What caused this low rpm knock ?

[wesam]

Could any body help me with this low rpm knock ?
it seems to happen frequently at 2180-2200 rpm

[WeathermanShawn]

You have several issues.

1. You are attempting to tune with incredibly high IAT's and your ECT's normal temperature are well over 200F. I understand your climate is hot, but fuel tends to detonate very easy when you have such incredibly high DYNAIRTMPS. You need to lower both your ECT (thermostat) and IAT's when attempting to tune.

2. I don't understand your B3605. Why are you commanding such lean AFR's, especially when it is hot.

3. Your actual AFR's (wideband) are way too lean. I would assume Table B0101 needs re-worked.

4. Why not run COS3. You are a perfect candidate for Trims. From your post(s) you have been attempting to tune AUTOVE for several months. I would not risk running so much KR on that engine. Do you have to run OLSD?

Overall, you are just way too lean...

[wesam]

WeathermanShawn I did not touch the commanded AFR when in open loop i think my tuner put it the way it is
you stated that i'm too lean in which map you noticed this because when i'm doing the AutoVe i check each time the Wide band map and try to reach as close as possible to 14.63 and it looks good now
what do you think i should fix first ?

[WeathermanShawn]

WeathermanShawn I did not touch the commanded AFR when in open loop i think my tuner put it the way it is
you stated that i'm too lean in which map you noticed this because when i'm doing the AutoVe i check each time the Wide band map and try to reach as close as possible to 14.63 and it looks good now
what do you think i should fix first ?

First, Wesam don't take my opinions or criticisms to personal. Tuning is not easy. OLSD is very challenging to get correct on the first few attempts. So, here is my best analysis.

Note: I am assuming your Wideband AFR (serial) is as indicated on your log..

83308331

Attachment (1): You are several AFR points richer than Commanded AFR. That is a very significant offset.

Attachment (2): Your Non-PE Commanded AFR vs Actual. You are several points leaner than Commanded AFR.

I would ask the EFILive Tuning community for their opinion(s). It just looks like your overall B0101 is not correct. Whether the Table has been overly smoothed, or not filtered as accurately as possible..it is hard to say. When I look at your overall AFRBENS..they really seem way off.

.8332

Perhaps others can suggest whether going to a COS3 would be better for you. Its harder to adjust for temperature bias without a COS, and you can utilize ST Trims with some of the newer COS's.

I suppose if you stay with your current tune, what about just using the Commanded AFR Values in the AUTOVE. Perhaps you have had conflicting advice..so let hear from others.

[wesam]

ok i will use the values of the B3605 in the AUTOVE
and also i want to know what advantages i will have if i go to COS3 over COS1 ?
do i need to tune every thing from scratch ?
sorry for th noobe Questions

[WeathermanShawn]

Wesam:

I'm kinda hoping 5.7ute or Joecar or others will jump in. Everybody, including me have been a little busy lately.

I believe COS3 will allow you much greater flexibility to adjust how the VE Table values react to ambient temperatures. Especially in your climate. I know it also allows you to run dual-spark tables. I'm not sure if COS1 does or not.

For my perspective, if you were to utilize Trims for several runs it would help. You could use those to tweak your VE Table. You can always go back to straight open-loop at any point.

So, stand-by and we will see if we can get a COS3 expert to help you out...

[wesam]

one last question until joecar and ute5.7 step in
my B3605 data is shown in units of lamda how could i change the data units to AFR or EQV as in the picture you sent ?

[WeathermanShawn]

one last question until joecar and ute5.7 step in
my B3605 data is shown in units of lamda how could i change the data units to AFR or EQV as in the picture you sent ?

In the Tune Tool..under EDIT..Properties.

There is a pop-up that lets you change units..:)..

[wesam]

:) i tried this today but did't closed the tune file so the changes come effective
thanks Seani will do the commanded fuel in open loop as in the autove tutorial and waiting for others to step in

[wesam]

I changed the B3605 as in the AUTOVE tutorial and took a short run
ummm for some reason the car is running rich now but it seem that the knock is gone and the AFR is steadier
what that mean ?
i inserted the new log file

[WeathermanShawn]

Cooler IAT's, Cooler ECT's..richer B3605.

Probably all good things, but still indicates air temperature is probably effecting your fueling.

Seems to be going in the right direction though..:)

[joecar]

I've been a bit busy, I'll take a closer look later today...

Just make sure you're sufficiently rich at any significant load, something like 12.6 at WOT.

For AutoVE make sure you verify failed MAF (DTC present) and OL (LTFT's zero, HO2Sx1 voltages don't switch, no STFT trimming).

COS3 has the IAT VE correction table which I think COS1 doesn't have.

[wesam]

So i should re tune the VE again now :(
Joe I will not go WOT at the moment waiting for you to have a closer look at the tune
here is my last tune file after changing B3605

[WeathermanShawn]

wesam:

Just for the record..Do you physically still have a MAF and narrow-band sensors? More curious whether you have the O2's. I would be curious to see how this tune might Trim with a COS3 at stoich (non-PE portions)..

[wesam]

No i don't have MAF its phisicaly removed
for the O2 sensors i just have one because i put the WB sensor in the place of the other O2 sensor
for the O2 there are disabled

[joecar]

You might want to recalculate B4001 IFR table... it doesn't look right.

[wesam]

Now after i put the CFOL as in the tutorial i noticed when i start the car even when the ECT is over 80* the Commanded AFR start at 12 then climb to 14.5 why it not start directly at the commanded AFR in B3605 ?

[joecar]

Now after i put the CFOL as in the tutorial i noticed when i start the car even when the ECT is over 80* the Commanded AFR start at 12 then climb to 14.5 why it not start directly at the commanded AFR in B3605 ?I noticed this when going MAF-less until it either warms up or has been running for the minimum time... the COS's have a table to correct for this.

[wesam]

I noticed this when going MAF-less until it either warms up or has been running for the minimum time... the COS's have a table to correct for this.

which table is this ?

[joecar]

COS3 has table A0008.

COS1 doesn't have that table.

[wesam]

it seems that the only way to solve all my problems is to install COS3
to solve the air temperature problems
any hint before i install it ?
should i put a 3 bar MAP sensor or the 1 bar will work ?

[wesam]

Ok i installed the COS3 successfully now i want to know how could i know the correct multiplier for the IAT VE multiplier ?

[joecar]

You mean the ECT VE multiplier...? Log with wideband during warmup, and create a BEN map to match A0008, and apply to A0008 (no filtering required since throttle will be steady state at closed)...

I don't remember, but I seem to recall that A0008 may be inversely defined... if that is the case, the the map would be 1/BEN (i.e. make a new calc pid).

[joecar]

If you are running boost then yes you have to run a 2-bar or 3-bar MAP sensor (and edit the scaler to match).

If you are NA, then the 1-bar MAP sensor is all you need.

[wesam]

You mean the ECT VE multiplier...? Log with wideband during warmup, and create a BEN map to match A0008, and apply to A0008 (no filtering required since throttle will be steady state at closed)...

I don't remember, but I seem to recall that A0008 may be inversely defined... if that is the case, the the map would be 1/BEN (i.e. make a new calc pid).

No i meant the IAT VE multiplier A0014 which will multipli the numbers in the VE table by a multiplier so the car will have constant AFR if the IAT changes

[joecar]

oh, if you can find a means of varying the actual air temperature, then you can use a BEN map... the trick here is how to manipulate actual air temp...

or log at different times during the day where air temp is different... build up accumulated BEN data.

[wesam]

Joe i have just one last problem
i have repeated engine knock after 2170 rpm
i don't know why
the AFR is good in the range of 14.5 and it seems that its not ECT or IAT dependent

[WeathermanShawn]

Big cam. Lot of CR and DCR.

Have you tuned for spark (dyno)? Try running off Low-octane Spark and see if you get any. High CR engines usually do not need much Spark Advance.

[wesam]

From which table could i retard the spark when i view the log fill i could not find the specified table which will give me the same spark in the log even when i calculate the multipliers
could any one look in my tune file

[WeathermanShawn]

Just take your High-Octane Spark Table B5913 and subtract -3 (degrees) from every cell from .32 g/cyl -1.20 g/cyl (see attachment).

Do exactly the same for your Low-Octane Spark. You always want to keep a good spread between High and Low Octane (in case of KR).

Save tune, re-flash. Now your High-Octane Spark is -3 degrees lower where you need it. If you still get KR, you can adjust it. If no KR, case-closed...

[joecar]

Wesam, do have a log showing that knock...

[wesam]

here is it
it was in the first page actually

[WeathermanShawn]

Odd..Sometimes you are showing KR when you take the foot off the pedal..(Frame 195-200).

Do you hear anything clanging? Metallic sounds, exhaust banging, etc. We may have asked before, but I was sure you had access to some High Octane Gas..

[joecar]

Can you also log DYNCYLAIR_DMA or DYNCYLAIR.

It looks like something is loose... each instance looks like a single intense event is detected, does not necessarily appear to be knock... can you hear it when it happens...?

[wesam]

No i don't hear it and the weird thing that it will not increase when i increase the throttle
i will log the DYNCYLAIR_DMA , DYNCYLAIR but could you tell me what those two PIDs will show ?
also does my spark table look high ? I'm running 97 octane fuel

[WeathermanShawn]

The DYNCYLAIR Pids are Airflow Pids that help determine your Spark Load among other things.

Wesam, it really does not look like the classic 'sawtooth' KR you normally see with 'real' KR. Seems like on one of your logs, I did seem timing being mysteriously pulled. I assume that was 'burst knock'. But in order to not complicate matters, I would leave that to another discussion. There are tons of threads on 'burst knock' on this forum you can search.

On 97 Octane you ought to be able to run the Spark Table you have (see attachment). I run 11.01:1 CR heads..best gas in this part of the U.S. is 91 Octane. In summer with the higher IATS I run a very conservative High-Octane spark. If our fueling is right, high CR heads do not need a lot of WOT Spark. I have included mine in an attachment for comparison.

[wesam]

ok here is another log with DYNCYLAIR_DMA selected
and after retarding the high and low octan tables by 2.5 degrees in this area
and still the knock is happening some times !!

[WeathermanShawn]

Very odd.

I don't think it is detonation KR. Most likely something is slipping or making a noise.

What year Knock Sensors do you have..Also, wesam what gear are you in when you get the most KR. Is it any of the Overdrive gears?

[wesam]

i have 2003 LS1 knock sensors
and it happened in the 3rd gear mostly
i don't feel that there is any noise also AC turned off
i have new stock engine mounts so there is no vibration

[WeathermanShawn]

I don't think I have ever seen KR like that..You run 97 Octane Gas and you lowered your Spark.

Wesam, your car and tune are very different, especially with your conversion. So, I don't know if it is a software issue (copies not filed over) or what. I am in no position to tell you to ignore it.

I know Joecar has talk about sophisticated techniques like utilizing an oscilloscope to determine knock. My opinion at this time is that it does not appear to be detonation but mechanical. Whether that is nothing or something is beyond my tuning knowledge at this time.

I guess I would continue working on the rest of your tune and investigate further..

[wesam]

Thanks Shawn i appreciate your help
some body told me today some big cams could make this knock at low RPM and it could be fixed by give it more timing in the low RPM where the knock is occurred
is this possible ?

[WeathermanShawn]

A lot of cam timing and spark research is a little over my head at this time. My cam has 'only' 10 degrees of overlap, so I suppose as you get into more and more overlap the possibility of an 'incomplete' fuel burn and KR at low loads might exists.

I carry a lot of timing (up to 44 degrees) at very low loads as my engine seems to like it. You might experiment, or confer with an engineer. My opinion is that at loads >.32 g/cyl I become more worried about KR. From your log, yours does not resemble any KR I have ever seen. But, wesam..check it out. Its just an opinion at this point.

Regards..

[wesam]

Shawn thanks again
my cam has 12.5 degree of overlap
i hope that joecar will have a glance before i will give it a try today
i will try to advance the area from .44 to .56 which seems to have the knock by two degrees then i will smooth it with the adjacent cells
i will report back

[joecar]

I can't see anything (load is low, timing is reasonable) that would be responsible for knock/detonation... it seems to coincide with the wideband AFR jumping rich... I have heard of rich knock, I can't tell if this is what you're seeing...

I take it you can't hear it knocking.


Thanks Shawn i appreciate your help
some body told me today some big cams could make this knock at low RPM and it could be fixed by give it more timing in the low RPM where the knock is occurred
is this possible ?Sounds plausible, give this a try.


Shawn thanks again
my cam has 12.5 degree of overlap
i hope that joecar will have a glance before i will give it a try today
i will try to advance the area from .44 to .56 which seems to have the knock by two degrees then i will smooth it with the adjacent cells
i will report backYes, try this, let us know.

I'm sorry, I seem to be running backwards today (actually it's now tomorrow already)...

[WeathermanShawn]

I wonder if that is 'rich' knock. It does seem to coincide with excessive fuel.

Wesam, you notice from Joecar's screen-shot you were at 12.6 AFR when you develop KR. That is a lot of fuel for such a low RPM and TPS. Why are you so rich there?

In another thread a tuner had very significant KR at Idle due to excessive fueling. I wonder if the engine bogs..and that sets off the KR sensors? You might have to dig through 10 pages of thread, but here is his dilemma and how leaning it out solve his KR..http://forum.efilive.com/showthread.php?13979-how-do-i-get-my-car-not-to-idle-so-high

[wesam]

ok i will try to lean this area before i give it more timing and will give it a try
i don't like to change to many things at once

[wesam]

Report back
after leaning it a little it seems the knock is gone
thanks Joecar and Shawn i'm happy to be among knowledgeable people like you
i hope the weather will come better so i will give my VE the final touches

[joecar]

Glad it worked... :cheers:... post a new log.

[wesam]

Well... its not completely solved because the amount of knock is decreased a lot but still exist
here is a log
i noticed that the AFR is perfect (frame 3045) but the spark is high i don't know why it appear as 29.5 while i commanded 20.1 i tried to find what multiplier caused this in all tables without lock
8515

[WeathermanShawn]

wesam:

I think your conversion may be causing you to run off the wrong Spark Table. I swear it looks like you are running off your B5932 Base Spark in Gear.

I wonder if the wrong Table got copied over? No wonder you have KR. Try lowering B5932 to the exact same values as your High-Octane.

That is what it looks like it is doing. It is certainly not running off of your High or Low Octane Spark...Why it would do this is beyond me???

[Abdulla]

Wesam, try to take one of the spark plugs, clean it, try if you can see the piston surface if coated with carbon (or the manifold or your throttle body's plate backside), put the plug back, run your engine for some time, re-examine your plug for carbon deposites. It could be excessive exhaust gas reversion due to an over advanced ICL cam grind + overlap (or wrong degreeing of camshaft during installation- over advanced)

Generally max exhaust gas reversion occurs just below @ <3000 rpm for most of cams grinds intended for high rpm range combined with large exhaust pipework " poor scavenging @ low rpm", the reversion will heat up your intake charge excessively = higher tendency to knock at low rpm

Give it a try unless you are 100% sure it is tuning problem, It may save you the time & frustration and let you concentrate on your tuning

When this knock started to happen? after specific mechanical upgrade or tune upgrade? How is your engine temprature rise rate at those rpm levels?