skibumm100 Posted July 2, 2006 Share Posted July 2, 2006 Hi guys, I've got a 98 Legacy GT, 2.5L w/auto. I bought it used a couple years ago with about 125K on it and started having a CEL light after about 2 months of ownership. I'm getting a CEL saying bad catylytic converter. I've changed the front O2 sensor (put in a Denso unit) and run Chevron techron fuel injector cleaner through it. New air filter, etc. I have OBDII scanning software on a laptop and don't see anything out of place. Both O2 sensors seem active. I reset the CEL periodically and it comes back, sometimes less than an hour later, sometimes days later. I think the code P420. I do remember it's for the front cat. When the car is first started, it runs really rich for a few minutes. Raw fuel smell. I know it should run somewhat rich when cold, but it seems REALLY rich. This is the first Subie I've owned so I don't know if it's "normal" to smell raw fuel when cold. I can watch the ECT as the car warms up and it seems to be reading correctly. The thermostat also looks to be working right. Are there any secrets to making this beast run right? I guess it's possible the front converter could be plugged up. I just don't want to throw a bunch of parts at it unless I know I'm headed down the right path. That's why I came here to seek out expert advice. BTW, as far as Subie message boards go, this one seems head and shoulders above the rest if you drive a Leggy. You guys rock....in advance. Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 4, 2006 Author Share Posted July 4, 2006 Does anybody have any suggestions? Sorry to bump but I need to get this fixed. Quote Link to comment Share on other sites More sharing options...
dxrflyboy Posted July 4, 2006 Share Posted July 4, 2006 The P0420 code simply means that the cat isn't doing its job. If it was plugged up, the car would barely run, if at all. The only way to fix it and keep the CEL from coming back on is to replace the cat. You will get some HC odor on startup, but it shouldn't be very much or last too long. Having a cat that isn't working will add to the problem. It's hard to tell from your description whether it is actually a problem or not. An injector cleaning may help, but there's no guarantee. If the odor goes away once the system goes into closed loop, there may not be much that can be done, aside from replacing the cat. Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 4, 2006 Author Share Posted July 4, 2006 Thanks for the reply. I didn't want to jump into changing the cat because a lot of times codes like that can be a symptom, not a cause. I just want to make sure I cover all the bases before I change it out. I've never owned a car that had to have a cat replaced and most cars I drive well over 150K. Might not be the case with this one. I don't smell gas once it's in closed loop. Quote Link to comment Share on other sites More sharing options...
nipper Posted July 4, 2006 Share Posted July 4, 2006 replace the second o2 sensor. you need to chase down the running rich before attacking the cat problem. The first o2 sensor, was that an Oe sensor or generic. that one has to be OE, the second one can be geric. Running rich you can burn up the cat. Also check to make suree the ECU temp sensor is reading right. Soobies dont run rich, its rare so something is wrong someplace. A cat wont make the car run rich, but running rich can kill a cat. When you say the o2 sensors are active, what are they reading? Also are you sure your software is truly calibrated properly (ie non ebay software). Even when the car is stone cold it doesnt run rich. Dont replace the cats yet, we need to chase this down nipper Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 4, 2006 Author Share Posted July 4, 2006 Nipper, I replaced the front O2 sensor with a Denso "OE replacement" 3-wire sensor. Do you know who manufactures the OEM sensor? When I said they are active I meant that they respond fairly quickly and bounce from rich to lean. They aren't slow to act. The front one is more "active" than the second one, as it should be. I still have the original one, assuming it's an OEM sensor. I didn't see a huge difference in the way it acted with the new one so the old one might be fine. I will put the laptop on it and check the mV readings, I don't remember the numbers right now. I need to check who I got the software and cable from, also. It's been a few years since I got it. The desktop icon says "Car Code OBD II". The ECU temp sender read around ambient temp if the car sits over night and then rises gradually to around 190 deg F when it's warmed up. It sounds about right to me. I was hoping it was the ECU sensor that was bad. It doesn't look like it's the culprit. Gregg Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 4, 2006 Author Share Posted July 4, 2006 I found the info for my OBD II scanning software. It's made by Alex Peper. Here's a link to the site for the software. http://www.obd-2.com/ Quote Link to comment Share on other sites More sharing options...
Commuter Posted July 5, 2006 Share Posted July 5, 2006 This particular generation of engine / ECU does have fuel maps on the rich side. They can be aggrevated by a sluggish O2 sensor (that is, run more rich). Twice, I've had issues with some carbon build up in my 97 OB 2.5L auto, and I do mostly highway driving. I'm not the only one, there have been other reports. Some years ago, Emily of CCR posted that these engines look like carburated engines from the 70's when they tear them down. There is that much carbon build up happening. Just FYI Commuter Quote Link to comment Share on other sites More sharing options...
dxrflyboy Posted July 5, 2006 Share Posted July 5, 2006 This particular generation of engine / ECU does have fuel maps on the rich side. They can be aggrevated by a sluggish O2 sensor (that is, run more rich). Twice, I've had issues with some carbon build up in my 97 OB 2.5L auto, and I do mostly highway driving. I'm not the only one, there have been other reports. Some years ago, Emily of CCR posted that these engines look like carburated engines from the 70's when they tear them down. There is that much carbon build up happening. Just FYI Commuter That would explain why we change so many cats on these cars... Quote Link to comment Share on other sites More sharing options...
nipper Posted July 5, 2006 Share Posted July 5, 2006 hrmm i guess im blessed. My 97 has the original cat at 185,000 miles. Thousg h i did make a cash sacrifice to the sooby gods on the AWD. nipper Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 5, 2006 Author Share Posted July 5, 2006 Guys, thanks for the replies. I'm gonna replace the 2nd O2 since it would be the one setting the code and I haven't replaced it since I've owned it. Sounds like I need to find a replacement cat also. Are there any troublesome parts in the vapor recovery system that commonly fail? That could also lead to a fuel smell on start-up. Any good afternarket cats out there? If carbon is a problem I will probably run some Seafoam or GM Top Engine Cleaner through it......before I replace the cat. Quote Link to comment Share on other sites More sharing options...
nipper Posted July 5, 2006 Share Posted July 5, 2006 there is the canister purge valve, but that is pretty good at throwing a code if its not working. nipper Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 9, 2007 Author Share Posted July 9, 2007 Hey nipper....the evap solenoid finally threw the code. PO 443. In the process of troubleshooting the code, I found that I had no vacuum back at the solenoid. I checked the line at the engine compartment and it had good vacuum. I checked the line where it comes through the floor under the rear seat and it had good vacuum. Nothing on the other side at the valve. I replaced that section with emission/fuel hose. Still have the code. After thinking about it for about 15 seconds I realized that the computer doesn't know if the evap system is working or not. It can only tell if the solenoid is shorted or open. Well..... it's neither. I checked the resistance with an ohm meter and it was in spec. I also had 12VDC at the plug with the key on, engine off. It's still throwing the code so I figured it must not like the valve or the connector is corroded. I bet the fuel smell is happening because there was no vacuum back there. If the valve is cheap, I'll get a new one. If it's expensive, I'll try to clean the connector. Quote Link to comment Share on other sites More sharing options...
nipper Posted July 9, 2007 Share Posted July 9, 2007 Hey nipper....the evap solenoid finally threw the code. PO 443. In the process of troubleshooting the code, I found that I had no vacuum back at the solenoid. I checked the line at the engine compartment and it had good vacuum. I checked the line where it comes through the floor under the rear seat and it had good vacuum. Nothing on the other side at the valve. I replaced that section with emission/fuel hose. Still have the code. After thinking about it for about 15 seconds I realized that the computer doesn't know if the evap system is working or not. It can only tell if the solenoid is shorted or open. Well..... it's neither. I checked the resistance with an ohm meter and it was in spec. I also had 12VDC at the plug with the key on, engine off. It's still throwing the code so I figured it must not like the valve or the connector is corroded. I bet the fuel smell is happening because there was no vacuum back there. If the valve is cheap, I'll get a new one. If it's expensive, I'll try to clean the connector. Damn im so good i scare myself at times The computer can read if the solenoids are opening, closing, or are stuck. I bet yours is stuck. you can always get one from a scrap yard. nipper Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 9, 2007 Author Share Posted July 9, 2007 You know, I think you're right again. I had the key on, engine off and I unplugged the connector and plugged it back in and it didn't go "click". I tapped on it but still no "click". Are they pretty much the same from year to year? Quote Link to comment Share on other sites More sharing options...
Skip Posted July 9, 2007 Share Posted July 9, 2007 I need enlightened " The computer can read if the solenoids are opening, closing, or are stuck" In this particular case where an evap or purge solenoid is stuck. Do you mean electricly open as in windings open? Closed as in not passing gas?? How does the ECU know the vacuum is not being tranfered? Is there a vacuum switch in the circuit that I have not seen? Does it use the Baro-MAP sensor? Is this associated with the old gas cap loose check engine light? Thanks in advance for the reply as.. Enquiring minds want to know. Quote Link to comment Share on other sites More sharing options...
nipper Posted July 9, 2007 Share Posted July 9, 2007 When the ecu sends a signal to a solenoid, there is an electrical spike that is created by the movement of the plunger. Every solenoid ever made does this, its just physiscs. It is 100% repeatable. The computer looks for this feedback. IF the plunger is stuck, there will be no feedback. If it travels part way, the full spike will not be generated. Supposedly the next level of OBD will be able to comminuicate this information (the powers that be have yet to decide this as it will take more computing power). This has nothing to do with what ever the solenoid is connected to,its purely a function of electromagnetisim that the ECU is taking advantage of. This is how the TCU can tell you when you have a failed duty solenoid in the tranny. nipper Quote Link to comment Share on other sites More sharing options...
OB99W Posted July 9, 2007 Share Posted July 9, 2007 [...]When the car is first started, it runs really rich for a few minutes. Raw fuel smell. I know it should run somewhat rich when cold, but it seems REALLY rich. This is the first Subie I've owned so I don't know if it's "normal" to smell raw fuel when cold.[...] I realize we're past this part of the diagnosis, but the following might put to rest some of the concern: http://endwrench.com/images/pdfs/ExhaustInfo.pdf Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 9, 2007 Author Share Posted July 9, 2007 OB99W, Thanks for the link. Mine seems to smell like fuel year-round. Not so much when the car is warmed up, which falls in line with the endwrench article. I'm not sure I agree 100% with the sentiment of the article but I guess it gets people off their back. Excess fuel is just that, excess. I know it needs to be richer for a cold start but my other cars start, run and drive w/o stalling or stumbling...but don't give off that intoxicating scent. Sorry, worked at a gas station way back...before vapor recovery hoses...a little brain damaged ....but not too much.:-\ Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 12, 2007 Author Share Posted July 12, 2007 Pulled the solenoid off last night and cleaned it up. Squirted some PB Blaster into the ports. Then I hooked it up to the battery direct and the valve shifts and operates. Put it back on the car and it doesn't click when I have power at the plug, KOEO. Might be a bad plug connection. I'm gonna try bending the pins slightly to try to get better contact. This bugger should work! I cleared the code and it keeps coming back with PO443. My old catalytic converter code has been solved. Haven't seen it in quite some time. Quote Link to comment Share on other sites More sharing options...
skibumm100 Posted July 13, 2007 Author Share Posted July 13, 2007 I pulled the plug off the solenoid valve last night and checked everything again. I've got a solid 12+ VDC at the plug with the valve disconnected. The solenoid valve resistance is 27.5 ohms. The book says anything between 10 and 100 ohms is good. I put a 12V test light on the plug connections to see if the voltage is being dragged down under a load. The light burned bright. I am really stumped. This thing should work. Tonight I might try sticking a couple of straight pins through the two wires and checking the voltage with the solenoid connected. I tried to back-probe the plug yesterday but it's so small, I'm concerned that I will damage it. Maybe I should just it. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.