Rising Sun Member Forums  

Go Back   Rising Sun Member Forums > Toyota 4x4 > General Tech Forum

Reply
 
Thread Tools
  #1  
Old 09-24-2012, 08:13 PM
Rzeppa's Avatar
Rzeppa Rzeppa is offline
Cruise Moab Committee
 
Join Date: Aug 2005
Location: Kittredge CO, USA
Posts: 5,136
Default 95 Subaru Legacy Fails NxOx

Apologies in advance for the non-Toyota question, but my daughter has a new-to-her 1995 Subaru Legacy that she can't get through emissions. Her main ride is a 1984 FJ60 so please give her a pass...

I don't know the exact miles but they're fairly high. She got the car for $50, non-running. The symptom was it had died on the PO and wouldn't start, even after the PO had installed a new starter and battery. My daughter put in a new alternator and serpentine belt and it still wouldn't start. I came down and diagnosed bad battery cables. New cables and she runs great.

She failed emissions the first try, passed HC and CO, failed NxOx by about 7 out of 2.5. I followed the manual to test the EGR valve working properly and the EGR pipe is not clogged. The valve operates properly when commanded by the ECU and the engine stumbles properly when the EGR is manually operated with an external vacuum.

There is a CEL code indicating "open injector on #4". There is no obvious indication of any malfunction of the fuel injectors, and my reasoning is that even if she is only running on 3 cylinders that if the injector was not opening to let fuel in then it would not likely create NxOx. No combustion = no pollution, right?

We changed the oil and filter, and air filter last night and now she fails by 5 out of 2.5. My gut suggests that it might be a depleted cat.

My understanding of NxOx is that it is caused by excessively high combustion temperatures, and that the EGR reduces the combustion temperatures by diluting the fuel air mixture with spent exhaust.

So, my question is, IF (and that's a big if, the car seems to run fine) one of the 4 fuel injectors is open electrically and is not opening, could that cause NxOx failure? Or is it more likely a depleted cat? Or maybe something else.

Passing CO and HC seems like a huge deal in terms of eliminating a bunch of stuff.

TIA for any advice. Her paper tags expire on Friday and she's anxious to get this figured out.

__________________
Jeff Zepp
Kittredge CO USA
1971 & 1976 FJ40s, 1978 FJ45, 1987 FJ60
Rising Sun 4WD Club, TLCA #4063
http://american3dprinting.net/
Reply With Quote
  #2  
Old 09-24-2012, 09:42 PM
powderpig's Avatar
powderpig powderpig is offline
Hard Core 4+
 
Join Date: Aug 2005
Posts: 1,341
Default

It may be good to post the figures up on the HC and CO. It may be close to what Daniel's 80 was doing, were it passed HC and CO but failed NOX because the O2 sensors were not working well and leaning out the fuel mixture too much. So if engine is not getting enough fuel, it can create what you are looking at. If you have a infared temp sensor and you can test the temp before and after the cats to see if you are getting any temp difference that would be good. But if not, then the assumption one can make is the the cats may not be getting enough fuel to lite off to burn at the proper temps to clean up the exhaust going out the tail pipe.
But if you state that one cylinder's injector is possiably (being open) over fueling #3 cylinder, then the O2 sensor on that bank can be over whelmed with fuel and be the computer could be leaning out the other cylinders to the point that it can create a lean problem and then have nox go up as well as not having enough fuel to allow the cats to work properly. How many cats and how many 02 sensors does it have. 1 cat and one sensor or 2 cats and 3 sensors or some other combination.
So I would make sure you do not have a bad injector on #3 as well as test the O2 sensor to make sure they are working well, before I replace the cats.

Maybe this helps
Reply With Quote
  #3  
Old 09-25-2012, 07:00 AM
DaveInDenver's Avatar
DaveInDenver DaveInDenver is offline
Hard Core 4+
 
Join Date: Jun 2006
Location: Larimer County
Posts: 6,475
Default

NOx is a byproduct of imperfect combustion due to nitrogen present in the fuel and air, 'high temp' just means that the reaction does not happen at room temperature. An excessively lean running engine produces more than normal but all real world combustion forms NOx, which is scrubbed by the converter.

Injector open electrically could also mean it's not fully functional mechanically. Could be leaking, stuck partially closed or only partially opening. It's common for them to get clogged or wear out. If the injector is not working correctly for whatever reason the ECU tries to compensate, which would throw off the timing of properly functioning injectors or ignition. Does the engine ping?

EGR seems like a good thought. I'd check the O2 sensor and probably (on principle for me) replace the plugs, wires, cap, rotor. Check for air leaks in the intake after the AFM (or whatever air metering the Subaru uses). I'd guess the catalytic converter is old and replacing it will likely get it to pass, but that might not the root problem. Still, could end up being how you fix it when the alternative is a $500 AFM.
__________________
'91 Toyota Pickup
'09 Kawasaki KLR650
'12 Gunnar Rockhound 29

"They say the test of literary power is whether a man can write an inscription. I say, 'Can he name a kitten?'" -- Samuel Butler
Reply With Quote
  #4  
Old 09-25-2012, 07:12 AM
Rzeppa's Avatar
Rzeppa Rzeppa is offline
Cruise Moab Committee
 
Join Date: Aug 2005
Location: Kittredge CO, USA
Posts: 5,136
Default

Thanks guys. Quick question: Wouldn't the OBDII system report faulty O2 sensor(s) and or faulty AFM?

FWIW, there were two missing bolts of the four between the air cleaner housing and the hose that connects to the throttle body inlet, creating a fairly good sized leak on the clean side of the air flow. We replaced those bolts and fixed that leak.

I assumed that when the code said "open #4 injector" that it meant electrically open, therefore not energizing and opening to let fuel through.
__________________
Jeff Zepp
Kittredge CO USA
1971 & 1976 FJ40s, 1978 FJ45, 1987 FJ60
Rising Sun 4WD Club, TLCA #4063
http://american3dprinting.net/
Reply With Quote
  #5  
Old 09-25-2012, 07:19 AM
DaveInDenver's Avatar
DaveInDenver DaveInDenver is offline
Hard Core 4+
 
Join Date: Jun 2006
Location: Larimer County
Posts: 6,475
Default

Quote:
Originally Posted by Rzeppa View Post
Thanks guys. Quick question: Wouldn't the OBDII system report faulty O2 sensor(s) and or faulty AFM?
Yes, I would expect so. Even the OBDI on my 22R-E can tell failures at this high level. The only point I would make is that not meeting failure criteria does not equal perfectly running either. A marginal or misadjusted actuator/sensor could still be under the fail trigger. I dinked with my AFM trying to tune my EFI to my cam by tightening and loosing the flapper spring, which tricks the ECU into varying the amount of fuel it delivers. Technically my AFM was 'failed' compared to stock setting but it wouldn't show up as a diagnosable failure because the AFM was still sending a perfectly valid signal but would have made a 100% stock engine run poorly.
Quote:
FWIW, there were two missing bolts of the four between the air cleaner housing and the hose that connects to the throttle body inlet, creating a fairly good sized leak on the clean side of the air flow. We replaced those bolts and fixed that leak.

I assumed that when the code said "open #4 injector" that it meant electrically open, therefore not energizing and opening to let fuel through.
It does, the ECU only knows electrical conditions, but they are electromechanical devices and it's not unusual at all for them to leak as the seats and seals get old. Also 'open' just means the coil impedance is out of range, so it might not have just really opened but the solenoid is just weak and only partially actuating. The only way to really know is measure it.

That can sometimes be done without taking the injectors out. Part of the troubleshooting steps to check connections and voltages at the ECU first because a broken wire would look like a failed injector to the computer and soldering is a lot cheaper than a new injector. On the 22R-E the injectors are not individually controlled but it sounds like on the Subaru they are (at least it knows one has failed). So you might be able to measure each circuit at the ECU and at the injector, which is handy. FWIW my injector coils are nominally about 14 ohms and at the ECU output should measure a minimum of 10V.
__________________
'91 Toyota Pickup
'09 Kawasaki KLR650
'12 Gunnar Rockhound 29

"They say the test of literary power is whether a man can write an inscription. I say, 'Can he name a kitten?'" -- Samuel Butler

Last edited by DaveInDenver; 09-25-2012 at 07:48 AM.
Reply With Quote
  #6  
Old 09-25-2012, 08:57 AM
SteveH SteveH is offline
Hard Core 4+
 
Join Date: Aug 2006
Location: Colo Springs
Posts: 1,273
Default

I would get a live-reading OBDII reader and watch the two oxygen sensors in real time. The voltages fluctuate from 0-1, and if the cat is bad, each sensor will show about the same fluctuation in voltage. I'd also expect to see a PO420 to show the bad cat. You can purchase a OBD tester that shows live data on Amazon for about $140.

You can move the 'stuck injector' to another cylinder and see if the code follows it, but this is sort of a pain messing with o-rings and fuel rails.

I think you're on a good track to eliminate any/all vacuum leaks - I'd rate that as a higher likelihood right now.
__________________
Steve H.
Colo Springs, CO
Reply With Quote
  #7  
Old 09-25-2012, 04:41 PM
Rzeppa's Avatar
Rzeppa Rzeppa is offline
Cruise Moab Committee
 
Join Date: Aug 2005
Location: Kittredge CO, USA
Posts: 5,136
Default

Update: She took it to a shop today, and they verified that it is indeed running on only 3 cylinders, they pulled the #4 plug wire and observed no change to idle. They also observed a good strong spark there.

So I guess the next step is to pull the injector connector and ohm out the injector. If it ohms out good, I would guess there is a faulty connection to the injector. That would be super cheap to fix if it is easy to find and get to. If it's the injector itself I still have to imagine that would be a bunch cheaper than a new cat. My recollection of pulling injectors is that they can be a PITA and need a flare wrench depending on how they're oriented and what's in the way of getting a socket over them.

Thanks again for all the helpful suggestions!
__________________
Jeff Zepp
Kittredge CO USA
1971 & 1976 FJ40s, 1978 FJ45, 1987 FJ60
Rising Sun 4WD Club, TLCA #4063
http://american3dprinting.net/
Reply With Quote
  #8  
Old 09-25-2012, 05:06 PM
DaveInDenver's Avatar
DaveInDenver DaveInDenver is offline
Hard Core 4+
 
Join Date: Jun 2006
Location: Larimer County
Posts: 6,475
Default

Quote:
Originally Posted by Rzeppa View Post
Update: She took it to a shop today, and they verified that it is indeed running on only 3 cylinders, they pulled the #4 plug wire and observed no change to idle. They also observed a good strong spark there.

So I guess the next step is to pull the injector connector and ohm out the injector. If it ohms out good, I would guess there is a faulty connection to the injector. That would be super cheap to fix if it is easy to find and get to. If it's the injector itself I still have to imagine that would be a bunch cheaper than a new cat. My recollection of pulling injectors is that they can be a PITA and need a flare wrench depending on how they're oriented and what's in the way of getting a socket over them.

Thanks again for all the helpful suggestions!
Couldn't say on the Subaru, but getting to them is a major PITA, actually getting them in and out isn't hard. The fuel rail(s) might be pressurized, I pulled the fuel pump fuse and cranked the engine to release the pressure. Otherwise pretty straightforward. Replace all the copper crush gaskets and rubber o-rings, don't be tempted to save a few dollars. :-)

Might not be any cheaper than a cat. The injectors for the 22R-E list at almost $200 and the good price is about $165. This is for a new part. Hard to say what pick-n-pull or used might be, get a couple used and figure out a way to test them maybe? I just put a Walker cat on my truck, was about $75 from Amazon. If it's like the 80 with multiple special fit ones, then catalytic converters are expensive.
__________________
'91 Toyota Pickup
'09 Kawasaki KLR650
'12 Gunnar Rockhound 29

"They say the test of literary power is whether a man can write an inscription. I say, 'Can he name a kitten?'" -- Samuel Butler
Reply With Quote
  #9  
Old 09-25-2012, 05:59 PM
DaveInDenver's Avatar
DaveInDenver DaveInDenver is offline
Hard Core 4+
 
Join Date: Jun 2006
Location: Larimer County
Posts: 6,475
Default

BTW, Witch Hunter can clean and service (to some extent) injectors, had good luck using them. RC Engineering is supposedly also fine.
__________________
'91 Toyota Pickup
'09 Kawasaki KLR650
'12 Gunnar Rockhound 29

"They say the test of literary power is whether a man can write an inscription. I say, 'Can he name a kitten?'" -- Samuel Butler
Reply With Quote
  #10  
Old 09-26-2012, 04:35 PM
Rzeppa's Avatar
Rzeppa Rzeppa is offline
Cruise Moab Committee
 
Join Date: Aug 2005
Location: Kittredge CO, USA
Posts: 5,136
Default

OK, she brought the car over this afternoon and we verified open (or at least >200 ohms) at the injector terminals. New injector was $136.19 at NAPA, in-stock on the shelf. Spec was 10-18 ohms, new injector measured 12 ohms, and had new o-rings already on it. No gaskets are present.

R&R was super quick and easy compared to the 22RE injectors. There is a cast alloy retainer over the top secured with two phillips head screws. Then all you have to do is pull the injector straight out with a little bit of twist, the only thing holding it in at this point is the o-rings.

There were some hoses and stuff in the way but nothing that had to be completely removed, just held aside. The whole operation was like 20 minutes, would have been ten if I had ever done it before, was working slowly and carefully so as not to bugger anything up.

She is on her way to get re-inspected. The CEL didn't come back on after I cleared the code. I'll report back on the results.

Thanks again for all the helpful suggestions.

__________________
Jeff Zepp
Kittredge CO USA
1971 & 1976 FJ40s, 1978 FJ45, 1987 FJ60
Rising Sun 4WD Club, TLCA #4063
http://american3dprinting.net/
Reply With Quote
Reply

Thread Tools

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


All times are GMT -6. The time now is 01:59 AM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.