M1 Flex Fuel - Engine Efficiency keeps changing

Hello,

I have been tuning a package with flex fuel support enabled. For the most part things are looking ok except for fueling.

The issue:

1. calibrate the direct injector charge cooling gain for E10, E25, E50, E75 via the standard means (the final calibration for me is 55C (@E10) to 85C (@E75)).

2. if the engine efficiency is tuned at E10, fueling at higher ethanol %'s is off. I usually have to add fueling (increase engine efficiency), but then my tune for E10 will be off. I am using alternative fuel target lambda blend table and alternative fuel ign timing blend table but I do not see an alternative fuel engine efficiency blend table.

What are your suggestions? Just increase the charge cooling gain even more for higher ethanol % even though they are already properly calibrated?

Thanks,

Kevin

Hi Kevin,

Those charge cooling gains are higher than they should be, I have a range of 14C to 24C in my car. Have a look at your Complete Fuel Calculations under the Monitor workbook, there will most likely be a value in the first set of calculations that is reporting incorrectly, rectify this, and then redo your Charge cooling. The Engine Efficiency table should not need to be changed when the fuel ethanol content changes.

Thanks @BlackRex.

Are you talking about an engine that has Direct Injection? The charge cooling effect of DI is considerably higher than PI. In one of Andre's webinars on tuning the engine efficiency table he showed that on the GT86 he had to increase charge cooling to 60C on pump gas.

I will however check the Complete Fuel Calculations.

Thanks,

Kevin

The charge cooling gains on the 86 package are a little unusual in my opinion. The values that are generated if you go through the correct approach to tuning charge cooling gain are unrealistically high. There are some peculiarities with the fuel model on the 86 package however where for example you can't change the secondary contribution table without revisiting the efficiency table as the lambda changes. MoTeC have put this down to high frequency oscillations in the low pressure fuel system as a result of the operation of the DI pump's spill valve.

In theory the engine efficiency will remain the same as you change fuel because it's not affecting the airflow into the engine (to be 100% accurate we may actually see a tiny increase in VE on ethanol fuel due to the higher charge cooling affect increasing the charge density, but in reality any real change is likely to be so small as to not matter). The real world tuning requirements however will depend on the accuracy of the fuel model and the inputs into the fuel model so for the flex fuel system to work as intended it's vital for everything to be 100% correct.

If something isn't correct when tuning for a single fuel then while it obviously isn't ideal, we can tune around this and just create an inaccuracy in the VE table. When we're tuning a flex fuel system these sort of inaccuracies that we have burned into the VE table will result in the flex fuel system not working correctly.

One of the other considerations is that the injector flow will generally be a little lower on ethanol when compared to pump gasoline due to the difference in viscosity. If I remember correctly MoTeC allow you to account for this when tuning for an alternative fuel but of course it's hard to get solid data and the flow difference will depend on the injector.

As Andre has touched on, DI is quite different!

Typical DI Charge cooling numbers are similar to the PI however, I did a lot of the work on the Toyota 86 Package, as well as the race series here for the cars, and the DI Charge cooling numbers for those cars is 13.5. Other turbo and super charged 86's that I have worked on have numbers around 13 for Petrol, and up to 26 for E85.

If you want to load your package here, I can have a look at it for you.

@BlackRex. Sorry for the late response as I have been traveling. Thank you for your generous offer. I would very much appreciate any insight you have into the setting in my package. I have attached it.

@BlackRex, just bumping this once more to hopefully catch your attention as I would greatly appreciate any insight. Thanks.

Hi Kevin,

For some reason I didn't see your previous reply.

Having a look through the Package, one area sticks out, and that is that the Alternative Fuel Injection Timing values are all 0, these should be around 400 degrees as a starting point, and will affect the lambda values being read. Try changing theses values to 400 and see if it makes a difference.

@BlackRex

Thanks for your reply. I didn't think I needed to populate that table since the Alternative Fuel Injection Timing Primary and Secondary Blend tables are set to 0. So no matter what the ethanol concentration it will never blend any of these tables in with the pump gas tables.

I will try it nonetheless.

Thanks,

Kevin

What are your logged injection timing values right now? Is it following the main (E10) injection timing maps?

@Arghx7 injection timing values track the main E10 injection timing maps at any ethanol concentration.

@BlackRex setting Alternative Fuel Injection Timing values to 400 did not rectify the problem as the blend table was always set to 0 and my injection timing has always been following the pump gas tables. Please feel free to take this offline as I think this may be a specific problem with my package. If you wish please contact me at kevin.schweigert@gmail.com. Thank you so much.