Ok, so lately i've been experiencing some issues. I have a single turbo setup mid mounted on the car. My passenger bank (bank 1) runs down and crosses under the trans to my drivers side, merges, then from there the piping goes to the turbo. I have my haltech dual widebands in the pipes coming off the headers. They were dead on accurate with my vortech kit and since i've added this turbo setup there has been a difference between the 2 wideband readings.

Symptoms:

driving the widebands are pretty accurate
upon WOT the wideband 1 goes sometimes a full point leaner than wideband 2
Nearing full boost (14psi) they are accurate again

tests:
wideband 1 was on the passenger pipe (wideband that goes lean) and wideband 2 is on the drivers pipe (stays pretty constant).
i left the sensors in the pipes and just switched the plugs so that wideband input 1 is now on the drivers side and wideband 2 is now on the passenger side. If this was a piping issue (due to longer pipe length on passenger side) then i would see the same things going on except it would be lean on wideband 2 now. Well i didn't. After testing wideband 1 (which is now the drivers side) now shows lean. SO that means it's not the sensor and it's not the piping.

I have free air cal'ed the sensors multiple times. take them out of the pipes, turn adjustments all the way CCW, turn on car and let them heat up, then slowly turn them clockwise until they just start to blink. I did this before i tried anything else, then i did it again when i swapped the wires. If they were off i would think they would be off all the time and not just during boost building.

So this seems like it's a problem with my wideband controller. I guess the next step would be to put them both in the same pipe and see what happens.

Files 7_4 and 7_7 are the logs of the events. The wires have been switched as i noted in the comments. 7_4 is wideband 1 in passenger pipe, wideband 2 in drivers pipe. 7_7 is wideband 1 in drivers pipe and wideband 2 in passenger pipe.

Also, last night another weird thign happened. I haven't messed with the car for a few days. I get in and fire it up to do this testing (7_7 logs) and as soon as it starts to idle my o2 corrections go crazy. one bank is lean and has +15% o2 correction and the other goes rich and has a -15% o2 correction. I haven't touched anything else. I tried multiple times to do a reset on the ecu and nothing changed. I ultimately gave up and turned my o2 corretion off and the a/f on each wideband were rock solid again EXCEPT the problem above still remained (since it has nothing to do with o2 correction). What is going on with my brand new haltech? is something failing?



Oh, also note on the 7_7 log i had that drop in rpms. I think it was spark blow out. It was real bad when i had a high gap in spark plugs but it's better now. It hasn't happened for a while until this log. I have my knock thresholds all setup and as log as possible in safe boost timing and it shows zero knock along with the fact that my timing is safe. I think it's spark blowout but if anyone has anohter idea let me know.

 

Sounds like your banks are crossed.

 

what do you mean crossed?

i switched them on purpose after having that issue to test to see if it's a sensor problem, bank problem, or wideband controller problem. When i switched the wires the problem move with the wires so it doesn't matter what pipe or what o2 sensor, the wideband 1 wires goign to the controller are the ones giving me the lean symptom.

That also means it's not a specific cylinder bank because the problem would stay in the same exhaust pipe if a certain bank had issues. right?

 

Meaning that your Haltech Bank 1 is actually reading Engine Bank 2, and vice versa.

 

i got that covered by my tests. No sarcasm here, was i unclear in my post about switchign them around and testing to see if it was bank dependant or not?

If i was unclear then ya, i tried switching them back and forth and it always stayed on what the controller calls "wideband 1". It didn't matter which physical engine bank it was readnig, it was always lean.

and i swaped the o2 sensors and it still read wideband 1 as lean, on either of the sensors i have. Could the controller just randomly go bad?

 

I was talking about your idling / o2 correction issue, not your original problem with one reading leaner than the other.

 

ooOO, i get ya.

well on the o2 correction is based on my haltechs so 1 is always 1. I guess when i get home tonight i could use my oem narrowbands and see if that changes anything.

but for now my o2 correction is ran off the same sensor that is my wideband through the haltech. I thought about that one for a second also.

 

I remember running into the same correction issue as you, one bank off the charts rich, the other off the charts lean. Turns out I had the white/blue wires crossed at the Haltech / WB controller connection (mine was hardwired, no plug). When it tried to correct the mixture, it was in fact correcting the wrong side, which quickly spiraled out of control into the overly rich/lean condition you are seeing at idle.

So, by you just swapping connectors under the car, it is essentially doing the same thing. Once you swap them back, your o2 corrections should go back to normal.

 

hmm, weird, i didn't think it would make a difference. I think i get it now though. So sensor 1 is in bank 2, so it's signal is making the computer adjust bank 1.

that makes perfect sense! thanks chris!

Now tell me what's up with my issue on the lean condition. THat's what i've been battling for a few weeks now. I did the wire swap to see if it was bank dependant and it's not so im' baffled.

 

NP

I'm not familiar with the o2 bung layout on the BP kit. Where on the piping do your Haltech wideband sensors get their readings? Are they appx the same distance from the exhaust manifold?

 

They are both a few inches upstream from the merger of the 2 pipes. So the drivers bank has less distance between header and o2 than the passenger. Sasha has his in the same place without any issues. I'm running 14psi of boost so i suggested maybe a delay on the bank 1 due to increased distance but he doesn't think that it would be an issue. It's funny that it's only when i'm building boost which makes me thing delay from backpressure. At that point though, which do i tune from? the drivers bank or the passenger bank?

I think it's "phantom" lean on teh passenger bank, but i would hate to be wrong.

 

Hey Binder & Chris - I think I'm having a very similar problem with the two banks reading differently. I've been working with Tom (TCode) via data log/email tuning to get my tune close before I dyno tune and my haltech widebands are acting strange.



Sensor 1 starts high and creeps down while sensor 2 seems pretty consistent. I've got 3 data logs taken over the past few days and it seems like the widebands are acting in this consistent (inconsistent) manner

Hat tip to Tcode Tom out in Croatia!

 

That looks like a different problem. I had issues going through rpm range slowly in the vac map. My widebands have never been significantly off when idling.

Looks like something is going on with your manifold pressure. Why is it going up and down like that when idling?

The a/f doesn't even start at stoich. Things I would do first:

free air calibrate your wideband sensors after letting the car sit all night. (i think this is probably the issue)

After doing that turn off the o2 correction on your map. Start the car up and see what your a/f readings are without correction. They should be equal on themselves with no changes.

Then make adjustments to get it idling after warming up very close to the 14.7.

Turn o2 correction back on and see what it does. If the a/f starts creeping up on one and down on the other bank then you most likely have the wideband for bank 1 and 2 switched. Meaning bank 1 sensor is in the bank 2 pipe and vice versa. This isn't an issue when you turn off correction but becomes a huge issue with correction on.

 

Thanks for the response - I pulled the o2 sensors out last night and cleaned them up - they didnt look muck'd up or anything - been disconnected since last night and I'm planning on calibrating here shortly. I'm working through the idle search among a few other things (hard start, stalling issues)...one step at a time.

Thanks for the insight - I'll update as things progress.

 

What's the build info on this car? recent changes, etc?

Tackle one thing at a time. Make sure the dead values (latency) on the injectors is correct first. With tuning idle always turn off the o2 correction. O2 correction should be off for all tuning then turned on only after everything is set up.

After the injectors are setup properly then you can worry about starting issues. The prime pulse is where you will want to tune if it's not wanting to fire when cranking. Find the temperature column that you are first starting the car and adjust the pulse. Up is more fuel and down is less. Once you get the cold temp area set you will most likely need to increase the fuel in the cells warmer so your follow up starts will be in tune.

Then move on to post start enrichment. If it fires up and then dies the post start enrichment needs work. With cold start I target a little more rich than 14.7. I start at around 12:1 then as the engine warms slowly pull it back to stoich.

If you can get it started and idling I would get those tuning cells close to 14.7 before going forward with fine tuning the start up.

 

why is your TPS going all crazy? Tiny changes to the TPS will make your AFR swing widely at low load/ rpm, esp with big injectors.

you are getting intergral windup in the closed loop with your TPS moving around like that.

 

^thanks Binder/Str for the input - I'm having issues recalibrating my widebands - I gotta chase this first but I will get through it and will look into the tune next.

greddy 18g twins, IPP 3 crate, DW 600, walbro 255, jwt c2, CJM stage 2...Tcode/Tom is working with me on the tune. Tom was saying that one of my axis was set on coolant temp (insert forehead on palm) and I know what he is talking about (completely my fault) so I'm thinking of grabbing Haltech's starter maps and starting completely from scratch.

Another thing he mentioned was fuel dampeners - the newer CJM rails have them built it - I don't have those newer rails - is there a good way to add them?

 

new widebands installed and they are calibrating per the instructions. let the problem solving resume!

 

There isn't a good way to add dampers except for purchasing the new rails. If you already have upgraded rails but without dampers I wouldn't worry about it until you get issues. The resonance issue only affected a small portion of people. It seemed to me it was mostly in the g35 community and I assume it was from the 8" longer fuel line.