1990 Corvette (Manual)
1994 Corvette (Automatic)
1995 Corvette (Manual)
To be more specific, fueling is based on airflow, how much air is going in the engine.
Knowing how much air is going in, the pcm can calculate how much fuel is needed.
The ratio is how many part of air to one part fuel
AFR
Air to Fuel Ratio
If we we are taking in so many lbs of air, we need so many lbs of fuel to mix the ratio.
Does that make sense?
No need to re-invent the wheel. But we can make it better
most ecms work like this
1. you have an airflow measurement, either with a MAF or speed density (a volumetric efficiency map, in other words a chart that says 'in this condition the engine flows this much air')
2. you have the fuel flow metrics, that end up calculating something like 'if you open the injector for this long, this much fuel comes out'
3. you then try to hit your AFR target vs airflow vs injector metrics to figure out how much fuel to inject
this is both in open loop and closed loop.
there are auxiliary calculations like closed loop trims, transitional fuel needs like accel enrichment, and some other small calculations and constants, but those modify things in the above three steps. without step three, you aren't going anywhere.
in other words, without step three, how would you know that you don't inject one part air for one part fuel?
Good in depth simple explanation stevo. I hope he understands.
I think there are a lot of people who think fuel control is from O2 sensor only.
Last edited by ralmo94; 04-06-2022 at 05:16 AM.
No need to re-invent the wheel. But we can make it better
I don't think fuel control is from O2 sensors only; I'm just used to a completely different computer environment since I primarily worked with Japanese cars instead of American ones. Or rather, my Corvettes are the first American PCMs I've mucked with. In the Japanese world you have actual "fuel tables" that are set to deliver a certain amount of fuel for a specific load cell (RPM vs airflow). There is no "AFR Target" setting because the AFR will depend on how much fuel is put into any given load cell in the fuel map.
Now that steveo has explained that these PCMs don't use "fuel tables" in the traditional sense but rather have calibrations for airflow measurement, it makes a lot more sense. Thanks to both of you!
1990 Corvette (Manual)
1994 Corvette (Automatic)
1995 Corvette (Manual)
I'm glad you understand what you are adjusting, it's hard to do something you don't understand.
Just to be clear I didn't mean to insult your intelligence, and I'm sorry if you felt that way, I haven't seen a Japanese calibration that was setup the way you described, I can imagine I would e confused if I looked at one.
No need to re-invent the wheel. But we can make it better
No offense taken! Here's a calibration from a 1994 Skyline GT-R as reference.
bnr32_normal_ecu.jpg
bnr32_normal_ecu_hioctane_fuel.jpg
It's in Japanese, but the adjustable values in the ECU are as follows:
K Constant: This is a catch-all constant that's used for all fuel calculations. It appears to function similarly to the "Injector constant" on $EE.
Injector Zero Time: Calculation for injector latency since larger injectors can 'lag' in response.
Feedback Switch: Set target temperature for closed loop transition.
Rev Limit: Self-explanatory.
Speed Limit: Stock JDM vehicles have a hard fuel cut at 180 km/h.
TTP Min/Max: Minimum and maximum injector pulsewidth vs RPM. This is usually used as a 'boost cut' for turbo vehicles.
AF_LIMIT: Not used.
Acceleration Enrichment: Self-explanatory.
VQ Map (not shown in above images): Voltage Quantifier. This is the table that converts MAF voltage into grams/second. It's the "MAF Calibration" table on $EE.
Fuel Table: Y-axis is RPM, X-axis is TP. "TP" or "Theoretical Pulse" is a calculated load value based on the K Constant and the VQ map. Each cell in the fuel table represents a TP index and an RPM. You change requested fuel based on where in the table you end up.
It wasn't until the R34 GT-R in 1999 that Nissan switched to using VE tables and doing things more "traditionally."
To further get your noggins joggin', here's the fuel data from a 1992 Subaru SVX:
svxtune.jpg
So yeah. I'd just never seen any of my tools used for 90s JDM cars refer to "Target AFR." It's always just fuel maps of load VS RPM. I can confirm that my buddy's 2005 Subaru Legacy uses target AFRs, for instance. I'm more of a 90s guy myself though.
Anyways, sorry for dragging this a bit off-topic, and again I really appreciate the discussion. It helps a lot!
EDIT: After staring at values for a while it occurs to me that these tables are probably actually AFR corrections. Target AFR is not a thing in any of the interfaces I'm familiar with, but if that means the ECU is always targeting 14.7, and thus the fuel tables are only mapping deviations from 14.7, then they make a lot of sense. Now I'm kind of impressed I was able to work with these things without figuring that part out. And it just makes me that much more impressed with how feature-rich $EE is for being a computer from 1994. Wild!
Last edited by NomakeWan; 04-06-2022 at 10:16 AM.
1990 Corvette (Manual)
1994 Corvette (Automatic)
1995 Corvette (Manual)
in other words, whether they integrate the AFR target into a table value or a separate value, it is always part of the mathThere is no "AFR Target" setting because the AFR will depend on how much fuel is put into any given load cell in the fuel map.
That is still airflow based,
The other table is what I believe is called ALPHA N fueling calculations, based off of throttle and rpm,
As far as I understood, early Subaru only used this for MAF failure, and rely heavily on MAF, unplug the MAF, no start
MAF failure, really rich, once the error code sets.
Although it occurs to me that a blending of some sort might be incorporated between alpha n and MAF.
To tune one of those, I would probably alter the MAF table to the required fueling.
Also, in the JDM ECU's just because you don't see it, doesn't mean it's not there, it may be just undefined, unless you have disassembled the code yourself and can confirm it's not there, but there should be a ratio somewhere in there, it's math, has to have it, may be eq ratio or lambda
No need to re-invent the wheel. But we can make it better
Bookmarks