Batman we have lean burn! 49.6 mpg!

[cfg83]

cheapybob -

Originally Posted by cheapybob

Sorry, but real world results trump theory.

The 98 Trans Am, with its 5.7L V8 just ran the same 22 mi 60 mph test loop that I was using for my Saturn before. At 60 mph on cruise, it got 32.3 mpg. EPA ratings for the car when new were 17/26, and the best it has ever done before since the day it was new was 28 mpg on the highway.

BTW, if you noticed, my spark is not retarding. If it were, my fuel economy would drop. If the water temp or actual air temp gets too high, it will retard, and lose both power and economy. As I see it, the engine sensors did not detect preignition, therefore they did not retard spark. The engine is obviously in closed loop. My guess is that the programming is self tuning as it runs, and that it isn't sensing any problem, so it just tunes as best it can on the basis of what it reads from the sensors. It thinks the air is hot and less dense, and injects less fuel by reducing the pulse width on the injectors, just like it would if it were actually hot outside. The result must not be seen as a problem at the O2 sensors, or it would be trying to compensate.

Why are you so sure the IAT temp is not looked at by the computer? If it wasn't, it would have no effect. On my Saturn, if I change the switch from 100 degrees to 241 degrees or the other way around while driving down the highway at steady throttle on level ground, fuel economy changes by 10 or 15% up or down within 5 seconds, and it stays changed.

Is this change coming from the scangauge? There is the theory that the scangauge is being fooled and therefore reporting bogus MPG (the scangauge doesn't have access to the 02 output, so it has to infer the MPG from other inputs). However, you have also reported lower HP to go with the HAI+IAT resistor mod, so that would imply that there is something else happening to support what you are seeing.

My test is going to be the following :

Use a digital A/F ratio gauge to monitor the 02 sensor. Try each resistor at steady throttle on level ground at same speed. If the A/F ratio gauge reports the same value for each resistor, then it is reasonable to claim that the scangauge is being fooled.

Do you think that is a good test?

CarloSW2

[cheapybob]

I see no point. I don't care if or why the O2's don't get the computer to inject more fuel. All I wanted was better mpg out of it and "acceptable" performance.

The car took almost exactly the amount of fuel the scangage said was consumed two tanks in a row. That tells me the scangage is calculating consumption from injector pulse width correctly.

If its actually using what it says its using and going as far as it says it went (I tested it against interstate milemarkers), then mpg must be correct also.

Show me how that could be wrong?

[cfg83]

cheapybob -

Originally Posted by cheapybob

I see no point.

The car took almost exactly the amount of fuel the scangage said was consumed two tanks in a row. That tells me the scangage is calculating consumption from injector pulse width correctly.

If its actually using what it says its using and going as far as it says it went (I tested it against interstate milemarkers), then mpg must be correct also.

Show me how that could be wrong?

Ok, that's the info I wanted to know and was waiting to hear. I'm still gonna do the test for my own purposes, because I have other uses for info.

CarloSW2

[cheapybob]

Does your scangage report the correct amount of fuel used and correct number of miles traveled now? As long as both of those are correct, and avg mpg shown calcs to that when I fill, I see no point questioning it.

If not, why haven't you calibrated/adjusted it?

I know very little of O2 sensors other than how to replace them.

I looked at your mods, and they are similar to those on my Saturn with exception that on mine I can override the IAT value going to the computer.

[cheapybob]

By "digital A/F gauge" do you mean one of the gizmos with about a dozen leds marked from Lean to Rich that you watch to see how many leds light up? If so, yes, I did have one of those connected to the O2 sensor originally until my wife disconnected it one day. It did show it running a bit leaner with the HAI and IAT faked, but it didn't go way into the lean area.

I don't think it was real accurate because just looking at leds you have no idea what the values are, but I got it free from an automotive engineer friend, and beggars can't be choosers.

I also have an old Heathkit analog exhaust analyzer with a sniffer tube that has an A/F ratio on it, but I've never tried using it.

[cfg83]

cheapybob -

Originally Posted by cheapybob

Does your scangage report the correct amount of fuel used and correct number of miles traveled now? As long as both of those are correct, and avg mpg shown calcs to that when I fill, I see no point questioning it.

If not, why haven't you calibrated/adjusted it?

I know very little of O2 sensors other than how to replace them.

I looked at your mods, and they are similar to those on my Saturn with exception that on mine I can override the IAT value going to the computer.

Some of my mods are originally based on yours. You were the first MPG junkie on saturnfans that I avidly read, and then I found gassavers.

I don't use the scangauge to calibrate my MPG, I only use it to assess relative gains. I use a GPS to calibrate the odometer and take my MPG calcs from that. I've always been filled my tank from 1/4 to 3/4, and I just can't get myself to change that habit .

CarloSW2

[cfg83]

cheapybob -

Originally Posted by cheapybob

By "digital A/F gauge" do you mean one of the gizmos with about a dozen leds marked from Lean to Rich that you watch to see how many leds light up? If so, yes, I did have one of those connected to the O2 sensor originally until my wife disconnected it one day. It did show it running a bit leaner with the HAI and IAT faked, but it didn't go way into the lean area.

I don't think it was real accurate because just looking at leds you have no idea what the values are, but I got it free from an automotive engineer friend, and beggars can't be choosers.

I also have an old Heathkit analog exhaust analyzer with a sniffer tube that has an A/F ratio on it, but I've never tried using it.

I have (finally) built one from a kit that has the bars and the digital character readout :

http://www.jaycarelectronics.com/pro...Max=&SUBCATID=

I am crossing my fingers that the steady-state condition will give me a reliable reading or at least a trend. In the best of worlds, the A/F ratio would have a linear relationship to the MPG jumps I am seeing when I swap in resistors.

If I was a car junkie going into this, I would have splurged for one of those data logger/analyzer gizmos that gives me data curves over time, but that's water under the bridge.

The sniffer sounds cool.

CarloSW2

[cheapybob]

Mine is different. Its not programmable and just has a dozen or so LED's, and as the mixture changes more of them light up.

Does an O2 sensor just measure the temperature of the exhaust or have some fancy way to actually measure O2?

[cfg83]

cheapybob -

Originally Posted by cheapybob

Mine is different. Its not programmable and just has a dozen or so LED's, and as the mixture changes more of them light up.

Yes, I'll bet yours looks like this :

Mixture Display Kit For Fuel Injected Cars
http://www.jaycarelectronics.com/pro...Max=&SUBCATID=

Does an O2 sensor just measure the temperature of the exhaust or have some fancy way to actually measure O2?

Here's a good article on the subject :

In-Car Air-Fuel Ratio Gauge Buyers' Guide
http://autospeed.drive.com.au/cms/A_3032/article.html

Here are details on different kinds of 02 sensors :

Five Types of Oxygen Sensors
http://www.forparts.com/Bos02update2.htm
(Specifically, I think this describes our narrow-band sensor that is in the exhaust manifold)

Unheated Thimble-type O2 Sensors (LS)

Bosch introduced this design in 1976 for feedback fuel control on automotive engines. The zirconia ceramic "thimble" is encased in a protective tube which extends into the exhaust manifold. Slots in the protective tube allow hot exhaust gases to reach the thimble. Reference outside air for the interior of the thimble comes from a hole in the sensor shell, or through the wiring connector. Unheated O2 sensors rely only on the heat of the exhaust gases to reach operating temperature, therefore they might cool off while the engine is idling and revert back to a fixed air/fuel ratio setting. This type of sensor generally has a single wire connector, though some have two.

Theory : When I see "reference outside air", I am wondering if this is another mechanism to spoof the 02 sensor. That is to say, if the reference outside air is also hot and dry (aka enclosed engine compartment), then maybe that is facilitating the lean-burn condition.

Here's a good article on the wide-band sensor that we don't have in our Saturns :

Planar wide band O2 / Lambda sensor
http://www.airfuelmeter.com/english/lsu4_en_sensor.htm

CarloSW2

[cheapybob]

The device I have is an Ultraflow III Stoichiometric Air/Fuel Ratio box with 10 LED's going from Lean to Rich. When I did the HAI and IAT mods, we added an output from the O2 sensor which goes to the interior where I can clip onto it an alligator clip from the Utraflow III box.

I guess its similar to the device you pointed out as far as how it works. It did react to the resistance being changed when I'd flip the switch. I didn't try hooking it up on the T/A because the O2 sensors on it are not convenient to get to like the Saturn's.

With the Scangage, if you watch IGN, you'll see it drop and feel the engine get real sluggish if actual air or water temps get above about 210. I think that's because it was retarding spark significantly when it detected preignition. It also seems to happen if I try to run it tricked to 242 degrees before the intake air is up to 80 or so, so I usually run it straight from the sensor a few miles to warm up before switching it to 242.

Interestingly, the T/A seems to have no problem starting and running happily with it at 224 to 231, on 87 octane regular, no less, which is surprising given its higher compression ratio.

Did you try your test?

One other comment. Without filling the tank to the same level at each fill, you can't really tell if the scangage has fuel usage calibrated correctly, and therefore, nor can you tell what your actual MPG's are. I try to go to the same station, same pump, and aways fill to the 3rd time it shuts off, with the last couple times filling very slowly. It doesn't overfill, and I get consistent results that way.