Ready to set it on fire and roll it down a hill.
Thread Starter
|
Starting Member
Joined: Mar 2020
Posts: 107
Ready to set it on fire and roll it down a hill.
ok.. short background.. bought this '94 s10 blazer (W code 4.3) for $400 not running.. turned out to be a broken power wire to the ecm.
ended up to replacing the entire fuel system. (tank, sender, pump, lines, filter) and started driving it.. one day it started running REALLY rough.. barely made it home. code 45, exh rich. so I put a new spider in it. (couldn't locate the plastic fuel lines at the time)
it ran like a swiss watch.. i drove it for 10 mintues to lunch.. had lunch. came back to a crazy hard start.. and running like CRAP. ( no SES)
threw some new plugs and wires at it just because. .(it's a '94, with 98k miles, no idea when it last had plugs and wires) the cap and rotor looked fine. still missing. i looked at the ignition wave forms with my oscilliscope and inductive pick up leads.. saw some issues with cylinders 3,4 and 5.. so I went ahead and put a new cap and rotor based on advice from the ign wave forms. waveforms look perfect.. still a hard miss at idle. so I decided to go ahead and replace the plastic lines in the manifold just to be 100%
still hard miss.
just for grins and giggles, decided to look at the tech1 and engine data again.. noticed Short term fuel trim was at 80, Long term at 109. (the leanest they will go) 02s was reading 1100mv w/o moving. i reset the long term fuel trim and it ran a lot better.. until it pulled it back to 109 again.
shut engine off, reset both fuel trims.. it ran well for a few minutes unitl it went closed loop and pulled the trims back again.
so i followed the diagnostics in the gm manual for a code 45 (even tho it wasn't setting one, which I believe was from it not reaching 190°F engine temp) unplugged the o2s, tech one showed 1096mv. grounded ck412 per the manual, tech1 shows 4mv. so I went ahead and replaced the o2s. (a miserable job), reset the fuel trims, and now the o2s is moving up and down like it should.. but the ECM is still trying to yank all the fuel out.
took it for a hop around the block.. and managed to get into some of the WOT learn blocks... it ran FANTASTIC the first time.. but then it pull the fuel out of those blocks and it just ran awful.
for some reason, this ecm is convinced it's running filthy rich.. yet it doesn't appear to be...
as a side effect of changing those plastic fuel lines, I removed the lower intake manifold.. put brand new felpro gaskets in.
timing set to 0° with a light..
i'm lost why it keeps pulling fuel trims back so far.... now this time i noticed it pulled long term all the back to 109, then the short term floated back up to 128.
at this point.. i can only see 2 possibilities.. the brand new spider is deflective or the ecm has issues...
i'm open to questions/comments/advice.. .because I'm about ready to set it on fire and roll it down a hill.. and trade it for a coupe deville!
reset the fuel trims.. it ran well until it pulled the fuel trims back again. so I
ended up to replacing the entire fuel system. (tank, sender, pump, lines, filter) and started driving it.. one day it started running REALLY rough.. barely made it home. code 45, exh rich. so I put a new spider in it. (couldn't locate the plastic fuel lines at the time)
it ran like a swiss watch.. i drove it for 10 mintues to lunch.. had lunch. came back to a crazy hard start.. and running like CRAP. ( no SES)
threw some new plugs and wires at it just because. .(it's a '94, with 98k miles, no idea when it last had plugs and wires) the cap and rotor looked fine. still missing. i looked at the ignition wave forms with my oscilliscope and inductive pick up leads.. saw some issues with cylinders 3,4 and 5.. so I went ahead and put a new cap and rotor based on advice from the ign wave forms. waveforms look perfect.. still a hard miss at idle. so I decided to go ahead and replace the plastic lines in the manifold just to be 100%
still hard miss.
just for grins and giggles, decided to look at the tech1 and engine data again.. noticed Short term fuel trim was at 80, Long term at 109. (the leanest they will go) 02s was reading 1100mv w/o moving. i reset the long term fuel trim and it ran a lot better.. until it pulled it back to 109 again.
shut engine off, reset both fuel trims.. it ran well for a few minutes unitl it went closed loop and pulled the trims back again.
so i followed the diagnostics in the gm manual for a code 45 (even tho it wasn't setting one, which I believe was from it not reaching 190°F engine temp) unplugged the o2s, tech one showed 1096mv. grounded ck412 per the manual, tech1 shows 4mv. so I went ahead and replaced the o2s. (a miserable job), reset the fuel trims, and now the o2s is moving up and down like it should.. but the ECM is still trying to yank all the fuel out.
took it for a hop around the block.. and managed to get into some of the WOT learn blocks... it ran FANTASTIC the first time.. but then it pull the fuel out of those blocks and it just ran awful.
for some reason, this ecm is convinced it's running filthy rich.. yet it doesn't appear to be...
as a side effect of changing those plastic fuel lines, I removed the lower intake manifold.. put brand new felpro gaskets in.
timing set to 0° with a light..
i'm lost why it keeps pulling fuel trims back so far.... now this time i noticed it pulled long term all the back to 109, then the short term floated back up to 128.
at this point.. i can only see 2 possibilities.. the brand new spider is deflective or the ecm has issues...
i'm open to questions/comments/advice.. .because I'm about ready to set it on fire and roll it down a hill.. and trade it for a coupe deville!
reset the fuel trims.. it ran well until it pulled the fuel trims back again. so I
Beginning Member
Joined: Jul 2021
Posts: 20
When you say brand new spider.. it most likely is a reman assembly right? Those are pretty rare to find brand new. Hows the fuel pressure when priming and running? Any leaks? Drop in pressure when holding? You probably reset ecm after all of the work you did right?
Thread Starter
|
Starting Member
Joined: Mar 2020
Posts: 107
I did pressurize the system with the upper plenum off the check for leaks, none were found.
i have reset the ecm several times. even left the battery disconnected for 20 minutes just be sure.
BF Veteran
Joined: Jan 2007
Posts: 4,625
From: Florida
An article on fuel trims. I wrote this in ODBII speak. For reference in your case, 102 = -20% and 154 = +20%:
https://blazerforum.com/forum/2nd-ge...-trims-102906/
Always add ST and LT together for a given bank, more than 10% is marginal, more than 25% is bad and usually sets a code. Your values show that the ECM is commanding lean for a PERCEIVED rich condition IF the ECM were to follow the other sensor values and use the ECM fuel cell look up table. IOW, the ECM looks at all the other (non O2) sensor values and calculates the theoretical fuel delivery value and then compares that to the actual fuel delivery value which is being commanded by the O2 sensors and the ECM, assuming closed loop operation. If your trims are excessively large all you know is that there is a big difference between the theoretical calculation and the actual delivery, you don't know what's wrong yet. It could be a wonky sensor causing in improper theoretical calculation. The O2 sensors are always holding the A/F mixture as close to 14:1 as possible, unless they are bad or their wiring connectors are bad or the bias voltage is wrong or the ECM is broken. There also can be an actual functional problem like ignition or fuel pressure or compression, etc. If your ECM and O2 sensors are working then the truck is not running rich but the trims are telling you that it would be if the ECM followed the other sensors and look up table so something is wrong. If the O2 sensors are pegged high or low and they are not defective then that's telling you that the system cannot command rich or lean enough to compensate for whatever the problem is. If they are switching properly and the ECM is OK then the truck is running at stoichiometry (14:1) even with excessive trim values.
Is there a difference in how it runs before and after the transition to closed loop?
George
https://blazerforum.com/forum/2nd-ge...-trims-102906/
Always add ST and LT together for a given bank, more than 10% is marginal, more than 25% is bad and usually sets a code. Your values show that the ECM is commanding lean for a PERCEIVED rich condition IF the ECM were to follow the other sensor values and use the ECM fuel cell look up table. IOW, the ECM looks at all the other (non O2) sensor values and calculates the theoretical fuel delivery value and then compares that to the actual fuel delivery value which is being commanded by the O2 sensors and the ECM, assuming closed loop operation. If your trims are excessively large all you know is that there is a big difference between the theoretical calculation and the actual delivery, you don't know what's wrong yet. It could be a wonky sensor causing in improper theoretical calculation. The O2 sensors are always holding the A/F mixture as close to 14:1 as possible, unless they are bad or their wiring connectors are bad or the bias voltage is wrong or the ECM is broken. There also can be an actual functional problem like ignition or fuel pressure or compression, etc. If your ECM and O2 sensors are working then the truck is not running rich but the trims are telling you that it would be if the ECM followed the other sensors and look up table so something is wrong. If the O2 sensors are pegged high or low and they are not defective then that's telling you that the system cannot command rich or lean enough to compensate for whatever the problem is. If they are switching properly and the ECM is OK then the truck is running at stoichiometry (14:1) even with excessive trim values.
Is there a difference in how it runs before and after the transition to closed loop?
George
Thread Starter
|
Starting Member
Joined: Mar 2020
Posts: 107
An article on fuel trims. I wrote this in ODBII speak. For reference in your case, 102 = -20% and 154 = +20%:
https://blazerforum.com/forum/2nd-ge...-trims-102906/
Is there a difference in how it runs before and after the transition to closed loop?
George
https://blazerforum.com/forum/2nd-ge...-trims-102906/
Is there a difference in how it runs before and after the transition to closed loop?
George
I don't know if it makes a diff.. but my blazer is a 1994 OBD-1 model. I'm not sure how much help that 95+ article will be...
BF Veteran
Joined: Jan 2007
Posts: 4,625
From: Florida
Yes I saw that you have an ODBI vehicle and that’s why I made reference to the translation values between counts and percentages. All of the concepts are still the same. The fact that the truck runs good in closed loop says you’re able to deliver proper fuel and spark, most likely. When you transition to close loop fuel management is handed over to the O2 sensors so this is likely where your problem is. Do you have the ability to graph your O2 sensor output? If there are no scanners that can graph the O2 function and you have to do it with a discreet scope it would be best to catch it up at the PCM input. Also have you checked for exhaust leaks?
George
George
BF Veteran
Joined: Jan 2007
Posts: 4,625
From: Florida
Some thoughts on your problem.
Your ODB system has one main job and that is minimum emissions by maintaining a 14:1 air/fuel ration and it does that by measuring the exhaust content with the O2 sensors and adjusting the injector pulse width, when in closed loop operation of course. The O2 sensors have no idea if there are any functional problems elsewhere and have one blind objective - stoichiometry. The point of fuel trims is to tell us if there are functional problem such that the fuel delivery that the O2 sensors are requesting is different from the fuel delivery as indicated by all the other sensors in the vehicle. If that difference is large, something is wrong because the O2 sensors should be calculating minor tweaks from the open loop values. We don’t know what’s wrong yet but the possibilities are:
The O2 sensors themselves have failed
The O2 sensor wiring or connectors have failed
The ECM has failed
Another sensor or component or its wiring has failed
There is some other functional/mechanical problem
The first question to be answered is are the O2 sensors properly reporting and the ECM is doing its job interpreting fuel delivery for stoich which implies that the O2 sensors are fighting some other problem or is it really a closed loop problem?
One thing to keep in mind is that the truck has to warm up to go closed loop so the fact that it’s running crappy when warm may be the transition to closed loop and the O2 sensor circuit or it may be something else warming up like an ICM or ignition coil for instance. That would produce the same result. Ignition problem, unburned fuel, exhaust goes rich, O2 sensors go lean to compensate in a failed attempt to get the truck running perfect and you have a misfiring engine running way too lean in the cylinders that are firing.
So the first thing is are the O2 sensors and their heaters running OK, reporting OK and the ECM is reading OK which includes the sensors, bias voltage, wiring and connectors. If yes then we need to find out why the truck is running rich when it’s warm. If no then we fix the O2 sensor circuit.
Stated another way, all we know now is that the closed loop system is compensating for a perceived rich condition and this happens when the vehicle warms up. We don't know yet if the truck is running rich or if the closed loop system is faulty.
Question: When you replaced the ECM power wire, how did it run? Why did you then "replace the entire fuel system"?
George
Your ODB system has one main job and that is minimum emissions by maintaining a 14:1 air/fuel ration and it does that by measuring the exhaust content with the O2 sensors and adjusting the injector pulse width, when in closed loop operation of course. The O2 sensors have no idea if there are any functional problems elsewhere and have one blind objective - stoichiometry. The point of fuel trims is to tell us if there are functional problem such that the fuel delivery that the O2 sensors are requesting is different from the fuel delivery as indicated by all the other sensors in the vehicle. If that difference is large, something is wrong because the O2 sensors should be calculating minor tweaks from the open loop values. We don’t know what’s wrong yet but the possibilities are:
The O2 sensors themselves have failed
The O2 sensor wiring or connectors have failed
The ECM has failed
Another sensor or component or its wiring has failed
There is some other functional/mechanical problem
The first question to be answered is are the O2 sensors properly reporting and the ECM is doing its job interpreting fuel delivery for stoich which implies that the O2 sensors are fighting some other problem or is it really a closed loop problem?
One thing to keep in mind is that the truck has to warm up to go closed loop so the fact that it’s running crappy when warm may be the transition to closed loop and the O2 sensor circuit or it may be something else warming up like an ICM or ignition coil for instance. That would produce the same result. Ignition problem, unburned fuel, exhaust goes rich, O2 sensors go lean to compensate in a failed attempt to get the truck running perfect and you have a misfiring engine running way too lean in the cylinders that are firing.
So the first thing is are the O2 sensors and their heaters running OK, reporting OK and the ECM is reading OK which includes the sensors, bias voltage, wiring and connectors. If yes then we need to find out why the truck is running rich when it’s warm. If no then we fix the O2 sensor circuit.
Stated another way, all we know now is that the closed loop system is compensating for a perceived rich condition and this happens when the vehicle warms up. We don't know yet if the truck is running rich or if the closed loop system is faulty.
Question: When you replaced the ECM power wire, how did it run? Why did you then "replace the entire fuel system"?
George
Last edited by GeorgeLG; Aug 1, 2021 at 11:19 AM.
Thread Starter
|
Starting Member
Joined: Mar 2020
Posts: 107
Some thoughts on your problem.
So the first thing is are the O2 sensors and their heaters running OK, reporting OK and the ECM is reading OK which includes the sensors, bias voltage, wiring and connectors. If yes then we need to find out why the truck is running rich when it’s warm. If no then we fix the O2 sensor circuit.
So the first thing is are the O2 sensors and their heaters running OK, reporting OK and the ECM is reading OK which includes the sensors, bias voltage, wiring and connectors. If yes then we need to find out why the truck is running rich when it’s warm. If no then we fix the O2 sensor circuit.
Stated another way, all we know now is that the closed loop system is compensating for a perceived rich condition and this happens when the vehicle warms up. We don't know yet if the truck is running rich or if the closed loop system is faulty.
Question: When you replaced the ECM power wire, how did it run? Why did you then "replace the entire fuel system"?
George
George
the fuel system... the blazer was hot-dogged on freshly chlorided dirt roads before it died..and then was parked in a grass field for about 3 years. the chloride rotted everything underneath. the return line was leaking fuel at the tank. due to the condition of everything, I opted to replace both the fuel lines and fuel sender. and since I was doing that, i went ahead and put a brand new Bosch fuel pump in it too. while I had things out, I discovered rust holes in the gas tank just above the 1/2 tank mark, so it got a new fuel tank too. then I replaced the Spider, and plastic fuel lines.. a new filter was part of the new lines as well.. so the only part of the fuel system I haven't replaced, is the steel lines running up the back of the block.
it also got new lower intake gaskets as I ended up removing the lower intake plenum to get the plastic fuel lines out
I have an antique data logger that was originally used for engine development by my father who was a GM powertrain engineer .. Dear God I miss that man. (RIP) I could really use his help on this. so I figure I'll configure it to read the o2 voltage every 0.5 second and log it.. then I can make a graph of whats going on.
I do have a digital storage o'scope.. but I don't think the resolution would be very good with the time base set to several seconds...
I'm also going to put a fuel pressure gauge on on it today and see what's going on there.. just in case the reman spider has issues. (assuming the o2 is right and it IS rich.. I suspect the reman spider)
exhaust leaks...there are no leaks upstream of the converter.. but the exh pipe is rotted through at the back of the converter... so I don't THINK that's affecting anything at the minute...
My apologies for being MIA on this... it's been too dang hot. I don't work outside when it's 90+ and humid A.F!
Last edited by kc8oye; Sep 4, 2021 at 10:38 AM.
Thread Starter
|
Starting Member
Joined: Mar 2020
Posts: 107
Just a note... that broken wire to the ecm... the blazer was originally purchased at a buy here/pay here place for people with bad credit... it had a remote kill switch installed while there was a lean on it. it was removed when it was paid off, and I believe that's where the power wire was broken... it wasn't spliced back together correctly. I just ran a jumper over to the fuse block.
BF Veteran
Joined: Jan 2007
Posts: 4,625
From: Florida
A few things.
If there is no stored history in the ECM for fuel trims then the moment the vehicle goes closed loop no automatic correction is applied (long term trims or block learn for your ODBI system) and as soon as the system sees a perceived rich condition it starts calculating the new Injector pulse width values to try and keep the exhaust at stoich. As that moment all the correction is in short term trims (integrator value for you) and no correction in long term trims are calculated yet. As time goes on if the ECM sees a consistent need for this correction then it starts storing it in long term trim values to be applied automatically at the start of closed loop operation. When this happens the short term correction Drifts towards zero (128 counts) and now you have a semi permanent fuel delivery correction always applied in LT trims and short term corrections are once again showing you any small fluctuations in engine performance as they drift around zero. You always add ST and LT trims together for a given bank for the net total correction being applied at any given time. Every time you reset the trims the ECM has to go back through this process: ST immediately changes, LT drifts toward that value while ST drifts towards zero correction or 128 counts for you. If you correct the problem and do not reset the trims then ST immediately swings to offset the now unnecessary LT correction, LT slowly starts drifting towards no correction and ST drifts in unison until the ECM learns the new situation.
The fact that it runs good in open loop (I mixed up closed/open loop in a previous post) and runs like crap in closed loop implies that something is still wrong with the ECM/O2 circuit. I just noticed that when you unplugged the sensor you measured a bias voltage of 1.1V. This should be 0.45V, or at least it should be in an ODBII GM vehicle of our vintage. You seem to possess the manuals for your vehicle, what is the proper bias voltage for your O2 sensor? You spoke about data logging and a digital scope. What does Your upstream O2 sensor waveform look like in closed loop? At one point you said now it’s moving up and down as it should but you also say that the trims are as lean as they can go which would imply an O2 output to the ECM around 1V which is roughly your open circuit bias voltage.
Testing fuel pressure is a good idea but since the system thinks your running rich that is unlikely your problem.
George
If there is no stored history in the ECM for fuel trims then the moment the vehicle goes closed loop no automatic correction is applied (long term trims or block learn for your ODBI system) and as soon as the system sees a perceived rich condition it starts calculating the new Injector pulse width values to try and keep the exhaust at stoich. As that moment all the correction is in short term trims (integrator value for you) and no correction in long term trims are calculated yet. As time goes on if the ECM sees a consistent need for this correction then it starts storing it in long term trim values to be applied automatically at the start of closed loop operation. When this happens the short term correction Drifts towards zero (128 counts) and now you have a semi permanent fuel delivery correction always applied in LT trims and short term corrections are once again showing you any small fluctuations in engine performance as they drift around zero. You always add ST and LT trims together for a given bank for the net total correction being applied at any given time. Every time you reset the trims the ECM has to go back through this process: ST immediately changes, LT drifts toward that value while ST drifts towards zero correction or 128 counts for you. If you correct the problem and do not reset the trims then ST immediately swings to offset the now unnecessary LT correction, LT slowly starts drifting towards no correction and ST drifts in unison until the ECM learns the new situation.
The fact that it runs good in open loop (I mixed up closed/open loop in a previous post) and runs like crap in closed loop implies that something is still wrong with the ECM/O2 circuit. I just noticed that when you unplugged the sensor you measured a bias voltage of 1.1V. This should be 0.45V, or at least it should be in an ODBII GM vehicle of our vintage. You seem to possess the manuals for your vehicle, what is the proper bias voltage for your O2 sensor? You spoke about data logging and a digital scope. What does Your upstream O2 sensor waveform look like in closed loop? At one point you said now it’s moving up and down as it should but you also say that the trims are as lean as they can go which would imply an O2 output to the ECM around 1V which is roughly your open circuit bias voltage.
Testing fuel pressure is a good idea but since the system thinks your running rich that is unlikely your problem.
George
