Auto Tuning with narrow band

[Bluehaze]

Hi Folks
I've so far managed to get the motor running fairly well in the driveway, and on Friday I took the car out of the shed for the first time for a drive up and down my street. I'm so far fairly happy with the way it runs, albeit a tad rich, but that it deliberate, on my behalf so as to err on the side of caution. The next big step for me is to go for a proper drive. This I plan to do next Friday, as I have an appointment with an exhaust (noise) test facility, one step toward registration.
Anyway, I was hoping this would provide an opportunity to use the auto tune function in TS. My concern is,

-will I get an effective tune with the NB sensor?
-should I flatten out my AFR table so as the O2 feedback is more accurate? I.e closer to 14.7.
-leave AE on?
-leave O2 correction on?

I have done a lot of reading on the forum, and get the impression that almost all you guys use a WBO2 sensor for tuning. Please note my project is not high performance, it has a stock cam and is naturally aspirated. It will be a daily drive for the wife.
Running V3 board with
2.905 software.

[Bluehaze]

Right,
I did yet more reading, and it seems the auto tune function us only good at no load conditions. So I decided to shoo the moths out of my wallet and lash out on the MTX-L unit from those great folks at DIY auto tune . In fact after reading the PDF for the unit, it looks like money well spent. I even welded a second bung into my exhaust header (just behind the last collector) as provision for a second sensor, way back in the build phase. Long term, however I plan to only use the WB unit for tuning, then just run with the NB once happy.
I hope the use of the WB will be like switching on a light in a dimly lit room, as far as tuning under load is concerned.
Any tips comments re the MTX-L unit?

[Bluehaze]

Ok. I know they say the first sign of madness is when you start talking to yourself, or so my TV tells me . So here I go.
Got my LC1 unit with analogue meter in the mail the other day. Fitted, calibrated and tested and I'm chuffed. I was right! It is like switching in a light in a dim room. My logs are showing me the misses I get when I give some right foot action are actually lean misfires, not rich, as I suspected.
I had a crack at the VEAL thing, just free revving, but either I'm still a way off or my settings regarding the correction factors need work. Luckily I saved the tune just before trying the VEAL (geddid-I'm here all week ), because the fuel map started getting jumpy, which led to oscillations in the motor RPMs, and it just got wobblier and messier. I'll try VEAL some other time, but for now, it's back to using the data logs, my AFR gauge and my ears to guide my shaping of the fuel map. If anyone has any comments or suggestions, just jump right in!

[abecedarian]

I don't think talking to yourself is such a bad thing and neither do I. It's when you find yourself asking yourself "What did you just say?" that you have a problem. ;)
What?

Oh... well anyhow, I'd like to keep an eye on your progress so please keep talking to yourself.

[BottleFed70]

Ok. I know they say the first sign of madness is when you start talking to yourself, or so my TV tells me . So here I go.
Got my LC1 unit with analogue meter in the mail the other day. Fitted, calibrated and tested and I'm chuffed. I was right! It is like switching in a light in a dim room. My logs are showing me the misses I get when I give some right foot action are actually lean misfires, not rich, as I suspected.

Careful. A miss always shows up as a lean spike regardless of why the miss occured (rich, lean, or ignition problem).

[Bluehaze]

Just did a few laps around the block. drive logged and checked at the end. So far I have found much improvement when increasing the relevant cells in the VE table where the feedback starts to show lean. Each successive drive I have noticed better power, better up hills, and no more misses! It seemed the misses were due to starvation rather than over-fuelling. I'm a lot more comfortable with AFR staying at or below stioc most of the time. I'm not there yet, but definite heading in the right direction.
Also checked the plugs, after all this high tech stuff, the plugs still provide the info I need. (Tan/grey electrodes, slightly off-white insulator,-now I'll sleep well)
The (log) output from the LC1 does seem erratic under speed, and I refuse to accept the AFR is jumping around that much. The analogue gauge is agreeing with what my ears are getting from the motor.
Any comments regarding the log are very welcome.


MSII V2.905
V3 Board

[LT401Vette]

Have you Calibrated your AFR table under the tools menu.

BTW, VE Analyze does work with a narrow band, but a WB certainly does improve results.

[Bluehaze]

Yes, although the "calibrate AFR" function didnt give me the same satisfying reply I get from making other changes, so I took it on faith that it took. I've decided not to get too hung up on lean spikes or wobbly readings. The motor runs pretty well, and I reckon with more fine tuning and a bit more time using VEAL the readings from the O2 sensor will settle down. Also when I get the correct Lambda delay and O2 correction factors correct.
Just one question Phil, when you say a NB is OK with VE analyse, do you mean the Live version or the one you do post logging?
Another idea I was toying with was using the second bung I installed in the header and re-employing the NB sensor as a back up for the sake of redundancy. That is, if the cross-over from the NB corresponds with the 2.5v point, then I can be confident the WB is maintaining accuracy. I've read in several places now that while the WB is great for tuning across a broad range of AFR, the accuracy cannot be assured.

[LT401Vette]

Just one question Phil, when you say a NB is OK with VE analyse, do you mean the Live version or the one you do post logging?

Both..

Yes WB accuracy can vary from so many conditions, they are generally most accurate near stoich