Re-scaling load in fuel and ignition maps.......

[Mo]

How do you go about this?

My current scales cover upto 5.6 but I'm seeing closer to 8 in logs taken today, so I really need to rectify this sooner rather than later.

Thanks in advance,

Zak

[calvinc]

if you're seeing load values of 8 then your engine must be moving a huge amount of air? what are the maf readings? are you sure you are monitoring the correct engine load address?

calvin.

[qoncept]

I've thought about doing something like this with Enginuity -- you tell it the range you want your axis to go to and it'll fix the rest of the table. I'm not sure how I'm going to do it yet, though, and it's behind a long list of other things I need to do, so don't hold your breath.

[FrSTi]

if you're seeing load values of 8 then your engine must be moving a huge amount of air? what are the maf readings? are you sure you are monitoring the correct engine load address?

calvin.

You're right Calvin
Something above 7.0grs already is a serious breathe if you run an approx 1.50b boost.

[Mo]

2.5 short engine with forged pistons and JDM heads so it should be moving a lof of air when I say nearer to 8 it actually peaked at 7.7, I don't have the laptop with me just now so can't check the logs for the MAF readings but will confirm tomorrow.

So can anyone explain on how to rescale the load; with many of us remapping our ECUs now it would be good to have it documented somewhere?

[calvinc]

attached is an example spreadsheet. it has 3 columns - the 2 outside columns contain the known axis and value data. the middle column you just have to put in the new desired load value. it will work out the % of the new load value and fill in the data values. this is assuming the data progression is linear.

now what load values do you change to - you should choose load values according to where you want the finest control. if you're having timing/fuel issues at a certain point then make smaller load adjustments in that region for finer control of the table data. making custom changes like this is pretty specific to your setup.

when I say nearer to 8 it actually peaked at 7.7

your car must be bags of fun then
i'd like to see the maf log when you get it.

calvin.

[Mo]

Thankyou for the that Calvin, I'll report back how I get on with the re-scale. I still don't have my laptop back so used a friends to get this pull today........

[Kha0S]

Holy ... 22psig of boost @ 12:1 AFR and 15 degrees of timing? What kind of fuel are you on? Using water injection?

[Mo]

98RON pump fuel with 10% methanol, no WI. I dont' usually run the meth but added it yesterday whilst out testing on track.

[crazymikie]

It looks like your boost is spiking quite a bit- I would try to adjust your wastegate duty cycle to smooth that out a bit.

Also, I'm pushing 18psi on a wrx stock turbo, 12.5:1 afrs and 15 degrees of timing at peak torque on 93 octane gas. The fuel quality we have in New England seems to be very good. I taper to around 11.5:1 at redline.

[calvinc]

the load value is high because of the low RPM and high boost. i'm still trying to find the section of code in the ROM where it calculates the engine load value, then we'll know for sure.

i've got the exact same setup as you but on a 2lt STI. 22psi, slightly richer AFR's, no WI, stock turbo (VF35). my engine load only jumps to 5.7 though. i can only imagine it on a 2.5lt, good going!

the boost curve looks ok. only thing is the slight dip just after the initial boost. you'll find this is caused by having too high a wastegate setting in the lower RPM. whilst this is good for spool-up it also creates that dip because the ECU see's too much boost after spool-up and then brings down the rest of the map. you can get around this by either lowering the wastegate duty in the beginning of the map or lowering the RPM at which it is set. either of these will likely affect your spool-up (which is great right now) but will aid in getting rid of the dip you see afterwards.

another way to try resolve the dip is to raise the expected boost levels but DONT change the wastegate duties. this way the ECU will be happy with the amount of boost and not cause the drop but the wastegate duties will prevent you from actually reaching the new 'inflated' levels of boost.

you should graph your wastegate/boost maps against what you have in the ROM. this will help to see the trends and make it easier to make changes. attached is another sample spreadsheet to show what i mean.

calvin.

[Mo]

Thanks Calvin, the spreadsheet is a great way to associate changes in boost settings and I very much appreciate you sharing this information but I'm still stuck with the scaling issue, and in particular what values I need to change in relation to your scaler example; are you saying I need to change the value of the load headings in each table?

--------------------------------------------------------------------------------------

All things being equal, E***** Road Dyno function indicated 340lb/ft on the 2.0ltr, and now over 400lb/ft on the 2.5 and that's with only adjustments made to the base fuel and ignition maps, once I have access to the rest I can make adjustments to extend the power band.

As an aside, isn't the engine load a reference to VE hence why it's higher at low revs?

[calvinc]

but I'm still stuck with the scaling issue, and in particular what values I need to change in relation to your scaler example; are you saying I need to change the value of the load headings in each table?

yes thats exactly it. all your tables are now going off the end of their scale and i'm not certain yet as to what the ECU does when this happens. does it scale further down or just apply the last column settings?

you will need to rescale in the following maps:
- low detonation fuel map
- high detonation fuel map
- ignition base timing map
- ignition correction timing map
- vvt intake cam timing map

you should also set new limits to the following:
- fine ignition learning range (engine load)
- coarse ignition learning range (engine load)
- knock learning load thresholds

you should also keep an eye on the [CEL threshold (MAF sensor)], this will effect fuel cut if the MAF sensor reads too much air. traditionally this is set at about 4.7v which should be about 300g/s.

calvin.

[Kha0S]

yes thats exactly it. all your tables are now going off the end of their scale and i'm not certain yet as to what the ECU does when this happens. does it scale further down or just apply the last column settings?

It appears to simply apply the last column settings, in my experience. I was running off the edge of my maps for nearly a month... albeit only by 0.2 g or so. I rescaled to add a column at 4.49 g and it works pretty well, though I've increased my boost a bit and am now seeing loads as high as 4.65.

you will need to rescale in the following maps:
- low detonation fuel map
- high detonation fuel map
- ignition base timing map
- ignition correction timing map
- vvt intake cam timing map

There's not much need to rescale the high detonation fuel map, as your car will not be seeing high loads when it's failed over. Ignition maps are good candidates for a rescale, but depending on how long you have the table pegged, you probably don't want to be varying your ignition timing much anyway... since peak load ~= peak torque, unless you're very pegged (ie, 1+ g or so), it's probably of limited utility.

I don't have VVT, so I can't comment there.

you should also set new limits to the following:
- fine ignition learning range (engine load)
- coarse ignition learning range (engine load)
- knock learning load thresholds

What sorts of changes do people usually make to these? I've thought about tweaking mine for more high-RPM listening, but didn't want to end up with noise artifacts.

/Andrew

[calvinc]

What sorts of changes do people usually make to these? I've thought about tweaking mine for more high-RPM listening, but didn't want to end up with noise artifacts.

high rpm listening isnt too much of a neccessity but adjusting the load parameters is. peak cylinder pressure occurs at peak torque, this is also where you're most likely to get det so keep the ECU listening to the stock knock sensor incase you're running maps that are too aggressive. det at low rpm is the most damaging.

calvin.