http://www.omninerd.com/articles/Imp...uel_Efficiency

ran across this earlier today...

 

I think that guy is off to a good start on studying fuel economy.

He has made a few mistakes, but none are really serious:

His equation for mass air flow is wrong, but the equation he gives is proportional to mass airflow...he just has not been shown how to match units to make sure an equation is not giving nonsense.

When he was trying to measure the Ram Air effect, he does not seem to realize that his throttle blades were between the air being 'rammed' and the MAP sensor he was monitoring and snuffing out any effect. It is interesting to note his Ram Air graph though, because it shows that if you put a less restrictive air intake on an engine and drive that engine back up to the same mph speed....the throttle blades just close a little more and restrict away any improvement from the higher flowing air intake. This is because it is always MAP that determines horsepower output at any set rpm, and it is horsepower that determines what speed you go. He writes this in his article earlier when he talks about MAP and its effect on engine load, but he does not seem to have come yet to realize it applies to his Ram Air experiment.

His statement you can connect an electrical potentiometer (aka volume control rheostat) to an O2 sensor and 'fool' its output is wrong. Added resistance to the tiny current from the O2 sensor can only lower the voltage...not raise it. You can falsely fool a PCM computer into thinking the air to fuel ratio is too lean by making the O2 sensor voltage lower, and the PCM will then over-richen....but you can't fool the PCM into thinking the afr is falsely rich, and then make it lean the mixture out because it would take 'negative resistance' on the potentiometer to do this, and that is impossible. Resistance is only 0 or greater on a potentiometer.

His statement about disconnecting the O2 sensor is wrong. When the O2 sensor is operating it 'toggles' the air to fuel ratio back and forth around 14.7 so that it varies between about 13.5 and 15.5
If you disconnect the O2 sensor, the engine will default to a very rich air to fuel ratio of about 12 (safe for the engine because no danger of detonation) and the fuel economy will get very bad compared to the 14.7 ratio. If the ratio were raised to the 16 to 19 afr area there would be a 5-10% improvement in fuel economy if the sparkplugs did not start misfiring...but you can't get there by disconnecting the oxygen sensor.

People here on Dodge Forum would do well to look at his
fuel consumption versus gears 6-5-4-3 graph
...that kills off all the nonsense about 3.92/4.10/4.56/4.88 gears improving highway cruise MPG.

The key good thing the guy is doing is MEASURING.

He is not a WAG'er (wild *** guesser)

By measuring and doing experiments himself he will eventually come around to improving his MPG...and eventually agreeing with all the other people before him who did careful measurements of how engine works.

99% of the time
American engineers, Japanese engineers, German engineers, and Russian engineers have all done similar experiments on engines and found similar results.

But that 1% disagreement is important...that is sometimes where progress will come from.....

 

I pretty much agree with your assesment Hank, except for the above statement. You can decrease the resistance seen by the ECU by placing the potentiometer in parallel. True by placing it in series would increase resistance, however, placin the "pot" across the O2 sensor would create a lower resistance, because the sum of parallel resistance is less than the amount of the smallest resistance in the circuit.

The question I have about this though would be if you fix this resistance, at some point to make the ECU think you are in a rich running condition and it trys to lean out the engine, the fact that the now fixed recsistance of the O2 circuit will not change, so wouldnt the ECU try to continually lean the engine to meat the parameters it is looking for until either you hurt something OR you get some kind of a CEL?

It would seem you would have to have something that looks at what the ECU is doing and at some point make the ECU believe it is doing what it is "supposed" to be doing in order to maintain the "performance" you are looking for. HMMM, sounds alot like these little "magic" boxes many people buy to make their trucks go fast.

 

OOPS< I think I just answered my own question by re-reading my own question.
The resistance of the O2 sensor would be the smallest resistance in the circuit and would change based on operating conditions, so as this increased in value the total resistance of the circuit could increase to a point where the ECU actually seen what it wanted to see.
So the resistance added in parallel would be a fairly large value (in relation to the O2 sensor value) to make a fairly minimal resistance change, so that you would still be within the range of the O2 sensors "normal output" to the ECU.

I love when I answer my own questions...now if those voices in my head would just quit trying to help.

 

You can make a relatively simple electronic circuit
to fool the O2 sensor to read falsely rich so the PCM will lean out afr
by placing in series with the O2 sensor voltage output leads
about a dozen thermocouples in series (aka 'therm-ionic generator')
and then placing those series coupled thermocouples
in an area where you can vary the heat.

The thermocouples are like little batteries that can add the amount of milli-volts you wish to the 0 to 1000 millivolt output of the oxygen sensor.

The oxygen sensor is a little
"Fluorite crystal structure cerium oxide concentration cell"
called a Nearnst Cell after the German who invented it
that outputs a small voltage (and even smaller current)
depending on the oxygen concentration difference
on either side of it.

I have never worked as a professional on Dodge pickups
but I have worked as a professional inventing
unusual Fluorite Crystal Structure electrical devices like fuel cells and concentration cells, and have a couple of US Patents awarded to me on that subject.