69 Ghost

iTrader: (1)

I know this has been covered but if you look at tuning for WOT there are 2 schools of thought. One is to set your desired AFR then adjust your VE tables to match the desired AFR the second is to tune both independently. It seems to me that the quickest is
to use the PE method. The more accurate way is to tune by editing the VE tables after you figure out your desired AFR. With WB tuning you can get it close before you head to the dyno then fine tune from there. I was thinking of setting my AFR's to 12.2 across the board then use my datalogs to edit the VE tables and bring them in line. Once it is close and pretty linear, knock has been taken care of, etc then head to the dyno and fine tune.

Grumpy

IMO, it's easiest, to find Stoich first, that way you can match the WB, and NB's to make sure you haven't missed the tune. Then with that information start looking at the other things you want/need to do. ie have the higher K/Pa elements of the VE table go slightyly rich, so that depending on you TPS PE enables, you don't have to use PE as much.

Detonation is such a finicky beast, I never consider it handled.... And,while you're talking about knock, gotta remember pre-ignition is a quiet killer.

69 Ghost

iTrader: (1)

Grumpy can you tell me more about what you mean about finding Stoich first? Datalogs show the desired AFR and there are 2 outputs for the O2 WB and NB which come from the same sensor. I noticed that the NB reads around 14.2 to 14.3 when the commanded AFR is 14.73 or Stoich.

Some people discuss what you are saying about having the VE tables slightly rich say BLM's of around 124 across the board so that when you tip in you don't go lean. In fact that is how VEMaster is setup to set the BLM's to 124 as the default BLM when you run the program unless you change it. With WBO2 tuning I expecting to be able to tune for that because you can see what is happening on tip in and tell if you have enough AE.

You mentioned in the past about getting the VE tables roughed in to about 80kPa then go for the upper tables using first and second gear runs. I am assuming anything above 80kPa at 80%+ TPS will probably be in the PE mode unless the motor is running very high RPM's. I am also taking your advice and taking notes to spot trends. Acceleration is watched closely in the datalogs for lean and rich conditions. Anything over or under AFR's of 16 and 11 are looked at and adjusted to rough in the AE and PE as long as the VE tables are fairly close.

Grumpy

By finding Stioch, I meant 14.7:1 but by using a NB, and WB. Using a single sensor defeats the purpose. The idea is that by using both you can eliminate the timing error that some folks fall into, ie too much timing will indicate lean, and too retarded, rich. using a NB helps confine you to the best average.

16s under anything other then possibly cruise, and overrun are definite no-nos.

You can up the PE limits *some* but the close the tune is, the less forgiving errors are. With practice, you can get pretty darned close, but intially it's better to have the PE kick in a little too early, rather then late.

69 Ghost

iTrader: (1)

Thanks Grumpy. I am hoping to get more people involved in this thread because I think that there may be something useful to come out of this. My original intent was to do exactly what you mentioned. Datalog using 2 separate sensors. I even bought new header collectors so that I had O2 bungs on both sides. I then did a LT1 mod on my throttle body to drill a hole in the IAC area to eliminate a possible split BLM issue that is common with the LT1 crowd but the TPI users are not very aware of. I believe that it is not a known issue for one reason. The older setups only use 1 O2 sensor so there is nothing indication of a split BLM unless they happen to do something like a dyno run and have 2 AFR sensors stuck in the pipe. It sounds like I really need to log both sensors and for the WB sensor log both the NB and WB outputs then compare the NB output with the NB sensor used by the ECM for control. Z69, JP, are you reading this????

The timing issue you are talking about? I can set the output on the Innovate to have a different or slower timing for a better average. I assume this is because the WB sensors react much quicker than a normal NB?

Grumpy

Ignition timing.
You can bump the timing up, and then show you're lean.
You can get to where you have too much. or too little timing, and still be at 14.7:1 AFR, with a wideband. On one *tune* I had a WB 14.7:1 AFR, and was at ~280 mv on the NB.

Z69

It would be a minor mod to the wb patch to allow you to log both sensors. But there is only one NB amp on a 730 so you couldn't hook it up correctly.
But I'm not sure that would help you with the simulated NB signal. The difference between the Stock NB and simulated
is the sensor. The wb sensor should still read the same regardless. It will just output it at a different voltage scale and slope to match a std NB curve.

69 Ghost

iTrader: (1)

Grumpy question I have not modified my timing tables for the last few runs. When you talk about timing are you talking about changing the tables then datalogging? I spent time getting my tables in line with very little or no knock using my NB sensor. I did this in conjunction with VE tables. I was not intending on modifying the tables unless I ran into knock retards that are larger than 2 degrees. My intent was to visit this after getting my VE, PE, AE tables somewhat in line then see how that affects the knock. Also you mention rich and lean. I have always assumed that a knock condition is mainly caused by a lean condition. How does being rich affect knock?

You imply that the guys running a WB sensor with the simulated NB in place of the original sensor should not be replacing their existing NB sensor. Before I start any WB tuning I should keep my existing NB in place and run the WB in conjunction with the NB and also verify the accuracy between the 2. This also implies one engine side unless you can verify that split BLM's are not an issue or ignore idle and low RPM differences.


Z69. If this is a minor mod to turn both datalogs on then for the WB the AFR is desirable for the NB all I would want would be the voltage? I think that if say the LC-1 is set for 1-2 volts range or whatever it is the stoich (14.7) can be checked against the mV of a NB sensor ie 1.47 volts is the same as 450mV on the NB. Grumpy am I on track here?

JP86SS

iTrader: (1)

Would be nice to see just how well the simulated output "simulates".
I've been curious on this since earlier WBs had issues with them.

Grumpy

Then you *should* be close.

The extra fuel can put the motor into sort of diesel/ auto igniton/ pre/igntion mode, rather then having the spark cascade the combustion event. While it's not much of a problem generally in the N/A world, the turbo guys can spend alot of time fighting it. Fuel once past Stoich, can take on many roles about what it's primarily doing, IMO.

I'm saying that you want, or if you can, using two sensors makes for having a sanity check. Reading the same output from one sensor two ways, justifies nothing (in this case).

Different WBs have different v's at stoich so I can't say your example is correct.