Idle - Strange AFR values

Hello!

If I use the base fuel and ignition map according the road tuning course, the engine start and die after 2-3 seconds.

I success to Idle it just when I decrease the AFR values approx 4-6 or when I increase the VE table values to approx 160 - 180

I used thous strange values to idle and I locked the ignition to 10 degrees and check with light gun to fix trigger angle and trigger offset at around 1500 rpm.

I was needed to change it 1 teeth offset and 10 degrees trigger angle to get 10 degrees.

I thought the problem was fixed and return to the base fuel and ignition map from the course but still the results same.

All sensor values are good, MAP show when engine start values approx 30-40

I am without wideband, I am wait ,I need to get it this week, but just want to make the engine idle and the values are really strange.

What can it be?

Thanks!

Have you correctly set your values for injector flow? If the injector flow rate data is inaccurate then this will mean you need to compensate this in the VE table in order to achieve your target AFR. Also confirm the base fuel pressure is correct - You've set this to be MAP-referenced and the base fuel pressure to 303.9 kPa absolute which is around 200 kPa gauge pressure which is very low.

Yes the injector flow rate is correct according to the car manual and what is the fuel pump rate should need to be in units of absolute kpa?

if I know its 3 bar fuel pump and regulator

Thanks

Absolute pressure = gauge pressure + atmospheric pressure.

So if your base fuel pressure is 303 kpa absolute, it is the same as 203 kpa gauge + 100 kpa atmospheric pressure.

Your fuel pressure regulator works in gauge pressure, a 3 bar (300 kpa) regulator = 4 bar (400 kpa) absolute pressure

Yes the injector flow rate is correct according to the car manual and what is the fuel pump rate should need to be in units of absolute kpa?

if I know its 3 bar fuel pump and regulator

Thanks

Actually the fuel pressure was 404 abd i reduce it to 300 to see if there is influence but it seems that there is no.

So I dont think this is the problem but I change it for 404 abs kpa.

If there is more options to solution I am be glad to hear.

Thanks

Update:

I have been back to the base fuel and ignition tables from the road course and set the fuel pressure to 400 abs kpa.

The engine still not idle but I had noice that when I change the injector flow rate from 180 cc/min (from manual 45 cc/15 sec) to approx 15 to 30 esspacially on 20 cc/min the engine idle well and not die.

I am really dony know how to solve it the engine mechanic and timing are good and it idle and run properly with the factory ecu.

Can you post up your current calibration so we can have a look please? Also can you give us some details on the engine and any modifications you've made.

Ok, Its Suzuki Jimny G13BB engine totally stock.

I want to learn and get experiance on the stock engine and after that it will be 1.6 G16B engine with turnocharger but its not matter now.

I chose that because I know how this engine need to work properly and I build parallel wiring loom that I can disconnect all the stock ECU plugs from the sensor and connect the Haltech loom so I can test and driver the car and not need to park it a lot of time because I use it.

I must say that I didnt connect the idle air control valve because I didnt see option for it in the Haltech sprint 500 loom, and I am thinking I should compenstate on that with the compesation maps that I disable now for the first tuning step, should I need to disconnect it totally when use the haltech ECU or leave it connected and controled from the stock ECU?

One more thing is that I warm up the engine with the stock ecu and then plug the haltech so the tuning process always happens when the engine in working temperature.

I had read now a little bit on the injection firing angle and that is my next step to check because I thinks maybe it is cause the problem ,even that I try in the begining of the tune the stock Haltech values on this injection firing angle table and then change all values to zero and now I know it is not good.

I attache files of the current calibearion and I really want to solve that problem (notice that I disable every compensation table, prime table , dead time table because I dont know which values to set) so maybe one of them is important and I dont know that.

Thanks for help!

Somebody? Please...

I failed to solve that issue and I can't start the tuning process.

Can you please post your calibration on here, not just screen shots.

One thing that I'd check is the polarity of your trigger sensor. It's typical for most reluctor sensors to trigger on falling edge unless they are specifically wired in reverse polarity. That won't affect your idle fuelling but it will certainly affect your ignition timing and timing drift as rpm changes.

I had check the polarity with oscilloscope when I connected first the VR sensor so I am sure its not in reverse polarity.

Can you confirm that when you have the ECU powered up but the engine not running that the MAP sensor is reading 100 kPa (absolute) or 0 kPa (gauge) +/- about 3 kPa? You're not using a coolant temp correction table so this will mean that when the engine is cold you need to compensate through the VE table. This doesn't explain your issue however I'd suggest you add a correction table to your calibration.

I'm interested to know what the injection time being reported by the ECU is at idle when you have the engine running? Can you let me know that and the VE numbers you need to put in the base table in order to get the engine to idle properly please?

Yes, when the ECU powered up but the engine not running the MAP sensor is reading 100 kPa (absolute).

You will see that in the video.

According the course I disabled first all the compensation maps as you explained so its not influence for the first start-up and tune so I add that maps later, but now it is not relevant so much because I warmed up the engine before start up and try tune with Haltech ECU.

You will see that in the video.

https://www.youtube.com/watch?v=DDYRZKNTGgI&feature=youtu.be

Injection time and VE numbers - show in video

https://www.youtube.com/watch?v=Plk0MEJ6i1U&feature=youtu.be

Thank you very much!

I've run this past Haltech while I was at WTAC and have an answer for you. The 'semi-sequential' mode you're running uses two injection events per engine cycle instead of one. This means that it's providing twice the fuel and so to get somewhat realistic VE numbers you will need to halve the injector flow. If the dead time data is wrong you will also find this can have a larger influence on the VE numbers sine the dead time error will be applied twice.

Ok great I am so happy that the problem found!

So now what should I config in the injection mode to solve the problem if I use now just 36-1 crankshaft sensor.

Should I stay with semi-sequential and change the injector flow to half i.e from 180 cc/min to 90cc/min?

or change injection mode?

Thanks!!!

The option is really up to you. You can run in VE mode with the injector flow halved or run in injection time mode. Either option will work. In the VE mode the model is still likely to not be perfect since you don't have accurate injector dead time data.

With the trigger mode the engine runs you're limited to either semi-sequential or batch fired. Sequential would require a home signal.