I just picked up a Fit. Getting use to the stick shift after countless years of driving stick on a 97 Tercel. Also getting use to the additional feedback from the tacometer and fuel efficiency monitor. I'm using 2000rpm as the threshold for shifting down, and 3000 rpm as the threshold for shifting up. I'm a bit puzzled by the fuel efficiency monitor. Before I had one, I would simply floor the accelerator on the Tercel. Of course, it responded in the way that any subcompact would, by ignoring the flooring and speeding up in its own good time. Because of this casual response, I assumed that nothing bad was happening -- I assumed that the engine was operating in exactly the same way as it does at the point in the pressing down of the gas pedal where it stops responding anymore.

Now that I have a fuel efficiency monitor, it shows that this is far from true. It shows the efficiency dropping markedly as I continue pressing the accelerator beyond the point at which the engine stops responding anymore. How is this possible, I thought -- the sound of the engine isn't changing at all! I did some web surfing and the general explanation seems to be that the algorithms controlling the fuel injection responds to the flooring by dumping tons of fuel into the combustion chamber, even though it doesn't actually help. Can anyone confirm with any confidence if this accurately reflects the "ground truth" with the Fit? If so, it means I can't just blindly floor it and rely on the irresponsiveness of the acceleration as a natural limiting mechanism on harmfully overdoing the acceleration.

Up until the time I found that (proposed?) explanation on the web, I assumed that that dramatic drop in fuel efficiency (in the absence of any change in acceleration) was because of simplistic monitoring algorithms. That is, the efficiency is shown to drop simply because in general, it is unproductive to floor the accelerator. But now I wonder if in fact the distance per gallone/litre is actually being monitored accurately moment by moment.

 

The ECU supplies as much fuel is needed for the requested throttle input and adjusts the timing according to the knock sensor. If you go WOT the ECU goes into open loop.

Read this:
http://www.autoshop101.com/forms/h44.pdf

The MPG meter basically shows your engine load. Less load = less fuel used.

 

Hmmm. OK. Thanks. I read the paper, but I think the key part is answered in your response. If your throttle is wide open, there is no feedback. Would it be accurate to say that under those circumstances, the mixture is richer than needed? I surmise this because the engine doesn't sound like it's working harder with more pressing of the accelerator, so if fuel/air mixture increases, it certainly doesn't increase the power or rpm. If that is the case, is there a way to avoid this threshold? It doesn't feel like a sharp limit to responsiveness, so it's a bit hard to avoid the threshold.

 

Yes, the mixture is richer than needed in open loop, it provides a bit of a safety net under high load where detonation is likely to take place. They didn't program it like that for no reason, referring to this:

AFAIK the ECU refers to a preset (base) fuel map during open loop. Data posted on this site has led me to infer that the ECU is actually pretty smart, and alters this map as you drive depending on conditions (altitude, fuel quality/octane, intake air temp, etc.

Best mileage is achieved by not going WOT and avoiding open loop. You can accelerate spiritedly using the gears' torque multiplication to your advantage while accelerating, any situation, not just from a stop. Engine load, not RPM, is key to efficiency.

Lower the load, use your gears (downshift when needed).

The Tercel's motors are all very fuel efficient, even the carbed versions, you could probably go WOT everywhere and still get mid 30s The car didn't weigh anything, motor was lowish compression, didn't make much power, didn't need much fuel. I used to have one too.

 

This is pretty much on the money.

I would only really specify that, even in closed loop there are plenty of scenarios under which fuel is added under load.

It's not that it is unnecessary, it's because that more fuel both in response to airflow and as a proportion (Richer Air:Fuel Ratio) is required to slow the burn and keep it controllable.

 

I get better power, response, and efficiency when I DO NOT go full throttle. It pulls so much timing back its useless.

I actually achieve about a 0.15 or so difference at the dragstrip dpeending if I go roughly 75% throttle versus flooring it.

 

Nailed it on the head. Over 80% throttle the ECU runs off pre-set maps, the ECU ignores the sensors. This is true for any fuel injected car. Above 80% throttle the air fuel mixture should be richer for performance. The Air/Fuel sensor keeps the mixture at 14:1 Air to Fuel when below 80% throttle.

Do you have mods?

The blue line at the bottom is stock ECU maps at 11.5 A/F (too rich) losing performance. Like the guys said Honda does this to protect the engine from knocking and keeping head temps down. The red line is the adjusted A/F ratio at 13.3 for best performance. AJ did a great job keeping that line flat!

 

So, basically, you have to drive the Fit differently because of the way the computer controls the engine, right? I was thinking of my old 1986 Civic Si. It had a computer that controlled fuel flow, but it was mostly the way I drove it as to what MPG I got. The computer didn't have the "control" on the engine that it has today. My MPG was about the same as the Fit. 1600 CC, 91 HP, 5 speed stick, but the car was smaller and lighter than the Fit. It's really interesting to me how the cars have become computers with wheels. Is that good or bad? heh

 

Wanderer: Thanks for the confirmation about open loop. I assumed that WOT meant Way Over the Top, or something similar. Is this right? As for lowering gears, I'll keep that in mind (and I also like to keep my RPMs below 4000 just because of the loudness). However, I usually want the strongest acceleration when accelerating from standstill, in first.

13fit: Wondering if you could elaborate on what you meant by pulling timing back?

De36: If any fuel injection ignores sensors over 80% throttle, that must mean my 97 Tercel did this too. Which means I was just wasting fuel (and harming the environment) when flooring it. Sigh. How many years did I do that?.... I have no mods. About the graph, assuming that we're looking at the top graph, there are two reds, two blues, and two vertical axes. For the point being made, am I right in assuming that we are suppose to ignore the flatter red-blue pair and focus only on the peaky red-blue pair? I get the fact that this means overthrottling gives you less HP and expends more fuel. Funny thing is that beyond a certain point in the pressing of the accelerator pedal, I don't experience a drop in HP as suggested by the graph.

siguy: I am ambivalent about the computers on wheels approach, since it means I can't trust my own senses. However, I wouldn't be surprised if it leads to more efficiency in the end.

 

Not sure if serious, internet sarcasm meter is broken today but WOT = wide open throttle (or pedal to the floor).

Another good reason to stay at 4k or below is to stay out of VTEC, the camshaft switches to a more aggressive profile and more fuel is added to account for this. I usually shift around 3500 if I'm in self made "Eco mode" and keep throttle between 25-50% when accelerating.

 

One thing about today's cars is that they are waaaay more efficiant than even cars in the 90's. Same or better HP from smaller engines, better MPG, etc. And, the computer has a lot to do with it.

 

In 1996 it was mandated that all cars are OBD2. So the Tercel was "similar" to the cars now. The present cars Run off a CAN (very fancy electrical system) that your Tercel didn't have among other advances.

The dyno graph was to make this point; focusing on the bottom graph (Air/fuel graph) the blue line is a stock map, and it runs rich.

There isn't a HP loss on a stock Fit per-say, the OEM tune is 11:1 (rich). If it was pulled back to 12.8~13:1 you would see a HP gain, but you'd have to run higher octane fuel for protection.

Honda has it set to 11:1 so its possible to go WOT on cheap gas safely.

13Fit said he has better 1/4 runs at 75% throttle. Which is very possible and he will notice that drop. What he should do is run an A/F controller (with ignition adjustment) then he would be making more power above that 80% throttle mark and would be able to go WOT.

 

^I was researching AFCs yesterday... I'll bump an old thread instead of posting in this one but that was my thought process as well.

 

What was the name of the thread?

 

@Wanderer
---------
No, just wanted to confirm, I was not being sarcastic. But I suppose that here's not much difference between Way Over the Top versus Wide Open Throttle. Good to know that 4Krpm is a sensible limit

I said that 4Krpm was my limit, but my actual cue to preparing for shifting is 3Krpm or a bit below. Which means that actual shifting takes place at 3Krpm or above. The only time I reach 4Krpm is when I want downshift for engine braking. I do single-clutch shifting since the syncros are new, blip the accelerator to get about 3000-3500rpm to shift down, then blip again before releasing the clutch pedal. The tacometer ends up anywhere between 3Krpm-4Krpm upon clutch engagement, so I assume that I'm blipping the the right degree (though I do wonder if the double-blipping might be redundant by one blip). However, the fact that there is a more aggressive camshaft profile at 4Krpm suggests to me that engine braking is also reduced if I end up with approximately that RPM upon down shifting. So I should delay my downshifting. (Yes, I'm trying to mooch some subtle driving wisdom).

Oops, just ran into the final user comment at How to drive in a manner that increases your fuel economy - Everything2.com, which explains that engine braking may save the brakes, but consumes more fuel because the computer maintains the fuel/air ratio. This smart fuel injection is sure changing conventional wisdom.

You said you keep the throttle at 25-50%. Does this roughly translation into pressing the accelerator pedal 25-50% of its full travel? I don't suppose you rely on any other indicator?

@De36
-----
I'm a bit confused by how to extract the intended message from the graph. For background, I googled fuel systems, and found this to be the most informative for non-car-buffs: Fuel system basics. Together with Wanderer's link (http://www.autoshop101.com/forms/h44.pdf), I have the simple picture that throttling hard puts you in open loop and enrichens the air/fuel ratio (specifically, enrichens the fuel and causes the ratio to decrease), giving more engine power at the expense of fuel economy. I understand that the bottom strip of your graph shows air/fuel ratio, but it is the top that shows what HP results at any given RPM. The richer mixture (blue) shows less HP at all RPMs, in contrast with the above explanation that richer means more engine power. I'm not sure which it is that I'm experiencing because beyond a certain point of pressing on the accelerator (say, some point "x"), there is no difference in acceleration compared to if I left the pedal at point "x".

Anyways, if more theory is needed for this to make sense, I'm perfectly happy to drop it. For example, your comment about higher octane gas probably needs more knowledge about the chemistry in the combustion process and the closed loop feedback, but I'm perfectly satisfied going with the empirical rules for not stomping on the accelerator beyond a certain point -- 25-50% is what I gather from this thread thus far, presumably 25-50% of the travel of the pedal.

 

This is incorrect. When engine braking, the fuel injectors shut off, and it uses no fuel. This has been verified with Scangauge.

VTEC does not engage on decel, so you don't really need to worry about where the revs land on downshift (except don't overrev lol) if you land at 5k on the downshift it's not hurting your fuel economy except on the rev match where you have to spin the motor that high to keep it smooth. Kind of irrelevant because if you're driving for economy this should rarely happen. Just let off the throttle in gear and coast, remember you're not using any gas while doing this. If for some reason you need to accelerate again, that is when you downshift, and you should be at a lower RPM at this point anyway.

Yes when I say 25-50% this is just an estimate, i'm not looking at a TPS signal or anything. I'm not that scientific.

 

OK, thanks.

I get that you're engine breaking by letting up on the accelerator pedal without downshifting. I'm going to give it extra attention in the next little while. Apart from gas usage, however, it just seems that downshifting cause more engine braking, but that might be just a perception.

By the way, I was wondering if you could comment on the double-blipping that I described. I do it alot, so if it's not ideal, it would be good to change the habit. Basically, I step on the clutch pedal, shift into neutral without letting the clutch pedal up, rev up to (say) 3500rpm, quickly let up on the gas and shift into the next lowest gear, rev up to 3500 rpm, quickly let up on the gas and shift into the next lowest gear, and finally let the clutch pedal up. The reason why I rev up twice is because I have only a vague mental picture of how an engine works and I'm not sure which if the two revving ups are important (maybe they both are).

 

This is a synchro trans.. just clutch in, shift, rev-match, clutch out.

I have to say though, I have read your OP and this thread several times, and for the life of me can't understand how you figured more throttle input would not equal greater fuel consumption.

Let's consider what a throttle does.. it basically chokes off the airflow into the engine. All this talk of Air:Fuel ratios should tell you that even at a fixed ratio, more air needs more fuel to burn it.

What you've done when you have floored the car but found it won't accelerate faster is just increase the load on the engine. This is measured as air mass per cylinder per revolution.

As you increase load by opening the throttle and decreasing manifold vacuum (increase in incoming air density) to make more power, it's done as a result of heat and cylinder pressure increasing. Basically to make more power, you need to increase the force of the explosion in the cylinder. The expanding gasses pressurize the chamber and force the piston down on the rod, the rod transmits this to the crankshaft which causes an angular acceleration. We describe this as torque.

Torque is directly related to cylinder pressure. Cylinder pressure is a result of heat/energy conversion. So with more heat comes greater pressure, which makes for more torque. More torque at any point in the rev range will give you more power. Torque is an instantaneous measure, when we measure instantaneous torque at a given crank speed we get Horsepower.

So this additional heat can lead to spontaneous detonation/knock unless you are running a very stable fuel, so the car adds proportionally more fuel for cooling the chamber and slowing the burn.

"Octane" ratings are actually an abstract measure of how resistant to ignition a fuel is. More octane is more stable, and less prone to spontaneous combustion.

Even on more stable fuel it is possible to impart enough heat to knock under certain circumstances. So under greater load it is generally understood that you can expect to use more fuel than would be strictly necessary, especially in a powertrain designed to run for over 100,000 miles. While that may seem "inefficient" on the face of it.. it's done to prolong the life of your engine, a scheme employed by pretty much every non-carbureted engine you will come across.

So instead of ratio of 1lb of fuel per 14.7lbs air you are now operating at say 1lb fuel per 11lbs of air. Or 25% more fuel to burn the same amount of air.

An easy way to learn how all this works and improve your gas mileage would be to invest in a Vacuum gauge and a Wideband Oxygen Sensor kit.

Maintaining Lambda 1.0 or an indicated ~14.7:1 Air/Fuel Ratio (on pure Gasoline, ~14.1 for E10 Gasoline) on the WB Gauge with the most Vacuum being pulled for a given driving situation will yield the best fuel economy.

 

In order to understand your single rev-matching, I found the following videos on synchros very helpful:


Now I know why, with the clutch pedal down, it doesn't help to blip the gas while in neutral, before shifting into the lower gear. The clutch is not engaged with the flywheel, so you're not affecting anything in the gear box. No revving you do will affect the moving parts in the gear box, and there is no "magical" way that decreased wear on the synchros can be achieved (I'm talking in the context of single-clutching).

You said that I figured out that more throttle would not equal greater fuel consumption. If I said that, it would likely have been in my muddled state (which might be my normal state). What I meant in my OP was that, beyond a certain point (say, some point x), pressing further on the accelerator did not result in any noticable change in the acceleration. I might as well have kept the accelerator pedal at the point x beyond which there did not seem to be any effect.

From this thread, I get that, even though I don't notice any difference, I'm driving the system into open loop. So there must be some change, just not noticable from the driver's seat.

About richer fuel to cool the cylinder under load, I found Fuel Ratio and Thermodynamics to be very helpful. I also found it helpful to read about the difference between controlled combustion and detonation. However, it isn't clear to me why greater cooling is needed. From various other readings online, I got the impression that, under proper comubustion, compbustion and removal of the resulting gases is so quick that hardly any of the heat transfers to the cylinder walls (relatively speaking). I didn't find anything online that explained why combustion under load would lead to hotter cylinders. Not that I necessarily need to know, just summarizing my knowledge trip today.

Thanks!