O2 insanity...
04-12-2011, 08:14 AM
#1
O2 insanity... I thought I'd throw this out to the community for opinions, feedback, etc. I'm having trouble getting a working pair of sensors for my 94 A4 project car. I first tried a pair of the new Delphi ES20317, then a pair of AFS75, which didn't look right, like they'd redesigned it. Here's my results:
Delphi ES20317: read 4mV @ KOEO, both very low, single to double digits only in mV, doesn't throw low voltage codes, but BLMs go straight to 160 in about 1 min. of runtime.
AC Delco AFS75: Thought these would be my magic bullet, but results are nearly the same as the Delphis. Consistently very low like the Delphis, though start at about 250mV @ KOEO. BLMs go straight to 160 as the Delphis do. Right side set a low voltage code. These don't look right. They look like a SMP sensor, though still sealed in their bags with part # on it. Maybe they revised the design and... (?) WTF?
***Here's the kicker*** of the 3 original AFS65s I had on hand, I have one good one that works perfectly on either bank I install it on. It reads consistently in the 4-500s at idle, goes to 900 @ WOT, BLM on either bank I put it on stays consistently at 126-128. I found a used one for $15.00 on FleaBay and it'll be here in a couple of days, for comparitive purposes. If it works as well as the other one, I'm using it!
Please tell me I'm not crazy.
Delphi ES20317: read 4mV @ KOEO, both very low, single to double digits only in mV, doesn't throw low voltage codes, but BLMs go straight to 160 in about 1 min. of runtime.
AC Delco AFS75: Thought these would be my magic bullet, but results are nearly the same as the Delphis. Consistently very low like the Delphis, though start at about 250mV @ KOEO. BLMs go straight to 160 as the Delphis do. Right side set a low voltage code. These don't look right. They look like a SMP sensor, though still sealed in their bags with part # on it. Maybe they revised the design and... (?) WTF?
***Here's the kicker*** of the 3 original AFS65s I had on hand, I have one good one that works perfectly on either bank I install it on. It reads consistently in the 4-500s at idle, goes to 900 @ WOT, BLM on either bank I put it on stays consistently at 126-128. I found a used one for $15.00 on FleaBay and it'll be here in a couple of days, for comparitive purposes. If it works as well as the other one, I'm using it!
Please tell me I'm not crazy.
04-12-2011, 12:02 PM
#4
TECH Resident
iTrader: (21)
Join Date: Aug 2006
Location: Wichita KS
Posts: 947
Likes: 0
Received 0 Likes
on
0 Posts
Once you figure this out please post, i havent logged my car but i keep getting o2 sensor codes on mine and I need to replace them. Thankfully I dont drive the car much at all and ill keep an eye on this thread.
04-12-2011, 12:20 PM
#5
You running OBD2 with rear O2 simulators, or actual rear O2s? Our cars have the same front O2s, I just don't have rears (OBD1). Which ones are showing up as bad for you? Are they low voltage codes? Which O2s are you using for front/ rear?
04-12-2011, 02:30 PM
#6
TECH Resident
iTrader: (21)
Join Date: Aug 2006
Location: Wichita KS
Posts: 947
Likes: 0
Received 0 Likes
on
0 Posts
04-12-2011, 02:40 PM
#7
Making a list of all the O2s that I can find that list for front position on 94-97 cars:
NTK (NGK) 21006
Bosch 15703
Denso 234-4012, 243-4000
Delphi 20317
AC Delco AFS75
BWD OS1008
I've read others on the forum saying they've had good luck with the NTKs. Have heard mixed talk about the Bosch, but mostly negative. Not a lot of options. It would suck if they've revised the design of the AFS75. If I didn't have a perfectly working 16 year old AFS65, I'd say I still have mechanical issues. Damned odd.
NTK (NGK) 21006
Bosch 15703
Denso 234-4012, 243-4000
Delphi 20317
AC Delco AFS75
BWD OS1008
I've read others on the forum saying they've had good luck with the NTKs. Have heard mixed talk about the Bosch, but mostly negative. Not a lot of options. It would suck if they've revised the design of the AFS75. If I didn't have a perfectly working 16 year old AFS65, I'd say I still have mechanical issues. Damned odd.
Trending Topics
04-12-2011, 09:40 PM
#8
TECH Fanatic
iTrader: (26)
Join Date: Mar 2010
Location: Milledgeville, GA
Posts: 1,909
Likes: 0
Received 0 Likes
on
0 Posts
I'm running the Bosch 15703's. I bought them BEFORE I read all the bad reviews about them on here. LOL So far, I've had good service from them. About 6,000 miles on them now I'd guess. I did have to replace one, but it was fouled by a failed Aeromotive AFPR. I could have tried cleaning it first, but it was under warranty, so I just had it swapped out.
04-13-2011, 01:34 PM
#10
Since it's stock for the moment, I'm not in a rush to go to an open loop tune. I'd at least want a wideband to monitor AFR if I went nuts...
And... reading up on O2 sensors, I may be rushing to judgement. I might still just need to eliminate any source of stray air they might see, (like tight left side valves) since they will respond to that much faster than an old sensor. Keeping the BLMs down so the O2s can hit closed loop/ several cold > warm > cold cycles and get normal behavior is the hard part at the moment. The AFS75s, when heated up and reading correctly in closed loop should go from as low as 50mV to 950mV rapidly... if I understand correctly. Though it looks and feels more comfortable to see them between 350-550mV, near a stoichiometrically perfect value of 450mV.
Higher than expected/desired IAC counts (still between 70-110 @ startup) make me think about looking for a restriction somewhere, etc. Maybe I need to crack the throttle blades 1-2 RCH. I had the luxury of carefully assembling the engine outside the car, so I was meticulous about everything. And the one older sensor is very happy on the right bank right now with a BLM of 128. Maybe it's actually slow and just looks good, though its voltage ranges look right- from about 300 to 900mV.
So before I do anything else I'm going to go back and adjust the left bank's valves with 3 flats past zero lash like I did on the right bank. I just want to make sure I'm eliminating all possible mechanical sources of errant air flow before messing around randomly.
Also looking at gm./sec readings for MAF, baro values for MAP to see if I can spot anything else that might yield a clue. The car is at 394 ft above sea level, so I should see 29.513 "Hg / 99.5kPa / 4.85V MAP readings. MAF should be around 8gps at correct stock idle speed of 650, maybe as low as 6 if I have a vacuum leak, as high as 9-10 at 750RPM before it warms up and tries to enter closed loop and drop to 650.
So, lots of thinking, taking baby steps to find the answers.
And... reading up on O2 sensors, I may be rushing to judgement. I might still just need to eliminate any source of stray air they might see, (like tight left side valves) since they will respond to that much faster than an old sensor. Keeping the BLMs down so the O2s can hit closed loop/ several cold > warm > cold cycles and get normal behavior is the hard part at the moment. The AFS75s, when heated up and reading correctly in closed loop should go from as low as 50mV to 950mV rapidly... if I understand correctly. Though it looks and feels more comfortable to see them between 350-550mV, near a stoichiometrically perfect value of 450mV.
Higher than expected/desired IAC counts (still between 70-110 @ startup) make me think about looking for a restriction somewhere, etc. Maybe I need to crack the throttle blades 1-2 RCH. I had the luxury of carefully assembling the engine outside the car, so I was meticulous about everything. And the one older sensor is very happy on the right bank right now with a BLM of 128. Maybe it's actually slow and just looks good, though its voltage ranges look right- from about 300 to 900mV.
So before I do anything else I'm going to go back and adjust the left bank's valves with 3 flats past zero lash like I did on the right bank. I just want to make sure I'm eliminating all possible mechanical sources of errant air flow before messing around randomly.
Also looking at gm./sec readings for MAF, baro values for MAP to see if I can spot anything else that might yield a clue. The car is at 394 ft above sea level, so I should see 29.513 "Hg / 99.5kPa / 4.85V MAP readings. MAF should be around 8gps at correct stock idle speed of 650, maybe as low as 6 if I have a vacuum leak, as high as 9-10 at 750RPM before it warms up and tries to enter closed loop and drop to 650.
So, lots of thinking, taking baby steps to find the answers.
04-13-2011, 03:01 PM
#11
TECH Fanatic
iTrader: (2)
Join Date: Dec 2010
Location: Oklahoma
Posts: 1,084
Likes: 0
Received 0 Likes
on
0 Posts
Deleting the o2's is the dumbest idea there is....(unless its a track only car with no closed loop functions)
the Bosch are peices of junk... well at least with headers. With headers the exhaust can't keep the o2 hot enough... bosch are very picky about running o2 temps..
I've had best luck with NTK's...
and about the whole you should have ,such and such, vacuum/gps at idle/650 rpms.... this just isn't always the case. Increasing the idle timing can give you more vacuum and less gps. same with A/F leaning the mixture out(to a point) has the same effect.
get some good NTK's and go from there..
the Bosch are peices of junk... well at least with headers. With headers the exhaust can't keep the o2 hot enough... bosch are very picky about running o2 temps..
I've had best luck with NTK's...
and about the whole you should have ,such and such, vacuum/gps at idle/650 rpms.... this just isn't always the case. Increasing the idle timing can give you more vacuum and less gps. same with A/F leaning the mixture out(to a point) has the same effect.
get some good NTK's and go from there..
04-13-2011, 11:48 PM
#13
Good thing I checked the valve train on bank 1. both studs on #1 had started to back out but they hadn't started to make noise yet. Dodged a bullet... big one. So I pulled all the rocker studs and put red loc-tite on them and torqued to 51 ft lbs. then reassembled. The spare used AFS65 O2 I ordered off of EBay came today. I plugged it in and everything was near perfect, both banks at 128-130 in CEL 16, 17, 18. IAC was down to 34.
As the new-old O2 heated up, if became apparent that it wasn't going to go above 380mV, regardless of throttle. So I still need a good pair of O2s. Maybe a coincidence that I got a bad AFS75 right out of the box (low voltage, bank 2 the other day, stayed around 4mV-50mV while the other bank at least moved around a bit).
So... 90% solved, still the slightest hint of stumble on tip-in but that probably can be chalked up to fuel trim moving from CEL 16-18 due to the slow/low mV on the new-old O2. Now sounds and feels like it has POWER! Upgraded my phone to one with better video last week, but the sound still cracks pretty bad when I rev it.
http://www.youtube.com/watch?v=kXH6kAY9Lyg
As the new-old O2 heated up, if became apparent that it wasn't going to go above 380mV, regardless of throttle. So I still need a good pair of O2s. Maybe a coincidence that I got a bad AFS75 right out of the box (low voltage, bank 2 the other day, stayed around 4mV-50mV while the other bank at least moved around a bit).
So... 90% solved, still the slightest hint of stumble on tip-in but that probably can be chalked up to fuel trim moving from CEL 16-18 due to the slow/low mV on the new-old O2. Now sounds and feels like it has POWER! Upgraded my phone to one with better video last week, but the sound still cracks pretty bad when I rev it.
http://www.youtube.com/watch?v=kXH6kAY9Lyg
05-12-2011, 10:51 AM
#18
I did hang onto the AFS75s. One is bad out of the box. Have to return it under warranty to Advance. Using a pair of 65s, including the old one that doesn't go over 380mV while I burn off the coating on the headers. Hate to ruin a good pair of high dollar new ones...
So- AFS 75s, though they look like they altered the design, work ok. The Delphi 10009s on my green car bounce comfortably from 400mV to about 750mV in closed loop. Wish I had another pair of those, but they don't make 'em anymore.
So- AFS 75s, though they look like they altered the design, work ok. The Delphi 10009s on my green car bounce comfortably from 400mV to about 750mV in closed loop. Wish I had another pair of those, but they don't make 'em anymore.
