Page 1 of 2

suggest me a guideline please

Posted: Tue Dec 11, 2012 3:16 am
by alollini
=> engine related bolt on list :
IK22 Denso spark plugs
HKS RSR intake (red 2 layer wet filter element)
Spoon Throttle Body
T1R downpipe
AEM exhaust
GE8 intake manifold plenum
motul 0w20 300v racing oil / mugen oil filter element

Bisimoto level 2 camshaft
Heavy duty valve springs and titanium retainer (enable higher revs)

once we'll have solved my software problem (see other post in FlashPro Manager),

i don't want my car to run too lean or too rich

with bolt on i fear to be leaner.
however i suspect in some circumstances it runs too rich. (when floored, it accelerate less than pedal at 90%, sort of)

my first intention is to drive, datalog, and then draw conclusions from those datalogs.

Is AFM calibration the first step ?

can you suggest me a plan ?

Re: suggest me a guideline please

Posted: Tue Dec 11, 2012 10:26 am
by Spunkster
MAP based tuning would be much easier as a starting point.

Re: suggest me a guideline please

Posted: Tue Dec 11, 2012 1:51 pm
by alollini
thanks for getting back to me!

1/ So in the Calibration > Fuel tab i check "Use MAP fuel tables ?

2/ So now AFM flow calibration is completely useless ?
i started there because i read somewhere is important for cruising/fuel economy and part throttle
my car is mostly for daily use, will play on track less than 1% of time.

3/ I understand i want for idle and cruising/part throttle AFR equal 14.7 flat;
what AFR do i want when full throttle ?

4/ I see both "WOT lambda adjustment" tables all filled with same 12,38 number, is it too rich ? (i run the car NA)

5/ I read that MAP is "racing" calibration, or is it just better ?

I did several passes of drives 15mn, datalog it, and corrected the AFM Mass flow (g/s)
its curve still look smooth
i have still zero knock count

i have now :
when i plot X=AFM.v Y=S.TRIM the closed loop Mean is flatter and under +2% and -2%
when i plot X=AFM.v Y=L.TRIM the closed loop Mean is very flat and inside +1% and zero.

6/ should i keep this "corrected the AFM Mass flow (g/s)" table when starting with MAP ?

7/ is making fuel right with normal driving relevant or EVERYTHING should be made on the dyno ?

=> i wish to do as much as possible myself, doing it right, and shorten my time on the dyno to narrow to the required only.

PS: i use french fuel "SP95 E10" ( i think is equivalent to USA 91 octane and has 10% ethanol in the mix)

Re: suggest me a guideline please

Posted: Tue Dec 11, 2012 1:56 pm
by Spunkster
AFM tuning no longer matters if you tune via MAP.

Re: suggest me a guideline please

Posted: Wed Jul 17, 2013 2:33 pm
by alollini
it took some time but I unchecked AFM and fuel modification have an effect now.

anyway i find another inconsistency :

if i do an XY graph WOT checked
X = rpm
Y = A/F

I get two completely different A/F behaviors for the car : one seem to follow map fuel table (maybe too lean in some areas) around 13.5
and another one probably too rich. around 12.5

during pulls, the A:F seem to jump from one behavior to the other, randomly, and with a very sharp change. 12.5 seem to make big power loss.

I thought it was air conditioning, it is not.

i verified the car is at wot at all points (no closed loop) no s.trim.

i thought this was jumps from column 9 to 10 but it is not.

with those jumps, impossible to make a knolegable decision on fuel.

Re: suggest me a guideline please

Posted: Thu Jul 18, 2013 12:41 pm
by alollini
i'm disappointed :
on two pulls one is correct and have AF ok btw 13,1 and 13,4
the second one is dull, no power and the AF drops from 13 @4000rpm to 11@7000rpm

i see no reason :
same altitude, same temps, same hour, same map
the car is in the same column 9 of the table.
Ltrim stays the same
Strim zero

and fuel status is "open driving" in both.

maybe i have a failure somewhere in the fueling system of the car ???

Re: suggest me a guideline please

Posted: Thu Jul 18, 2013 1:41 pm
by Spunkster
When reporting problems you must post the calibration used as well as datalogs showing the problem.

Re: suggest me a guideline please

Posted: Sat Jul 20, 2013 10:30 pm
by alollini
here is a graph where i firs spotted the problem :
you see in the middle the A:F jumping down ?

does the lambda overheat ?
I think i run in fact too lean... anyway i dont trust measurements because there are two trends instead of one.

see this xy graph showing at WOT A:F vs rpm
i upload my calibration and datalog
good idle
good low cam
(10.32 KiB) Downloaded 330 times
altitude from 1100m to 700m
exterior temperature 31°C
(1.35 MiB) Downloaded 317 times

Re: suggest me a guideline please

Posted: Sat Jul 20, 2013 11:47 pm
by alollini
here is a much better datalog

there are two 3rd gear pulls, with a cooler weather ambient temp 19°C

and two 3 rd gear pulls on a rather flat road, one reads lean the second looks rich ???
(1.12 MiB) Downloaded 325 times

Re: suggest me a guideline please

Posted: Mon Jul 22, 2013 1:14 pm
by Spunkster
In the second run you can see that when the K.Control goes to 100 that is when the additional fuel is being added. YOu may want to try to reduce your timing in this area as it is trying to keep the engine running safely.

Also the intake you ahve is known to be the most troublesome we have ever tested with. You may be better off using the stock intake instead.

At part throttle due to the interior diameter it can cause a reversion of air and disrupt the AFM readings.

Re: suggest me a guideline please

Posted: Tue Jul 23, 2013 7:13 am
by alollini
Thank you very much for this precious information.

- I will now monitor K.Control ===> is it I need to reduce advance, or knock sensitivity due to my camshaft ?

- I have unchecked "use AFM" ===> this should end the problems of the intake ?

If not then i will go back to stock intake.

Re: suggest me a guideline please

Posted: Tue Jul 23, 2013 2:19 pm
by Spunkster
You may have to do both as you may be getting valve train noise picked up, but not registering as knock.

Not using the AFM may help with the intake you have, but it is not a good design to begin with. Stock might actually be better for you. This is something you can test back to back.

Re: suggest me a guideline please

Posted: Tue Jul 23, 2013 3:16 pm
by alollini
in the area where i see K.Control go to 100% (it always starts at 55% ...)

in what increment should i reduce advance and knock sensitivity (up = reduce sensitivity) ?

advance -0.5° ?

sensitivity +1% ?

do i reduce only the specific -blue- field, or a group of 4 fields -blue and white near fields- ?

Yes I will go back to stock on the intake, in september.
The reason why i changed the intake in the first place was for look and noise.
At that time i thought the stock one was "restrictive".
Now i understand ithis HKS is bad design and suck hot air = all bad.

Re: suggest me a guideline please

Posted: Tue Jul 23, 2013 3:28 pm
by Spunkster

K.Control adding fuel...

Posted: Wed Jul 24, 2013 12:57 am
by alollini
i'm on it,
but i have not understood enough to explain it to the tuner properly (things are similar on his cosworth, but strategies looks different)

sure he decreased advance in most of the main ignition tables.

i thought the knock control tables were showing their effect by making "knock retard" so the "add fuel" behavior was a surprise.
These are our tuning recommendations:

• Main ignition tables - decrease the ignition timing in the main (MBT) ignition tables at rpm and load points where the knock ignition limit tables are negative. This will then allow us to use the factory knock control more effectively.
- should i decrease by exactly the negative number in degrees ?
- the tuner already decreased main timing and observed the torque going up.
• Knock ignition limit - use low values where the engine is likely to knock.Note that you should always adjust the main ignition tables if increasing values in the knock ignition limit table.
this has not been modified from the "stock equiv"
• Knock sensitivity - increase the knock sensitivity (reduce the values) in columns 4 and above. For engines with noisy cams, headers etc, you may need to reduce the sensitivity (increase the values).
I have spotted a square from 1750 to 2500 rpm and from column 6 to 10 where K.Control goes up.
I decreased advance of 0.5° and increased number for sensitivity by 1% in that area.
plan to do it once more if i see no change.
• Knock retard - use a smaller knock retard than stock - 4 to 10 degrees. The lower columns (6 and below) will have little effect on knock retard as the knock ignition limit should be much higher than the MBT ignition timing in these areas.
those tables were containing a flat number = 10
I have tried 6 and now it is 4.
The fuel recommended for the CRZ stock in France is SP95
one month before doing the camshaft I started using only SP98 with is a little superior octane and is known for offering six more degrees of advance freedom. I did that as a security measure. Even if i don't need more advance i think better fuel is ... better ?
also i try to get it always at the same station, except when traveling.

I still see some knock retard in datalogs, up to 11° from 1000 to 3000 rpm and none above.
This retard seem to be less often at WOT.

Throttle tip in retard (gears)
i looked if it is near throttle tip in or release, and i don't think it is.
My up-shift is much better now that i have "-100" in the tip in retard for 2nd gear up. do you think that can trigger K.Control too ?