Engine/ECU mod ideas for mileage

[jcantara]

On my last car, I had rewritten some of the code in the ECU to help with mileage (and performance). I had grand plans to keep developing further, but unfortunately I had an unforeseen mechanical problem (are any problems foreseen?) involving belts breaking, pistons and valves getting intimate.

Anyhow, here's what I had actually done before the car died, with some explanations, and cautions:
Advanced timing in the low load / low rpm region of the timing maps. I was only able to do this because I constantly (using my palm pilot mounted to the dash) monitored the engine for knock and could be on the ragged edge of safety without actually going over it. Factors like brands of gasoline and intake-temperature greatly affected the ability to advance the timing, so I had to keep a close eye on it. I would say that most cars in most reasonable circumstances could advance the timing a few degrees without having to worry or monitor it; but I was running 10+ extra degrees advanced at times.

I leaned out the low load / low rpm region of the fuel maps. This obviously helps with fuel consumption, because the car is using less fuel for a given load. This does have some environmental ramifications though - even though the mileage was increased and fuel consumption overall decreased, some elements of the emissions were actually increased (I didn't measure this) over normal. I forget the specific emissions, I believe it is the CO2, but I don't have the info in front of me. Again, I monitored the knock with my palm pilot, and also had a wideband O2 sensor installed, and an EGT (exhaust gas temperature) gauge to keep everything within limits. I also had to instruct the ECU to not compensate out my lean-burn changes, and run in open-loop, otherwise it would compensate out my changes in closed-loop. Basically, I used my brain and extra gauges to manually run in closed-loop, haha.

Both of the above things didn't really require any active participation aside from monitoring everything to make sure nothing was breaky. At this point I could get 30 MPG easily (from a turbocharged AWD sportscar). Not bad for a car rated 19/26 city/highway.

I had a few ideas I wanted to try though to improve fuel economy further, but never got a chance to:
I wanted a special extra-lean-burn mode of the car for the highway. I wanted to be able to open the throttle body completely (to remove that restriction) and let the turbo spool up (extra efficiency gained from exhaust energy scavenging). In order to do this, I'd obviously need to limit the power somehow, so I wouldn't keep accelerating. I had a few ideas on how to accomplish this: I could cut fuel to certain cylinders, so I could run on 3, 2, 1 or even 0 cylinders depending on how much power I needed (I believe GM does this on their SUVs); I could also run extremely extremely lean, injecting barely enough fuel to keep the car moving forward, while being too lean to cause egt problems (egt soars when running lean, but then starts to drop if you continue to get leaner, like way leaner). Also, I could combine the two methods if necessary. Another method would be to have the ECU look to keep the car at a certain speed, and if I was over that speed, cut fuel, if I was below, just inject fuel as if at WOT (wide open throttle). This would effectively work the engine more like a diesel, controlling the fuel delivery rather than the air delivery. I would need a separate input to the ECU to control fuel-flow (since the throttle would be fully open all the time; but I had that sorted out already (used a potentiometer to control my 2-step rev limiter for drag racing). This obviously would require a lot of monitoring to keep track of the state of things, and a lot of fiddling while driving to keep the car at a certain speed; but I believe a large mileage increase could be seen.

Another idea in conjunction with the above would be to keep the turbocharger wastegate fully open while in the above mode. I know that (on my car anyway) with the wastegate completely open, I could still build significant boost and it helps open up the exhaust path.

-Jesse

[JohnnyGrey]

You'll probably see great gains from keeping your turbo's wastegate open during lean cruise. You can open that throttle wider while relieving the exhaust path.

I'm running a closed loop 17.4:1 ratio right now and it's working out pretty well. No pinging to speak of and power delivery is smooth enough. I'm using an AEM wideband gauge with some custom circuitry to lie to the ECU. I can train the ECU around two different variable AFRs with the flick of a switch.

Don't worry about the emissions. Even during lean burn, your 2.0L will produce fewer NOX compounds than your average V6 family car. If you look at total NOX compounds produced instead of PPM, they're rather low.

[jcantara]

Originally Posted by JohnnyGrey

Don't worry about the emissions. Even during lean burn, your 2.0L will produce fewer NOX compounds than your average V6 family car. If you look at total NOX compounds produced instead of PPM, they're rather low.

Ah, that's great info! Does the NOX actually decrease (total sum) in lean-burn, even though PPM wise it increases? Or is it around the same amount as with stoich-burn?

-Jesse

[JohnnyGrey]

Here's a nice chart from Toyota:



Notice that at the very worst, NOX is 20% higher than the stock tune, which is nothing to worry about. Conceivably, since we're all driving with a light foot we're probably releasing fewer total gases than the average driver despite the 20% lean-burn NOX increase. Also notice that if you go even leaner, NOX starts to come back down to stock levels.

GM's ECUs have a "lean cruise" mode, but it's disabled in all of their US cars to comply with US law. If you multiply that 20% NOX increase out to every Escalade, Yukon, Navigator, you can see why it wouldn't be a great thing in crowded cities. However, I'm not going to let anyone make me feel bad because my modified 2.2L might be producing almost as much NOX as their stock 3.0L.

[sc2dave]

Originally Posted by jcantara

On my last car, I had rewritten some of the code in the ECU to help with mileage -Jesse

Do you know how to rewrite code for a 95 Camry ecu? also, what size resistor can i use to trick the ecm into thinking that the intake air is hotter?

[micondie]

For a good primer on tricking your ECM, read this Auto Speed Series http://www.autospeed.com.au/cms/A_110897/article.htm
What you want to do is trick your ECM into thinking that the air is colder which will cause it to advance timing.

[sc2dave]

Originally Posted by micondie

For a good primer on tricking your ECM, read this Auto Speed Series http://www.autospeed.com.au/cms/A_110897/article.htm
What you want to do is trick your ECM into thinking that the air is colder which will cause it to advance timing.

air is colder and it will waste more gas.

[Phantom]

Originally Posted by sc2dave

air is colder and it will waste more gas.

No the ECU thinks the air is colder so it advances timing. The MAF/MAP measure the amount of incoming air and adjust fuel based on that not air temp.

[PaleMelanesian]

My car gets its best mileage in the middle of the hot Texas summer. Hot air is less dense, to it takes less to fill the cylinders. It takes less fuel to get the right mix with the air. Basically, it behaves like a smaller engine.

[TheEnemy]

Originally Posted by JohnnyGrey

You'll probably see great gains from keeping your turbo's wastegate open during lean cruise. You can open that throttle wider while relieving the exhaust path.

I'm running a closed loop 17.4:1 ratio right now and it's working out pretty well. No pinging to speak of and power delivery is smooth enough. I'm using an AEM wideband gauge with some custom circuitry to lie to the ECU. I can train the ECU around two different variable AFRs with the flick of a switch.

Don't worry about the emissions. Even during lean burn, your 2.0L will produce fewer NOX compounds than your average V6 family car. If you look at total NOX compounds produced instead of PPM, they're rather low.

Apparently the lean ping is more of a myth

From an aviation website
Pelican's Perch #43:<br>Detonation Myths
I figure they should be a good source concerning mixture since they tend to adjust it on the fly.

Understanding compression on a pump gas street motor - Engine Masters

The popular hot rodding article has quite a bit of good information, somewhere on the site they have a good article on mixture as well that confirms the avweb website's statement about leaner than stoch reduces pinging, as does richer.