CA18DET - Base setup help

[CamsX]

Hello, first time posting and first time attempt at tweaking an ECU. I'll try to explain my setup and where I'm currently at.

List of relevant parts:
ECU 23710 - 44F07 with Nistune Type 1 fitted
1993 Latinamerica 200SX with CA18DET - 4ports intake manifold - No O2 sensor (some sort of EDM version I think)
CP Forged pistons with 9.0:1 compression ratio
S15 Turbo @ 10psi
Tomei Poncams and Valve Springs
OEM MAF (Z32 MAF on the way)
Factory 370cc injectors
Evo8 560cc injectors

So I got Nistune earlier this month, and after installing it the car was running wonderful. Was finishing up the break-in mileage after the engine overhaul using the car with an H-Dev Stage 2 chip, and as soon as I got Nistune installed I simply made the entire fuel map more rich to compensate for the higher compression ratio. Not sure if this was the proper course of action, but as I said the car was running perfect.

As my targets go to whatever HP I can get @ 17psi on the S15 turbo, worried I would max out the stock injectors, I decided to go ahead with the Evo 8 injectors installation, and this is where I've hit a brick wall.

As I wrote to Matt by email yesterday I was having a hard time getting it to idle properly. As Matt recommended I wanted to adjust the Short Term Trim and the Long Term Trim, but I noticed that after "unchecking" the O2 Sensor Off option in Feedback Switch 1 I was unable to get a proper read, since my car doesn't have a factory 02 sensor. At this point I'm not sure if this can be done with the AFR from my Wideband or if I'm simply screwed. Consult view always show FTS @ 0% and no data for FTL.

Also as suggested, tried again with the 44F07 base map and after resizing the injectors it changed my TP value from 183 to 120. Reading a bit further on everything I could find over the internet I got the car to idle properly by decreasing all the values in the Min & Max Tpulse Width tables by 0.66 (As injector resize decreased the K Constant TP value also by 0.66), which worked OK.

I'm pretty sure one of my biggest issues right now is not knowing the proper latency for the injectors I installed. According to the Mitsubishi forums Evo8 Injectors values are something like this:

voltage ms
4.6875 3.312
7.03125 1.68
9.375 1.032
11.7188 0.672
14.0625 0.432
16.4062 0.264
18.6768 0.144

513cc/min

Played a little bit with the Latency values, trying anything from 430ms to 740ms, adjusting the K Constant accordingly to get car to idle at around 14.8afr (and even around 14.8afr @ 2000rpms), but it drives like sh!t. Super lean on cruise and 13~afr on load. Went to the Fuel Map to compensate like with the stock injectors, making the entire fuel table richer and readjusting the TP/Latency to get the idle closer to stoich, but not much success on the drive-ability aspect.

Heck, now the ECU is starting to detect knock while idling in my garage. Any ideas why this could be happening?

Again, no issues with stock injectors and a richer fuel map. Every suggestion or comment is welcomed!

[raddy]

hi there, adjust k to reach about 14:1 at cruise,not at idle. Adjust idle by ttp min,latency and map value.

[CamsX]

Hey Raddy, thank you very much for pointing out that the K needs to be set on cruise. Helped me a lot to get it driveable for a while.

Now my new (bosch) Z32 MAF has arrived and after installing it I'm having troubles getting things adjusted again.

I definitely need help setting up the basic stuff, since idle is @ 12AFR even with a 150ms latency. Kconstant is at 500, but I don't know how to adjust the TTPmin/max or the TP/Fuel Scales for the Z32 MAF.

I let nistune do this for me but I'm unsure they are adequate. If someone can take a quick look at it.

Thanks in advance.

[raddy]

dont play with idle for now,let it be as it is. Set latency to common value, 4D hex is good base value. Set K at cruise to jet afr of about 14:1. Then continue with load tune, aim for about 13:1 for boost transition and for about 11:1 at full boost. Check if TP scale is adequate, at full boost it should use last or one before last column (at right side of the map). Fiddle with K if needed. After that you should have rough fuel map. Be sure that ignition map is safe, I would say maximum 10deg for full boost and afterwards.

If idle is too rich with adequate latency,lower related map values to zero. If still to rich,enable O2 flag at idle area of map and go to negative values. If still rich,find map called ttp min,select all values and lower it in steps. Thats it. ca18det is very easy to tune,just need some practice and experience. Dont rely on base maps from forum too much,every engine and setup may require individual access. But to tune it correctly and completelly you still need adequate knowledge and experience. But again,ca18det is easy as cake to tune (and also to destroy:))

[CamsX]

Hi Raddy, thank you very much for your comments. Much appreciated.

I think I got the TP scale close, as my TP load at 6000rpms is around 109 and my max value in the scale is 120.

I have a couple of questions regarding your last post:

- You recommend not going over 10deg of timing. The last column values of my Timing Map are all around 16-20 (filtered Values), and I'm assuming more timing means a smaller numerical value, am I correct?
- What do you mean by "enable O2 flag at idle area of map and go to negative values"? My car doesn't have a factory O2 sensor, so I'll the tracing is done from the wideband.

A bit scared of the Destroy comment. Any safety precautions to consider?

Thanks again.

[raddy]

filtered timing values are real timing advance,so higher value mean more advance=more 'danger'. To dial afr safely you need safe ign.timing. I would use not more than value of 10 for full boost, 20 for boost of zero bar, 35-40 for off boost/low load cruise. After fueling is done,you can start to tune ignition with knock detection to find optimal full boost timing. Final result will be affected by used fuel mainly.
You can enable O2 feedback in fuel map regardless of oem lambda,only reason is to allow negative values in filtered fuel map that can help you to reduce injector open time. Its done by pressing O in map cells,check user manual for this.

[CamsX]

Thanks again for the pointers. Reduced the entire Ignition Timing Map (10° from stock) as you recommended and enabled O2 Feedback on the low portion cells similarly to the stock map.

Can a leaky injector(s) be the cause of my rich idle? My lowest Injection time at idle is 1.74, usually averaging 2.2 values, and can't get it above 12.8afr @ 1000~rpms. Funny thing it is fine for the first couple of minutes on Cold Crank, and after a while it gets very rich, even with 50ms injector latencies.

[raddy]

did you lowered map values to negative and/or lowered ttp min and still got too rich idle?

[CamsX]

yes, even with negative map values in the 16-17afrs it keeps correcting and enriching the mixture.

[raddy]

so decrease ttp min then...

[CamsX]

I've tried. But I can't get it to affect the idle towards a leaner mixture at all by reducing the values in the TTPmin table. If I increase them I can see the AFRs getting richer at the higher values.

I not really sure if I'm supposed to have the FTS value stucked at 0% and nothing registered on the FTL column. I'm thinking I might need to reset the ECU battery based memory (if it has one). If this doesn't work I'll take out the injectors and have them checked for leaks.

Unfortunately, I'm going out of the city and won't be able to test in the next 5 days.

[raddy]

fts and ftl doesnt mean anything here... Do you use stock fuel pressure?

[CamsX]

I was using stock fuel Pressure, but the last tests have been done with 43psi (standard fuel pressure for evo injectors) but not much improvement. @43psi on cold start, 380 Kconst, Latency 430 it is fine for while until it gets up to temp. Afterwards, when it goes to close loop I guess, it starts getting rich idle down to 12.5-13.3. The only way to idle at 14.7~afrs is with 160 Latency, but it becomes undrivable on this setting, with lots of cracks and pops from the exhaust.

I'll go back to stock fuel pressure to move up on the pulsewidth range.

I'm close to giving up and go back to stock injectors.

[louiswun]

Assume your Z32 AFM is in good condition. This method of tuning is only apply to low AFM signal voltage on idle on CA18DET. No factory O2 sensor connected.
Do this way only if you know how to tune your enigen.
Do this way only if you have a good fuel pump. only if you have no boost leak. and installed a re-circulate BOV (more accurate on injectors latency tuning).
Forget about the idle mixture at the begining of tuning.
Tune your K value on full throttle until it reach your target AFR at target boost (with safe timing first)
Tune your TP scale to suit your actual TP.
Tune your timing map
Tune your latency only by 100% throttle close right after mild boost, throttle close at around 3,000 rpm, pay attention to your AFR gauge while lifting up your right foot,
adjust until you see very short spike to mild rich then super lean (fuel cut). Those rich spike usually happen at the moment while idle switch from off to on.
Tune your VQ map AT LAST to bring your idle to proper mixture ratio, only tune it if you have done the K value and latency, don't try to adjust K value and latency for idle mixture, or it will ruin your previous tune, tune your idle mixture only at hot engine bay (warm AFM, warm intercooler pipe, warm hood) with hood closed.
(If you do not have O2 sensor signal feeding to your ECU, idle mixture (idle switch on) will richer than "idle switch off" at around 1 to 2 point of AFR , even if your AFM signal is the same. Injector pluse width cannot tell the idle switch extra fuel.)
idle mixture tuning usually need to tune 2nd and or 3rd column on your VQ map.
Check the AFM signal voltage while tuning your VQ map.
The first column on your VQ map is 1040mv, it will not affect you idle mixture. your ECU did not read the first column on your VQ map.
2nd column on your VQ map is 1120mv, if the AFM signal on idle is between 1040mv and 1120mv, it can be "tune". Your idle mixture will be richer if the the idle signal voltage is closer to 1040mv, the more it stay closer to 1040mv , the richer it will be.
If the idle signal voltage is closer to 1120mv, the less extra fuel add it will be.
Let say if your idle AFM signal is 1040mv and you have tried to tune column 2 (1120mv) to 14.7 AFR, once you have touch your throttle to idle switch off, it will instant goes leaner to 1 to 2 point of AFR.
Let say if your idle AFM signal is 1120mv and you have tried to tune column 2 (1120mv) to 14.7 AFR, once you have touch your throttle to idle switch off, you will see smooth AFR transition between idle switch on and off.
This extra fuel will disappear if you connect a good O2 sensor to your ECU.
As I know 44FXX ECU is anti-pollution with cat converter, it require O2 sensor.

For your reference, on my own CA18 with GTR35 factory 550cc injectors. 4 bar fuel pressure while engine off (Atmospheric pressure), K value is only 310.
Your EVO 7/8/9 injectors K value should be close to mine, try latency between 400 to 580.
I have just done a CA18 for emmission test passed on today. HCR32 AFM, Nismo 600cc injectors.
Full warm up (coolent temp 75C) cool hood, cool AFM, cool intercooler pipe, idle AFR 16
Full warm up (same coolent temp 75C, after 30 minutes of driving) warm hood, warm AFM, warm intercooler pipe, idle AFR 15.2
Full warm up (same coolent temp 75C, after 30 minutes of driving, 5 minutes of stand still idle) warm hood, warm AFM, warm intercooler pipe, idle AFR 14.7

[CamsX]

Thank you very much for your comments. Very detailed. I will definitely look into moving the idle MAF mV value closer to 1120 and see how it affects idle AFRs. By tweaking AAC, TPS and Throttle body stop screw I might be able to achieve this.

One question, why did you decide on having a 4 BAR base fuel pressure? Any particular benefit of such a high setting?

Cheers!