• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

Fuel/Spark tuning for LH 2.4/EZK with TunerPro!

Update, runs well, but is way rich in boost (9s). Should not be; on the same fuel table on other bins it stays in the 12s.

So obviously there is some constant or multiplier I missed. Hrmmm. Other than that, seems fine.

and this lc1 just makes me want to murder someone, the sensor is brand new and it keeps giving me the "timing error"... turn key off and on again and it's fine. Grrrrrrrr.
 
DNAsEqUeNcE said:
how far back in the stream is it?

you need that one at LEAST 3-3.5 feet
Click to expand...

Well, it's not far enough; it's in the stock o2 sensor bung because MrBill said "oh whatever it'll be fine" :-P

You know, thinking about it, HIS wideband is in the downpipe, definitely less than 3 feet and he has all sorts of problems :lol:

So, I was incorrect in my assumption that the super richness (in the 9s at WOT/boost, was in the 12s with the same bin and 016 maf) was caused by the recompiled 950 bin - i tried the 937 and 984 bins both and it was exactly the same. I guess changing to the 012 maf involves more than changing the table and the two multipliers/switch points immediately after the table. If it's funny, if anything I would have thought it would be lean. But at least that means that everything else I did worked! :)

Still seems kind of odd because the 950 bin was built for an 012 MAF if I am correct. Guess I need to continue my 984 disassembly effort to see what's different... Back to hunting through the matrix I guess... Still happy that it worked! :cool:
 
beepee said:
Still seems kind of odd because the 950 bin was built for an 012 MAF if I am correct. Guess I need to continue my 984 disassembly effort to see what's different... Back to hunting through the matrix I guess... Still happy that it worked! :cool:
Click to expand...

And bear in mind that the 950 box is a LH 2.4.2 !
 
I've found something in 984 bin that looks strange to me, and might be somehow related to immo function, as I think it is not present in other bins. In 950 bin these pins have following funcions:

P5.1 - unused, routed to where WOT switch should be connected on N/A, unused on 950
P3.2 - unused
P3.3 - unused
P3.4 - external watchdog, probably the same here
P4.1 - output, goes to CA3272 "C" input. Actually, on my ECU I don't have CA3272 populated, so it is unused too. But on 984 this chip is present, and as from pictures I have seen the PCB's are the same

Here is the code:
<pre>
cpl P3.4 ; Port 3.4 0199
code_19B: jnb P5.1, code_1C5 ; Port 5.1 019B
setb P3.2 ; Port 3.2 019E
lcall small_delay ; 01A0
setb P3.3 ; Port 3.3 01A3
lcall small_delay ; 01A5
setb P4.1 ; Port 4.1 01A8
lcall small_delay ; 01AA
lcall small_delay ; 01AD
clr P3.2 ; Port 3.2 01B0
lcall small_delay ; 01B2
clr P3.3 ; Port 3.3 01B5
lcall small_delay ; 01B7
clr P4.1 ; Port 4.1 01BA
cpl P3.4 ; Port 3.4 01BC
sjmp code_1C5 ; 01BE

small_delay: mov R5, #0x14 ; 01C0
loop: djnz R5, loop ; 01C2
ret ; 01C4


</pre>
And here is the gif animation of what it does:
984_pins.gif


The code gets executed depending on the state of green pin. The yellow is toggled, I don't know it's initial state, hence it is yellow, and just becomes black for a moment of toggle, i.e. flipping it's state from 1 to 0 or vice versa, depending on what it was before the toggle.
The red ones are red when set to 1 and white when set to 0
 
beepee said:
Well, it's not far enough; it's in the stock o2 sensor bung because MrBill said "oh whatever it'll be fine" :-P

You know, thinking about it, HIS wideband is in the downpipe, definitely less than 3 feet and he has all sorts of problems :lol:
Click to expand...

I didn't have any problems until I ran the car without the downpipe and sensor the detached - something broke in my LC-1 from being on for 10 minutes and not having a sensor attached :e-shrug:

ANDdddd mine is 3 feet - it's at inside radius of the lower bend. So blow me.
 
Last edited:
Did you change the injector constant? Here is what Ipdown says they are in these different bins. They are all the same in all the b230ft bin except the 984 and the 950 is different then the 8v Turbo bins.

B204FT/GT 2.0 16v turbo (950, 597 and 578 all the same) chips:
128, 128, 0, 10, 157

B230FT 2.3 8v turbo, 984 chip:
128, 80, 0, 9, 139

B230*T 2.3 8v turbo, 954 chip:
150, 128, 0, 9, 139

B230 2.3 8v N/A, 933 chip:
131, 128, 0, 15, 49

beepee said:
Update, runs well, but is way rich in boost (9s). Should not be; on the same fuel table on other bins it stays in the 12s.

So obviously there is some constant or multiplier I missed. Hrmmm. Other than that, seems fine.

and this lc1 just makes me want to murder someone, the sensor is brand new and it keeps giving me the "timing error"... turn key off and on again and it's fine. Grrrrrrrr.
Click to expand...
 
sbabbs said:
Did you change the injector constant? Here is what Ipdown says they are in these different bins. They are all the same in all the b230ft bin except the 984 and the 950 is different then the 8v Turbo bins.

B204FT/GT 2.0 16v turbo (950, 597 and 578 all the same) chips:
128, 128, 0, 10, 157

B230FT 2.3 8v turbo, 984 chip:
128, 80, 0, 9, 139

B230*T 2.3 8v turbo, 954 chip:
150, 128, 0, 9, 139

B230 2.3 8v N/A, 933 chip:
131, 128, 0, 15, 49
Click to expand...

Going to try again today, but the extreme richness seems tied to the recently installed 012 AMM... I tried changing the load axis today as well with no real result...

I even tried a stock bin expecting the 016 AMM and things still went stupid rich. I would have thought if anything it would be lean! :wtf:

Also, I've ran 2.4.2 and 2.4.4 bins with no issues IIRC.

When my money tree finally starts to bear fruit I will be looking into a SLC Pure Plus, it uses the same sensor. My plans right now include a holset, downpipe, and 3" exhaust so I probably won't move the sensor for now... It seemed to work quite well until all of this super-rich business. For now I will see if I can ghetto-fab some sort of heatsink for the sensor...

By the way, nice job on the immobilizer work, and excellent visual!
 
Turbosteen said:
I changed only the first injector constant to get the afrs right. Think something about 90 now (660cc injector on E85). Works like a charm so far.
Click to expand...

I did that and it worked great! But I don't want the swap from 016 to 012 AMM to add timing, but, I think I figured it out.

Changing the AMM table and constants works, but then the fuel map is all skewed. Original fuel map AMM axis went from 0-128. With the 012 AMM, things would end up super-rich. I took the base value (128), multiplied by 1.333, and got approx 171. Then divided 171/16 and decided to make the table step down in increments of ten, except the 15th cell which I made 11 (which makes the rest of the scale multiples of ten as well;-))

So basically here's what I have:

-950 bin compiled from ipdown's disassembly
-injector constants from 957 bin (b230ft)
-stock 012 AMM settings from 950 bin
-tuned fuel table from a 937 bin
-modified fuel table axis

and it runs like a charm! Also tried the disable checksum checking flag and it seems to work, so no more CEL! w00t!

AT 17psi and low 12 AFRs this f*cker slams you in your seat like a horny cheerleader :oogle:. yeeehaw! :-D Can't wait for holset!
 
Guys, this stuff is awesome. I know in some point of time I'll play with it. One question comes to my mind right now (maybe it was already posted - and I missed it) - is there a localized constant/map for e-fan control? Basically I'm thinking about lowering trigger threshold for low fan speed.
 
beepee said:
I did that and it worked great! But I don't want the swap from 016 to 012 AMM to add timing, but, I think I figured it out.

Changing the AMM table and constants works, but then the fuel map is all skewed. Original fuel map AMM axis went from 0-128. With the 012 AMM, things would end up super-rich. I took the base value (128), multiplied by 1.333, and got approx 171. Then divided 171/16 and decided to make the table step down in increments of ten, except the 15th cell which I made 11 (which makes the rest of the scale multiples of ten as well;-))

So basically here's what I have:

-950 bin compiled from ipdown's disassembly
-injector constants from 957 bin (b230ft)
-stock 012 AMM settings from 950 bin
-tuned fuel table from a 937 bin
-modified fuel table axis

and it runs like a charm! Also tried the disable checksum checking flag and it seems to work, so no more CEL! w00t!

AT 17psi and low 12 AFRs this f*cker slams you in your seat like a horny cheerleader :oogle:. yeeehaw! :-D Can't wait for holset!
Click to expand...

fun stuff there
 
DNAsEqUeNcE said:
Nice,

update OP With that special bin?
Click to expand...

You really can't accomplish anything with it than you can't accomplish in other bins. This one is pretty much a learning exercise.

I do hope to find the checksum flag in other bins and update the XDFs but i've been kinda burnt out on it this weekend, and am coming down with a cold today... and to top it off my dishwasher has instructed me that today is housecleaning day so... yeah. :e-shrug:

Edit: i'll post the ASM source and compiled bin i'm using, but i need to switch it over to a non-commercial fuel table first lest I become a giant hypocrite.
 
First big step. Im buying a laptop. Hope to join in on the fun soon. I have mikes chips, i can rewrite these correct?
 
You must log in or register to reply here.
Back
Top