[dirty_d]

Has anyone tried anything crazy like a "lean burn" mode for cruising on the highway? I figure if you disabled closed loop mode, lean the mixture way out and inject water into the incoming air for cooling and knock prevention that would help out the fuel efficiency. The idea is to get the engine to put out the low 15-20 hp for cruising without having to have the engine waste maybe another 15hp(guess) internally sucking air past the nearly closed throttle plate. ie you push your foot down a lot further to get the same power out of the engine while using around the same amount of gas. I don't think 30-35 mpg is too far fetched on a 5.0 if this was done right. I think that would be pretty sweet to have a 20 year old 300hp car that gets 30mpg... I'm not really in any rush to burn a hole through a piston though

[copermn]

that seems like a lot of work to do on such an old engine, you know? if someone has done that sort of thing before that's really cool and power to them, I just think it'd be really hard to do and take a lot of time.

[dirty_d]

I was thinking just to put a voltage divider using a potentiometer on the MAF output, so you can trick the computer into thinking less air is entering the engine. It would have to be in open loop mode though. I'm not sure how you would force the computer into open loop mode. Maybe unlpugging the O2 sensors? that might put it into some other failure mode though.

[fogged306]

Open loop dumps more fuel. You want to really do that you're going to need a tuner like a quarterhorse, the software, and a wideband. Then you can lean out your cruise and idle and add a ton of light load timing to pick up MPG. But it's gonna take you a long time to re-coup the money spent on the parts and software in fuel savings.

[302Army187]

by time you would mod the car with simple things, tuner software, and such, your going to spend quite a bit of money.


You can get a 4cyl beater car for $2,000, as much or less then you would spend on the modifications, and get 35mpg in almost any car like that.

[dirty_d]

Hmm this is true. I can envision an el cheapo method of doing this. a switch that switches the O2 sensors from the computer to an analog voltmeter mounted in the car. and a potentiometer wired to the maf signal. Switching the O2 sensors to the voltmeter should put the car into open loop mode, then fiddling with the potentiometer and watching the voltmeter you can dial the mixture to around 14.7:1 . Then using a little math you should be able to calculate at what resistance the potentiometer should be to put the mixture at some desired ratio. Then you can just mark the ratios down around the knob. Another potentiometer could control a servo motor that rotates the distributor to the desired timing, just kidding .

[WickedSnake00]

First off, this is overly complex, and assumes extended steady state. Switching in and out of this mode would be quite difficult.

But here's my thought-
You would need a stand alone wideband to watch. You'd need to actually amplify the upstream HOS signal. (1V = rich, 0V = lean, so you'd need it to think it's richer than it is.) I would not mess with the MAF.

To counteract knock, you'd need at least 2 things. I'd suggest using a great deal of EGR to effectively reduce the displacement of the motor. You'll need to be careful, because if you use a great deal of EGR you'll need an EGR cooler, which will increase heat rejection, and possibly require a new radiator. You'd also need to be able to pull lots of timing.

All of that is based on my knowledge working in the natural gas engine industry. Keep in mind natural gas is more suited to lean-burn due to it's higher knock rating. See the Cummins BG+ and CG+ engines for that. Also note that the newer NG engines have gone stoich, so what does that tell you?

And of course you'd need a way to control all this simultaneously. That is the difficult part. Overall I'd say the time and effort required is not worth it.

[fogged306]

Narrowband o2's don't measure AFR. They switch between rich and lean constantly. The more often they switch the closer the mixture is to stoich and that's how the computer knows where to keep the mixture. Also, once you put the car into open loop, the computer ignores the o2 sensor readings, so forcing open loop and then doing anything to the o2 sensor readings will get you nowhere. You should pick up a book on how the EEC works and then draw conclusions.

[dirty_d]

Right, I didn't want to alter the O2 sensor signal, just disconnect them to force open loop. Altering the MAF signal was my plan to change the AFR. When it goes into open loop it should already be just about dead on 14.7:1 because it will still be using the adaptive tables that were being updated in closed loop. A change in the MAF signal based on the transfer function should create a proportional change in the AFR. The last thing I wanna do is destroy my motor, but at very light loads(steady 65mph) I don't think there would be enough heat or pressure in the cylinders for detonation to even occur would there?

[WickedSnake00]

Originally Posted by dirty_d Right, I didn't want to alter the O2 sensor signal, just disconnect them to force open loop. Altering the MAF signal was my plan to change the AFR. When it goes into open loop it should already be just about dead on 14.7:1 because it will still be using the adaptive tables that were being updated in closed loop. A change in the MAF signal based on the transfer function should create a proportional change in the AFR. The last thing I wanna do is destroy my motor, but at very light loads(steady 65mph) I don't think there would be enough heat or pressure in the cylinders for detonation to even occur would there?

Still sounds awfully risky. Sure you can trick it into flowing more air than it should, but keeping it at a specific lambda that way or in open loop isn't likely to work. There's a reason we have closed loop, and that is to adjust everything based on feedback. Without the feedback we're way off. It can go both ways depending on conditions. If it goes rich you're losing the benefit, and if it goes lean you'll find yourself saying "why does my motor sound like marbles in a can?"

[fogged306]

I mis-read where you said you were putting the potentiometer on the MAFv wire. In theory, yes that would work to a point. But like wickedsnake said, going to be tough to keep the AFR where you want it. If you lean it out a certain percentage at light throttle it's not going to stay the same AFR at every point on the MAF curve. Believe me. With quarterhorse software, you can actually take your MAF curve and multiply the whole thing or parts of it by percentages. That can change your AFR's from all spot on, to some very rich/lean and others only slightly more rich/lean. To get the curve where you want it to be it takes a lot of datalogging and a lot of fine tuning. The very basic way you're trying to do it is going to be extremely tough to pull off with good results I think. If you're going to do it, post up how it worked for you.

[dirty_d]

Ok I will, but it won't be for a while. I'm still working on my gt40p install and plan to enjoy them for a while before I take any risks, lol. Also I was thinking of wiring a simple analog voltmeter to a fuel injector. The reading on the meter should be the average voltage. a higher voltage should represent the lowest fuel usage since the EEC grounds them to turn them on. That would take the guesswork out of tuning for the best mpg if you maintain a constant speed. To get the actual MPG I would still just use the odometer and gallons at the fuel pump.

I'm not trying to get this to work for various speeds under any conditions. Basically I just wanted to be able just "turn it on" when I'm already going 65 on flat ground and keep the throttle steady. If i had to speed up or slow down or something I'd just turn it off, or have it automatically do that based on a range of TPS voltages. This isn't necessarily something that's going to be practical, just an experiment.

[fogged306]

You'd be better off setting it for AC instead of DC and looking for lower voltage.

[WickedSnake00]

Originally Posted by dirty_d Ok I will, but it won't be for a while. I'm still working on my gt40p install and plan to enjoy them for a while before I take any risks, lol. Also I was thinking of wiring a simple analog voltmeter to a fuel injector. The reading on the meter should be the average voltage. a higher voltage should represent the lowest fuel usage since the EEC grounds them to turn them on. That would take the guesswork out of tuning for the best mpg if you maintain a constant speed. To get the actual MPG I would still just use the odometer and gallons at the fuel pump. I'm not trying to get this to work for various speeds under any conditions. Basically I just wanted to be able just "turn it on" when I'm already going 65 on flat ground and keep the throttle steady. If i had to speed up or slow down or something I'd just turn it off, or have it automatically do that based on a range of TPS voltages. This isn't necessarily something that's going to be practical, just an experiment.

For all the effort you'd be putting into this, you'd be better off running Megasquirt.