As i am driving around, usually when datalogging.... my ecu will just "reboot" on me....
like as if i quickly turned the key off then back on.... all the dash lights come on for a second.... HPT gets disconnected, and cant reconnect itself...
same exact thing as when you turn the key on... or when you finish flashing your computer... you all know what i mean....
but why does it do this randomly in mid-drive???
Ususally if i am driving for any length of time, i have the laptop and HPT going... just to keep an eye on things and to see what happened if things act up. So i cant really say if it does this without hpt hooked up or not.... but i know it does it alot when it is hooked up and logging.... and it is frustrating... cuz you cant pick up where you left off... you have to save the log and start a new one... and then try to hand merge them later.... (average out the histograms and such)
A few things that may be related....
1. it seems to happen the most at freeway cruise speeds... like around 65-70mph... dont see it very much below 50mph... and above 70 it has yet to do it.... alot of times it has done it after a few min of higher speeds.... then slowing back down to cruise speeds... like as soon as i hit 70 "oh good lets reboot now"
2. I do have 2 randomly occuring CEL codes... 621 and 622... both generator terms... the one i have heard is related to the horn circuit... my horn did ingest some water at one point.... and didnt work after that... but i recently replaced it... but it still sounds very weak... and sometimes still wont honk. the fuse is fine, the relay works (you can hear it click when you hit the horn, even if the horn doesnt blow. but even if it does blow it is a kinda lame oooga and this is the highway blaster which is supposed to be deafening. I dont quite get why or how the horn would have anything to do with the ECU or anything... but worth mentioning... The other code i cant find squat on.
But what i do find interesting... is that HPT actually has a control for the "generator L-term" which is what one of the error codes is about... So can anyone tell me what the L-term is? or why HPT gives us a switch to turn it on and off... it must be of some use in tuning... and that could clue me in to other possible causes. In searching the web, i found lots of people asking about these "generator terms" but no one answering them with useful info... or just lots of generic one line definitions... "Generator L-term Error: fault in the L-term circuit" gee thanks! how useful that is!!
3. My Car will occasionally choose not to start up when warm.... (takes 30 min at rest, or a quick spritz of starting fluid to start up again) I posted on this before here on jbo... and people here, and many others told me "sounds like a fuel pump issue" so i replaced my fuel pump with a racetronics yesterday.... and went out to re-tune for the new pump (and new-found proper fuel pressure at high rpms) and sure enough... pulled over shut down car to upload some changes.... and sure enough... wouldnt start again. just like before on the old pump. so i am guessing it was not the pump.
4. I do have low-impedance injectors, run via inline-resistors, to my high-impedance driver. they seem to be working fine... but it is worth mentioning anyway.
I had this happen once. I think it was related to my laptop, not the ECU.
FU Tuning
Same thing with me ken, when i was doing mine , it would reboot alot. Nothing ever happened from this tho.
Same with me...happened only when the laptop was plugged into the cig lighter and running HPT. If the laptop was running on battery it was fine. I got a new battery and my alternator checked out fine...hope I can get it tuned on tuesday when I head to halifax and have no more issues but I'm not too optimistic.
I'm sure I'll have to fight with everyone tooth and nail (since thats how it usually goes with HPT threads)
but this is common. something with the way class2 serial communicates with our ecus.
I brought it up to Chris on HPT forums and he said its a normal side effect that cannot be fixed.
it happens to me too all the time (when logging) but my HPT usually reconnects when the ecu recovers
Yeah, mine reconnects as soon as the guages/ecu starts reading right again....when it shuts off HPT says something like lost signal/no signal (something like that) and when everything comes back online it says reconnecting...although the computer froze once and we lost 45 minutes of logging which kinda sucked but we managed to get enough....I'm glad to hear I'm not the only one having this issue cause no one seemed to know what was wrong that looked at my car.
Quote:
something with the way class2 serial communicates with our ecus.
Something with the way the ecm/pcm is set up to handle communications, and something with the way the laptop is programmed to communicate. I have rarely run into these types of issues using the factory Tech2 scantool. But using aftermarket scanners and laptops, communications issues happen all the time. I've got $15k worth of Snap-On scanner at the shop and I can't check codes in a 96 Grand Gm without setting an ABS code for communications failure with the network. But it's not just these cars, it's frequent across all makes and models.
As far as the no-start, isolate the problem more. Is it only with the laptop connected? If so, can you disconnect the laptop and make a change? Is spark present? Is fuel present? Does the situation change if you enter clear flood (press your foot to the floor and hold before cranking)? How about if you try several short bursts of cranking? IOW, do some tests and the results may help pinpoint this issue.
-->Slow
DaFlyinSkwir(LS61) /PJ/ OEM+ wrote:I'm sure I'll have to fight with everyone tooth and nail (since thats how it usually goes with HPT threads)
but this is common. something with the way class2 serial communicates with our ecus.
I brought it up to Chris on HPT forums and he said its a normal side effect that cannot be fixed.
it happens to me too all the time (when logging) but my HPT usually reconnects when the ecu recovers
yeah it has done that to me on rare occasion before. Usually it was just a gauge light show. and everything else proceeded as normal. It would rarely loose connection, and if it did it always would reconnect. but this is different...
now it does it all the time... like i have yet to make it thru a complete tune drive without the interuption. and now when it does it, it will never reconnect... it sits there trying till it runs out of re-tries. it is like it completely lost the car. but if i stop the scan, and restart it... (escape key, then space bar) it will wipe my log clean and reconnect.
but sometimes even after doing that... it will almost instantly reset itself again... i usually have to disconnect and leave it off for a minute or so before reconnecting... and then it is fine for a while.... But it is likely to do this to me several times in an hour.... i took a 2 hour drive the other day, and it rebooted on me at least 7 times.
I was running on a cig lighter... i can try on battery power only... see if that is any better.... but my laptop sucks the juice and batterys are $$$$ for it... so would rather not have to tune on battery power.
but i also cleaned up my "generator terminals" (Generator L and F Term are the 2 wires that come out of the front of the alternator... one goes to the gauge cluster, the other to the ecu... i guess if these arent working you might be getting a charge from the alternator, but the ecu will think something is wrong.... and it may react accordingly. So i am hoping that maybe that is a source of both of my issues... the computer loosing signal, and the no start... (perhaps the ecu is resetting things because it thinks the alternator quit)
slowolej wrote:As far as the no-start, isolate the problem more. Is it only with the laptop connected? If so, can you disconnect the laptop and make a change? Is spark present? Is fuel present? Does the situation change if you enter clear flood (press your foot to the floor and hold before cranking)? How about if you try several short bursts of cranking? IOW, do some tests and the results may help pinpoint this issue.
-->Slow
Trust me... i have done plenty of isolation and trial and error... i just didnt think anyone wanted to hear all my grief....
no... the no start is totally laptop independant.... it did it to me yesterday and the laptop was not even in the car... (mind you i keep the hpt module in the car and hooked up, i just plug the usb cord in to the laptop when using it... but i have disconnected the hpt box after it would not start, and that had no bearing on anything.)
when the laptop was with me when this happened... at first i thought it was a corrupted tune... or a bad upload... (first few times it did this was when i stopped and shut down to upload my changes) so i would re-flash the tune or drop back to the previous working tune... but this never helped...
I never pulled a spark plug to check for spark... but i would imagine there is plenty of spark, as the thing fires right up and runs like a champ if you give it a lil shot of starting fluid. if there was no spark, starting fluid wouldnt do anything...
is fuel present? hmmm that is a tricky one... i would guess no... as i thought this was my fuel pump... but replacing the fuel pump and wiring harness for it did not solve it... and the fact that it fires up and runs instantly with a tiny spritz of starting fluid... makes me think there would need to be fuel pressure present... otherwise it would burn off the ether and then die or sputter as is struggled to get it s own fuel... but this isnt the case at all.. the tiniest hint of ether (starting fluid) makes the problem totally go away and the car starts right up and runs like nothing was ever wrong... BUT... oddly... you can smell gasoline when it isnt starting.... like it is getting gas but not burning it... So that further backed up my fuel pump theory... i figured it was getting gas, but while the starter was sucking all the voltage, the dying fuel pump combined with minimal voltage was creating such low pressure it was not able to atomize.. and was just dripping from the injectors.... so it was there... just not enough, and not able to burn.... until the starter shut off and gave the fuel pump its full power back... at which point things were fine again.
clear flood does nothing... when it does this... i have tried everything... i have feathered the gas... held it to the floor... pumped it... everything you can think of... the only thing that seems to do anything is if you let OFF the pedal suddenly... it acts like it tries to fire for a brief second. but never actually does.. but short of the sudden throttle close, it just cranks and cranks with no sign or desire to fire. I have disconnected the ignition and cranked it.... i have pulled the fuel pump relay and cranked it... you name it i have probably tried it...
I have cranked it solid for like a minute straight... nothing... i have done several short bursts in a row (fearing for my solenoid and gears the whole time) nothing... multiple long bursts... nothing seems to change it.... only things that have any effect on it are time... letting it sit for 30min or so... or a squirt of ether... or the sudden throttle close makes it surge, but not start.
any thing else you suggest i should try?
I plan to hook up a fuel pressure gauge here soon, so i can see how that is in situations like this.
I also want to see if it will roll start... i am willing to bet that it will start right up by rolling it and poping the clutch... but have yet to try.
I am hoping that my recent tinkering with the alternator L and F terminals may have helped this situation out too... not sure how it all works, but thought that maybe that if the computer thought there was not enough power that it might do something to add to this problem.
Ken, I have been having similar problems to you in many cases. I do appreciate you asking the questions, that way I figure out my problems too! As far as the ECU restarting while I am datalogging, mine does this too, as does my friends car. I didn't think it was too relevant, and is only a slight nuisance so I've been living with it.
But the warm start issue is something I have asked about before and no one really seemed to have had the same problem as me. I've since bought a port fueler so I don't have to deal with the woes of having the larger injectors firing all the time, but I am still desperate to figure out why it was happening to me. Once the car was warmed up, if I turned it off and then tried to turn it back on again, I would get severe bogging/hesitation and then it would stall. The only way I could get it to start would be to do some fancy footwork to keep the revs up. It seems that my case is a bit different than yours because you can't seem to get it to start at all.
I do know that every time I do fire it back up, I get a giant cloud of black smoke - I am assuming the engine is flooding or something. My speculations were always aimed at the injectors. I tried changing every tunable parameter in HPTuners but to no avail. My injector constant was very good, and I spent hours upon hours getting it to a good spot so I don't believe it had anything to do with that. It was suggested that I check the FPR, to see if the diaphragm inside was still okay by pulling the line to it while the car was running and seeing if any fuel leaked out.
I considered the coolant temp sensor, as well as the IAT incase it had something to do with the actual temperature of the engine not reading correctly. I swapped out the IAT since it was fairly painless, but the ECT sensor was not as quick of a fix and it was just a hunch so I didn't want to do it for nothing so I never tried.
I doubt much of that helps you all that much, but I hope you can figure it out.
" To study and not think is a waste. To think and not study is dangerous. "
I had warm start issues (with 440cc injectors), then I changed my injector constant.
no more problems.
here's a hint... formulas and scaling doesn't seem to work right, my number is a guesstimate.
DaFlyinSkwir(LS61) /PJ/ OEM+ wrote:I had warm start issues (with 440cc injectors), then I changed my injector constant.
no more problems.
here's a hint... formulas and scaling doesn't seem to work right, my number is a guesstimate.
This may not apply in PJS situation, but upon start the injector pulse is being modified by the "Zero" IPW Multiplier column. This means if you have used your IPW multiplier to add fuel in for boost you are also causing a rich condition on startup.
___________________________________________________________________
Hahn Stage II - Mitsu TD06-20g |3" Turbo-back Exhaust | 61mm Bored TB |
HP Tuners | Innovate WB02 | Spec Stage 3 | Team Green LSD | TurboTech Upper | Full Addco Sways | Sportlines & Yellows |
DaFlyinSkwir(LS61) /PJ/ OEM+ wrote:I had warm start issues (with 440cc injectors), then I changed my injector constant.
no more problems.
here's a hint... formulas and scaling doesn't seem to work right, my number is a guesstimate.
I wonder if its possible if you have the calibration with the "Cranking VE" Table that you could correct that warm start issue without modifying injector constant.
http://www.hptuners.com/forum/showthread.php?t=23483
I wish I could figure out if that calibration is only 04+ I would go to the junk yard and find one to test with.
___________________________________________________________________
Hahn Stage II - Mitsu TD06-20g |3" Turbo-back Exhaust | 61mm Bored TB |
HP Tuners | Innovate WB02 | Spec Stage 3 | Team Green LSD | TurboTech Upper | Full Addco Sways | Sportlines & Yellows |
LD9's do not have a cranking VE table... if we do, HPT won't let us see it.
Cunts.
-Chris-
-Sweetness-
-Turbocharged-
Slowly but surely may some day win this race...
Ok, so it sounds like Ken's problem is insufficient fuel during crank. Clear flood doesn't help, and adding fuel does. Are there voltage based corrections for the injectors? There should be. It takes more time for an injector to open when voltage is low, so the calculated "on" time should be longer. And the amount of time needed is dependent on the injector's characteristics. Have you ever tried a jump start during this condition? Even though the engine cranks ok, this might help keep system voltage up and would make for a good test. How about looking at the connection point? If the power for the injectors and the ecm are taken off the starter, can you move them to the battery to decrease system voltage drop? What about trying a battery with a greater cold crank amp rating?
Also, if you're running a lower fuel pressure that will result in poorer atomization at the injector and can cause starting issues. Poor atomization can cause the seemingly impossible condition of a vehicle needing a shot of ether to start yet giving indications of too much fuel upon startup. Too much liquid fuel is injected into the chamber and there isn't enough heat generated during compression to allow the spark to start combustion. More fuel pressure is better if you can do it.
I think I'm with PJ if there are no injector corrections available. Lie to the ecm and tell it the injectors are slightly smaller, or turn up the pressure if it's not already and leave the injector constant smaller. You'll be making compromises in your tune, but that's part of what this game is all about.
-->Slow
so far i am running stock fuel pressure... but i am running the racetronics fuel pump now, so i have the availability of higher presure... but i figure why mess with it if it works.
The starting issue was VERY random and intermittent. most of the time it started like a champ. but then some days it just would not start at all. it was very odd. I would think if it was a tune issue, it would be present all the time the tune was there.
I am not opposed to making a few lil tweaks to the tune to try and help this... but i would rather not redo everything just to fix a random issue.
Actually... it has not happened to me at all since i put in the new fuel pump.... well... no... i take that back... it happened twice... but both times were immediately after shutting down to flash the ECM with a new tune. And i never understood that, but it seems common... that the computer takes a few minutes to "get used to" a new tune. It seems like the car always runs like @!#$ after a flash... and i always think "man what did i screw up that time" and then after a few min it works itself out and runs fine from then on.... SOOO i am going to write of those 2 non-starts as simple "computer wasn't acclimated yet". Normally i would get those non-starts just for no reason, just out of the blue (and alot after a flash). So until it fails to start for me for no reason... I am going to hope and assume that the fuel pump fixed that issue.
BUT... what i would really like to know right now is what is up with my idle under open loop... the thing runs, idles and coasts down perfectly in closed loop. i could have it at redline, and take my foot of the gas and she purrs right down to target idle speed and sits there nicely... but as soon as i take it out of tuning mode and go back to the O2 sensors... it drops past target idle speed then revs way up past and jumps around for a bit till she evens out.... I thought it was like one cell way out of whack.. and when it hit it it would get whacky and try to compensate. But that should be ever MORE noticable in OL. not completely gone.
I got her tuned in better than before... she used to bounce till she stalled... now she doesnt stall... just bounces. But i just found it soooo wierd that there were no signs of this in closed loop. but as soon as i set it to open loop, bang there it was again. Makes me think it is either something that i am NOT tuning for (table i have not messed with, or something that got adjusted automatically that i didnt know about)... or something i cant tune for (screwy sensor or physical issue)
What did seem to help was dropping my offset back down a little.... i was using a 50% offset, as i had a few cells that were like 142 and such... and was just thinking that i never had the idle issues this bad till i bumped the offset up. the offset of 50% really helped the overal running tune, but killed my idle... so i dropped back to 40% and just let the 142 chop down to 139. and reset my VE tables for the 10point difference.. (i have only ever adjusted the main 3 tables for VE... double fire single fire and high rpm. i dont touch the idle or coastdown. i leave those however hpt adjusts them. So as the offset moves the coastdown and idle accordingly, i let them move... and just adjusted the main 3. So dropped the offset to 40, bumped the main three by 10 points... and things seems alot better with the idle... not great.. but better.... but then when i did some closed loop tuning i just found it odd how perfect the idle was. and then the issues came instantly back under open loop. so it just made me wonder... I know the fuel trims are disabled in CL... but are there any tabels that ARE IGNORED in CL, and only used in OL? that might help me to find my issue... as it seems very much loop related.
Also... I found out that those check engine codes were refering to the 2 wires coming out of the alternator. so i cleaned them all up real good, and since then i have not had either code return. So i am thankful for that... But it hasnt been THAT long... so i am not writting it off as case closed just yet.... but... oddly... my new horn (which never sounded as strong as it should have, more a weak groan than a blasting honk) now has also completely died on me.... i can still hear the click when i honk, but no honking... just a kinda odd and random thing. (i bring it up now as somewhere i read that the horn shares a circuit with something related to the above engine codes, and a shorted or bad horn can trip the CEL and throw a generator trim code)
I really appreciate all your help thus far... i am sorry to write novels... but just trying to give all the details, as the solution to these types of things rarely is sitting on a pedestal marked with a neon sign.
For the ECU and horn problems, check the grounds for both. If the fuse for the horn is good and you hear the relay click, it's safe to assume that it's getting power (but check anyway, I'm sure you know that). A bad ground would/could also make the ECU reboot if the voltage was too low due to a drop in the ground. Start with cleaning up the battery cable ends and then at least the ground for the horn and the ECU. Having the laptop plugged in may be just enough extra load to raise the voltage drop on the ground.
----------------------------------------------------------------------------
'04 Cavalier
2.2 Ecotec Supercharged w/ 2.9" pulley
Charge air cooler
Poly bushings - lower control arm and engine/trans
ECM\PCM\VCM***
Chris
'02 Z-24 Supercharged
13.7 @102.45 MPH Third Place, 2007 GMSC Bash SOLD AS OF 01MAR08
Quote:
but i figure why mess with it if it works.
Well, that didn't stop you from modifying things in the first place, did it?
Quote:
The starting issue was VERY random and intermittent. most of the time it started like a champ. but then some days it just would not start at all. it was very odd. I would think if it was a tune issue, it would be present all the time the tune was there.
I would not make this assumption. The only way to know for sure is to rule out everything which isn't the tune. I've seen one corrupted bit make all sorts of heartaches. But if it's not giving you trouble anymore, that's a good thing.
I don't use HPT so I don't know how "tuning mode" works in that software. If they're using a factory mode, it's possible it adds a fixed amount of timing which can greatly affect your tune. Also, it sounds like you have closed loop and open loop backward. Closed loop is when the O2 readings are used to adjust the engine fueling and the short and long term trims.
-->Slow
Taetsch Z-24 wrote: ECM\PCM\VCM***
Chris
Blah blah blah... Every manufacturer has a different name or code for it... As long as we all know what it means.... what is the difference!?!?!
slowolej wrote:Quote:
but i figure why mess with it if it works.
Well, that didn't stop you from modifying things in the first place, did it?
Well, yes, actually it did... but that is a long story, longer than my normal longwindedness. but in short... most parts i dont upgrade until there is a need to. didnt built my engine till i spun a bearing.. I just dont like wasting good things... so i wait till the good goes bad, then I get better. seems a waste to tear apart something that is perfectly good.
Quote:
Quote:
The starting issue was VERY random and intermittent. most of the time it started like a champ. but then some days it just would not start at all. it was very odd. I would think if it was a tune issue, it would be present all the time the tune was there.
I would not make this assumption. The only way to know for sure is to rule out everything which isn't the tune. I've seen one corrupted bit make all sorts of heartaches. But if it's not giving you trouble anymore, that's a good thing.
I don't use HPT so I don't know how "tuning mode" works in that software. If they're using a factory mode, it's possible it adds a fixed amount of timing which can greatly affect your tune. Also, it sounds like you have closed loop and open loop backward. Closed loop is when the O2 readings are used to adjust the engine fueling and the short and long term trims.
-->Slow
And yes... you are right... i did have them backwards.... I used to confuse the terms all the time... but then for some reason i had it in my head that "closed loop" was closed to all external input, and relied entirely upon the programmed data. (Like being "close minded" = not open to new input, stuck on what is already in your head. or as would make sence in electronics, a closed circuit TV, means it only uses the specified camera to feed the TV, nothing external) But i looked it up, and you are right. Closed loop is using the sensors and trims and all... Geesh.. why they gotta make things so confusing...
Anyone have a good way of remembering this? since my lil mnemonics obviously reminds me of the wrong answer.
Or at least an explaination of WHY they named them as such?
Quote:
Anyone have a good way of remembering this?
Closed loop refers to the controller making a change and receiving feedback. An open loop system makes changes blindly. There's no way to know if the change does what's desired. A closed loop system can, say, adjust fuel richer then look at an O2 sensor to see if the engine is actually becoming richer. If the loop of change -> feedback -> change is broken, it's open. What goes out doesn't come back. If the loop of change -> feedback -> change is working, it's a closed circuit.
HTH
-->Slow
