Hi, I have a project that has come together and is running for the first time.
Car has an 88 turbo motor, adjustable fpr, 720cc x4 injectors, rtek 1.8

It starts pretty well and initially runs pretty smooth, but slowly starts to run worse, until it seems like it's missing or running on one rotor or something. It seems the more the engine warms, the worse it runs. My wideband gauge shows that it's running quite lean, it usually reads 22.4 which is the max lean reading. However I never get to see the ratio right at startup because the sensor is still heating. Because of this I can't really get it to idle to set timing properly or get it all the way warmed up. My fuel pressure is up around 60psi, I could go a bit higher but I feel like I have a fuel injection sensor that's out of range or giving inaccurate information to the ECU. Does anyone have an idea of what to test? I have heard of the coolant temp sensor causing problems like this but mine is new. TPS is adjusted, IAT sensor is also new... what else may cause a problem like this? Is there a good way to test the MAP sensor or AFM? Or would that be barking up the wrong tree?

Thanks for any help!

 

The water thermosensor, if out of whack, would more likely create a situation of too much fuel and not the opposite but you could take the reading at the ECU to verify it is okay if it would make you feel better about it. Pin 2I at the ECU houses the Green/White wire for the sensor. W/key to on and engine cold it will read 2 to 3 volts and the voltage drops to about .5 volts when the engine is fully warmed.

 

first, i wouldn't trust your wideband reading because i have never seen a rotary engine actually manage to want to stay running beyond around 17-18 AFR, maybe 20ish with a smog pump and light deceleration and that is off idle, idle requires even more fuel.

 

Great info guys, thank you. I was wondering that about the AFR myself. What might be the problem with my wideband setup then? I've checked for exhaust leaks and the ACV and split air passages are all blocked off. Sensor is mounted about 26" past the turbo as instructions state. If I rev the engine lightly, 2k to 3k or so, the wideband gauge will fluctuate a bit, down to 15-18 AFR but then comes right back up. I've calibrated it twice now, having that working would be a big help.

I replaced the o2 sensor, thinking the ECU isn't going into open loop, and that seemed to help things - I got it all the way warm and it will run well at 1500-2k. But I am pretty confident the motor is not getting enough fuel, even if I can't trust my wideband yet. The car won't start at 40psi, it tries but it's obvious there's not enough gas. And when I started it last I tried turning the fuel pressure down while it was fairly warm.. engine speed dropped and it got choppy like it was missing. It runs best between 70-80psi, what gives? I have checked fairly well for vacuum leaks but I could always check again.

It also seems like it's running a bit sluggish even when it's smooth, takes more pedal movement than I'd imagine it would to rev it. This is my first rotary car though so I could be off base. Does the ECU take time to learn and dial in the fuel curve, or is it already set and always outputs the same values?

Any suggestions you guys can throw out would be helpful. Thanks again.

 

maybe post a video, perhaps you're only running on one rotor.

 

And perhaps you have an issue w/the primary injectors.

 

stop tryin to steal my thunda!

lol, but yeah, sounds like possibly one of the primaries isn't kickin.

 

Yeah that would explain the lean reading... How would I test an injector? Can I just spin the CAS and listen for it to click with a stethoscope or something? I have a pair of stock injectors, unknown condition, that I could swap in there to test. I'll try to get a video tomorrow.

 

Spinning the CAS to listen for clicking works for the clicking test.

 

Keep in mind that unburnt fuel wont register on the wideband. Just sayin.

 

Ok, I've uploaded a video to youtube since attachment size limit is pretty low.
It doesn't seem like it's running on one rotor, I can get it to idle decent, but it seems to want way too much fuel as well as advance.

I am unsure what to think of the timing on this car. I decided to find TDC for myself earlier in the process of getting it to run and after viewing the apex seals in L and T holes of rotor 2 and marking the pulley, finding the center point, etc., I marked a spot that is a little over 20* advanced from the stock leading mark (so 15* advanced from the factory TDC). Is this normal? Which mark would you advise me to trust?

When it idles, around 1k rpm, that mark of mine (the 20* advanced one) is about the lowest timing mark I can dial it down to, and it is very shaky. It smooths out if I advance the timing more, it will take what looks like 50* of advance easily. If I jump a tooth on the CAS to try and retard it a little more, to try and get it back to the stock mark, it jumps way back and I can never really hit that sweet spot. Is there a way to lock the base timing so I can set it easier?

And still, it runs best (fastest, smoothest idle with most vacuum) around 70-80psi of fuel pressure. I can sometimes dial it down to 50ish, depending on where the timing is set, and it still runs fine but soon enough it starts to miss or run choppy like you may see in the video.

Also the car doesn't hot start very easily, and not at all if the fuel pressure isn't way high.
The car started easiest with the engine somewhat cold and the "base" timing around 10* (based off the stock mark), but then after it started I couldn't keep it steady for a little while - it would keep jumping down rpms, like from 2k to 800prm every few seconds and kinda be all over the place. That was right after I jumped the CAS back a tooth. Is this the ECU adjusting to timing changes?

I also noticed the ECU advancing timing to keep the engine running when I was trying to get it to idle lower. The advance would drop as RPM dropped, but then the computer would advance the timing a few degrees and the engine would speed back up and smooth out.

I'm not sure what to trust or how to tell if things are set right, besides how it sounds and vacuum readings and such. At best it pulls about 12psi of vacuum.

Ok I'll stop there, thanks for your help everybody!

watch?v=C0W1rX3ilqg&feature=youtu.be

 

cant find your video, link is bad

your description of having to turn up the fuel pressure is an indicator that the engine is in fact running lean. increasing fuel pressure just makes the injectors push in more fuel, but it isn't really the way you tune fuel curves.

your wideband may be working but i suspect it is reading leaner than the car is actually running at by at least a few points.

 

The link works for me, did you type in youtube.com/ before you pasted the link? I wasn't sure I'd be able to post links uet so I did it that way.

And yes that is my conclusion as well.. do you have any idea how to properly adjust the fuel mixture? Is the variable resistor worth messing with, or would you say it sounds like I have a sensor out of range?

 

If you're willing to depin the fuel injector's ECU wires you could switch the primary wires w/the secondary. This will cause the ECU to treat the secondary injectors as if they are the primaries.

 

the variable resistor only affects idle, it won't affect cruise or off idle lean issues. you may want to look up how to either pressure test or smoke test the intake for leaks, most common places for leaks would be a cracked TID tube at the turbo, if there's any cracks in the plastic it will run lean at all speeds.


edit: watching the video it sounds like it is running on one rotor, if you can make a video while you try to rev it i could confirm this better since the exhaust could just be making it sound that way. i'd suggest doing this first to confirm.

corrected link:
https://www.youtube.com/watch?v=C0W1...ature=youtu.be

the pulses sound too slow for both rotors to be firing and i only hear distinct pulses from one rotor when you brought the RPMs down.

 

Alright update: This seems to definitely be a timing/spark issue.

I started the car and revved it to make the video

and then after I shut the video off, I pulled rotor 1's leading plug wire to see if I could determine which rotor it's running on. Pulling that wire didn't do much, if anything, so I put it back and pulled #2. Same effect, maybe a little bit more pronounced. Then in a stroke of boldness I pulled #1 at the same time and the motor kept running on just fine. So it is running off of the trailing plugs, and maybe not even off both. By that point I was too shocked (literally and figuratively) to keep pulling wires so I shut it down.

The plugs are all new, I've pulled them to test for spark and it has always looked good and strong, I know my coils work, the wires are going to the correct plugs, etc. I am thinking this is a timing issue.

Before I started the car this time, I lined up the CAS and set the marks even with the 5*BTDC mark on the pulley, the stock leading mark. I am using the lower pickup in my CAS, I've marked the tooth that lines up with the lower (toward rear of car) pickup when the 2 marks on the bottom of the CAS are aligned and aligned that mark with the pickup while the leading pulley mark is even with the timing pin on the front cover. That's the best I know how to do. I can line it up with my found TDC mark if you want, but I'm not sure how accurate it is and it probably won't make much of a difference.

Is it possible the timing is just that far off? Like the timing is so advanced that the trailing plugs are the only ones to catch the mixture? Is it possible to assemble a rotary so that the marks on the pulley mean nothing?

Thanks for helping me out, I know it's a bunch of info.

 

Check for 12v at the leading coil

 

There is definitely 12v at the coil. The spark shocked me and I saw it arc out both sides of the coil when I pulled the wires. I have pulled both (actually all 4) plugs and grounded each and watched for spark that way as well, like I said it looks strong. I will test for 12v but I really don't think the coils are the problem, or the wires or the plugs for that matter. I could be wrong but that is the impression I get.

 

Confirmed 12v at coil

 

If you're not getting fuel then having spark isn't going to change things much.

 

But if I'm not igniting the mixture with the leading plugs, even the perfect mixture won't get it to run right. How is it that pulling the leading coil wires has little to no effect on how it runs? Is that normal?

EDIT: Like Fuhnortoner said, unburnt fuel won't register on the wideband. It may be getting plenty of fuel but not igniting it properly to give a proper reading.

 

Did you not state that when you check for spark you have it? Are you suggesting that you have spark when you manually check it, but when you screw the plugs into the rotor housing you suddenly don't have spark? Doesn't that seem more than odd?

 

pretty sure your problem is that youre only running on 1 rotor. Most likely either primary injector wiring or a clogged primary injector. Id check the injector wires and make sure you dont have them backwards.

If those look ok do this:

Relieve fuel pressure

disconnect all power from both coils

Take the UIM off

Unbolt the primary rail

Zip tie or wire tie the injectors to the primary rail

Turn the car to run and restore fuel pressure

With the injectors out of the block, have someone turn the car over and see if fuel is squirting out.

 

definitely only firing one rotor.

from here you have a 3 way fork to go:

1)check compression(though when you were cranking it did seem fairly even and had SOME compression on both rotors, a dead rotor or one with really low compression is easy to spot by the lumpy uneven cranking pulses)

2)double check your ignition to see which, if any plugs are not firing. new plugs doesn't guarantee that they are working, but youd have to be really unlucky and have 2 faulty ones on just one rotor, which is very unlikely unless they got oil soaked somehow. the coils arcing to ground also doesn't guarantee that they are arcing the plugs. fouled or faulty plugs will still not arc even if the coil is trying its damndest to get them to try.

and last but not least

3)check fuel by doing what satch recommended and depin the primaries and try starting the car on the secondary injectors with them wired into the primary pin locations to the ecu.

you can figure out which rotor is firing by disconnecting the trailing coil connectors underneath it then pop the lead plug wires off the leading coil with an insulated pair of plug pliers.


i have a strong feeling one of your primary injectors isn't turning over, it could be the main engine harness is crispy or it could be a stuck shut injector.

don't mess with the timing or even bother with that, if it was off even a small bit it wouldn't even run on 1 rotor.

another way i test to see which injector is dead is by spraying carb cleaner into a lower intake manifold nipple that goes into a intake runner to the rotor. the engine will either stall or it will immediately pick up and run like normal for a second. spraying carb cleaner into the intake duct doesn't always give you a positive result and it also doesn't tell you which rotor isn't running properly.

 

I guess I don't understand why pulling the leading plugs wouldn't kill the motor. On trying to start it the last time I got it to idle with the timing right at the 5*BTDC mark and the wideband was registering in the 19's. Still running rough as hell. But then smoke started coming out from behind my gauge cluster and soon enough my tach stopped working. I'm at the end of my rope with this, I feel like it's sapped every bit of me that I've put into it and given so little back.