Your Long Term fuel trim fluctuates between -10.2 and +12.5 at idle? That sounds crazy, it should be stable about 0. Can you correlate the fluctuation with fluctuations in the airflow measurement and with reported AFR? 1:10 AFR at idle is also crazy especially if LTFT is negative.

It sounds to me like you have a sensor that's producing erratic readings and the ECU can't make heads or tails of it. The key sensors would be MAF, intake temp, front 02, throttle position and eccentric shaft position.

Having a high AFR could also be caused by incomplete combustion, such as if a coil, spark wire or spark plug have failed or connected incorrectly. Not sure if we ever confirmed your ignition was in good shape.

Good idea to disable LPG for now and get the car to run right on normal fuel first.

 

Long Term Fuel Trim is constant 10.2 on idle. It's on 7% when I drive.
The AFR value that is calculated from the Lamda value is jumping between minimum 14.53 and 15.39, while the commanded AFR is 15.01

Today I disconnected my battery before I started, and after I reconnected it I started the car, and confirmed that the fuel trims were back to 0%, both LTFT and STFT. The start was like it was always, roughly 4 cranks and the engine started. It drove amazing, I felt the strong compression in the engine, I felt power, sharp response, it drove perfectly. Then roughly 8km later my LTFT started climbing back up to 10%, turned the car off, tried to restart, and we were back at the starting problems again. The engine is now warm, I stopped for 5 minuted, and disconnected the battery again to reset the ECU yet again. I will no go do some groceries nearby, and will see how it starts warm with a reset ECU.

10%+ percent LTFT should indicate according to this website: Fuel Trim?

 

Full Topic here: https://www.rx8club.com/series-i-tec...oo-lean-78270/

After I reset the ECU I had the problems of the Topic Starter to which 9krpmrx8 replied. Engine stalls, need to apply a bit of throttle to keep it running, very hard starts. The ECU seems to correct this by adding fuel so the engine runs on idle. It adds up to 10-13% on idle, and 7% in normal driving conditions. My symptoms are less severe, but it seems that there is a big vacuum leak. But it still beats me why the car has no problem starting after I reset the ECU.

My MAF was dirty, from the side where the air flows against, it was black, while the other side was nicely bronze transparent as it should be. A new MAF didn't solve the problem for the OP.

I still bet on a vacuum leak (which I need to find with smoke), or a failed plug/coil, but I will need to diagnose it, not guess.

 

Just took the car for a drive, I started the car with a reset ECU and STFT and LTFT still on 0%. Cold starts, half-warm starts and warm starts are no problem. Just went shopping, engine still warm, starts like I was used to it. Engine kept dying at idle. Then the LTFT started climbing from 0 to 12 again, and the starting got gradually worse and worse. But the car was now becoming able to keep the engine running at idle.

 

Oh good, so it is consistent. That's better

Agreed with the vacuum leak diagnosis. Loose hose somewhere or coupling somewhere after the MAF sensor. A common culprit is the little VFAD line that runs from underneath the throttle body to the VFAD solenoid beside the air box.

Coils wouldn't hurt in any case, if yours are approaching 25,000 miles or have unknown mileage.

 

Once I thought it didn't work, but that was because I didn't disconnect the battery long enough and the LTFT was still stored according to my DashCommand app, hehe. Really, once the ECU is reset, it's like it's reborn. It reminds me of my moped where I didn't have this needle in the carburator, so I needed to add like 20% of throttle so it would start, otherwise it would flood itself . Got a new needle a couple of days later in my mail.

VFAD Line, VFAD solenoid, will check it out. Where else can I look?
I tried taking off the airbox so I can check out all coils, but didn't manage to take the airbox out. Is there a trick to it? I already got it out of the 2 pins that are close to the bumper tho...

My LTFT is only 10% when the engine runs idle, and 7% when cruising over the highway, which according to this website means it's a vacuum leak at idle: http://www.aa1car.com/library/vacleak.htm

 

What is this exactly?




I looked around, and found this to be still connected, but looked like it wasn't connected correctly. I pushed it up, heard and felt a click. Still doesn't seem to be completely in, but it got in deeper.

Those are not my pics. My phone is a piece of garbage that uses flash, but disables it the moment it takes the picture, so I ended up with useless pics.



I also found this from the first LPG garage that failed to install a BRC LPG system (2nd garage did a good job and succeeded). They mentioned they used a wire to read the RPM, but said they fixed it nicely. This looks like complete crap to me.


Update:
15:45 - Was reading the fuel trims while I was fiddling with the VFAD line. When I open it, the STFT goes up. When I chocked it off with my thumb, it got quite lower, and got the LTFT as low as 1.6! It went back to 3.3, and after I restarted the car it went as high as 15,6% on LTFT, gradually dropping back to 7%. When the fans of the radiator started blowing I noticed that the LTFT went down pretty quickly (i think it also went up once). Maybe it's just something like that that cools the engine and corrects the LTFT in intervals? I am removing my airbox, finally did it, and I can check also the coils.

 

The coils look pretty good, but how can I see if they are the originals? This week I will change the plugs and see how that works out. Unfortunately I didn't have a multimeter to check the resistance of the coils, I will do it the next time.

 

Check the cranking RPM, and the starter cranks the engine between 220rpm and 240rpm. Cut off fuel supply and cranked for roughly 10 seconds to get these values.

According to: The Garage - For RX-8 specialist services and professional rotary engine help | Rotary Revs
So my starter is still good, as it falls within the 220-250rpm range. Would ordering a 2kw starter be advisable?

 

My battery died today. I cranked the engine and within a couple of seconds it dropped to clicking. I will recharge the battery today and see what that will do. Sigh...

 

I can't speak knowledgeably about LPG systems.
Imo, unless there is some practical reason that you must have it, I would convert it back to stock.
8's are a PITA to keep healthy under normal conditions.
You said earlier your battery was new?
It wouldn't surprise me if the installer jerry rigged sone wiring and you have a parasitic draw draining your battery, the alternator may be bad, or you have a bad ground connection somewhere.

 

The practical reason is, it's 1 euro per liter cheaper. In 10.000km I already saved 1400 euro. In another 20.000km it has paid for the entire car.

 

Today I disconnected the L1 coil and started the engine. It made no difference in starting performance or idling. I will test all coils this weekend, L1 seems dead.

 

iirc rennys idle on the trailing plugs

 

According to this video they seem to run on both Trailing and Leading (except for the one being dead ofcourse). I think we can believe Oklahoma Rotary .

 

Google rotor balance test.

 

I googled it, but what should it tell me?

 

How to do it.

 

Replaced coils with new ones. Car runs smoother, but starts remained the same. Due anyway as they have 80k miles on them.

Tested the battery at the local garage. Battery is charged 100%, Starting Performance 78%.
CCA listed on the battery is 540 and 62Ah.
Battery failed the test, presumably because it passes without load but fails under load.

Could it be that the startmotor pulls so much from the failed battery than the coils simply don't ignite?

EDIT: The CCA was rated at 420 according to the tester. Battery is under warranty and try to get it replaced by a stronger battery if they acknowledge the battery failure. According to the tester, it's a shorted cell.

 

80k on the coils???????? if you have a cat check it now. that battery isnt big enough and needs to be replaced. what starter is in the car?

 

Will check the cat once I have the opportunity. When I hit the gas I hear some rumbling like some small plastic pieces in a tube I would say (opposite to marbles in a can). I only hear that with the window open and doesn't come from the dash.

Not only is the battery not big enough, it even failed. No wonder why the starts were so slow at the beginning.

I still have the 1,3kw starter, I have a 2kw starter ready to be installed.

 

Today I'm going to install the new 2kw starter, and we will see how it starts then.

LTFT is still anywhere between 10-17%, (day to day).

 

So I replaced the original N3Z1 2kW starter with an Alanko 2kW. Yeah, I guessed it was the 1,3kw, but I was wrong. I was hesitant at first as they are seemingly both as powerful. Replaced the starter anyway.

With the orignal N3Z1, I got to 240 cranking rpm max. With the new starter I got up to 300rpm! No idea what kind of starter it really is by code tho. Costed me only 80 euro, and the question is how long it will last.

Furthermore I have replaced the ignition cables with those from NGK. They seem to have a resistance of 5,5 over 3,5 over the original ones, irrelevant, just a random fact.

So far:

Replaced all coils
Replaced all plugs
Replaced all cables
Replaced starter

Where it usually didn't start unless I gave a bit of throttle, or didn't start at all hot no matter what I did, the car starts now hot within 3 seconds without giving any throttle at all. The starts significantly improved!

However, I only masked the real issue with this. The problem that the car start "tough" still persists. The car drives fine after resetting the ECU when its cold, and the car started cold without any issue within 2 seconds. It's only after the engine starts to learn from the o2 sensor, LTFT increase, and from that moment on the car starts like total crap. With the new starter, I can atleast rely on it that I can restart the car and not push it across the intersection another time to get out of the way (seriously..).

I had to do the coils, plugs, cables and starter anyway, so there is no lost money in that.

Given that the car starts great after resetting the ECU, and starts poor as soon as it started learning the idle, gives away that its a problem with a faulty sensor which comes into play once the engine warms up and car gets into closed loop. So my first guess is the o2 sensor. It has been 80K for this sensor over 13 years, so I take it that it is toast, but I'd rather find out how I can test whether it is really the o2 sensor and not anything else.

As for another fun bonus, the speaker whine is completely gone it seams after I replaced the starter. Funny...

 

Hmm. What's your airflow g/sec at idle?

 

My car starts almost the same.... and it helps too if I press the throttle a bit. No LPG, replaced coils/wires/plugs with Bennetbuilt D585 set, and premixing now.

Sometimes on a cold start I get a bang from my exhaust (cat is empty)