mark kibort

kind of off topic, but lately with the gas prices going well over 4 bucks a gallon, i started to think of how to improve gas mileage.

A friend of mine got all excited when he was reading about the Water4Gas websites. Sorry buddy, you cant extract enough hydrogen out of water with 12volts to use by our engine at WOT, let alone at cruise. So, i started to think of ways to get more efficiency out of or motors or mainly, out of my truck!

So, thinking about airplanes, when you cruise, there is a technique called "lean of stoich" that is used for better efficiency. why cant we do something like this in our cars on the hyway on long trips or part throttle? First of all, most of us know when you run too lean at idlel, your emissions go nuts in NOX. too rich, and its CO. But, if planes can do it, why cant we?

It might be some kind of mechanism to cut out the 02 sensor, go open loop, and then back the fuel pressure WAY WAY back. 17:1 afr range. maybe reducig the voltage to the fuel pump to 6volts or something. better if it could be done by shark tuning so that when you are at part throttle or less, mixtrues goe to 17:1, unless you are idling or warming up.

I bet the fuel efficiency would go up 20%

hacker-pschorr

Mercedes is working on just that - for other countries. They say US fuel has too much sulfur for their new setup:

http://www.autoblog.com/2011/03/17/m...icient-engine/

dr bob

Plenty of modern engine management systems make serious adjustments to fuel and spark maps for better cruising and other low-load fuel economy. Almost exclusively thes are drive-by-wire systems. The engine management system sees the input from the pedal, changes maps, and only then actually opens the throttle for you. A pure reactive system like ours could start out lean with lots of advance, and not directly see that dump of extra air as you open the throttle. It would go way lean first , ping and miss with the lean load demand with high advance, have a bit of a bog as the fuel tried to catch up to the new air flow. Too much pollution, driveability would suck bigtime. Drive-by-wire throttle is a huge benefit for the economy and tailpipe folks. Better ones are almost unnoticeable when driving. Still lots of room for improvement in them though.

John Speake

Fuel prices here have just reached £7 GBP an Imperial Gallon...... about 12 USD.

JWise

Mark - As you mention, lots of piston powered aviation engines get run "lean of peak" at considerable fuel savings and increase in range. However, that's generally in a steady state operating environment, much different from that in which a motor vehicle engine operates. I also wonder if avgas is perhaps lower in sulfur than what we use?

IcemanG17

& we complain about $4.....wow

However a 20% increase in economy would take a 20mpg cruise to 24mpg....at $4 per gallon thats only 3.3 cents per mile improvement....but at what risk?

jcorenman

You can, but you will need a SharkTuner ...
I've experimented with this, and it works-- with a number of caveats: I think continuous lean-running will wreck cats, and with/without cats the NOx is going to through the roof with lean mixtures. So you're trading an environmental kindness for an unkindness of a different sort. And I have no idea if there are any issues with running these engines with a lean mixture.

Bob, with ST you can limit your re-mapping to light load, mid-RPM cells-- the highway cruising part of the map. Doing that, it seems fine on transients-- increased throttle immediately moves the LH to a high-load row on the map, back to normal fuel plus the acceleration enrichment. And even a slight grade on the highway moved it ou of the lean area.

Cheers,

mark kibort

how did you do it? why would it wreck cats? too much NOX? the EGT's should be lower, thats the point of why it is done in the avaition world. lean of stoich, contrary to popular believe, EGTs and CHT's both go down. just until it starts to miss and richen it up a little. near 17 or 18:1 probably at least 20% lurking there. who cares about the environment. you think a few 100 928s are going to effect anything. The airplanes are doing it, so why cant we! PLus, it will allow me to run the 928 on long trips, burn less gas and in the end put out less polutants, so I have my own moral justification all worked out.

Yeah Bob, it seems if you did this with the chip , you would only modify the light load settings where there would be no chance of pinging when the load was increased suddenly. PLUS, you can still operate lean and use higher power settings and not have issues, you just dont get the power potential as you do on the other side of stoich. again, i was thinking a setting that only works to lean the mixture way out, on cruse settings or throttle positions less than 35% full throttle.

PorKen

You could do it with a wideband O2 with dual analog outputs. Just switch from one analog to another programmed for 17:1 (instead of 14.7) when the idle switch is off. (Idle will be lumpy if lean.) The O2 loop is slow by design, so there won't be a sudden lean spot when you switch. The loop programming will hover ± around 17:1 in it's normal sine wave pattern(s).

According to the internets, 15.4 makes the most lean power. (It's also the limit of a narrow band O2 sensor.) I have programmed my S3 chips to make the O2 loop keep a flat (no sine wave) 15.4 at cruise, and a flat 14.8 at idle (you can run the stock loop, too).

dr bob

At least where I live and drive, the opportunities for steady-state low-load cruising are rare at best. Closest thing is a run up the central valley, or trip through one of our near-b deserts. the part of the trip that gats us there is invariably hilly, so we'd be in and out of the low-load section of the map.

Mark, are you willing to risk your stroker engine to the experiment?

re: your reference to aircraft engines and lean-of-stoich operation. Aircraft engines typically start out with low mechanical compression ratios, and the cylinder pressures drop off as altitude increases. At normal GA cruise altitudes up to 10,000 feet AMSL, thinner air can reduce cylinder pressures by up to a quarter. Same effect you felt when you drove to Mammoth and decided to experiment with electric blowers on the intake. Anyway, the combination of low mechanical compression ratio and low air density plus the high octane requirements for avgas means that the probability of lean detonation is much lower than for our cars atsea level. So running leaner will cause lean misfire before it causes lean detonation. Yes, under that specific low-load condition, leaner mixture makes cooler EGT and CHT numbers. Your 928 engine sees low cylinder pressures like that right up to the point where engine load increases. Might be a steady-speed change with more throttle opening, might be a steady-flow change where RPM's decrease. Unless the engne management can detect the change before the cylinder pressure increases, you risk lean detonation. That's why drive-by-wire systems make this easy.

Jim, are you seeing a few knocks on the sensors when the controlers transit from light-load mid-RPM cruise into a higher-load area on the map? And how does the closed-loop mixture control handle the targeted weak mixture operation? Is it possible to put in a cruising map for touring, something I could switch in to using the low-octane loop jumper with a switch?

jcorenman

With a SharkTuner, and I don't know for sure that it will damage the cats.

I did not see any knock-retards, and didn't do anything special in that respect-- the LH seemed to get fuel in quick enough that it wasn't an issue. One of the things that happens on a stock intake is that, when cruising with light load, there is a fair bit of plenum with pretty high vacuum. When you open the throttle-- even a little-- there is a bucket-load of air goes past the MAF sensor, to fill the vacuum. You can't see the spike on the ST, but it certainly has to help the rise-time of the load signal.

With an '87+ car, the LH processor has two maps and the EZK has three, selected by the coding plug. The LH's "cat" map is intended for cat-equipped cars, and enables the NBO2 feedback and closed-loop operation under light/moderate loads. The "no-cat" map was originally for ROW cars without cat's, disables the NBO2 sensor even if present, and always runs open-loop. This is the map I used for my lean-running experiments.

And I agree, there's no advantage except on long, straight, boring highways-- which there seem to be a lot of, connecting the interesting bits.

dprantl

I've been doing this for months with my ST2. I just lean out the non-cat fuel map upper cells and force non-O2 sensor mode. I can dial ~18:1 and the car still runs smooth. I haven't gone for an extended highway trip like this, but I estimate I should get a little over 30mpg

Dan
'91 928GT S/C 475hp/460lb.ft

Tony

get these two...doing it for years in airplanes

mark kibort

yes! do a test . would love to see the results. 25% increase in fuel economy would be nice!
ha ha. now , how do i hook one of those mixture levers in to my 928.

Bob, my experiment is really for the chevy tahoe.

123quattro

If you change your cruising A/F ratio from 14.7:1 to 18:1 you would be using about 18.5% less fuel if I can still do math. However, the engine will make less power at that A/F so you'll have to run less throttled to regain the torque, which will increase efficiency. So, call it a 20% increase in fuel economy.

When not oscillating around the stoich point, the catalyst will quit working. It's a 3 way cat.

1. Reduction of nitrogen oxides to nitrogen and oxygen: 2NOx → xO2 + N2
2. Oxidation of carbon monoxide to carbon dioxide: 2CO + O2 → 2CO2
3. Oxidation of unburnt hydrocarbons (HC) to carbon dioxide and water: CxH2x+2 + [(3x+1)/2]O2 → xCO2 + (x+1)H2O

You wind up with tons of NOx, but not much CO or HC. You would turn your car into a gross poluter. You also run the risk of lean misfire. That can spike the temp in the cat since the combustion moves from in-cylinder to in-catalyst. It will melt in very short order if that happens under load. Then your car makes hardly any power due to massive back pressure restriction.