Have a problem that arose while driving around normally. The car has a RR stage 3 kit, headers, the normal stuff. not sure if that even matters. I did reach out to RR. No response as of yet.

The car barely runs at idle and is extremely hard to start. (it will run fine over ~1000rpm) and is full rich in bank2. I CANNOT clear that code P2198 no matter what I do. I replaced both O2 sensors twice with NTK sensors the wiring is fine because the sensors give readings. I replaced the port injectors thinking maybe E85 gummed one open. I disconnect the battery, clear codes with TIS, etc. It's there to stay, I can't get rid of it, even with the car not started.

I have found a few threads where the GS said a new ECU was required to fix this problem. Nothing with IS-F other than this code popping up with headers, which I have.

Please watch the videos before responding.

Per TIS I am either at:

a) needing a new ecu
b) needing to replace one of the direct injectors on that side (possible that running E85 fouled one of them?)

Here are the videos.

 

I read that it might have something to do with the fuel pressure. Did you try checking that too?

 

low pressure is set to 70psi from day 1 with the SC kit. high pressure reading around 1700 at idle in techstream. seems fine.

 

I had a Current Circuit Low but with Signal Stuck Lean. For me it was that the wire extension that came with the headers had melted a little. Not sure if that could be the problem but thought id share just in case.

 

I did inspect that passenger side pretty well and it was far away from the headers. I'll try to bypass the extension and plug it in directly. Tight fit with the headers in there.

 

My high side fluctuates, I see around 550-600psi at idle and 1700-1900psi at wide open throttle

I just had to replace mine I was getting a low pumping circuit code though(P2238, P2241), they were also around 2 years old. If you have equal length its extra tight, but not impossible to do without removing anything but the splash shield. Visually they looked fine, but once I got them out and next to new ones I could really tell how brittle they had become.

you can get them from PPE, or
I went with these: Extension for Toyota & Subaru Oxygen Sensor, 90980-10868 & 90980-10869

 

Cool, will definitely check that side extension first. Hopefully when I plug an O2 sensor directly into the original plug I will be able to clear that code.

 

What extension are you guys running? The primary or secondary cats? I have PPE headers and don’t have an extension on my secondaries. I’ve never checked my primaries.

I’ve seen a few posts about only running OEM O2 sensors. These cars don’t like anything else. They are cheapest on rockauto.

 

Primaries, otherwise the sensors won't reach. I've been running NTK for a year and a half and they have been perfect, then this happened.

 

Interesting. My info comes directly from PPE. I’ve had some CEL issues, so they gave me some tips. They said only use OEM sensors. YMMV.

 

I would definitely check the harnesses on both sides and sensors for good measure. Just my suggestion but instead of NTK sensors go with the oem Denso ones instead if you decide to replace them(got mine on rockauto). Also since you use e85 and I’ve seen what it does to injectors, lines, fuel tank, etc and I would just send them out to get cleaned and flow tested. If all that checks out I would say at that point it’s fair to say it’s the ECU but at least you tried everything else first.

 

This. I said above Bosch, but I meant Denso. Corrected for accuracy. Sorry.

 

Way back in the day I had a 1980 Honda 750. It came from the factory with NGK sparkplugs and ran perfectly. I put ND plugs in it - same heat range, same everything, just different manufacturer. They lasted 600 miles, and then the engine would not run over 6k rpm. Put NGK back in and never had a problem again. I have no idea why this happened. Ever since, I've been loathe to run plugs or O2 sensors from sources other than OE sources. When redline is at 9500 and you only get to 6000, you can imagine how gutless the bike was. Never wanted to have that happen again.

I've done a lot, and I mean a whole lot, of study on how O2 sensors work. I participated in a project called DIY-Wideband probably 15 years ago, and learned how wideband sensors work (which also means you get to fully understand how a narrowband sensor works at the molecular level). The current used to detect the relative oxygen content compared to ambient air is in the microamps range. It doesn't take much to upset these things, so it's critically important your connections are clean, tight, and weatherproof. It's also critical the air holes on the sensor itself are seeing ambient air because the sensor needs it to balance itself.

 

Problem seems to be resolved and the code is gone. I removed the passenger side "O2 harness extender" and plugged the sensor directly into the factory plug. The extender tested fine and tested at 0.3 - 0.5ohms across the connectors. I'm not sure why it stopped working now and I'm not sure why it worked fine for over a year and a half.

If I ever remove the headers again I'm going to snip and extend the factory O2 sensor plugs so they come down near the transmission. If the driver side extender ends up going bad I'm going to have to remove the header anyways to even access it.

 

That measurement is a guess at best. Sub 10 ohm checks need a completely different measuring methodology than any DVM provides directly. Besides, microamps are affected by milliohms, and the signal coming from the sensor is in microamps.