okay, i see. team must have been saying that he lost a few hp going from 14s at wot to mid 13s. lol. he's crazy.

i am anxiously awaiting completion of the new baseline. i can't even fool with the old one now that it's expired.

that said, i'm happy with my tune already. i just like to play and learn.

thanks again everyone. i'm learning a lot.

pretty soon i'll understand this better than all of you.

 

can you explain the reasoning here for me a little more? i'm not really clear on why it matters. if either way you're getting, say, 12.5 as the result when you want it, what does it matter if you're entering 12.5 or 10 in the tables? in either case, one is actively trying to get the car to spit out 12.5 when it's moving, right?

if two tunes get the same results, aren't they the same tune even if they get the same results with tables showing completely different cell values? is this a philosophical question? what is the tune? the cell values or the results?

 

thats the answer I finally came to myself while tuning... fact is for the actual afr to match the values in the map the following needs to be correct: all 3 injectors sized correclty, maf, and VE tables.

Not sure you can sort through 5 unknowns without some serious dyno time

 

well, clearly being able to just enter exactly what you want and get exactly that is ideal, but if that's impossible given the equipment, what is ultimately the difference between doing that and, well, "tricking the computer" if the results are the same?

maybe it's morally wrong or in violation of a tuner's code to trick it? maybe it's something more complicated i don't understand?

 

This is really interesting, can this PP02 work with my Unichip Q Piggyback?

I downloaded the program from the PPO2 homepage but it says something "this beta has expired, download a new version" but where do I find that?

 

you have to wait till he finishes the new version of baseline. it shouldn't be too long. i figure the more i pester him, the faster it will get done. lol.

should work with your unichip as long as you can log data and load maps with it.

it may be as easy as just copying in your logs and hitting a big red button for a tune. i think that's what he's going for.

 

Ok, waiting, that is the hard part

Logging and load maps is of course doable in Unichip Q.

Can't wait until I can test this, I think this is just what I needed, It would be perfect to make my own maps since im alwas playing with new parts and leaving my car for mapping to a tuner is expencive, I rather learn to do it myself (with this PP02 if it works good) =)

 

The results are what is important. If you ever go Fi you will see that it is even harder to get the cell values matching the actual . It becomes an excercise in chasing your tail . Better to get it reasonably close then just change the tables to get the right result.

 

Exhaust gas is moving at several hundred feet per second and the front O2 sensor is about 1 foot from the exhaust ports and its response time is measured in miliseconds.
The "latency" for the actual feedback loop is a fraction of a fraction second.

The fact of it is that 100ms is fast enough to be considered virtually instantaneous.

Now, the data rate for CAN is slower than this, but that is a granularity issue, not latency.

You can consider your WBO2S readings to be as good as the last combustion cycle.

 

Interesting... the best I have ever seen tested was the Innovate LC1 - and that was 200ms behind the exhaust sampled at the WB02 sensor.

 

Keep in mind I think of tunes in two states; steady state and transition.... transition is going to vary based on a ton of variables - while steady state can and should be dialed in to a very small percent - 5% or less is my goal typically.

That is why tuning by hand is such a PITA - it is very hard to get enough data to look at steady state only and rule out the transitional variables.


HOWEVER, if you are aiming for 12.5 and you need to set your base fuel to 10 to get it... then something is not right - and you shouldn't band-aid it; you should go and fix it.

 

We are talking about the OEM WBO2S.
The aftermarket sensor will be regulated by its distance down the exhaust system.
Even then, you are talking about 3 or 4 ms. The rest is coming from the latency in the controller, not some arbitrary "bottleneck" in the flow of gasses.

 

^^^ Yes, but when you look at the log - that time has to be accounted for.

 

would be interested to know how close your table are to actual ???

 

I'll pull my last tune up when I get some time; from memory I was averaging 3-5% in most areas and I had a few 10% areas in higher loads and low RPM's.

I started at like 22% - so over the course of a month and 5-6 tunes I got pretty close.

 

okay, let me think about this. on first blush, it seems not to matter to me whether you get what you enter provided you get what you want as output.

let A be a maf/injector scale and B be a cell value in the air/fuel table, specifically, 12.5. and let C be another maf/injector scale and D a different cell value in the air/fuel table, not 12.5.

it would seem that if when running the tunes, AB=12.5 and CD=12.5, then AB=CD. so AB and CD look like the same tune just looking at results.

however, under the surface, if you're running CD, you're going to have some parts of the car doing more than they should to account for other parts doing less than they should and vice versa. this is what I think Kane means or implies when he says it's a bandaid way of doing things. the fuel tables bandaid the maf and injector scales, and the maf and injector scales bandaid the fuel tables.

overall, the way i'm now thinking about it, if you tune in this bandaid way, you end up with a tune at odds with itself, one that sort of wants to come apart. Possibly such a tune would effect vehicle life.

do i have this right?

still, i wonder if maybe it's just a sweet spot, a transcendental plane of vehicle harmony, that we're looking for here. on this view, there are infinitely many ways of making her sing the same sweet song.

i think this is what rOtor is talking about in his thread when he says that at some point, tuning is just art.

 

Sort of....

The other issue with the band aid thing is that your MAF scale is a 2D map - so volts = xyz airflow. This is used to calculate load and access the fuel table to determine the value of the fuel to inject.

If you monkey with your fuel map instead of your scales and the scales are wrong - then you have a "lean" or "rich" spot in your MAF scale (airflow/volts) that can appear and re-appear seemingly randomly because load is a dynamic calculation.... that is why you want to get the scales right first.... know what I mean?

 

hmm. i was hoping i had drawn a valid distinction between street or hobbyist tuning on the one hand and race, ideal or professional tuning on the other -- the former tunes for results and the latter tunes for accuracy. both ways would acceptable here.

but it looks like that's not gonna fly. the distinction i've drawn, the way you see it, is between wrong tuning and correct tuning. is that right? and the idea is to as close to the ideal as possible.

okay, well it looks like under WOT i'm off by about 9 percent in the middle RPM range. what i enter is 9 percent richer than what i get. it's not bad, but it's not perfect. of course, the afrs look great

you spoke of "lean" and "rich" spots in the maf scale. is there an ideal spot in the maf scale like there is in the AFR? so 12.5 on the AFR is like what on the MAF scale?

 

2nd WOT afr comparison chart here. now that i figured out this chart stuff, expect to see charts for everything coming tonight. dynos, afrs, maf scales. everything. data data data data.

 

looking good man

 

Gotta sort out that APV spike.

 

I think it would be helpful to think through the process in order to understand what I mean. Ok I am going to eliminate ALL other variables except the MAF and the Fuel Tables for simplicity.

Step on is the PCM gets a voltage reading from the MAF - it then look up the table (MAF Scale) in order to determine the airflow at that voltage.

Ok so now the PCM knows that at XYZ volts - you are moving 200 g/sec of air; it then uses the RPM (etc) to determine the load on the motor (airflow actual / airflow possible) - based on the RPM and the load the motor has two values that together can determine where on that fuel map the target AFR is... and then fuels for it.

So if your 200 g/sec is actually 209 g/sec - but you are playing with the fuel map; that reading can manifest itself at various load and rpm cells in the fuel map.

This is why getting sensors right is the key.

See what I mean?

 

that's making sense. thanks dude.

here's some more charts. i've got 3rd WOT AFR compare and some AP dyno compares.

what would be the best way to chart the maf data? from a wot run or a cruise run or both?

 

what would you suggest?

 

Just chart g/sec of airflow in those afr charts to start with....