Resetting the ECU

[GraemeWi]

Originally posted by GTST

GraemeWi...

that's fair enough...

but what was the graph like BEFORE the ECU reset?

you have the initial run after reset and the next couple but what was the graph like before the reset?

was there a difference on the dyno with the before reset and after reset?

Shore thing - I'll attach a copy - though this one is a bit harder to read. The black line is the initial run before the reset. As you can see, there was a fairly large power loss just over 100 kph on the initial run. Post reset the red pen was used for 2 runs, and there is more power after 100 kph, then a green pen for another run.

I'll insert a disclaimer - my R33 wasn't a happy car when I first got it. It was only making a max of 5psi with everything stock. This is why I took it to a workshop with a dyno to see if they could help me out. After this dyno chart, the stock controller was binned and a Turbosmart 2 stage was installed - much better!

Cheers,

G

[GraemeWi]

Originally posted by rev210

GraemeWi,

by the looks of the dyno the first 'red line' shows an ignition fault similar to dirty plugs, after a few good runs particularly with platinum plugs they will be 'cleaner'. There are 'bumps' present still in both charts indicating this is the reason for the difference. Each subsequent 'power'  run will slightly improve the condition.

In 'Closed loop' operation , the oxygen sensor may be re-calibrated on a reset. But it's 'closed loop', meaning very light throttle only.  

Like I said it does nothing.

Hi,

in the first chart I posted the plugs were new items gapped and installed about 5 minutes before the run.

The second chart has a 'before' run, and that dyno session didn't involve any plug changes.

As far as I know the ingition timing is the only thing to be remapped. The oxygen sensor used in the Skylines is narrow band and not suitable to be used for any learning functions.

The ignition timing could be retarded in a ECU for a number of reasons - a tank of bad or low octane fuel, a drive involving high ambient temperatures and loads, someone hitting the block with a spanner while the engine is running....etc. So I'll stick to my recomendation that a ECU reset can have good results.

Cheers,

G

[TTT]

dude..

there is something seriously wrong there..

why is the whole graph all up and down all over the place?

AND, up until this 100km, the PRE ecu reset is producing more power all the way through the revs...

[rev210]

Originally posted by GraemeWi

Hi,

in the first chart I posted the plugs were new items gapped and installed about 5 minutes before the run.  

The second chart has a 'before' run, and that dyno session didn't involve any plug changes.  

As far as I know the ingition timing is the only thing to be remapped. The oxygen sensor used in the Skylines is narrow band and not suitable to be used for any learning functions.  

The ignition timing could be retarded in a ECU for a number of reasons - a tank of bad or low octane fuel, a drive involving high ambient temperatures and loads, someone hitting the block with a spanner while the engine is running....etc. So I'll stick to my recomendation that a ECU reset can have good results.

Cheers,

G

dude,

The engine both before and after looks like it exhibits an ignition fault or another kind of fault none the less. There are 'bumps' in the power curve although less obvious in the 'after' run, they are still visable.

The car clearly has a fault in the power delivery, that continues to be present 'after' your ecu reset. So we can't very well claim the reset fixes anything or even makes it better, given the fault is clearly still present. Intake and engine temps and all sorts of things can have bearing on the magnitude of the fault making this evidence for ecu reset rather weak.

I have not reset my ecu, since doing quite a number of modifications. My car is a fairly quick factory based ecu car. Would it be quicker after a reset? I can't be stuffed doing it because it won't be, why waste my time.

At the end of the day an ECU reset done properly is not going to hurt anything so if people want to try it go ahead. It may or may not fix a glitch, after all it is still an ECU and therefore not immune to error. So knock yourselves out ,better than throwing rocks at dogs for an afternoon I guess.

[Yakuza]

i heard you have to bypass something to reset the ecu ? is that true ? or just unplug the ECU for a night ?

[pushead]

All you have to do is disconnect your battery , step on the brakes 5 times to discharge any left over static and plug it back in. takes 2 mins to do .

[GraemeWi]

Originally posted by rev210

dude,

The engine both before and after looks like it exhibits an ignition fault or another kind of fault none the less. There are 'bumps' in the power curve although less obvious in the 'after' run, they are still visable.  

The car clearly has a fault in the power delivery, that continues to be present 'after' your ecu reset. So we can't very well claim the reset fixes anything or even makes it better, given the fault is clearly still present. Intake and engine temps and all sorts of things can have bearing on the magnitude of the fault making this evidence for ecu reset rather weak.

The purpose of posting the dyno plot was to illustrate the ECU learning ability. My car did have a power delivery problem - earlier in the thread I said it had a boost problem, plus it's an auto and gives a somewhat messy change to 3rd during the run. There had been a fair amount of discussion on the forums about the ECUs learning ability, so I asked the guys at the workshop specifically for a before and after plot of the reset. After that they then went on and found there was something wrong with the factory boost controller.

Of course I could have been lead up the garden path, as the guy in the workshop had partisipated in the forum discussion on resetting ECUs...who knows?

I have not reset my ecu, since doing quite a number of modifications. My car is a fairly quick factory based ecu car. Would it be quicker after a reset? I can't be stuffed doing it because it won't be, why waste my time.  

At the end of the day an ECU reset done properly is not going to hurt anything so if people want to try it go ahead. It may or may not fix a glitch, after all it is still an ECU and therefore not immune to error. So knock yourselves out ,better than throwing rocks at dogs for an afternoon I guess.

No worries - it's one of the great things about the internet, in that it allows people to exchange ideas and express differing opinions. (Except for hi-fi forums where if you can't hear the difference between a jug cord and a $500 power cable you are branded an amateur!)

Cheers,

G

[Knat]

Originally posted by pushead

All you have to do is disconnect your battery , step on the brakes 5 times to discharge any left over static and plug it back in. takes 2 mins to do .

Would disconnecting your ECU, rather than disconnect the battery, do the same thing?

If so, how do you properly disconnect the ECU?

[pushead]

look in the passenger side near the door below the glove compartment, there is a screw holding the panel. unscrew it and flick the panel out, it should be held down by two clips, takes abit of force but you will not break it. Thats where the ECU is. This is a timely process... I would just go to the boot of my car, undo just the 'red' cable connecting to the battery leave it out for 5 minutes and step on the brakes a few times , plug it back in and it will all be reset. Oh dont forget to disable your alarm when doing so

You will need to adjust your clock and tune your radio stations back it after the process.

Not sure with other peoples experience with resetting the ECU, car runs smoother and feels more power.. don't know might be a mental thing.. hope it helps.

[Knat]

Thanks pushead. Yeah I thought I'd give it a crack because after my last service, my car seems to be idling a bit rough/low when I come to a stop mainly. I've read up on the other threads about these common problems and thought this is something easy I can try myself. In the service I had just my oil, filter and fuel filter changed. Haven't had the plugs changed or the afm cleaned etc. yet. Suppose that's next...

[pushead]

you can always try. Give it a go dude, it doesnt 'hurt' your car

[semtex]

I did it without disconnecting the battery via the consult port and a paperclip... And the ECU told me that everything was fine too

Thanks Airzone. I read about that method somewhere. Do you, or does anyone else, have a pic of the consult port ? In an R33, is it a grey port in the fuse box in the cabin on the drivers side (where your right knee would be when sitting in the drivers seat) ?

I wouldn't mind hooking a switch up to it so that I can reset ECU from the drivers seat without utensils !

Thanks again.

[semtex]

I have found the consult port and reset my ECU with a paperclip. Even better than this was the fact I got the all OK flash sequence. I will post pics when I get round to adding an ecu reset switch.

[FATGTS-R]

My car always had a prob with idling at about 1100rpm. Tried everything, that screw at the rear of plenum was all the way in. Got a new exhaust fitted and it still idled high. I reset the ecu about a week ago and in that time the idle has come down to 800rpm. I have not noticed any other difference to the running of the car. Whether its a coincidence or not I dont know but before the reset noboby could get it to idle under 1100rpm and now its idling fine. Who knows?

[darkhalf]

the idling when neutral switch = ON and TPS = ON (no accelerator) the ECU adjusts to the previously stored pulsewidths which make it run stoich according to the ecu had stored.

this may be running it with higher pulsewidth adjustment prior to your ecu reset and now is running factory settings since you cleared the RAM

the ECU has capacitor backed up RAM (whether separate chip or inside MCU) . any method of removing power will allow the cap to discharge and you will lose the RAM contents of which the ECU will then rely on O2/knocking information to recreate this data with the current configuration of the engine