AFR mapping for emerald ECUs

Disclaimer

This program and any advice given are provided entirely at your own risk. If your engine blows up or you fail your emissions tests don't come moaning to me - if you change your ECU it's your fault and nobody elses.

Instructions

Emerald ECUs

From the datalogger save the datalog, calling it "logfile.lg1" and saving it in the same directory as the afr program.
Double click on the afr program, a window will pop up and disappear
The results from the run will be stored in a file called results.txt

Megasquirt ECUs

From megatune start datalogging, saving it in the same directory as the afr program and calling it datalog.xls.
Once you've finished your datalogging run, choose stop datalogging from the menu.
Double click on the afr program, a window will pop up and disappear
The results from the run will be stored in a file called results.txt

Interpretting the results

The data below is from one of my test runs. By default the software tries to map to 14.7:1 at low load through to 12.5:1 at high load. See the installation page for details on defining your own AFR targets.

General information

The first section is just some basic diagnostics so you can make sure your temperatures were sensible and that everything has worked ok.

Max RPM = 4718 Max load site = 12 Max MAP = 315Kpa Temperatures: Coolant=58-107 ChargeAir=22-59 Total results logged = 21455

Number of samples

The second section, headed "Number of samples" tells you how much data the program has to work from. The datalogger will have generated about 16 samples a second and the program requires a minimum of 20 samples before it will give you any suggested tuning tips. I would suggest that you should aim to achieve for somewhat more than that, for fine tuning 50+ samples would be better but for the first stages of mapping, when deltas are likely to be larger 20 samples is plenty. There will always be load sites that you won't be able to get data for, don't worry too much about them.

Fuelling fiddler ================ Number of samples ----------------- RPM 0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 11 538 2443 133 132 39 72 27 61 11 1 . . 255 590 13 46 19 5 7 1 2 . . 24 19 871 221 64 14 45 . 3 . . 5 26 55 47 92 71 76 2 4 . . 3 43 25 33 22 20 81 9 5 . . . 5 8 30 29 21 83 8 6 . . . 11 . 16 37 2 85 1 7 . . . . . 1 21 1 33 3 8 . . . . . . 7 4 5 3 9 . . . . . . . 11 9 . 10 . . . . 1 . . 14 11 . 11 . . . . 1 . 1 2 12 . 12 . . . . 1 . 38 64 19 . 13 . . . . . . . . . . 14 . . . . . . . . . . 15 . . . . . . . . . .

Observed AFRs

The third section lists the observed AFRs at each load/speed site, bear in mind that some of these values may be for sites that only have a couple of samples so it's a bad idea to use the AFR table in isolation.

Observed results ---------------- RPM 0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 16.6 14.7 13.6 17.1 18.2 17.8 15.3 14.9 16.7 14.0 1 . . 10.7 11.4 14.4 14.5 13.9 14.5 15.2 14.2 2 . . 14.6 14.6 11.2 11.9 14.2 14.3 15.7 . 3 . . 16.9 13.3 12.7 13.8 12.4 14.1 15.0 16.5 4 . . 16.6 13.2 13.9 13.2 14.1 13.6 15.2 15.6 5 . . . 13.5 12.3 13.7 13.7 13.9 14.4 14.2 6 . . . 15.8 . 13.9 13.7 13.7 14.3 13.2 7 . . . . . 13.9 14.6 15.5 14.9 13.4 8 . . . . . . 15.3 15.6 12.8 13.0 9 . . . . . . . 15.4 13.7 . 10 . . . . 13.9 . . 13.8 14.7 . 11 . . . . 13.9 . 15.3 13.1 14.5 . 12 . . . . 13.9 . 14.6 14.9 15.4 . 13 . . . . . . . . . . 14 . . . . . . . . . . 15 . . . . . . . . . .

Recommendations

The next section, headed "Recommendations" is the recommended actions for you to make, based on the results logged. The suggestions listed are what percentage fuelling change you need to make in order to run at the target AFR. By default the program is set up to aim for 14.7 at load site 0 and to ramp up to 12.5:1 at full throttle. Dots in the results indicate that no results were logged for that site, a "-" that there weren't sufficient results but that those that were recorded were indicative that it might be rich and a "+" that those results were lean. A # indicates that the results that were logged at the site looked to be about right but we didn't see enough results to be certain.

In the example case it's suggesting that at load site 0 many of the speed sites are running lean, this may well be because I have turned the "cut fuel on overrun" feature on in the ECU. The general consensus of results though is that at load site 2 from 2000-2500rpm we're running quite rich and 1500-3000 at load site 3 are a bit rich. At a first pass I would lean off 10% from the first set and 5% off the second set of sites and go out for another drive. That should get all the numbers within the right ballpack and from there I can make another set of smaller changes. When I see isolated +1 or -2 I'll generally leave those alone unless the results are from a particularly long datalog and I'm fine-tuning the map.

Recommendations --------------- %age change based on target AFR of 14.7 at site0 to 12.5 at site15 RPM 0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 + +0 -8 +16 +23 +21 +4 +1 +13 - 1 . . -27 -22 - -1 - - + - 2 . . +1 + -23 -18 -2 - +8 . 3 . . + -7 -11 -4 -14 -2 +5 + 4 . . + -7 -2 -7 -1 - +7 + 5 . . . - - -2 -2 -1 +3 + 6 . . . + . # -1 - +3 - 7 . . . . . + +6 + +8 - 8 . . . . . . + + - - 9 . . . . . . . + + . 10 . . . . + . . + + . 11 . . . . + . + # + . 12 . . . . + . +12 +15 + . 13 . . . . . . . . . . 14 . . . . . . . . . . 15 . . . . . . . . . .

Warmup wizards

The program includes two warmup wizards, one to set the fuelling, the other to help set up the idle valve. In my experience you want to get the fuelling as good as you can, then you can set up the idle position. Warmup is pretty hard to get right and will take a number of iterations, you'll probably do a couple of fuelling itereations, a first stab at the idle valve, then back to fuelling and round the loop a few more times before you're happy. Since each attempt will need the engine to be stone cold you get one shot at it each morning so you'll need to persevere.

AFR during warmup varies from engine to engine, the numbers the program produces are really just a set of approximations that should be about right for most setups. The AFR analysis should be pretty self explanatory, at each temperature step the "ideal" AFR is displayed, along with the results of the datalog and a suggested change.

Warmpup wizards =============== Warmup AFR analysis ------------------- Temp Ideal Current %Change -10 0 10 20 30 40 50 13.3 12.4 -6 60 14.0 13.1 -6 70 14.7 11.6 -27 80 14.7 14.9 +2 90 14.7 13.7 -7 100 14.7 14.7 0 110

The idle speed recommendation really is a guess as I have no way of knowing what proportion of air is flowing through the idle valve or where your idle stops are. My suggestion is that you only make a small proportion of any change suggested until you've worked out how the suggestions work on your engine.

Idle speed analysis ------------------- Only make changes here once your warmup AFR is correct Temp Ideal Current %Change -10 0 10 20 30 40 50 1075 765 +40 60 1012 793 +27 70 950 801 +18 80 950 1030 -8 90 950 971 -3 100 950 954 -1 110