Very Rich AFR's at Idle and Low Load

[Vermont]

I just installed my new Up-pipe and few other goodies. Well ever since I have had my wide band o2 (AEM UEGO) I have been seeing very rich AFR's on idle. I sit at 13.8 AFR at idle after warm up and 10.4-10.0 at light load cruising. In the OL tables I am hitting my target of 10.4-11.0 under boost head on every time. Only when the car is in CL is it running very rich. I have checked the CL tables and they look normal I cant seem to figure out why then are running so wacky.

I will be flashing on a high WGDC map to see where I am siting with fuel learning trims and taking a learning view snap shot of it. Any one have any ideas? Also I do not have my wide band hooked up for logging yet just using gauge right now. Going to be fixing that shortly just need to find my self a rs-232 cable.

More to follow as I find out stuff.

 

[HolyCrapItsFast]

Can you please post a log? What are all of your mods both old and new?

Without knowing any details the first thing that jumps out would be a bad O2 sensor or a dirty MAF sensor. It doesn't sound like a leak to me. Unless the intake was changed out for a smaller one or injectors were changes all with out a tune, I don't see a leak making it run rich. But it couldn't hurt to do a boost leak test.

Was this running fine one day and suddenly it's rich or did you always have this condition?

 

[Vermont]

Here is a link to a google spreed sheet log of my No WGDC.

https://spreadsheets.google.com/spreadsheet/ccc?key=0AlsR5qgX5JgYdENkaDBreWJYd3lnX1dOZDV1ME4tRGc&hl=en_US

I will be tossing on a new map here shortly and testing it out for waste gate. That map was with zero WGDC min and max, while using the rest of my stage 2 as base.

As soon as I saw 25psi I let off. Funny thing was no knock.

I have no logs of any AFR's as I need to hook up that UEGO. I am going to get a Learning view snap shot here in a little I will post that when I have it.

 

[HolyCrapItsFast]

The 25psi can only be one thing IMO and that would be the lines to the EBCS are screwy. This usually means that there is no pressure signal to wastegate what so ever so the wastegate is simply closed all the time.

As for the CL rich issue I will need a proper log for that but it would probably be prudent to figure out the boost issue first.

It is not a big deal to have the uego hooked up at this time. For closed loop we want to see what the front O2 sensor is doing. Once we figure out the boost issue I would like for you to log the following.

A/F Correction 1
A/F Learned 1
A/F Sensor Ratio
CL/OL Switch (Status)
RPM
Engine Load (Calculated)
MAF Voltage
Throttle Position

When you take the log you need to drive around town and on the highway for at least 20 minutes and use smooth throttle inputs. Try not to drive above 3600 rpm and try to keep your throttle below 50%. Otherwise drive normally.

 

[Vermont]

I had the WGDC all set to Zero so there would be no signal to the waste gate what so ever. I installed the EBCS just as GrimmSpeed had on directions. I will be flashing the new map at 0900 and then driving around till my IAM gets to 1.000 then I will do a WOT log.

The thing I am worried about on the EWG is that it did not open untill 25.25 psi. Which means it does not have a 1bar spring it. Which means import performance has their head up their collective ass's. As I cant tell the fucking springs apart that they sent me with the waste gate. When I called them they assured me it had a 1bar spring it already.

 

[HolyCrapItsFast]

You should try bypassing the EBCS altogether. Run a line from the turbo outlet to the wastgate directly with nothing between it. This will prove the spring pressure and it will reveal whether the problem is the solenoid set up or the wastegate.

As far as the EBCS is concerned, lower WGDC should result in lower boost

 

[Vermont]

So I flashed on a LOW wgdc map (sorry for mix up your right) and guess what.... Still over boosting. I am thinking now it is my EBCS set up I have it set up just like how grimmspeed says.

Also here is a Learning View shot I just took. This after a couple miles drive with many pulls. For some reason the IAM got re set to .8000 again. I did have it up to 1.0000 at one point then it re-set it's self.

Any ideas on a good way to set up the EBCS?

 

[HolyCrapItsFast]

Did you try what I suggested?

I will have to do some digging since I am not to familiar with the grimspeed unit. Is it a three port?

 

[Vermont]

It is a three port and I have been looking at how people are setting up the perrin units. Just cant figure out which side goes where. Here is a pic of the Grimmspeed directions that I used.

I have not tried Just routing the turbo outlet to the ewg yet. As I am now pretty certain it is how I set up the ebcs. I followed the picture there exactly to the letter. also I think the fuel problem could be due to the ebcs set up..... might be seeing the whole thing as a vac leek. Just an idea.

O and by the way your fucking awesome for helping me man.

Edit-Think I just found the problem..... I did not plug the ports on the Tial..... I did not know I even needed to. I also need to reduce WGDC by a lot more then I did.

 

[Vermont]

Going to head back to the shop to plug the holes in the Tial. Also going to add the rs-232 port on my UEGO. Just went and picked one up. I will send an update once I am done.

 

[HolyCrapItsFast]

Any time man. I'm here to help!

https://www.grimmspeed.com/install_guides/PDFs/boost_control_solenoid_Subaru_online_version.pdf

Hmmm, As suggested by the diagram, the top port on the EWG should be left open to atmosphere. This has always been how I do it. The EBCS looks like it is clearly labeled so I can't imagine you hooking that up wrong. As I understand it the ports are scribed with the port number. Port one (1) goes to the intake or to atmosphere. Port two (2) goes to the bottom port on the waste gate and port three (3) goes to the turbo outlet.

Definitely try hooking the turbo outlet to the bottom port in the EWG with out anything else hooked up. If you still hit 25psi then it is the EWG. If you hit 1 bar then it is the EBCS.

I really don't think the fueling issue has anything to do with the EBCS set up. I would be very surprised if that made any impact to CL fueling no matter how wrong it is.

 

[Vermont]

Fixed the over boosting. I had left all the other ports in the EWG open..... So in essences the EWG was one giant vacuum leek. Got that fixed and done (also installed my new hawk pads). Well I took it out for a log and the cable got cocked at an angle when I went WOT and my laptop lost connection mid log. But I was looking at my Boost Gauge and it was max of 23psi and then it would quickly drop down as the TD kicked in. Man EWG are fucking scary man. I scared my self the first time a heard it..... No joke. Also through 5-6 pulls all hitting 23psi only one value of -1.40 on the FBKC. Pretty sure its noise. Gonna lower the WGDC by A LOT like A LOT more.

Unfortunately the CL fuel trims are the same. Fixing the EWG did fix the neutral gear idle AFR. It is back to 14.7 and some times moves a little bit. CL driving is still 10.0 and clutch in gear is still 13.8ish. That number moves around a lot. OL is still good with at 23ish psi I was safely around 10.9-11.4.

Here is another Learning View I took. Once again the learned fuel trims are all wacky.... Wish I could figure out why. I am going to try and get a log in using the parameters you told me to use. I am starting to think at this point that maybe my front O2 is messed up. CL AFR's were never this rich before the install. I was also messing with the front O2 too get my up-pipe on and off.

 

[Vermont]

Might sounding more and more like I got to much PB blaster and or yanked on it to hard and is now dead..... Any ideas on how to test this theory with out pulling it?

 

[HolyCrapItsFast]

Try unplugging the O2 sensor. This will throw the car into open loop and you should see your A/F Sensor 1 monitor revert to 14.7. This doesn't indicate a bad sensor but at least it indicates its doing something.

Only a log will tell me anything at this point. Again I am not interested in any WOT pulls. Just a nice long log of normal driving... And not the kind of driving that Fuji thinks is normal :lol:

 

[Vermont]

OK I am gonna get hammered here with evo rules. I will give you a log tomorrow.

 

[Vermont]

So today when I was trying to go out and get a log with the STI. The car threw it's self into safe mode.... Here is a Learning View picture of the codes. I reset the ecu and parked the car. Also I have the AEM wide band hooked up now for logging.

 

[HolyCrapItsFast]

Oh ouch... Those are never a good thing. Does the car run smoothly or is the idle shaky. Have you done a compression test lately?

 

[Vermont]

Never had a problem before this one. Had a compression test done back in August of last year. According to my parents who took it in, all results were good. I am pretty sure the 2 miss fire codes are dude to to much fuel being dumped into the chambers at idle.... In the morning, I am going to try and pull the plug on the front O2 and see what it nets for a result like you said. Then I am going to get a full log like you wanted. Quick question, do you want me to add my AEM UEGO to the log? also you said a 20min log around town. You do realize that the log is going to be like 50 pages long right? Also what is a good way to test to see if this is being caused by a pre-O2 sensor leak exhaust leak?

Thanks once again for the help man.

 

[HolyCrapItsFast]

what year car? Are you CAN logging? Yeah... 20 min is what I want. It probably wouldn't hurt if you could add the AFR gauge to the log but also log the stock A/F sensor. I want to see what your O2 is doing.

I find exhaust leaks with a burning incense. I wave it around all the places it could leak and look for the smoke to be disturbed.

 

[Vermont]

........... Are you fucking shitting me about the incense dude...... I was thinking of just spraying some soapy water over the area where I think its leaking then, running the car and waiting for bubbles.....

As far the Logging. I have an 09 STI and I have been using a Textrix cable connected to the OBDII port. I am thinking at this point it is either an exhaust leak right before the O2 sensor. As I did take that whole header piece off. The gasket could be bad and not sealing it properly. Thus the car thinks it is running lean since all the outside air that would be in the exhaust and thus it dumps fuel in CL. I was reading that during closed loop and idle the engine pulses and thus air gets sucked into the exhaust track.