Resetting the computer: Myth or reality?

[Gary Allan]

I'll accept your explanation ..but would offer that if they were identical ..and functioned in an identical manner and were of identical construction ...what would possibly motivate an assembly engineer to go to the extreme expense of getting different connectors when it would reason that an assembly tech would just grab ONE identical sensor and put it in and connect it

They don't make harnesses long enough for the assembly tech to mistakenly plug the wrong connector on the desired sensor

There would be no advantage or sensible benefit to doing this ..either in production, assembly, or in after sale service. The exception being that you could recover the added assembly expense in after sale OEM parts sales ...which on O2 sensors, would be some very slim pickin's for the time frame of recurring sales.

..but I won't argue the point..you appear to speak with authority here. There are plenty of things that the manufacturers do that makes it appear that they were exposed to Pb in their childhood.

 

[jmacmaster]

To get back to the original question, he said:

"I changed out the 02 sensors . . . I wonder if I should've left the battery disconnected longer to 'reset' the computer. I've heard some say this is a myth, and then other folks swear you need to do this. Which is true? Does the computer need to be 'reset' or will it do this on it's own after I put the O2 sensors in?"

He's not asking about resetting the computer, by disconnecting the battery, to clear the codes. He's talking about the commonly held misbelief that after replacing certain parts related to things monitored by the computer, or doing certain modifications on something that's monitored by the computer (such as installing new spark plugs or an air intake kit), you HAVE to disconnect the battery and thereby reset the computer. In my opinion, this is not true. The standards for certain things monitored by the computer are set at the factory for the average driver and for average driving conditions. As any given driver drives the vehicle in any given set of conditions, the computer is constantly monitoring and changing certain settings to conform to that person's driving style, the weather conditions, the altitude, and other things. In today's newer vehicles this is done quite rapidly by the computer (much faster than the computers in older vehicles used to do it). There is no need to disconnect the battery to allow the computer to reset itself. The computer only needs to reset itself (actually, the correct words are "adjust itself") with respect to the part you replaced or modification you did, and it does so all by itself, and quite quickly, as you drive. If you disconnect the battery to reset the computer, its counterproductive, because the computer resets many settings for many variables to the factory settings for the average driver and driving conditions, and with respect to those settings, the computer has to then relearn and set a lot of them in response to your driving style and the driving conditions, even tho the settings are for things that have nothing to do with the part that you replaced or the modification that you did.

 

[Gary Allan]

jmacmaster

I pretty much agree with what you say here. In closed loop ..the PCM is using all sensors and constantly adjusting the fuel mix. In open loop, however, the system is designed to compensate for long term system degradation. The only system that I can think of (there can surely be others) is the 88-93 MAF system on some 5.0 Fords. I am led to believe that this was a complex 100% 24/7 self tuning FI system. Great for cams and whatnot that send other systems into distress. Good for managing up to 7 liters.

In most numb-dumb speed density systems ..in open loop..the PCM monitors the O2 sensor output. It doesn't adjust anything ...just records the rich and lean conditions. Too many leans ..it broadens the pulse width. Too many rich readings ..it narrows the pulse width. It also calculates the short term trim based on the long term trims.


If, however, you've had a lazy O2 sensor that has caused your adaptive state to evolve to very high ...levels of compensation (broadening or narrowing) ..the increment or decrement process can take some time to return to normal. This occured on my OBDI system. The O2 read bad ..it was replaced ..only to have a new code appear stating that it was always in a high voltage state since it was in open loop enough to trigger the code.

This condition, btw, wasn't remedied by just the battery disconnect technique. It needed a more in depth scan tool to reset the adaptive cells.

Admittedly, this would be an exceptional situation ..just adding it FWIW

 

[brianl703]

My guess is that the reason the downstream and the upstream sensors have different connectors, if not to prevent mixups during the assembly or repair process, is so that someone can't (easily) do what was suggested in this thread, which is to swap the upstream and downstream sensors.

Why would they want to prevent that? Here's my theory:

On Ford vehicles at least, the catalyst efficiency is determined by the ratio of upstream sensor switches to downstream sensor switches. The slower the downstream sensor switches and the faster the upstream sensor switches, the higher the catalyst efficiency is calculated to be.

As we know, a sensor starts degrading--switching slower, responding slower--as it ages. A sensor in the upstream position degrades a lot faster than the one in the downstream position, according to what I've read.

If a degraded upstream sensor and a not-so-degraded downstream sensor were switched, in theory the calculated catalyst efficiency threshold would be higher than it really is. Perhaps this might be enough to get rid of a P0420 or P0430 code.

It's interesting to note that one method of permanently getting rid of a P0420 or P0430 code is to install a resistor-capacitor combination which functions as a low-pass filter between the downstream sensor and the computer. This causes the computer to see a slower switch rate than what the sensor is actually sending.

 

[brianl703]

As far as open loop and closed loop, this is what I know to be true of Ford EEC-IV and EEC-V vehicles:

In closed loop the computer is using output from the oxygen sensors to adjust the long and short term fuel trims. The long and short term fuel trim tables apparently take into account engine speed and engine load, so they are a 3-dimensional array: Load, RPM, and fuel trim. The fuel trim is a correction applied to the fuel injector pulsewidth which has been calculated from the MAP or MAF sensor reading and the coolant temperature sensor reading.

In open-loop the computer is not using input from the oxygen sensors BUT it is still using the fuel trim that it learned during closed-loop. If any data is missing, I believe it can extrapolate the fuel trim from the data points it does have to build the 3-dimensional array of load, RPM, and fuel trim.

With OBD-II, the check engine light is supposed to come on once the long-term fuel trim reaches plus or minus 25%, although my 1996 Contour had an idle fuel trim of +24% (went down as engine load went up) and it was running fine. It turned out that the hose between the PCV valve and the intake manifold had a fairly large leak in it. Still, it ran fine.

 

[brianl703]

Yes, I have heard that it can adjust individual injectors too.

It really wouldn't be hard to confirm with a dual-trace scope or a couple of period counters (shows the pulswidth in mS).

I will say this though...it IS pretty intelligent.

The thermactor air diverter solenoid was stuck downstream, causing it to throw a self-test lean code for both banks, which makes perfect sense because that's basically air being pumped into the exhaust manifold which is causing the oxygen sensors to read lean.

I ordered new diverter and bypass solenoids (I figured if one had failed the other would soon fail too), but in the meantime the new injectors I ordered because the old ones were leaking externally got here.

I put the new injectors in, started it up, re-ran the self test, and instead of giving the lean code, it gave a code for "thermactor air stuck upstream".

I can't figure out how it knows the difference, nor can I figure out why it can't tell the difference with old (likely unevenly clogged) injectors.

There is the possibility that there was a slight vacuum leak or something that I fixed when I replaced the injectors, which was confusing it, since I did replace a couple of PVC/vacuum hoses when I was in there. EDIT: A slight vacuum leak that is not enough to trigger a lean code, as I verified that the lean code went away when I disconnected and plugged the vacuum line to the thermactor valve.

If I wanted to be really scientific about it, I could have put the old injectors back in and re-ran the test. But I was already irritated at having to do the job twice since the first time I just replaced the o-rings which did not fix the fuel smell.

 

[Sid]

Focus owners who add modifications dealing with the intake always reset the computer to compensate for different airflow although they say this shouldn't be needed since the computer is constantly monitoring airflow and should compensate by itself eventually. Still some people seem to solve their troubles just by reseting the computer. Cheap insurance I guess.

 

[brianl703]

Sometimes resetting the computer will cure drivability problems even if no modifications were made.

 

[Gary Allan]

The Ford system is allegedly more "intelligent". It has been asserted that it can actually adjust pulse width for each individual injector ..by routinely modifying the pulse width on one injector and mearsuring the change in O2 reading. Polling them, if you will. My DC system is not so cerebral.

I can't find the section on the adaptive cells at this time ..but the simple decription of the open loop mode is that it throws out the O2 reading and uses a plotted fuel curve that is either incremented or decrimented based on recorded O2 cross overs (as it is termed).