reactor2

I'm trying to "dial in" my A/F ratio in...on the street. I'm only making WOT passes in 2nd gear (converter unlocked) since I am doing this on the street. The issue I'm having is I don't know which output device to trust. HP Tuners (Actual AFR by RPM and MAP) is telling me my WOT AFR is around 11.0 where as when I'm using my digital gauge it is telling me 10.3-10.6.

BOOST: 6psi max
Wide Band System: PLX Devices

Questions:
1) Which do I trust (PLX gauge or HP Tuners)?
2) How much leaner can I reasonably expect in 3rd gear (converter locked and greater load on engine)?

I don't want to go lean (stock bottom end) but I don't want to leave too much power on the table.

umrjmac

For the first one, if you have the device configured correctly within HPTuners, you should have a pretty close match of values between the gauge and the data log. I am not familiar with the PLX device, but on my NGK AFX, when the device starts it throws a sequence of 3 known output voltages that represent the upper, lower, and stoich values. Logging that can be used to help determine ground offsets so that the transfer function that you're using to define the PLX device within HPTuners reads correctly.

I know there is a pretty good thread or two on their forum that walk through this, so I would head that way. Alternatively, I am guessing that you can search here for "hptuners wideband ground offset" and get something worthwhile.

hope that helps and good luck.

Someone more experienced will need to chime in on your second question.

directnosfogger

i would make sure you are looking at the correct PID. i tune in lambda as to not get myself confused, but you should be looking at the input from your wideband and not the commanded a/f. i wouldn't trust the gauge itself over hptuners. the gauges can sometimes be delayed and it would be bad practice to try and follow a gauge vs logs that can be reviewed for tuning.

PE tuning is agnostic to gear. if you set your PE for lets say 1.3 straight across the board and your stoich is set to 14.7, you are commanding 11.3 once you enter PE based on the threshold you set. if you want to step your PE table to reach a certain A/F at RPM adjust accordingly.

reactor2

Thanks for the response!

The PLX WB input was already configured in HPTuners shipped. I remember verifying the PLX Lambda as follows:

PLX Devices spec:
Wideband = linear 0-5v
0V = 0.68 lambda
5V = 1.36 lambda

The computed transfer function:
y = (1.36-.68)/(5v-0v) + .68
y =.136 + .68 = y

HP tuners form results in:
volts/(1/.136) + .68
volts/7.353 + .68

I did this last year but I seem to remember verifying the actual input "Wide Band O2 - PLX".

Anyway, driving around with the scanner definitely seems to jive with what I see on the PLX Gauge (I don't have them connected at the same time). However, what I see at WOT on the gauge does not seem to jive with the AFR histogram. Seems to be .5 - .8 off.

When I go WOT and look at the gauge I see 10.3-10.7, but the cells show me an average of 10.9-11.15.

If the input is configured correctly which is more accurate, the gauge or the few hits I got in the cells? The cells are at the 400 rpm granularity, whereas I would assume the gauge is continuous.

Is anybody else seeing this type of discrepancy at WOT?

reactor2

Thanks for the response. Yeah I'm looking at the actual, not that commanded. Since I'm only running 6psi boost I'm gradually increasing fuel as boost builds. I'm shooting for 12ish low rpm to 11.0 at peak rpm/boost. At 3000 rpm, for example, there really is isn't any boost so I treat it as NA (12ish) but as rpm/boost increase I increase fuel (more or less linearly) to 11.0.

Today, after work I made a WOT pass and the gauge settled down around 10.3 (I was expecting closer to 11 based on the tune). The engine was hot (30 mile drive) and the intake temps were also hot as it was 88 and a little humid today.

This is a little apples/oranges comparison as the last log I ran after "finalizing" the AFR was last year in late October so at max, the temp was 60*...probably not even that. That's when I saw 10.9-11.15 in the logs.

directnosfogger

if it was as hot as you say, you should take a look at the additional tables which adjust af. add vs iat and add vs ect would definitely increase the seen af ratio for safety precautions. if you are dead set on keeping your command a/f you could set those tables to 0.

i like stepping my pe table as well and shoot for 11.3 at peak. i do the same thing with my supercharger. with meth, 10.7.

umrjmac

I apologize, I should have been more clear. What I was getting at is that your theoretical 0V and theoretical 5V values may not be the actual min and max voltage as read by the HPTuners box. You have to be cognizant of ground offset (0V not actually being 0V due to the fact that your wideband ground is not in the same exact location as the PCM ground). Looking at your math, you've defined the transfer function strictly based on what the wideband manufacturer's manual said, which is the right place to start, but you need to validate the actual min and max voltages as recorded by the HPTuners device during the wideband's startup sequence.

If you do this, you should end up with the gauge reflecting the logged value very closely.

RonSSNova

For WOT tuning, I like to use the chart display.
I set up one pane to show both commanded and actual from the gauge. Same min/max for each of course.

It gives me a better picture of what is happening.

Then use the AFR error readout in the table, or even graph it and make changes to the VE table by hand. I find that multiply by approx 80% of the error works pretty well.

I have an AFX and never look at it's display. I also tune in Lambda

Ron

directnosfogger

Log the input as Voltage to see what comes up when you start your wideband. I use an AFX as well.

I see a printout from logging as 1.001, 3.98, .020, and 4.85. The important thing to note is that it prints lambda, lowest possible value, and highest possible value.

onspeed

I have an Innovate LC1 w/ DB gauge + HP tuners wired into the ECM EGR line, and I use the HPT output over the gauge. I'm almost certain my gauge is not accurate as it reads around 13.5 when my car is running closed loop, while the HPT output and the LC1 serial output is reading around 14.7. Also, you can change the initial test voltage when the unit first receives power, and my DB gauge seems to be reading about a 1 volt less than the output actually is (confirmed with a multimeter).

If you can connect the WBO2 to a computer as you can with most Innovate products, I'd compare that reading to what HPT is showing. Might also use a multimeter to check gauge voltage versus actual display value on gauge.

directnosfogger

Once you know the appropriate voltage you can adjust your formula for your wideband input in hptuners and make sure you are receiving the correct translation of volts to afr/lambda