smitty2919

stevo: You suggested in a thread I started to use your program to diagnose a spark retard issue I was having. You suggested to pull timing to verify if the knock was false or not.

I am confident that it is false knock, but is your program capable of diagnosing WHY I'm getting false knock vs TTS Datamaster logging system? Everything I see in DM is operating like it should (O2 sensors turning off at WOT etc).

steveo_

don't misunderstand what false knock is.

if it was a sensor issue it'd be real knock, not false knock.

there is no real way for a datalog to tell you "why" you're getting false knock, since false knock is a physical thing (banging or harmonics); and there's no sound sensor or camera feed to the ecm.

you can tell if it's false knock or not usually just by the behavior of the knock system. it'll max out knock retard, and you'll see it try to recover, but it'll just jump back to max knock retard, and knock counts will continue to increase. using eehack's accelerated logging rate really helps with that due to almost 1.5x sample rate. real knock caused by excess spark advance should be completely eliminated when the stock knock retard of 9 degrees hits, not keep incrementing knock counts... if it's not, it's usually false.

smitty2919

Understood thank you. After learning about Knock Recovery Rate and that false knock comes on instantaneously then tapers off as recovery rate kicks in, this is exactly what my car is doing.

I see .3" of retard around 4100rpm, by 4500rpm it spikes to 10* then will step it's way down until redline....every...WOT...pull in 3rd or 4th gear. 1st or 2nd gear WOT pulls the issue is not there.

I never see my knock counts "stop" every time KR is occurring the knock counts keep increasing.

steveo_

the best test is to use eehack to remove like 3 degrees of timing at a time and see if the knock goes away.

an lt1 probably wont knock at 25 degrees no matter how hard you hit it. eehacks recent versions allow you to hover over a knock event to see what the advance is as the event occurs. if its nonsense, like 23 degrees at 4000rpm, you can bet its false.

do keep in mind that knock can occur at 2000rpm but log at 3000rpm depending on your logging sample rate and how fast you're revving... its not totally accurate as all you get is a count and an rpm/map, the ecm doesnt correlate the two, and a tenth of a second isnt very high resolution sometimes.

assuming its false....

valves too loose? too tight? exhaust hitting frame? broken inner spring? knock sensor overtorqued? engine mount broken? ujoints failing? trans mount broken? stuck lifter?

all these things have happened to me personally and first appeared as knock counts.

fyi I run zero knock retard at wot below 5000rpm and 2 degrees max above, but im super confident about my afr, timing, and consistent fuel quality. I would reccommend trying everything to eliminate false knock before going so far.

Prototype007

iTrader: (2)

You sir are a blessing to the LT1 world.

Thank you and donated!

steveo_

thanks, appreciate it!

will have a new version out soon, but try my beta..

steveo_

4.4 release: http://fbodytech.com/eehack-4-4-release/

Snorkelface

Now I'm starting to wish I converted to OBD1. It's hard to even find a local tuner. :/

steveo_

it's not particularly harder to tune OBD-II, it's just more time consuming and expensive, isn't it?

what's expensive is the flash ability, right?

i wonder how hard it'd be to reverse engineer a flash tool for that thing.

if someone got me a tech tool, obd-ii interface, and an ecm, even on extended loan, i'd be tempted to give it a shot.

it can't be that crazy hard, can it?

hell, the key to the 'secret unlock code' on obd-i was broken by a simple binary NOT on an 8 bit value.. and before we had that figured out, i just read the raw challenge/response from memory to break in.

if a flash tool could be written that just leaves writing a tunerpro definition, right? im sure the timing and fuel tables could be found quickly enough..

does anyone have a dump of an OBD-II bin?

AINT SKEERED

iTrader: (2)

I have tunercat, will the cables I have for it work with this set up?

hrcslam

Like I said in the other thread, yes.

If the cables work with TunerCats it'll work with this too. I have TunerCats too (that's where I started), I use EE Hack, Scan 9495, Tuner Pro RT, and TTS Datamaster with the same cable and computer no issues.

AINT SKEERED

iTrader: (2)

thanks, data logging was what I have been missing all along to tweak my dads tune at the track.

Snorkelface

If I had the stuff, I'd definitely send it! I think you should post a separate thread on that. I'm sure someone would chime in!

smitty2919

So just to clarify, I already have TTSDatamaster which is a datalogging program ONLY. If I was to stick with TunerCat programs I would need to download WinFlash along with their definition file.

But I can switch to:
TunerPro+EEX definition file as a tuning/editing program
EEHack as a datalogger and flashing program

Sorry for the ignorance between definition files/tuning programs/dataloggers. I was overwhelmed with "tuning" but this phantom knock I am getting is annoying me and I don't want to give my tuner another $100 to diagnose this. I could get access to a dyno and look into this myself.

hrcslam

Yes, EE Hack is great for tuning. If you need something for troubleshooting, get Scan9495.exe. I use Scan9495 for troubleshooting issues, EE Hack for logging for tuning, and Tuner Pro RT for making changes in the tune. I started with TunerCats, so I still use that to load the tune.

fbody_brian

I was still using winflash to load the tune, but now that $EEhack is so optimized it flashes the bin in half the time or better. adjustments to just e-side, like fuel ratio, etc just flash that side of the bin. So from the time you press flash to the time it's done is roughly 50 seconds. Ain't going back to flashing with anything else now!

smitty2919

I just downloaded tuner pro, eehack and the eex definition file. I'm hoping to test out that it connects to the car and does a datalogging session later tonight. Then move into pulling timing on the fly to test for false knock like Stevo_ suggested.

From there....I don't know lol MUCH more reading will need to happen to correct my issue in the tune if I am to do it via setting a WOT spark retard value.

Sucks I didn't get into this earlier after paying $100+ to TTS....

I'll be donating for sure after I give it a test run and all goes well.



EDIT:
I tried monitoring my car. Opened EEhack, connected the car and laptop, KOEO, selected Com3 since Com4 wasn't working and all it does is say "Connecting" "Sending Silence Request" and the debug screen keeps repeating "SERIAL: WAITED_TOO_LONG in serial_read_packet" as the error counter keeps going up.

Edit 2:
I needed to leave it on Com4 as DataMaster saw Com4 and also uncheck the "Corvette/B Body BCM" option in Settings.

steveo_

yeah you do have to pick the right serial port (unless you have a single serial port, eehack picks that automatically). it should remember your selection.

the corvette/bbody option doesn't matter. disabling it simply speeds up connecting on an f-body which doesn't have a body control module to tell to shut up.

a hundred bucks for a 90s era datalogger is kind of a ripoff in this day and age. even scan9495 walks all over it.

free tuning software is what the lt1 deserves, since it's an ancient platform.

smitty2919

After messing with EEHack I realized you need to plug laptop into the car. Turn key to KOEO or start the car THEN open the EEHack program THEN click "Connect". This was it automatically finds the serial port upon opening and works fine.

I was opening the program, plugging into the car, turning car on then hitting "Connect". It did not like that.

fbody_brian

I leave it open, then connect to the car no problems. I just go to the settings after plugging into the car and refresh the com port.
It should take you to the settings screen automatically if the com is invalid.