12P Flex Fuel Discussion

[Holden202T]

I think the thing people need to comprehend is the oxygen sensor is showing the percentage of oxygen left in the exhaust, for any given fuel with the perfect burn you should have the same amount of oxygen content which is why lambda 1 = 14.7 for petrol, 6.4 for methanol, 9.7 for e85 etc .....

the thing is, to get that perfect burn you need a significantly different amount of fuel injected with 1 part of air to get the same lambda 1 result.

right or wrong, regardless of how you tune, if you don't change the AFR tables in the tune to suit the fuel you are using then you are not 100% adjusting everything properly..... and by saying your not doing it properly if that only means the numbers don't look right and there is no other effect to the tune then its still wrong.

the amount of times i have seen tunes where the AFR table is still petrol and its a methanol tune or whatever is just amazing, and the tuner doesn't give a shit, but anyone else that gets the tune doesn't know whats been done and to get it back to "correct" its really a matter of starting from scratch.

now im not saying i am perfect, but i always aim to try and learn the right way to handle the tuning and to do that i believe you need to understand the right way to offset the right tables to provide the right real world results..... if others don't want to take the time to learn that then i honestly don't care, thats your choice, but i am simply trying help you understand what the right way to do things is and why.

[vlad01]

Well said and this is also they way I always aim to do it.

[antus]

The ecm works in afr so its a hack not to update the target afr, lambda makes more sense for tuning but its a pain to adjust all the adx to display it.

[Chuff]

The ecm works in afr so its a hack not to update the target afr, lambda makes more sense for tuning but its a pain to adjust all the adx to display it.

Exactly what changes to the ADX would be required?

[antus]

Change the equasions for everything afr to calculate lambda instead.

Although im not sure target afr to lambda would make sense, but would be needed if you displayed wideband as lambda and wanted to overlay the two. Hmmmm. Might need to create a new item for target afr (e85) target afr (methanol) etc so you can still see the pcms real target afr and a hacky one by fuel type you can oberlay with wideband lambda. Yuck, maybe its too far outside of how the ecm was designed to be worth it.

[VL400]

There has actually been a bit of discussion behind the scenes on what the best way to go about this is. Will just discuss fuel here but spark will also requires some changes but no where near as drastic.

As most will agree the most accurate (and correct) way is to change everything AFR in the ECU to operate as lambda, calculating a final target lambda and then from a lookup table of ethanol vs stoich AFR create a final target AFR. For example a calculated final lambda of 0.95, lookup table says that with 70% ethanol stoich is 10.7AFR, 0.95 * 10.7 = 10.16 target AFR.

Unfortunately there are a number of reasons that this is impractical. Being in assembly there are tricks to make the code run faster at the expense of ease of modification. So with AFR stored as X / 10 allowing 0-25.5AFR it wont store a lambda value with any sort of usable resolution, this means a new format allowing 0-2 lambda or X / 128 (this gives lambda resolution slightly better than 0.01 which is acceptable). Every single calculation then needs to be changed - so all the target AFR tables, warmup offsets, lean cruise, idle etc etc. Trying to tune with the now variable AFR after these changes would be a nightmare, so everything would be (including the wideband reading) done in lambda, just a new item added to the ALDL stream for the variable stoich AFR (a calculated target AFR can then be done in TP if the user wanted to know this). This means for the handful of users wanting flex fuel it effects everybody.

The last point is my major reason for not wanting to go down this path. There wont be a huge take up for flex fuel, and at this stage it will likely require extra hardware to get an 808 and 424 reading an ethanol sensor. There have been minor changes to 11P and 12P over the years (two come to mind, 11P knock counts and 12P baro compensation) causing so much confusion and extra support time that this change to lambda would really require a fork in the 11/12P code - ie two versions, you need flex fuel and lambda or you dont. Again more work for very little takeup.

Right now the most practical solution is still a pulse width modifier based on ethanol content with a new data stream item showing said modifier. This is how most aftermarket ECUs operating in AFR work (even some lambda ECUs). It also means nothing changes for the vast majority of users, they would just see the modifier at 0% much like 1 bar users see the boost multiplier. But its not strictly the correct way to go about this...

Discuss away!!

[Holden202T]

your the one that has to make the code work so happy to go with whatever works for you.

also i agree, having to re-write the whole code for a few people is not really practical, plus everyone already knows it how it is and it also works in with how pretty much all other delco's work so makes sense to keep them all similar if we can.

at the end of the day, whichever way you do it will give the desired outcome which is to allow you to go from straight petrol to straight e85 and anything in between in a seamless fashion and thats really all the flex users need.

[Dylan]

Your idea sounds like the go. One code base for everyone is the go.

[0081]

The 2nd option does sound easier.

I think more people would consider the 808 and 424 in conversations if it was capable of E85 also.

Im surprised E85 is not more popular in general with newer cars

[vlad01]

Im assuming you mean flex fuel functionality not e85 itself as its totally capable of running on it.


e85 isn't as popular as there is little public info that isn't propaganda against it or completely misinformative.

So many people I have mentioned using e85 to just bash it and call it rubbish and none with have actually used it. I have come across one guy who tried it in his car without any idea that it has to be tuned for it lol.

People I know who actually use it and tune on it love it, some that have fully converted to using it on all their dailys.