PDA

View Full Version : Dreaded knock.... cant seem to get my tune right



Dopey
June 2nd, 2010, 12:00 AM
Hey Guys,
Kudos to the great forum and masses of information it holds :)

I have been lurking on here for ages trying to learn as much as I can to get my tune right, but I feel that it is above my head at the moment... The more I play some things become spot on, but then I start having other problems.. lol.

Any way, I have been messing around with my tune (attached) which was tuned by someone else but has always been problematic, knock and afr's jump all over the place. I feel that I have got the VE pretty much bang on, however I have been very cautiously playing with the 02 Dynamics to try and get rid of the spikes and knock on throttle stabs.

I have attached my tune and a decent sized log where you can see there is knock all over the place, yet when I compare my timing to others and even stock it seems very conservative?

*Note* please ignore the AFR spikes to 10 and >16 ever since I changed my PLX TTL to 232 circuit from my external MAX232 board to the small board off ebay featured in one of the stickies it has started spiking... I think the regulator on the board hasnt got enough puff to power the ttl converter and the PLX unit.

Any tips or help would be greatly appreciated :hihi:

Car:
Holden VX SS M6

Alloy OTR
85mm ported throttle body
LS6 manifold with ported neck to suit TB
Magnecore 8mm leads
Tri-y headers
2x 3" High flow metal cats
Twin 2.5" Xforce cat back
4.11 Diff
PCV vents via a filter (ditched the catch can setup)
Copper NGK plugs

WeathermanShawn
June 2nd, 2010, 02:59 AM
If you can re-post your tune (as a tune file), I could look it over.:)

Chevy366
June 2nd, 2010, 03:08 AM
Can't we view .ctd files ? .ctd is what is used by EFILive now for the BBF (black box flashing) file extension .

WeathermanShawn
June 2nd, 2010, 03:16 AM
Hey Chevy:

Maybe I am the 'Dopey', but I can not do it on my end. I have the latest EFILive Scan & Tune, and BBL downloaded..but I still utilize laptop and Scan & Tune as my tool of choice.

It is up to the OP. Its hard enough to get people to look at your Tune and Logs..so it up to him.

I can see some legitimate KR issues at WOT (I can read log). I assume he is running SD Open-Loop..but since he is logging his O2's also (one bank is bad)..I am just guessing.

OP, up to you..:)

Chevy366
June 2nd, 2010, 03:25 AM
Hey Chevy:

Maybe I am the 'Dopey', but I can not do it on my end. I have the latest EFILive Scan & Tune, and BBL downloaded..but I still utilize laptop and Scan & Tune as my tool of choice.

It is up to the OP. Its hard enough to get people to look at your Tune and Logs..so it up to him.

I can see some legitimate KR issues at WOT (I can read log). I assume he is running SD Open-Loop..but since he is logging his O2's also (one bank is bad)..I am just guessing.

OP, up to you..:)

Open and save as .tun file , then it is as was .
I am on my Linux OS so can't try downloaded .ctd file , will have to boot into Winders and see , maybe you can't ?

joecar
June 2nd, 2010, 03:31 AM
I can open his .ctd...

Change the Windows file association for .ctd to be the tunetool instead of S&T...

WeathermanShawn
June 2nd, 2010, 03:33 AM
Thanks Chevy & Joecar..that did it. Took a few more steps, but now I can open it.

Always good to learn. I'll look it over along with you guys..

Thanks for the tip..

EDIT: Looks like he is running Open-Loop..but MAF enabled? C2901-C2904..

joecar
June 2nd, 2010, 03:34 AM
I can directly open his .ctd without converting it to .tun.

joecar
June 2nd, 2010, 03:37 AM
See image...

Chevy366
June 2nd, 2010, 03:38 AM
I can directly open his .ctd without converting it to .tun.

+1 on no conversion .
Right click on Icon , choose "open with" > "EFILive Tuning Tool" .

WeathermanShawn
June 2nd, 2010, 03:40 AM
O.K..now I get it!

Man, you guys are smart..Now what up with his tune?:).

Thanks (I mean it..not being sarcastic..)..

Chevy366
June 2nd, 2010, 04:28 AM
Ha , already on another OS Android (phone) , so cant view his tune and log .
Will leave that to you , "grasshopper" . ;-)
Just so I dont get in trouble, grasshopper was what was used by a priest to describe a young student in Kung Fu TV series. :-)

WeathermanShawn
June 2nd, 2010, 04:37 AM
Dopey:

So far, the main issue I see is that you are running Open-Loop..but you have not defaulted the MAF C2901-C2904..Since you have B0120 set to 4000 RPM, you are running Open-loop, but you are not running OLMAF or SD. Basically, it is impossible to do.

This is verified by the fact that you are still running off the High-Octane Spark Table.

You either need to default the MAF (SD-Tune), or set B0120 to 400 RPM (OLMAF), or set B4205 to stock Temp Values (Closed-Loop). I would imagine that your airflow values are jumping all over the place, and hence your Spark values are just as erratic. If you get the Tune parameters set correctly, you will find your fuel and spark values will fall in line.

Thats my take. Good luck..

Edit: Dopey..maybe its the new forum look or I just need clarification. Is this a Custom OS? That may require some further investigating (on my part) concerning closed-vs open-loop parameters. I.E., my answer is premised on a non-custom OS tune..If it is custom..then it I would have to rethink my review..

Thanks..

WeathermanShawn
June 2nd, 2010, 04:38 AM
Ha , already on another OS Android (phone) , so cant view his tune and log .
Will leave that to you , "grasshopper" . ;-)
Just so I dont get in trouble, grasshopper was what was used by a priest to describe a young student in Kung Fu TV series. :-)

Yea, I got you. I am old enough to remember the T.V. series..:grin:

voda1
June 2nd, 2010, 06:06 AM
Dopey:

This is verified by the fact that you are still running off the High-Octane Spark Table.

Thanks..[/I][/COLOR]

What info tells you that?

WeathermanShawn
June 2nd, 2010, 07:04 AM
813681378138

814181408139

I linked his DYNCYLAIR PID to both the High & Low Octane Chart.

His log shows he is running off his High-Octane Spark..I just do not know what type of Tune he is running. A Custom OP will let you do what he is trying to do..

Does not explain the KR. Could be airflow..fuel..and even summer IAT's. It it kinda all over the place.

Anybody else see anything?

voda1
June 2nd, 2010, 07:46 AM
I linked his DYNCYLAIR PID to both the High & Low Octane Chart.



How do you link them?

WeathermanShawn
June 2nd, 2010, 08:36 AM
In the EFILive Tune Tool, I check the following (see attachment)..

There are a few other rules..I think you want your Pids and Tune Units to be the same..I.E. Metric etc..There have been a few recent posts (Tips and tricks)..but that is how I do it.

In this case, DYNCYLAIR (g/s)..is how I tracked it..

joecar
June 2nd, 2010, 09:56 AM
I've been busy, but I'll catch up on this thread later tonite.

Dopey
June 2nd, 2010, 10:15 AM
wow! I definately did not expect this many replies so soon, haha. :hihi:

Sorry, I posted it up late last night and forgot to include the OS its COS3 01290003 I believe (hence I didnt think any of those MAF paramaters were applicable). But it would be music to my ears if it was something so simple :cucumber:

I am wanting to run OLSD and ditch the 02's (and keep the wideband in permanently). One thing I noticed, is I had set B3601 to 14.7 after reading about the AutoVE being skewed by the W02 having a different lambada value vs the tune, however I just re-opened it and it has gone back to 14.68 Im guessing this is the bit depth used to store this variable? I have checked that there are no instances that the commanded is Lambada in b3647 so it should not enter semi-OLSD.

I have re-attached the .ctd as a .tun file just in case any one else is having issues (I have changed the default program for .ctd to the Tune tool, let me know if anyone wants screen caps on how to do this in Win7).

*Edit*
A bit more info, I also thought it was not using the low spark tables, but I dont know why? Ambient temps were around ~16*C (60*F for some) during the log last night.
Also, I have disabled BKR by zeroing out B6212 so I dont think it has anything to do with that either... I have gone back to the base tune so many times, I may have left something out... Although I did get the tune done by a respected tuner here who has got nothing but rave reviews from others, so theoretically it should have the correct settings in the base tune.

WeathermanShawn
June 2nd, 2010, 10:23 AM
Alright, makes sense now with it being a Custom OS.

Forget about the MAF Parameters then. Your right, under these circumstances I do not believe they are applicable.

I'll look it over again this evening. May need an extra set of eyes on it to figure it out. I agree, it looks very unusual for KR with a relatively low spark being commanded.

Will get back to ya..

5.7ute
June 2nd, 2010, 10:50 AM
Spark does appear low for your set up.
Have you verified the knock with knock phones? Checked for the exhaust hitting the chassis?What are the road conditions like in your area?

Dopey
June 2nd, 2010, 11:48 AM
Roads were smooth, I was on a freeway as I dont like to do any tuning on suburban roads. I havent checked anything with knock phones, to be honest I wouldnt know where to get them...:doh2:

The knock always seems to be apparent when there is any hint of wheel slip, I dont know if this has anything to do with it.

I have replaced both knock sensors and looms (got oem delco ones from the states)

I do get the usual LS1 lifter noise. I cant see anywhere where the exhaust would be hitting anything, although I do get a backfire on back off when just the headers are installed (took the cats off to check the headers for slag around the 02 bungs). Just in case anyone was wondering.... it sounded great! :grin:

It just doesnt make sense as I can do a 0-100 run with 0 knock and then the next run in the same conditions is plagued with knock. :confused:

5.7ute
June 2nd, 2010, 12:14 PM
Very odd. I havent had wheelspin induce false knock but have got it from a bad road surface.
I would log some of the EST spark pids for traction control, torque limiting etc & see if they match what the KR pid is seeing. (Just in case)
If the knock is false you could try desensitize the knock sensors. Be very careful here though as if you desensitize them enough they wont pick up real knock. I use a mates knock phones whenever doing this to make sure.

WeathermanShawn
June 2nd, 2010, 12:24 PM
I'm with Mick..I have had the bad road surface do it..but wheel-spin? Anything is possible.

The question I have is short of employing the more sophisticated audio techniques to determine legitimate KR..have you tried the true and tested method. Take out some timing..or find some 98 octane gas. I know from your post, you say it is erratic.. :confused:.

This may sound ridiculous, but can you hear it pinging or knocking. My car is moderately loud, but if I roll the window down..turn off the radio..if I hear pinging (marbles or metallic sound) I log KR 99% of the time. I can hear it. Can you hear any pinging?

I desensitize the KR sensors in regard to recovery time. I increase 5-10 fold how quickly it recovers. If it is legitimate KR it comes back. That way it only takes your timing down for 2-4 seconds instead of 10 seconds or more.

Unless you have oil leaking into the CC chamber and are really carbon-ed up..Thats all I can think of for now..

joecar
June 2nd, 2010, 01:52 PM
You still need to get a MAF DTC otherwise it will run from the MAF (or not from the VE)... try an experiment: zero your B5001 MAF table and see what happens.

What injectors do you have...?

WeathermanShawn
June 2nd, 2010, 02:16 PM
You still need to get a MAF DTC otherwise it will run from the MAF (or not from the VE)... try an experiment: zero your B5001 MAF table and see what happens.

What injectors do you have...?

Well thats the ultimate question. Are you wanting MAF enabled or not?

Dopey
June 2nd, 2010, 03:04 PM
Thanks for the quick responses everyone :)

I have only ever put PULP 98 RON in her (even though she only gets 18L/100kms)

I cant say I can hear any pinging, but the exhaust is rather loud I will ask a mate to come for the next log session and see if he can hear anything... Im not confident enough to desensitize the knock sensor myself as I dont want to blow the engine up if I cant hear it pinging.

@5.7ute, good idea, I'll do a log on the way home from work and see if I can marry up the KR with any of the traction pids

@WeathermanShawn What kind of values are you putting in your recovery rates if you dont mind me asking?

I already thought about the carbon build up, so I ran the Subaru top end cleaner a couple of weeks back (kinda like SeaFoam). Half a can through the inlet manifold and the other half was sprayed in via the spark plug holes and left to soak overnight.... Strange thing was it barely smoked at all, just a slight grey tinge through the exhaust... I havent had the heads off, but the rocker assembly is squeeky clean (Castrol Edge 10w60) cant see any hint of carbon getting up there.

I have been trying the old fashioned way by pulling timing where ever I can find knock, but where do I stop? The car already struggles to spin the rears with 4.11's which should be a piece of cake (or so I read)... The most troublesome spot has already been lowered to 15 - 16 and since I have lowered that section, all around it starts to knock so the square is growing every time I log it :help2: (see attached)


Im begining to think it may be a stuffed lifter tripping the knock sensors, I cant find any PIDs to log the individual knock sensors.. Are we able to monitor them individually? I figure if its stronger on one compared to the other it would indicate if it is the front or rear of the engine..... Or maybe im just getting desperate? haha...

Dopey
June 2nd, 2010, 03:06 PM
wow you guys are quick ;)

@joecar I will try that, I would absolutely love it if it was that simple *fingers crossed* Injectors are stock (28.8lbs I believe)

@WeathermanShawn I am wanting to run completely no maf (OLSD)

WeathermanShawn
June 2nd, 2010, 03:33 PM
wow you guys are quick ;)

@joecar I will try that, I would absolutely love it if it was that simple *fingers crossed* Injectors are stock (28.8lbs I believe)

@WeathermanShawn I am wanting to run completely no maf (OLSD)

Dopey:

You are going to have to Default your MAF to run OLSD..That may be part of the problem. It is using MAF..and you might have been making all your changes to the VE Table..

I drastically increase my KR recovery Rate (see attachments). Just to be clear..It does not desensitize the amount..just 'quickens' its departure..if it is real the KR will come back immediately!

I have attached a log where I had a very brief KR event lugging it in 5th or 6th gear..I maximized the scale to display it..

Dopey
June 2nd, 2010, 04:19 PM
So what happens if the maf isnt installed? Sorry, left that part out as well ;)

Thanks for the screenies, I'll increase the recovery rates similar to yours to see if it gives the dreaded saw tooth pattern or just spikes here and there (hopefully indicating background noise or a false positive)

WeathermanShawn
June 2nd, 2010, 04:30 PM
Mmm..thats kinda a 'trick' question..

Seems like from a previous thread, that you still are not going to run SD as long as you have not failed it in the software.

Now, how it is calculating CYLAIR..no MAF Frequency..Thats an excellent question.

I like Joecar's suggestion. Zero out your MAF Table and see what happens. Sounds like it still pulls out a MAF airflow if that Table is active..:confused:.

Great question..

Edit: I agree, a lot of your 'KR' looks mechanical. One big spike..then nothing. I think I saw a couple of lower RPM 100% TPS that looked more real (sawtooth). The KR on my screenshot was real..I good hear some pinging when I lugged it on a hot day going up an incline. So, that gives you a pretty good idea of what 'real' KR look like on a scan..with a very fast Recovery Rate..

joecar
June 2nd, 2010, 05:34 PM
wow you guys are quick ;)

@joecar I will try that, I would absolutely love it if it was that simple *fingers crossed* Injectors are stock (28.8lbs I believe)

@WeathermanShawn I am wanting to run completely no maf (OLSD)When you zero the MAF table:
- if engine runs, then PCM was using VE table all along;
- if engine does not run, then PCM was using MAF all along;


So what happens if the maf isnt installed? Sorry, left that part out as well ;)

Thanks for the screenies, I'll increase the recovery rates similar to yours to see if it gives the dreaded saw tooth pattern or just spikes here and there (hopefully indicating background noise or a false positive)If MAF is not installed, this is what happens:
- PCM detects MAF failure and sets one of the MAF DTC's (P0101, P0102, or P0103);
- the presence of any of those 3 DTC's causes the PCM to ignore the MAF table and run exclusively from the VE table.

i.e. you are not in SD mode unless a MAF DTC is current.

You can also get the MAF to fail by editing the failure thresholds (e.g. setting C2901 to 1 Hz will fail the MAF).

joecar
June 2nd, 2010, 05:48 PM
The key is: when a MAF DTC is current, the PCM is ignoring the MAF and running solely from the VE table (i.e. SD).

Then if you disable long term trims, short term trims, closed loop, semi-open/closed loop, then you have OL.

Running both gives OLSD... when VE table is corrected, runs good like this (instant throttle response), but will most likely damage cats in long term usage (so replace with pipes)...

If you want to run cats, first correct VE (in OLSD) and/or MAF (in OLMAF) or both, and then run either CLSD or CLMAF... you want the front O2's to switch to allow the cats to function correctly (which prolongs their life)... the rear O2's will monitor the cat function/efficiency.

joecar
June 2nd, 2010, 05:52 PM
More info, see my profile, scroll down to where it says this:



PCM calculates airmass from:
- above B0120: MAF,
- below B0120: MAF (steady throttle/airflow) and/or VE (transient throttle/airflow),
- in SD (MAF-less): VE.

Dopey
June 3rd, 2010, 07:19 PM
Sorry for the late reply guys, I tried what joecar suggested and zeroed the MAF tables out and the car started up and ran fine...... :confused: So it looks like the person that did the tune before just had a different way of putting it into OLSD....

Just as a precaution I changed the tune and followed the AutoVE steps to put it into SD mode, but I still cant figure out what the knock is... Unfortunately its pooring today so I am not to keen to go out logging.... I will keep playing with the tune over the weekend and post up the results.

One thing I thought about was setting the timing table to 11 across the board (every cell) and seeing if I still get knock... Surely 11 is more than conservative...

Does anyone want to share a pic of their High octane spark map as a comparison? Im trying to figure out if my timing is just too aggressive?

Dopey
June 3rd, 2010, 09:26 PM
Another thought, how are people getting figures for B6240 - B6251? After looking through a few stock HSV (australian) tunes they seem to vary from one to the next...

WeathermanShawn
June 3rd, 2010, 11:50 PM
Another thought, how are people getting figures for B6240 - B6251? After looking through a few stock HSV (australian) tunes they seem to vary from one to the next...

Dopey, those are the ones I do not mess with..keep them stock. Perhaps you are asking which one is actually stock? Again, I leave those alone.

Here is my High and Low Octane Spark. Rarely get KR..only on very hot days..lugging engine etc. Thats on a 11.0:1 CR..91 Octane (Colorado-Western U.S. Highest Octane Gas).

Good idea to set 11 degrees across the board. That should help.

joecar
June 4th, 2010, 07:12 AM
Dopey, are you getting any MAF DTC's at all (i.e. any of P0101, P0102, P0103)...?