Huh?

What?

What are you guys talking about?

You can wait for me or you can futz around doing it yourself and wait even longer (and take the risks as well).
Choice is yours.

 

What is the Fuel (PW) Column of the Logfile measure? It is not duration...I was told that there was a bug and it wasn't an accurate measurement at all.

Feel free to correct me. I just need to know what it is.

 

Average injector PW.
The OBD-II spec (or, at least, the way it is implemented in the RX-8) apparently doesn't have a way to show staged injector setups.

 

So average PW for all of the injectors, just one? Is it like the PFC where is does a PW duration based on the size of the Primary Injector (Like 25ms; where actually everything over 12ms available goes to the secondary)?

 

I like to "futz" around

 

Average of all, IIRC.
The staging is funky.
The primaries will go out to 80% or so and then drop back to 60% when the secondaries kick in at 20% or so.
Then, the primaries gradually ramp back up again and the secondaries stay sorta flat.
When the P2s kick in, the primaries fall again, the secondaries spike up and the p2s gradually rise.
Its kinda stupid.

 

So; how would you calculate the volume of fuel being injected off of that? Are doing it in Duty Cycle (as you mentioned %'s) or Duration (what I thought it was).

My Dashhawk has a PID for ml/fuel per second - so somehow the calculation is being done. If it is a % of total fuel available - total fuel period... that is fine too - but the Logged Fuel (PW) of "31000" as a value isn't helping me any.

 

You can't


The PCM knows how much fuel it is injecting.
But the CAN data can't represent it, so you get those two values that are in the spec.

 

Well.... Sh*t. That is not cool.

Rum and Coke time....

Time for plan B.
- Get drunk
- Soil myself
- Regroup and come up with plan C.

 

Must be cheap Rum to have to mix it with Coke. I like it straight.

 

Way to kick a man when he is down...

 

Don"t worry Kane, you're not alone. I started drinking an hour ago Took a break from hard liquor. Drinking wine today.

 

two margaritas here... I'm not tuning shiat... although I am taking my schweet ash tym putting a half quart of oil in my 8... I've got so much foam I almost went upstairs to have a "talk" with the wife, doh!

 

Poor Kane... remember what I said in your Baseline Automated Tuning thread?

 

Dollar Bet?

I can tune around it - it just makes my sweet MAF scaling program a lot harder.

 

lol... almost anything can be done with enough work. I was just saying that you're likely to end up with a different branch of code for every style of EMS out there

 

No - I refuse. :p

 

In my efforts working with the accesstuner I honestly believe the ease of tuning comes down to getting the MAF scaling as good as possible.

Otherwise it's just a shot in the dark as far as the tuning is concerned. Hopefully soon I'll go out and take some MAF data logs and start tweaking it by hand. I'm convinced that 50% of the work is in the MAF scaling but we'll see.

Kane I think that if you were able to take a CSV of grams/sec, MAF voltage, and your STFT/LTFT output you could develope a program that scales the MAF accordingly. No doubt there's more logic involved but it seems like a great starting point.

 

If MAF scaling is 50%, then I believe that the VE table is the other 50%, with Timing making up the other 100% for a grand total of 200%!!.

Seriously though, I had tried everything with this software to try and get my AFRs to stop leaning out @6k rpm. Then I went and made some small(I thought) changes to the VE table for when I was in boost, and all of a sudden I was too rich above 6k rpm.

-Yambo

 

I'll be writing about this in more detail soon; the Fuel Trims can be used to help scale the MAF; but it is not nearly as clean as my MAF program... but mine uses the Fueling as the "known" variable. Fuel trims don't always work as the MAF isn't exactly linear - and in open loop you aren't getting that feedback from the motor.

 

Welcome to my world.
Everything affects everything else. Nothing operates independently.
I don't know who's fault that is - Cobb, Denso or Mazda - but there are a whole slew of interdependent factors that you have to learn before anything you do actually has the desired effect.

 

The fuel VE table won't change your AFR in the long run. It can be used in addition to MAF calibration to minimize fuel trims for safer tuning. For example, if at a given RPM/load you have +15% fuel trim, you could in theory set your fuel VE to 1.15 for that cell to eliminate the trim, although I wouldn't really recommend it.

You may already realize this, I just want to clarify for others that this cannot be used to adjust your target AFR. It is only a compensation for what could be an incorrect MAF reading, incorrect injector scaling, incorrect wide band reading, or all of the above plus several other factors.

EDIT: Kane, you will need to have the user input this table (in addition to the AFR targets table) to get accurate fueling information if you end up using fuel trims to scale the MAF in your software.

 

Furthermore, the fuel Ve table goes to **** once you are in positive manifold pressure.

 

I wonder why that would be. It's just a simple fuel scaling... are the OE wide band readings erratic when you start pushing that much air through the engine?

 

Because its NOT a simple fuel scaling.
Its an ABSLoad multiplier.