winestone

Default

There must have been something turned on, like a small light, for that high of a current reading. Normal current draw, while systems are in the sleep mode, should be around 20 milliamps. Even 100 milliamps is too much current draw. For things to go into the sleep mode it may take several minutes after the power is connected up and the car is in the parked mode. When the sleep mode is entered you will see the current load drop, while you are monitoring the current.

Last edited by Cougar4; 08-09-2019 at 06:23 AM.
Cougar4 is offline   Reply With Quote
Old 08-09-2019, 06:45 AM   #5
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

That makes sense. I'll re-test this after the car has been off for half an hour and will make sure the doors are shut etc to see if it makes a difference. I think your theory with this is correct though, as the battery doesn't drain itself if the car has been sitting, which I assume would happen if there was a constant drain on power occurring.

Stay tuned for updates after I test your suggestion.

P.s. It feels strange to be talking to someone in Alaska about my o2 sensor hahaha. The wonders of the internet, eh? Whats the weather like there? Cold i'm guessing haha. Hats off to ya if you work on these cars for a living in those temps!
rorygtb is offline   Reply With Quote
Old 08-09-2019, 08:42 AM   #6
Cougar4
Scooby Guru
 
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:
2001 LL Bean Outback
Winestone

Default

Your current meter needs to stay connected up with no interruption to the power in order for things to go into the sleep mode. I don't think it will take more than a few minutes for things to go to sleep but it may take more time.

We are having the best summer this year I think I have ever seen. The temps are mostly in the 70'sF. We actually need some good rain to get some moisture back into the ground. I don't work on vehicles for a living, but I like to fix things when needed on them and help others keep their cars running, especially with electrical issues.
Cougar4 is offline   Reply With Quote
Old 08-09-2019, 12:51 PM   #7
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

Thats really not too bad! Much the same as a regular Scottish summer haha!

Have just finished doing some testing and here are my findings


Firstly, I tested the amperage between -ve terminal & ground strap again with all doors closed etc... 0.12amps! Happy days, so no parasitic drains/open short circuits it seems.*



Resistance between signal wire at loom-plug and ecu input pin - 1.0 ohm



Resistance between;*

+ve heating wire and ground = infinity*

-ve heating wire and ground = 1.9ohms*

Signal wire and ground = infinity*



Testing loom plug voltage IGN: ON engine: OFF**



+ve heating wire = 12.2v

-ve heating wire = 0.36v*

Signal wire = 0.70v



Testing loom plug voltage ENGINE ON*



+ve heating wire = 14.0v

-ve heating wire = 0.65v

Signal wire = 1.0v



Thanks again for your help, it truly is much appreciated*
rorygtb is offline   Reply With Quote
Old 08-09-2019, 05:02 PM   #8
Cougar4
Scooby Guru
 
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:
2001 LL Bean Outback
Winestone

Default

From your test results it looks like things are fine but there is a couple of things I question. The resistance to ground on the -ve heating wire seems odd to me. The heater circuit is controlled by the ECU through that wire. When the heater is OFF then the resistance to ground should be high, not 1.9 ohms. There other thing I question is the signal wire voltage going above .9 volts, the upper end of the normal operating range. Both of these abnormal things point to possible problems inside the ECU.

I would like to know what the voltage is on the -ve heater wire after the car has been running at normal operating temperature. The heater should be turned off then and the voltage should be close to the input voltage of the heater, 14 volts. This means no current is flowing through the heater resistance to create a voltage drop across it.

You earlier stated that the signal wire at the sensor end was showing a normal output voltage but you didn't see the same voltage at the ECU end of the same wire. In order for that to happen that wire would have to be separated somehow, or you mistakenly took the second reading on a different wire. From your resistance testing results, it shows no problem with that wire. Recheck the voltage at both ends of the sensor wire again while the engine is hot and running normally, and while things are connected up normally. The voltage should be the same at both ends of the wire, between .1 and .9 volts.

The .12 amps of current draw you saw is too high, unless something isn't in the sleep mode yet. Something is drawing excessive power. It could be some added accessory on the car causing that. Normal current draw should be around .020 amps or 20 milliamps.

Last edited by Cougar4; 08-09-2019 at 05:39 PM.
Cougar4 is offline   Reply With Quote
Old 08-10-2019, 06:16 AM   #9
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

Hello!

Okay so i've just finished carrying out the tests you suggested.

Here are the findings:

Drove the car up to operating temp (around 20 mins or so)

So with the engine ON...

-ve heater wire (at the plug) is reading 1.09v
+ve heater wire (at the plug) is reading 13.7v

I forgot to check resistance on the ground wire unfortunately, however I think the low voltage output proves your theory about something being amiss here. I understood that the negative element should now be reading closer to 14v as there is to be no more current passing to the heating elements after the sensor is up to temp.

Signal wire voltage (with engine on)

at plug - 0.15v at idle. Raises in tandem with revs. I revved the car using the throttle cable in the engine bay and watched the voltage spike to 1.02v. The higher the revs, the higher the voltage output

At the ecu input (plug plugged into the ecu and with the engine still switched on)

voltage 0.15v (same as at plug). Voltage also raises to 1.02v at the ecu side when throttle is pressed.


Some other background info which might help:

I have a digital voltmeter installed in my dash. It typically reads 14.0v with normal load (side lights and radio for example being used). When accelerating, I sometimes notice it drop sharply to about 13.4v then raise back up when the throttle is let off. Not sure if this is related or not. Only does this sometimes and appears to be sporadic.

One last thing: the car has always (may this support the ECU hypothesis?) struggled with cold starts. Even with the OEM o2 sensor, it always used to break up on acceleration and would eventually clear up by the time the car was hot. I read a thread a while back where a man detailed the EXACT symptoms. He explained that the only thing which fixed his issue was a replacement ECU. His car was the same year and model as mine (using the same Y9 ECU).

Thanks again,

Rory
rorygtb is offline   Reply With Quote
Old 08-10-2019, 06:24 AM   #10
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

Its maybe also worth letting you know that on my initial test of the ECU plug signal voltage, the reading was taken on a cold start. The Voltage at idle was 0.85v. When the throttle was pressed, it spiked to 1.15v and then was 'all over the place' dropping to around 0.35v then back up to 1v. It seemed totally random.
I understand the o2 sensor cycles itself quickly from .1-.9v to heat the sensor up, however the voltages being produced were completely random, sometimes falling outwith .1-.9
rorygtb is offline   Reply With Quote
Old 08-10-2019, 07:10 AM   #11
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

http://www.uklegacy.com/forums/index...ting-findings/

here is the link to the old thread I read where the guy had the same issues as me. Not sure if related but provides some good context to my problem from the start
rorygtb is offline   Reply With Quote
Old 08-10-2019, 07:26 AM   #12
Cougar4
Scooby Guru
 
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:
2001 LL Bean Outback
Winestone

Default

Good job on the testing. It appears the O2 heater circuit isn't turning off like it should after the sensor gets to the proper operating temperature. The signal wire is working good, but the signal voltage still seems too high when the engine is revved up. I am wondering if the alternator is generating some excessive AC ripple voltage. To check that out you can disconnect the connector going to the alternator and then start the engine. See what the sensor voltage does then. If it is still the same thing then it seems something inside the ECU is causing it to go higher than it should. I'm not sure that having the voltage a little on the high side is really a bad thing or not. The heater circuit issue isn't a good thing though. One thing to check is the what the engine coolant temperature is indicating for the ECU. If it isn't working correctly that would make the sensor heater stay on I would think. What does the software state the engine temperature is? I'm not sure what else the ECU looks at to know when to turn off the heater element, if there is something else.
Cougar4 is offline   Reply With Quote
Old 08-10-2019, 09:05 AM   #13
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

Thank you!

Gonna go home today and test the alternator. Might even swap it as I have another unit which I think is the same at the house. Can't hurt to check.

I drove the car again for another 15 mins. Pulled over, unplugged the alternator when the car was switched off and started it up. My multimeter read 0.8v on the singnal wire which then started to slowly drop .01v at a time.
I turned the car off to re insert the test probe at the front of the car and when I switched it back on (still with the alternator unplugged) it showed only 0.22v. When i plugged the alternator in, interestingly the voltage dropped to 0.09v.

The heater cable showed 11.8v (makes sense if the car is running off the battery only). This makes me wonder why the voltage DROPPED on the signal wire when the alternator was plugged IN. Surely the voltage would increase! The voltage on the heater positive wire went up when the alternator was plugged back in.

I will check coolant temp sensor when I get home to my laptop. IIRC however it always reads true. Sits around 90deg celsius-ish. I know there are two sensors though, so I might test the other as one controls the temp gauge whereas the other gives information to the ecu.

One thing I noticed (not sure if this is normal and due to the alternator being unplugged) was the idle sped up when the alternator was plugged back in.

Starting to feel as though I am running out of options here . Will try the alternator test when I am home and look into finding another ECU in the meantime

Thanks again bud
rorygtb is offline   Reply With Quote
Old 08-10-2019, 09:39 AM   #14
stevevetter
Scooby Newbie
 
Member#: 505136
Join Date: Aug 2019
Default

bumpity bump bump
stevevetter is offline   Reply With Quote
Old 08-10-2019, 09:40 AM   #15
stevevetter
Scooby Newbie
 
Member#: 505136
Join Date: Aug 2019
Default

I need three ....
stevevetter is offline   Reply With Quote
Old 08-10-2019, 09:40 AM   #16
stevevetter
Scooby Newbie
 
Member#: 505136
Join Date: Aug 2019
Default

replies so I can make a thread to show my ignorance and get majorly flamed. Sorry
stevevetter is offline   Reply With Quote
Old 08-10-2019, 09:46 AM   #17
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

hahahahaha thats all right mate, I was much the same. Joined only a few days back. Best of luck in finding what ya need
rorygtb is offline   Reply With Quote
Old 08-10-2019, 02:36 PM   #18
Cougar4
Scooby Guru
 
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:
2001 LL Bean Outback
Winestone

Default

Well the O2 signal voltage seems to be working a little better now that the alternator check was done. The sensor generates the voltage on the wire to the ECU in response to what the air/fuel mixture is in the exhaust, so the alternator voltage shouldn't effect that signal voltage unless something is happening inside the ECU to make it change. If the ECU temperature reading is normal then hopefully replacing the ECU, or having it repaired will correct the heater circuit issue.

One other thing you can do to check the alternator with your meter is rev the engine to around 1,500RPM and with the meter set to read AC voltage, measure across the battery terminals and see what the reading is. If your meter blocks DC voltage in the AC mode you should read less than .1 volt AC if the diodes inside the alternator are in good shape.
Cougar4 is offline   Reply With Quote
Old 08-11-2019, 06:14 AM   #19
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

Hi bud,

Good observation on the signal wire producing it's own voltage, I always forget it's powered by the Zircona element inside.
This would support the theory that the problem lies inside the ECU i suppose.

I did a test for AC voltage, however was a bit confused by the readings on my multimeter- maybe you can make sense of them.

On the AC voltage settings, there are only two options. These are labelled '200' and '750'. I am not sure if this means 0.2v and 0.7v or something else as there is no indication of what type of scale is used.

The battery terminals read 30.0.
I tested the current between the positive terminal and grounded the earth pin on the metal casing on the alternator- I got the same result.

Not sure if this is the equivalent of 0.3v or 0.03v

Looking into buying a new ECU currently. Found a couple for 70 so not too expensive if it solves my problem!

thanks again man, honestly do appreciate you taking the time to help me with this!
rorygtb is offline   Reply With Quote
Old 08-11-2019, 06:14 AM   #20
rorygtb
Scooby Newbie
 
Member#: 505063
Join Date: Aug 2019
Chapter/Region: International
Location: Dundee, Scotland
Vehicle:
2000 Legacy GT-B TT
Blue Ridge

Default

Hi bud,

Good observation on the signal wire producing it's own voltage, I always forget it's powered by the Zircona element inside.
This would support the theory that the problem lies inside the ECU i suppose.

I did a test for AC voltage, however was a bit confused by the readings on my multimeter- maybe you can make sense of them.

On the AC voltage settings, there are only two options. These are labelled '200' and '750'. I am not sure if this means 0.2v and 0.7v or something else as there is no indication of what type of scale is used.

The battery terminals read 30.0.
I tested the current between the positive terminal and grounded the earth pin on the metal casing on the alternator- I got the same result.

Not sure if this is the equivalent of 0.3v or 0.03v

Looking into buying a new ECU currently. Found a couple for 70 so not too expensive if it solves my problem!

thanks again man, honestly do appreciate you taking the time to help me with this!
rorygtb is offline   Reply With Quote
Old 08-11-2019, 06:35 AM   #21
Cougar4
Scooby Guru
 
Member#: 53443
Join Date: Jan 2004
Location: Anchorage, AK
Vehicle:
2001 LL Bean Outback
Winestone

Default

I assume the 200 is for 200 volts full scale and the 750 is for 750 volts full scale. Since you got 30 volts it means your meter doesn't block the DC component of the voltage. You would need to add a proper size capacitor in series with one of the meter leads in order to block the DC from the meter. Some meters don't block the DC when checking AC voltage. Fluke meters, along with some others, do block the DC. I wouldn't bother with this test until you get a meter that blocks DC in the AC mode.

Glad to be of some help.

Last edited by Cougar4; 08-11-2019 at 06:41 AM.
Cougar4 is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

All times are GMT -4. The time now is 09:25 PM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2019 Axivo Inc.
Copyright ©1999 - 2019, North American Subaru Impreza Owners Club, Inc.