lmdew Posted October 31, 2006 Share Posted October 31, 2006 CEL P1133 The front O2 sensor on the 99 Impreza I picked up is steady at 3.8 volts. I put in a new sensor and got the same results. All the wiring seems to check out. Blocked CAT???? It runs fine, just made a 2000 mile trip. The CEL is for the front bank no response. Ideas? Thanks Link to comment Share on other sites More sharing options...
ShawnW Posted October 31, 2006 Share Posted October 31, 2006 OHM it to the ECU and see if theres any breaks in the wires. Link to comment Share on other sites More sharing options...
Cougar Posted October 31, 2006 Share Posted October 31, 2006 Usually the sensor voltage runs between .1 to 1 volt so that is way too high of a voltage. Where are you placing the meter probes to get this reading and what kind of a meter are you using? Is it a digital meter with a 10 megohm input impedence? Link to comment Share on other sites More sharing options...
lmdew Posted October 31, 2006 Author Share Posted October 31, 2006 I probe the splices of the new sensor with my Fluke 83. I can also see the voltage with my computer through the OBDII connector. Yes the voltage is way high. When I checked the old sensor on the bench with the torch it read .5 volts. Link to comment Share on other sites More sharing options...
starsea Posted October 31, 2006 Share Posted October 31, 2006 What type of connector and software do you use? I have been wanting to get one. What is the torch test? Link to comment Share on other sites More sharing options...
Cougar Posted October 31, 2006 Share Posted October 31, 2006 I probe the splices of the new sensor with my Fluke 83. I can also see the voltage with my computer through the OBDII connector. Yes the voltage is way high. When I checked the old sensor on the bench with the torch it read .5 volts. Well you have a very good meter there. It sounds like the circuit to the sensor may have a high impedence on it or there is something wrong with the ECU for that circuit. Have you checked the wiring between the ECU and the O2 sensor for a connection problem by looking at the resistance of the wires between the sensor and the ECU, as Shawn rightly suggested? A bad wire connection to the sensor should generate a error code if the ECU is working correctly. Does disconnecting the O2 sensor cause the MIL to light? If not, then the ECU O2 sensor circuit is not working correctly or the there is something wrong with the warning light circuit. If the voltage stays at 3.8 volts with the sensor disconnected then the ECU is the problem in my opinion. You may be able to fix it by checking the pins internally that tie to the sensor. Perhaps a lead is not making good connection to the board. I have heard of other people fixing the ECU this way. To Starsea: By heating the O2 sensor with a torch the sensor is looking at the gases of the flame and putting out a voltage that corresponds to the gas flame conditions. Since the flame is burning clean the voltage is about in the middle of the sensor output. If the fuel mixture was rich the output voltage would be around 1 volt. Link to comment Share on other sites More sharing options...
Wayne Boncyk Posted October 31, 2006 Share Posted October 31, 2006 Larry - it's almost certain to be a problem with the ECU. Try testing the voltage with the O2 sensor unplugged, with the engine running and with the cat/exhaust hot. If you can probe the open circuit voltage with the engine running while you have someone vary the throttle, you should see fluctuations around one volt into your high impedance Fluke meter. If not, then probably the high voltage that is coming from the ECU has blown that sensor. If you get nothing from the existing sensor with a hot cat as you run from relatively lean idle to relatively rich open throttle, then you've gotta fix/replace the ECU before you plug in another O2 sensor, or you'll risk blowing that one, too. Link to comment Share on other sites More sharing options...
Cougar Posted October 31, 2006 Share Posted October 31, 2006 Larry - it's almost certain to be a problem with the ECU. Try testing the voltage with the O2 sensor unplugged, with the engine running and with the cat/exhaust hot. If you can probe the open circuit voltage with the engine running while you have someone vary the throttle, you should see fluctuations around one volt into your high impedance Fluke meter. If not, then probably the high voltage that is coming from the ECU has blown that sensor. If you get nothing from the existing sensor with a hot cat as you run from relatively lean idle to relatively rich open throttle, then you've gotta fix/replace the ECU before you plug in another O2 sensor, or you'll risk blowing that one, too. Wayne, I don't think he has damaged the new sensor but I do think you are correct about the problem being inside the ECU. If it is, hopefully the problem is just due to a loose wire that ties to the sensor lead and he can repair it. From the test results that Imdue got on his replaced sensor it sounds like it is ok also. Link to comment Share on other sites More sharing options...
Wayne Boncyk Posted October 31, 2006 Share Posted October 31, 2006 Wayne, I don't think he has damaged the new sensor but I do think you are correct about the problem being inside the ECU. If it is, hopefully the problem is just due to a loose wire that ties to the sensor lead and he can repair it. From the test results that Imdue got on his replaced sensor it sounds like it is ok also. Yup - you're right, I glanced over that part of his post. Sorry, Larry! Link to comment Share on other sites More sharing options...
sea#3 Posted October 31, 2006 Share Posted October 31, 2006 The CEL is for the front bank no response. What is the code ? There may be a recall on the ECM Sea#3 Link to comment Share on other sites More sharing options...
lmdew Posted October 31, 2006 Author Share Posted October 31, 2006 Yep, that was my plan. I'm going to disconnect the sensor from the harness and see what the ECU has vs the output of the sensor. I just need the time to get there. Thanks for the info! I'll update when I get it checked. Link to comment Share on other sites More sharing options...
hankosolder2 Posted November 1, 2006 Share Posted November 1, 2006 Yep, that was my plan. I'm going to disconnect the sensor from the harness and see what the ECU has vs the output of the sensor. I just need the time to get there. Thanks for the info! I'll update when I get it checked. A few other things to consider...if you have a grounding problem, the voltage drop may appear on a sensor lead. Measure the voltage drop between the ground you were using during your 02 sensor meaurement and the negative post of the battery. Also check the voltage drop between your 'test' ground point and the engine block. There should be negligable voltage showing on your meter. As others have said, it could also be the ECU. I think it's pretty much physically impossible for the 02 sensor to generate that much voltage by itself. It is possible that there is leakage between the 02 sensor heating element (Is this a heated 02 sensor?) and the actual 02 sensor output lead. Good luck, Nathan Link to comment Share on other sites More sharing options...
lmdew Posted November 2, 2006 Author Share Posted November 2, 2006 I disconnected the O2 sensor in the engine bay. It worked fine if read directly. At the same time the ECU still had 3.8 volts. I pulled the ECU cover and looked at the circuits and no defects noted. I'm going to try another ECU or pull the O2 sensor wires at the ECU plugs and then see if the ECU still has 3.8 volts. That will confirm its internal. Link to comment Share on other sites More sharing options...
lmdew Posted November 5, 2006 Author Share Posted November 5, 2006 I believe I have confirmed its the ECU. I pulled connector B136 pins 7 (white +) and 20 (black -) which are the O2 sensor input wires. The ECU still sees 3.8 volts and reading the O2 sensor direct its 0.001 to 0.8 volts which is right in the range is should be. How long is the Federal Specifaction Emissions Performance Warranty period? I thought it was 8 years, but have not been able to confirm it. I'll check with the Dealer to see if the ECU has any recalls. Looks like its a new ECU. Link to comment Share on other sites More sharing options...
Cougar Posted November 5, 2006 Share Posted November 5, 2006 I would check wires going the pins you stated with a ohmmeter. Make sure they are making good connection to the board first before you condemn the unit. If the wires are ok then the problem is most likely with the IC that ties to those wire connections. Since you have a voltage on the pins chances are the connections are ok and there is a actual circuit problem causing this trouble. If you would like me to fix the ECU for you let me know. It should be a pretty easy fix by replacing the right part. If you are handy with a solder iron you could try it yourself and save a lot of money over even a rebuilt one. You don't want to know what a new one from the dealer would cost you. Link to comment Share on other sites More sharing options...
lmdew Posted November 5, 2006 Author Share Posted November 5, 2006 I'd pulled the wires out of the ECU plug and the ECU still sees 3.8 volts. There should be 0 volts there with no input, I would think. The ECU is all solid state and I did not see any burned/shorted circuits, but I'm not a micro-miniture repair guy either. Thanks for the thoughts. Link to comment Share on other sites More sharing options...
lmdew Posted November 18, 2006 Author Share Posted November 18, 2006 HELP! Update: 1) I tried a used ECU and had the same results. 3.8 volts as seen by the OBDII readout. 2) I did resistance checks on the O2 sensor wires: - jumpped in the engine compartment (sensor-ground wire) 0.00 ohms at the ECU connector. - Individual wires to ground, OL (open line) With the ECU Connector plugged in and the key on: - Sensor wires 3.8 volts at the ECU and the engine bay connector (O2 sensor disconnected) - Heater wires 12.8 volts = battery voltage As I see it, the 3.8 volts is coming from the ECU. So either I have two bad ECU's or I'm over looking something. If you have the factory manuals with good wire diagrams on the ECU/O2 circuit please send them my way. Anyone with a 99 Impreza in Colorado Springs, I could hook my OBDII reader up to and check your O2 sensor voltage? Thanks, Larry Link to comment Share on other sites More sharing options...
Cougar Posted November 18, 2006 Share Posted November 18, 2006 Larry, I would do the same tests you did to make sure the circuit to the sensor is ok. It is clear that from your results that the wiring to the sensor is ok. It is hard to say what the open circuit voltage of the ECU is supposed to be normally since it will depend on how the circuit is designed. Measuring a good ECU like you want to do is the best way to find out that I can think of myself. Your test procedures to try and solve this trouble are very good I think. It seems to me that the replacement ECU must have the same trouble as the original. If you would like me to check out the spare unit for you let me know. If it is bad, I may be repair it for you. On units I have repaired before I have found that the output circuit device is damaged. If you can get another ECU for around $30 it wouldn't be worth the effort though due to shipping costs since I live in Alaska. Link to comment Share on other sites More sharing options...
lmdew Posted November 18, 2006 Author Share Posted November 18, 2006 Since the O2 circuit to the ECU is just getting input voltage from the sensor, wouldn't you think if the sensor was disconnected the wiring should be reading 0 volts? How do you check the internal circuits? Thanks for the info. Link to comment Share on other sites More sharing options...
sea#3 Posted November 19, 2006 Share Posted November 19, 2006 HELP! Update: 1) I tried a used ECU and had the same results. 3.8 volts as seen by the OBDII readout. 2) I did resistance checks on the O2 sensor wires: - jumpped in the engine compartment (sensor-ground wire) 0.00 ohms at the ECU connector. - Individual wires to ground, OL (open line) With the ECU Connector plugged in and the key on: - Sensor wires 3.8 volts at the ECU and the engine bay connector (O2 sensor disconnected) - Heater wires 12.8 volts = battery voltage As I see it, the 3.8 volts is coming from the ECU. So either I have two bad ECU's or I'm over looking something. If you have the factory manuals with good wire diagrams on the ECU/O2 circuit please send them my way. Anyone with a 99 Impreza in Colorado Springs, I could hook my OBDII reader up to and check your O2 sensor voltage? Thanks, Larry Ok this is only a theory Since you have replaced the ECM and have the same out come, I would think you might have a back feed through the harness somewhere. The most likely place is the B19/E1 connector (see schematic). May be some corrosion or water in the connector. If not that, I would disconnect the C7 wire and the C20wire at the ECM then see what the scan tool says. If it says 3.8v it’s inside the ECM. If its 0 then get a voltmeter and measure the wires on the harness. Since the B19/E1 connector share the front and rear 02 sensors there may be a back feed through there Hope this Helps SEA#3 Link to comment Share on other sites More sharing options...
Cougar Posted November 19, 2006 Share Posted November 19, 2006 Since the O2 circuit to the ECU is just getting input voltage from the sensor, wouldn't you think if the sensor was disconnected the wiring should be reading 0 volts? How do you check the internal circuits? Thanks for the info. Actually most circuits will have some sort of a open circuit voltage on the pins. When the sensor is connected the voltage should change since it is changing the circuit resistance when something is tied to the open circuit. The sensor changes resistance as the exhaust gases change and this makes the current that is flowing through the sensor change. Without really knowing how the internal circuit is designed it is hard to tell what is correct. One trick I use to help find trouble areas is to compare resistance readings of certain areas of the circuit I suspect trouble in. You need to have a good circuit to compare the readings with though. Link to comment Share on other sites More sharing options...
sea#3 Posted November 19, 2006 Share Posted November 19, 2006 The wiring diagram that was missed last time SEA#3 Link to comment Share on other sites More sharing options...
lmdew Posted December 14, 2006 Author Share Posted December 14, 2006 Well, I got a Subaru front O2 Sensor and put it in. It still reads 3.8 volts, but now responds as it should instead of a constant 3.8 volts. I have never seen an O2 sensor with that voltage signal, but the ECU is happy and has been for the last week. I guess you need to get the SUBARU O2 sensors for proper operation. In this case the bench check did not help. Link to comment Share on other sites More sharing options...
ShawnW Posted December 14, 2006 Share Posted December 14, 2006 Im wondering if you might have gotten a generic sensor where you had to crimp wires...and that maybe you got the shielding for the signal wire touching the inside of the signal wire somewhere and it was grounding itself out. It almost has to be a problem between the plug for the o2 sensor and the sensor itself if you ask me but truly good advice to get a genuine Subaru part for this...maybe even better to put a USED genuine on than a generic. Link to comment Share on other sites More sharing options...
Cougar Posted December 14, 2006 Share Posted December 14, 2006 Thanks for the update Imdew. Glad you got it fixed. Does the voltage across the sensor drop to a normal level (0v to 1V) when the car starts to warm up? Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now