Noticed the gear doesn't up shift immediately when I take my foot of the gas pedal...

[costermonger]

What's your mpg?

26, but I only have ~2000km on it (1200 miles). One fill up in particular brought my average way down, if I exclude it, I'm getting ~30.

I have a FWD though. If I take my time on the highway, it shows me about 43 mpg.

 

[TreyP]

But that's not consistent. One can satisfy the first part, and not the second, vice versa. First part first:



According to this, a (modern) car coasting in gear is not stalled.

What happens when you step on the pedal is entirely up to the ECU. As far as it's concerned, when you're not stepping on the gas pedal something has to turn the engine at an RPM equal to or greater than X (I haven't played around with it enough to figure out what the number really is, but I'd bet it's around 1000). Whether the engine is spun by the power it's creating or by backload transferred through the drivetrain from the momentum of the vehicle, the ECU doesn't really care. If it has to provide fuel to the engine to maintain the RPM at the threshold, it will, but if the RPM is 2000 because the engine is coupled to the wheels and the car is coasting, there's no reason to provide fuel at all, and so it does not. The engine isn't creating any power, but it is rotating and it is getting air and a spark*, just no fuel.

Now, if you step on the pedal, the ECU decides you want more power, so it alters the throttle position if necessary and commands the injectors to open as required. Instant power. Because the only thing missing was fuel, the 'non-running' (freewheeling, or windmilling, depending on terminology) engine is immediately developing power again because it was only missing 1 component of the internal combustion cycle.

So coasting uses no fuel, but the engine is maintained in a state where stepping on the gas will immediately yield power, since only fuel was removed in the first place.



By this definition, a car coasting in gear is stalled, because there's no combustion happening. There's no need for it. The engine will happily spin due to the rotational energy provided by the transmission, until the car slows too much, and then it's not coasting anymore and the ECU steps in with fuel to keep things running.

** I am editing my response because in a way I think we are saying the same thing in a different way. What I am saying is that a running engine is always using fuel, air, compression and ignition. If not, it is stalled. A coasting car in not a stalled car by definition, again, I don't know how you are getting these ideas from what I wrote. A coasting car could be stalled, but could also still have an idling engine, which is usually the case.

 

[TreyP]

Exactly! Its nice when people state verifiable facts, supported by logical arguments and using exact terms correctly, and not just stating you are wrong just because they "know".

For the record, I was an ASE certified auto technician for 10 years, so it is not simply a case of "I just know" inherently.

As I alluded to in the above post, I think we are getting caught up in the semantics of when an engine is considered to be 'running' and 'stalled'.

 

[costermonger]

Dude, I'm just trying to break it down into small enough bites that you can understand. I was a mechanic for 10 years, so I think I know how an engine works.

Yeah, I've been working with mechanics long enough to know that understanding every single nuanced function of an engine isn't exactly required to fix it.

Unless it is a hybrid, a running engine is always using fuel, air, compression and ignition. If not, it is stalled.

When you are in gear, at say 50 mph, and you let go of the gas pedal, the car detects that it's in an overrun condition (power requested is 0) and the ECU commands the injectors to zero.. The injector solenoid is closed. Fuel that is pumped to the injectors is returned, and no fuel is injected into the cylinder. No combustion occurs. One of two things happen next: you step on the gas pedal and the ECU commands the injectors to do their thing again, and combustion is resumed and you get immediate power (since the engine has happily been spinning away, pumping air this whole time), or you slow down enough that the engine RPM reaches it's idle RPM + a little bit and the ECU commands the injectors to open so it can smoothly establish an idle condition.

Now at various times in this thread, you've called the above condition "stalled" and claimed it never exists. It's impossible for a carbureted engine to function this way, but it's the norm for fuel injected engines of any degree of sophistication, gas or diesel.

A coasting car in not a stalled car by definition, again, I don't know how you are getting these ideas from what I wrote. A coasting car could be stalled, but could also still have an idling engine, which is usually the case.

Post #47, you're arguing that modern fuel injected, computer controlled engines continually use fuel. This is wrong.

And if your car idles while you're coasting down a hill, you (or it, I suppose) have decoupled the wheels from the transmission.

As I alluded to in the above post, I think we are getting caught up in the semantics of when an engine is considered to be 'running' and 'stalled'.

May be true; it's easier to discuss what the engine is doing in terms of power development and rotation.

 

[TreyP]

I posted this question on another forum which consists of members with a high knowledge of all things automotive. This is the best answer so far, and it is by an ASE master technician. He states that the injectors will cut off, just as has been stated in this thread, but also states that the cut off is for a very short amount of time to prevent the engine from stalling. In other words, he agrees with us all on some points and brings the whole idea into one package.

I remember something about this being brought up in another thread. Toyota, as well as other manufacturers have what is called DFCO or Deceleration Fuel Cut-Off. What this means is that when you take your foot off the gas, the fuel injectors will actually shut off totally.

Let's stick strictly with the general thread of an automatic transmission, for these operate slightly different from a manual.

For starters, unless you are in any other selected gear except D4-5, there will be no engine braking, thus you will not be able to "drive" the engine from coasting. Also, if the engine is not running at all, then you would not be able to have any engine braking either, as there would be no fluid coupling in the torque converter, nor would there be any fluid pressure from the pump, since the pump is driven by the engine.

Here is how the DFCO operates, and it is a VERY short lived operation: When you remove your foot from the accelerator pedal, the ECU will cut off fuel to all of the injectors. when engine RPM has reduced down to an idle speed (i.e. 800 RMP or so) then the ECU will begin firing the injectors to prevent the engine from stalling. While this does not seem like a very long time (Which it isn't) doing this over a long drive does improve fuel economy by a small amount over NOT using DFCO. The DFCO has been around for YEARS. Toyota, as well as GM and others have been using this program since before 2000. My 1997 Tacoma has it, but mine is a manual transmission, and does things slightly different.

There are certain requirements for the DFCO to operate. For one, your engine has to be at normal operating temperature, so it will not engage first thing in the morning or evening after work. On a manual transmission, SOME vehicles will not engage the DFCO unless your engine was above 1800 RPM when you removed your foot from the pedal. This prevents the engine from stalling out, as the engine drops rapidly in RPM in a manual trans.

On all vehicles if you return your foot to the gas pedal at any point, the ECU turns the injectors back on, and in an automatic transmission, if you shift the shifter into neutral, it will not engage.

Here is the thread, if anyone cares to follow along.
http://www.tacomaworld.com/forum/general-automotive/274728-rpms-help-settle-debate.html

 

[vroom]

For the record, I was an ASE certified auto technician for 10 years, so it is not simply a case of "I just know" inherently.

I'm not going down the road of spitting out my credentials on a forum. It's pointless as people can write whatever they want. And if so, good for you, I respect that. That means you know how an engine is built and how to fix it, but not necessarily how it actually functions and how the laws of physics come into play. The engeneers at Mazda actually design engins. They've studied for years on end to make these things their expertese. These guys I truly trust to know when fuel should be injected and not so I save the maximum of $$. From my experience with this car so far, they have done this very well.

As I alluded to in the above post, I think we are getting caught up in the semantics of when an engine is considered to be 'running' and 'stalled'.

I think you are partly correct, but as explained quite clearly by costermonger, your semantics just come out so full of contradictions its hard for us to get what you are trying to say. You can try going on debating if there is a difference between an engine "running" and one that is "turning" or on the definition of a "stall", but it's pointless. If you want to put your car in neutral when you coast, then be my guest, it's your money. I know I'll be trusting the Mazda engeneers, the laws of physics and comon sense, and keeping it in gear (well, the auto tran does that for me anyway) so it does not cost me a dime. The added benefit of engine braking might even save me a brake job in the long run!

 

[TreyP]

I'm actually not interested in discussing this any further because everything I say has been distorted to fit someone else's idea of what they want to debate. For example, I never once said anything about taking the car out of gear when going downhill. I'm not sure where the confusion is coming from, but it is like I am speaking a different language, I post one thing, and then it gets turned into something that I never wrote nor intended, so I'll just leave it at that.

 

[costermonger]

I posted this question on another forum which consists of members with a high knowledge of all things automotive. This is the best answer so far, and it is by an ASE master technician. He states that the injectors will cut off, just as has been stated in this thread, but also states that the cut off is for a very short amount of time to prevent the engine from stalling. In other words, he agrees with us all on some points and brings the whole idea into one package.

Here is the thread, if anyone cares to follow along.
http://www.tacomaworld.com/forum/general-automotive/274728-rpms-help-settle-debate.html

That's pretty much exactly what I'm saying. The only CX-5 specific I'd add is that since the skyactiv-drive transmission refuses to unlock the torque converter above 5mph, it behaves more like a manual and keeps the fuel cutoff for however long it takes the car to slow to the point that the engine needs fuel again to maintain at least an idle RPM. Could be 2 seconds, could be 2 minutes (if you find a big enough hill).

In Canada the car has a real time L/100km display rather than MPG and you can see (with slight lag) exactly when the car is or isn't behaving this way.

 

[vroom]

^Correct. The Toyota thread TreyP posted is correct for older (or common) auto transmissions. These would decouple the transmission when coasting (basically putting it in neutral), and gas would need to be injected at an idle regime to prevent stalling when idle RPM where reached. As the tread clearly states, engine braking is not possible with these transmissions, and the engine falls to an idle (around 800 rpm). What started this whole thread is the fact that this is not what happens with the CX-5. The first thread states that engine RPM is maintained (does not go to idle) while the car slows down. You can actually feel the car downshift as the car slows. With the CX-5, engine braking does happen, which means the car's momentum is what keeps the engine spinning, not gas. All of this meaning that the auto tranmission is designed to shift more like a human would shift a manual, great isen't it?!

So returning to the Toyota thread posted by TreyP, because the RPM is maintained above 1800 when you release the gas pedal on the CX-5, DFCO (Deceleration Fuel Cut-Off) is always engaged until the car is almost completely stopped. Conclusion: no matter how high the RPM when you lift your foot off the gas, it will not cost more gas.

 

[costermonger]

^Correct. The Toyota thread TreyP posted is correct for older (or common) auto transmissions. These would decouple the transmission when coasting (basically putting it in neutral), and gas would need to be injected at an idle regime to prevent stalling when idle RPM where reached. As the tread clearly states, engine braking is not possible with these transmissions, and the engine falls to an idle (around 800 rpm). What started this whole thread is the fact that this is not what happens with the CX-5. The first thread states that engine RPM is maintained (does not go to idle) while the car slows down. You can actually feel the car downshift as the car slows. With the CX-5, engine braking does happen, which means the car's momentum is what keeps the engine spinning, not gas. All of this meaning that the auto tranmission is designed to shift more like a human would shift a manual, great isen't it?!

My wife's Kia Soul was the first auto I'd ever driven that attempts to behave this way. It's amusingly low-tech (4 speed!) by comparison to the skyactiv but the computer running it wants to keep the torque converter locked up as much as possible. This has a huge impact on fuel economy (usually estimated as a little over 10%) and improves the driving feel.

I think almost everybody is using this technique now, Mazda just seems to have done a particularly good job with it.

 

[Buger]

Here is how the DFCO operates, and it is a VERY short lived operation: When you remove your foot from the accelerator pedal, the ECU will cut off fuel to all of the injectors. when engine RPM has reduced down to an idle speed (i.e. 800 RMP or so) then the ECU will begin firing the injectors to prevent the engine from stalling.

http://www.tacomaworld.com/forum/general-automotive/274728-rpms-help-settle-debate.html

Yes, what the ASE master tech guy on the url mentioned said is correct in that the ECU will cut off fuel when the foot is off the accelerator. You can see from what he wrote that the engine is not "stalled" while no fuel is being supplied to the engine. It is only when the RPMs go down to idle speed of around 800rpm that the ECU will begin firing the injectors again to prevent stalling.

This can easily be seen with diagnostic software connected to the OBD2 port. In fact, my old 2003 protege5 does the same thing so it isn't really anything new or specific to high tech hybrids or anything.

In any case, I don't think there is any disagreement over no fuel being supplied to the engine even though it is spinning. There definitely was a lot of confusion over people using the words "stalled" in a different context than most of us (including the ASE master tech guy quoted).

Being in a high gear can mean you can cruise for a very long time without using any gas. There are some minor pumping losses but even this is reduced by the late closing intake valves to reduce the compression ratio.

 

[CX-SV]

So 21 MPG average in first 1.5 months is what I read in first post.

Are we talking New York City traffic?