MAP direct correlation to FE?

[sword_guy8]

I just realized, that's all well and good to get a more accurate reading, but I'm really talking about relative readings anyway. They will all be relatively accurate to each other, so when I read more MPG, I'm getting more than I was, regardless of the actual numbers. That's all I was getting at. And I'm getting the highest MPG for any given speed at a MAP reading of 20 inHg.

Furthermore, if the readings are off, I would be willing to bet it's by a negligible amount.

Also, yes, it is a speed-density system. But that system does take into account the MAP.

[P-hack]

Originally Posted by sword_guy8

Also, yes, it is a speed-density system. But that system does take into account the MAP.

MAP != MAF, and I'm talking relative too, everything is relative, if even to the gas pump.

[P-hack]

bpw/duty cycle is only part of the story. The pressure drop across the injector (hence my question about fuel pressure regulator reference) is a major player. If your regulator is referenced to atmosphere, then it will deliver more fuel/time as map decreases. If it is referenced relative to manifold, then it fuel delivered is basically time dependant.

The other effect is pulse width distortion which has larger effect at lower duty cycles.

Either of these, but especially the first should give you pause as to your conclusions about MAP using duty cycle.

Plus the limitations of polling computed values vs injector/vss events(interrupts) can throw it off a few percent FYI.

[sword_guy8]

The regulator is referenced to manifold vacuum.

Pulse width distortion, being the given pulse vs the actual pulse, is usually measured in nanoseconds. .01 milliseconds is equal to 10,000 nanoseconds. The highest value I could find regarding PWD for a fuel injector is 200ns. That's .0002 milliseconds. It would have a greater effect at shorter pulse widths because it would be a greater percentage of the pulse width. But let's say a pulse width is 3 milliseconds. The width, taking into account a 200ns distortion, would be 2.9998ms, which is 99.993%. Now let's say the width is 9ms. With distortion, it would be 8.9998ms, which is 99.997%. This is a completely negligible difference. Even if the distortion was 500ns, or even 1000ns, the difference would be negligible.

In this post: http://ecomodder.com/forum/showthrea...ate-21342.html The poster, toc, posits that pulse width is a very accurate means of measuring fuel usage, and therefor fuel economy. His comments make sense.

I'm not sure exactly what the problem is with my equation for MPG. It takes into account only vehicle speed, pulse width, and engine speed (and a few constants such as pulses per revolution, etc.). The equation does not use MAP to figure out fuel usage. It also doesn't use an assumed stoic mixture. It doesn't care about the mixture. It cares about how much fuel is actually being injected into the engine.

Where I'm coming up with my MAP theory, is that I'm reading the MAP while I drive, and cross referencing the MPG under different conditions.

Real world numbers (rounded): Drive up a hill on the highway at 70MPH with cruise on in 4th gear. As the car starts to climb the hill, the MAP increases with TPS (cruise trying to maintain speed). When everything stabilizes on the same hill with the same grade, the MAP will be 24, and the MPG will be 12. Now, downshift from 4th to 3rd, and after stabilization, MAP will be 20, MPG will be 14. Engine RPM is, obviously, higher. But MPG is also higher.
Now, same experiment with a less steep hill. Climb the hill in 4th and the MAP goes to 20, MPG is 18. Engine RPM is the same as the other hill (still going 70MPH). Downshift to 3rd and now the MAP is 16. The MPG is also 16. 20inHg always yields the highest MPG. Taking into account your other statement of limitations of polling throwing the number off by a few percent, the numbers would only be changed by 0.5. Because a few (meaning 3, just as a guess) percent of 14 is 0.5 (actually 0.42 which works in the favor of this being accurate). The fact remains that at 20inHg the MPG is higher than at 24 or 16. I've found the range of highest to be between 18 and 22 (still highest at 20, but by a negligible amount in that range).

To put it another way, if you load the engine too much, you're still pumping a lot of air in and not getting the power out, hence inefficiently using the fuel. Conversely, if you unload the engine too much, you're wasting energy just to keep the engine running. This is where the "lower RPM equals better fuel economy" thing comes in. Lower RPM raises MAP (all other conditions being equal, such as MPH, grade of road, etc.). However, if you were to lower your RPM to the point that MAP was very high in order to maintain vehicle speed, the MPG would start to drop again. In my engine, that magic MAP number is 20. It may be different in a different engine.

These are real world numbers that I have monitored over and over again just to make sure there wasn't a fluke. Same hills, same numbers. No matter how I approach it, the highest MPG will be reached with a MAP of 20. And once again, the MAP is not in the equation which gives MPG.

MAF, MAP, and even O2 sensor readings would not yield an accurate MPG calculation. MAF assumes stoic all the time which simply doesn't happen unless you have a wide band O2 (and even then, mixtures are changed to keep up with engine loads). MAP really needs a whole lot of other inputs.

So basically, I'm saying that there is an engine loading (MAP) in which the engine does the most efficient job of creating power.

[P-hack]

Originally Posted by sword_guy8

So basically, I'm saying that there is an engine loading (MAP) in which the engine does the most efficient job of creating power.

More specifically, there is a specific rpm and load (map) where an engine is most efficient, have a peek at bsfc charts (and of course the ecu and a million other variables will have their effect), I tend to think it is more sensitive to load than rpm, generally speaking, but some cars like a tdi are most efficient at full throttle at a certain rpm:
http://ecomodder.com/forum/showthrea...-got-1466.html

Edit, and one reason many cars do worse at full throttle is because of enrichment, otherwise lower pumping losses is good.

[sword_guy8]

I've looked high and low for a BSFC chart for my engine, and I can't find one. You'd think, given that the 3.1 was a very common engine, that there would be a chart *somewhere*.

I just posted in that other forum. I'm at a loss for data on this one... :/

[sword_guy8]

Strange, I got an e-mail saying there was a reply, but it seems to be missing. Anyway, it's a GM 3.1.

[P-hack]

Originally Posted by sword_guy8

I'm at a loss for data on this one... :/

I don't think so, map is a good indicator of load.

FYI, the general rule of thumb I use in an unknown car is 2/3 throttle and center the shifts at about 2200 rpm. You can probably narrow it down with more observation.

Also wanted to make sure you were aware of "advanced" driving techniques from there i.e. pulse and glide, engine kill switch for glide, basic hypermiling techniques like avoiding the brakes, paying a lot more attention to potential obstructions so you can time them, etc. etc. I was getting about 55mpg around town with my saturn, and accelerating at 2/3 throttle and gliding to stops was virtually undetectable as far as traffic flow (with a 4mph bump start if it is just a stopsign or leave it off for longer stops).

P&G is a pita at higher speeds though, need aerodynamics (which would compliment your gearing change). But around town it is quite natural and puts the hybrids to shame.

BTW, never heard of an 89 corsica with a 3.1. 3.1 is going to be a little big for best economy, especially on the hiway with the throttle closed (manifold vacuum=pumping losses), but it's what you got. P&G works because you only operate the engine with the throttle well open, and off the rest of the time (idling is %100 loss). A smaller engine will have a more open throttle at cruise.

[sword_guy8]

I modded it when my 2.8 crank broke in half. That's when I installed the 4 speed transmission that's in it now. Like I said, the 4 speed knocked my 70MPH cruise RPM from 3800 to 2100. HUGE difference. I wanted an even lower geared version, but they're hard to come by. I wanted the lower gearing in order to keep the RPM down and loading a little higher.

Using the app, I've determined the fuel cutoff parameters for my engine. If it's above 1100 RPM with the throttle closed, it will cut off the fuel after about 5 seconds. I've downshifted while going down steep hills and been able to go about half a mile without using any fuel. I use it to slow down when coming off the highway too. Works out pretty well. I'll take the 66% TPS into consideration when accelerating and try to see if my average improves. Thanks!

[betasniper]

In my opinion, 2/3 throttle is far to much throttle because that would almost always put you at atmospheric pressure. I have a vacuum gauge in the wagon and barely 1/4 throttle (estimated) would cause the vacuum reading to be 5"vac at low rpms and anything past 1/3 (estimated) results in 0"vac. The MAP is not linear with throttle position.

Edit: It would be interesting if you have a SG or UG to see both TPS and MAP on the same screen to see what the correlation is.