E85 is in the Tank

[JoshBoody]

Thought I would show my first E85 experience last night over here also.

Took the E85 plunge… only 20 miles from my house, so don’t know why it took me so long.

Below are logs off meth and on meth… back to back runs.

I’m running 30% E85 and my goal was to pick tuning that would be very compatible with or without E85… in case I can’t find a station. Without it, the tuning would be a little aggressive, but not too bad. And trims look very good off corn juice also.

The setup is 1.12 scalar with ATR with slightly higher load (around 145 peak) then stock. Procede open loop table is custom and lowish settings of 60 and 50%.

I have a very conservative setup, as there was no timing reduction remotely in sight in 100deg weather. I’m going to keep it here for now, until I finally get around to replacing the clutch… colder weather will probably make this a necessity.

This is roughly a 390HP setup in 100deg weather. Pretty strong I think, but also conservative (which I like).

E85 is great!
Attached Images

[dzenno@PTF]

Why are you upping fuel pressure using the procede? And you should really sto calling it open loop

[JoshBoody]

Why are you upping fuel pressure using the procede? And you should really sto calling it open loop

Why.... it is OL and directly changes fuel volume. Reason... cause i don't want to deal with boost alterations by the DME. And I can make changes with a laptop instead of reflashing... I want my base mapping as simple as possible.

[dzenno@PTF]

No matter what you do to that FP the car is always running closed loop fuel control...open loop confuses people as it makes them think it exists...all you're doing is upping fuel pressure...why not call it what it is

How's signal altering with a piggy simpler than just flashing your car? Map switching?

[JoshBoody]

what are you talking about. There's a fuel base and CL adjustments which have +/- authority from this base. I altered the base mapping called OL... set initially by Cobb with some bias by procede... works very well. I bet my trims are more stable then yours and I have more adjustment options.

far from map switching, you are flashing each time.

[dzenno@PTF]

what are YOU talking about? lol

Open loop is fuel control when closed loop is not active. This car runs closed loop 100% of the time, there is no open loop...again, you're just upping fuel pressure...if you're going off the procede labeling then fine its just a wrong label

For others reading this, just to clarify:

Open Loop
Open loop simply means there is no feedback of the result to the ECU. In our case, it means there is no sensing or measuring of the exhaust gas to see how the motor is running. The fuel injected is determined by the RPM and throttle position, derived from fuel injector pulse width numbers stored in the fuel maps, and is trimmed for environmental conditions due to air temperature, air pressure and engine temperature.

Closed Loop
Closed loop means there is feedback of the result to the ECU. In our case, it means there is sensing or measuring of the exhaust gas to see how the motor is running. This sensing is done by the probes in the exhaust which generate a voltage based on the gas around it. These probes are referred to as Oxygen sensors, Lambda sensors, O2 sensors, Exhaust Gas sensors and probably a few other names as well. I’ll call it a Lambda sensor.

N54 runs closed loop 100% of the time, or, as its said, it is fully closed loop.

So what "open" loop does procede let you configure? I know you know the difference, I'm just being blunt for the sake of clarity.

[DCAFS]

Seems simple enough.......if you raise fuel pressure then you are going to close that loop more quickly.

So it should have some influence.

[dzenno@PTF]

Seems simple enough.......if you raise fuel pressure then you are going to close that loop more quickly.

So it should have some influence.

I'm not saying changing or altering fuel pressure has no effect on fueling. It most certainly does. I'm just referring to the terminology being used saying that "open loop" doesn't exist on this car and can't be tuned. You can do whatever you wish with fuel (o2 bias, fuel pressure signal bias, flash fp targets in the flash, changing AFR targets using a flash), no matter what you do, the DME runs closed loop

You can run 2800psi pressure on the HPFP and you'll still be hitting the required AFR targets. If this were indeed open loop it'd affect your AFR and make it stupid rich if that 2800psi wasn't the pressure required to hit that target. Due to closed loop however, the DME will close that IPW down as much as it can (STFTs will go full negative to -34% max) to try its best to stay on its targeted AFR for given rpm/load

[JoshBoody]

I’m bored, waiting for the women, so let me correct you:

Open loop is fuel control when closed loop is not active. This car runs closed loop 100% of the time, there is no open loop...again, you're just upping fuel pressure...if you're going off the procede labeling then fine its just a wrong label

Wrong, n54 may run in open loop if it determines there could be a fault with primarily the O2 sensors (possibly other causes)… it will not throw a code right away and it’s possible during WOT, thus I prefer trims close to 0.

Open Loop
Open loop simply means there is no feedback of the result to the ECU. In our case, it means there is no sensing or measuring of the exhaust gas to see how the motor is running. The fuel injected is determined by the RPM and throttle position, derived from fuel injector pulse width numbers stored in the fuel maps, and is trimmed for environmental conditions due to air temperature, air pressure and engine temperature.

N54 has a base map, which is also the open loop mapping if needed. And this is NOT based on throttle position, but instead load determined from MAP, temp, rpm, etc (maybe throttle if you are beyond manifold MAP). Pressure is set from actual load and rpm, and IPW determined from actual rail pressure.

Closed Loop
Closed loop means there is feedback of the result to the ECU. In our case, it means there is sensing or measuring of the exhaust gas to see how the motor is running. This sensing is done by the probes in the exhaust which generate a voltage based on the gas around it. These probes are referred to as Oxygen sensors, Lambda sensors, O2 sensors, Exhaust Gas sensors and probably a few other names as well. I’ll call it a Lambda sensor.

This probe controller generates a current (not voltage) reported to the DME, which is basically the amount of O2 pumped in or remove from the exhaust gas relative to a stoich reference cell. But I never did get a good grasp of exactly how WBs work.

[dzenno@PTF]

Its ok no need to continue the discussion, cheers

[newb335guy]

dzenno

sent you a pm about meth was hoping you would look over it please and give some feedback.

[Terry@BMS]

I would zero out open loop on the procede side, move your lambda targets over to the flash side and zero out the o2 bias (set to 50 in their interface, iirc) on the procede side, then adjust the Cobb fuel scalar as needed to neutral the trims out. Maybe 1.12 for example. Basically, remove the fueling from the procede entirely. Now what lambda values to target under E85 are open for debate. I bought a 55 gallon drum to make fill ups easier (no station close by) and am doing some addl testing in that area now.

[Terry@BMS]

no matter what you do, the DME runs closed loop

It's true that the DME is programmed to run in closed loop mode most of the time, but if you unplug an o2 sensor, for example, the motor continues to run and target a close to stoich air/fuel ratio in fault mode. So it's clearly using a back end open loop table. The same table that our fuel trims are based off. Referring to fuel pressure biasing changes as open loop changes is not entirely accurate but it's not because open loop doesn't exist. FWIW we started using the term to avoid customer confusion between the JB4 and procede.

[dzenno@PTF]

It's true that the DME is programmed to run in closed loop mode most of the time, but if you unplug an o2 sensor, for example, the motor continues to run and target a close to stoich air/fuel ratio in fault mode. So it's clearly using a back end open loop table. The same table that our fuel trims are based off. Referring to fuel pressure biasing changes as open loop changes is not entirely accurate but it's not because open loop doesn't exist. FWIW we started using the term to avoid customer confusion between the JB4 and procede.

I don't disagree and do know that. There is an open loop table but the point I was trying to get across is that it is not used under normal operating conditions. There's also a failsafe Timing table but those tables are outside the normal operating conditions of the car and aren't being used unless there are unforeseen issues (o2 sensor going bad, pump dying, injector dying, etc).

Basically to reiterate, when the car is running 100% without issues it runs closed loop fuel control all the time. Changing fuel pressure does not equal tuning open loop. I guess it doesn't really matter we can call it tuning the hula hoop at the end of the day

[Terry@BMS]

The "open loop" changes we make on the piggyback side have the same net effect to the tuning as the fuel scalar Cobb recently added. Only instead of it being static it's dynamic more like the 335is fuel scalar map they added.

[JoshBoody]

I would zero out open loop on the procede side, move your lambda targets over to the flash side and zero out the o2 bias (set to 50 in their interface, iirc) on the procede side, then adjust the Cobb fuel scalar as needed to neutral the trims out. Maybe 1.12 for example. Basically, remove the fueling from the procede entirely. Now what lambda values to target under E85 are open for debate. I bought a 55 gallon drum to make fill ups easier (no station close by) and am doing some addl testing in that area now.

I disagree on the base fuel volume (is this ok for term DZ?). Keeping some offset doesn't limit you to have a matching load/boost profile which could change due to environment, or even on/off meth iats. Plus the DME drops boost target considerably in the top end. So i add just a slight offset with pressure for zeroish trims with no boost fluctuation. You could always add more scalar, but i found i can get more stable trims with the 2d mapping.

I do use ATR for lambda though.

[Terry@BMS]

Technically the best way to accommodate that would be to increase the flash load target (which is disconnected from boost control anyway) at higher RPM where needed. But I agree that as long as the lambda targets are set on the flash side the rest will work fine either way.

[JoshBoody]

Technically the best way to accommodate that would be to increase the flash load target (which is disconnected from boost control anyway) at higher RPM where needed. But I agree that as long as the lambda targets are set on the flash side the rest will work fine either way.

Im curious then how you set boost. Basically linear to throttle i would think... BUT also need to mirror the DME setpoint to some degree. So you can never match effective load exactly. And when consider IAT boost fluctuation and 2 maps (non & meth), you get farther from the fuel requirement. Anyway just curious about your approach.

[Terry@BMS]

Im curious then how you set boost. Basically linear to throttle i would think... BUT also need to mirror the DME setpoint to some degree. So you can never match effective load exactly. And when consider IAT boost fluctuation and 2 maps (non & meth), you get farther from the fuel requirement. Anyway just curious about your approach.

Ideally you would want the DME load to fall in the ballpark, so it's calculations are as close as possible, on the map you use most of the time. And if you really wanted to be a perfectionist you could match the throttle to load curve as well. There is no easy DME solution for meth on vs. off so you would optimize the DME for whatever you normally run and then use the piggyback side open loop adjustment for the alternative to keep the trims in check.

The reality is as long as you move the lambda targets over to the DME side and then add more room in to the fuel trims by upping the fuel scalar, reducing the open loop required on the piggy side, you are 95% of where you would want to be anyway. +34% on the fuel trims can take up a lot of slack and as long as they don't max out or bottom out it's not much of an issue. So it sounds like you are right there. I'm just pointing out additional steps can be taken to neutral those trims other than just the fuel scalar and lambda table setup on the flash side, and open loop on the piggy side. Like adding a bit of load target where needed. Before the fuel scalar a week or two ago this was all we could do.