Is cruising in neutral really saving me gas?

October 13, 2010 3:19 AM

I drive a 2003 Manual-Transmission Honda Element, and I've gotten into the habit of shifting into neutral at every opportunity, usually when I'm at the top of a large hill. I do this under the impression that this is saving me gas, but I'm curious about the specifics.

Yesterday I found myself at the end of my tank far away from the closest gas station, and my method for driving to the gas station was to gain speed (to about 55) and then shifting into neutral, cruising along until my speed dropped too much (usually about 40) where I would shift back into 4th gear, speed back up to 55 and shift back into neutral again. Is this worth it? Am I crazy?

I'm also curious about any tips or resources you have for driving in neutral to save gas. Thanks!

posted by ejfox to technology (24 answers total)

This page will tell you more than you need to know.

Coasting in neutral is considered to be rather dangerous, however, since (in some situations) it severaly reduces your ability to react to danger.
posted by le morte de bea arthur at 3:32 AM on October 13, 2010 [1 favorite]

What they ^ said.

Short answer: when you are in neutral, you are using gasoline to idle the engine and turn the alternator and the rest of the accessories. If you coast in gear, the car's momentum turns those things and you may be using zero fuel.

Your "pulse and glide" method is favored by the MPG people, but I think you need one of those scangauge things to get it tuned right.
posted by gjc at 3:50 AM on October 13, 2010

If you coast in gear, yes, momentum does drive the alternator etc, but your speed goes down quite a bit more ( you decelerate faster).

I do this on one uninhabited, dead-flat part of my commute. It's about a mile. I notice at the pump when I haven't been doing it. Just a little.
posted by notsnot at 4:05 AM on October 13, 2010

I doubt this is helping you much, if at all.

If you coast in gear, the car's momentum turns those things and you may be using zero fuel.

Well, sort of. You may be using zero fuel at that particular moment, but using the car's momentum to turn the engine slows the car down faster. If you need to slow down or stop, then that's great - free braking, zero fuel consumption. Not so great if you don't need to stop or slow, because it just means that you'll have to step on the accelerator again that much sooner. I don't think it saves fuel so much as change its consumption patterns.

There are a lot of variables here. The engine has an efficiency curve of some sort, and produces the most power per gallon at a particular RPM. If you're always accelerating and decelerating, the engine RPM is going to be at peak efficiency very little of the time. On the other hand, drag and friction sap less of the car's energy when it's moving more slowly. Somewhere in there's an RPM/MPG combination that's optimal. Driving slow is good. Choosing a gear that keeps the engine near peak efficiency is good. Constantly decelerating and re-accelerating -- I don't see how this helps.
posted by jon1270 at 4:12 AM on October 13, 2010

Yesterday I found myself at the end of my tank far away from the closest gas station, and my method for driving to the gas station was to gain speed (to about 55) and then shifting into neutral, cruising along until my speed dropped too much (usually about 40) where I would shift back into 4th gear, speed back up to 55 and shift back into neutral again. Is this worth it? Am I crazy?

Accelerating back to 55 probably uses more fuel than you save by decelerating to 40. In the same scenario, I would most likely try to drive steadily at the slowest rate I could to try and make it to the station.
posted by telstar at 4:25 AM on October 13, 2010

Previously.
posted by amf at 4:35 AM on October 13, 2010

jon1270: It works because engines get a higher BSFC at moderate/high loads (accelerating) than they do just cruising along with a very small load. So you get more "bang" for each "buck" of fuel, in a sense. But I'm pretty sure this only works if you turn off the engine during the coasting portion. You can't just leave it in neutral (or in gear).
posted by lohmannn at 5:22 AM on October 13, 2010

It works because engines get a higher BSFC at moderate/high loads

Yes, I can see that making sense as an isolated factor, but it seems awfully difficult for a driver to balance that against other factors and apply it usefully while driving. I guess that's why hypermiling is sort of a sport for some people.

When you're dealing with light bulbs, a lot of energy goes into heating the filament before any light is produced, so dimming a light bulb to the point that it's producing very little light means that most of the energy is being wasted as heat. Similarly, running an engine near idle means that a large fraction of the energy is being used just to turn the engine. But then, in both cases, we're talking about a larger slice of a smaller pie...

The part about having to turn off the engine to see that particular benefit sounds both correct and impractical.
posted by jon1270 at 5:46 AM on October 13, 2010

Coasting in gear turns off the fuel injectors in most (and probably all) modern cars and your top gear probably doesn't offer that much resistance so it would let you coast for quite a while. However, accelerating is the enemy of fuel efficiency so you would have been better off driving at the slowest speed possible in your highest gear.

On a side note, the link that Le Morte posted to the ecomilling sight provides some cool information. However, they are dead wrong about how to accelerate.

Accelerate from a standstill slowly and steadily – doing so will save you money.

He couldn't be more wrong.

He points out earlier in that section that even under modest acceleration, even small, fuel efficient cars get "3 to 5 times worse than it is while cruising at a constant speed" and says that he calculated a rate of 7.8mpg under modest acceleration in his Mazda 3 with a 2.3L engine. My wife owns that same car and I can tell you that it gets around 30mpg at 60mph (it actually is higher than that but I like round numbers). Lets just say, for the sake of the example, that at full throttle, you get 0mpg.

Lets look at two scenarios, in both cases you need to accelerate from a standstill to 60mph.
1. You accelerate as slowly as you can taking 30 seconds to get up to speed.
2. You accelerate as hard as you can taking only 10 seconds to get to 60mph.

In the first scenario, you've averaged 7.8mpg for the 30 seconds it takes to get you up to cruising speed. In the second, you average 0mpg for the first 10 seconds and 30 for the last 20 seconds for an average of 20mpg over the same time period.

I know it is counter intuitive, but you should actually accelerate briskly to save fuel. You actually don't want to use full throttle because the engine will use a richer air/fuel mixture (I forget the exact reason why) but something around 70-80% throttle is usually ideal.
posted by VTX at 5:52 AM on October 13, 2010

VTX, your math doesn't work.

Let's look at your two scenarios.

1. You accelerate to 60 mph in 30 seconds while burning fuel at 7.8 mpg. During that time you travel 1/4 mile (assuming straight-line acceleration) and burn 0.03 gallons of fuel. You've been at 7.8 mpg the whole time.

2. You accelerate to 60 mph in 10 seconds at zero mpg. During that time you travel 0.083 miles and burn an infinite amount of fuel. You then have 20 seconds to travel at 60mph, so you travel another 1/3 of a mile and burn an additional 0.1 gallons of fuel, for a total distance of 0.416 miles while consuming 0.1+infinity gallons of fuel

Clearly, the 0 mpg idea doesn't work. So let's sub in non-zero number for fuel economy during hard acceleration.

2b. You accelerate to 60 mph in 10 seconds at 0.1 mpg. During that time you travel 0.083 miles and burn 0.83 gallons of fuel. You then have 20 seconds to travel at 60mph at 30 mpg, so you travel another 1/3 of a mile and burn an additional 0.1 gallons of fuel, for a total distance of 0.416 miles while consuming 0.93 gallons of fuel, for a real average of 0.44 mpg.
posted by jon1270 at 6:45 AM on October 13, 2010

my method for driving to the gas station was to gain speed (to about 55) and then shifting into neutral, cruising along until my speed dropped too much (usually about 40) where I would shift back into 4th gear, speed back up to 55 and shift back into neutral again.

This is nonsensical. Keeping a constant speed is much more efficient than decelerating and speeding up again.

As mentioned, accelerating hard and gently are not good cases for the engine (and the monitoring electronics involved) in terms of efficiency, nor is changing throttle positions due to the delay in the control loop before the ECU finalises the correct fuel/air mixture for efficiency. So get up to speed and try and stay up there is your best bet. Losing anything that you have gained with a machine deliberately just to have to gain it again makes no sense.

Think of it walking around a huge cone shaped hill - you get up to a certain height and it took effort to climb up to that point. Why would it be easier on you to slide down some of the way while you rest if you have to climb back up again afterwards? It's easier to maintain your height than keep changing it.

It works because engines get a higher BSFC at moderate/high loads (accelerating) than they do just cruising along with a very small load.

Cite, please. There may be marginally SFC gains for under load than at steady state, but it is extremely unlikely (I'd go as far as impossible) that this will relate to modern cars in that a car sitting at a good speed for engine revs with a modern ECU would actually be using less fuel if it accelerated. It is possible that for a given fuel air mixture you get a higher SFC accelerating than lighter load, but that mixture is constantly varied in a road car on a feedback loop with a decent stack of variables. Either way, this heavier load would be easily replicated by changing to a higher gear or slightly touching the brakes - your logic suggests that a slight brake application (to increase the load) would produce a higher SFC and so better economy than leaving the brakes off. This is patently nonsense.
posted by Brockles at 6:59 AM on October 13, 2010

I'm also curious about any tips or resources you have for driving in neutral to save gas.

The best way to use neutral to save fuel is to not use it. As mentioned, fuel is still being used to idle your engine when in neutral, whereas fuel supply is closed off under in-gear deceleration (engine braking, of varying degrees).

Neutral is not a means to save fuel beyond the edge case of being able to coast to maintain speed down a gentle incline where being in top gear and off-throttle would have slowed you. Any incline down which you could still be in gear (but off throttle) and not lose speed would be more efficient than not being in gear. So, neutral is better purely on a really, really gentle slope that engine braking would produce a loss of speed. Anything that uses neutral or engine braking that means speeding up again beyond the unavoidable (stop lights!) is less efficient than maintaining speed.

Remove as much braking as possible from your journey - namely, whenever you want to slow down, do so through engine braking (off throttle, but stay in gear) and extending this distance by only shifting down when you have to (keeping mild engine braking and a long deceleration mode) is more efficient. It's annoyingly slow and needs a lot of looking ahead, but this is the best means to save - your aim is to reduce the amount of fuel going into your engine while also reducing the amount of load the engine needs to cope with. All acceleration is load, so anything that you can do to reduce the number of times you need to accelerate is a good thing from an economy point of view. When you do need to accelerate, do so relatively briskly as it is better for the engine's efficiency. On the same note, changing the state of the car by moving your foot around while accelerating or trying to keep a steady speed, is also bad. Engines and modern electronics find it much easier to pinpoint the fuel air mixture for max efficiency if you don't keep changing the rules - use cruise control and let it sort it's own life out rather than trying to out-clever it. Constant speed corrections will burn more fuel, even if the speed changes are tiny.

In addition, driving in neutral for any period more than fleetingly is borderline dangerous. The car works better dynamically (from a suspension point of view) when it is loaded and so will always react better to fast (ie emergency) inputs if the driveline is connected and loaded up. So coasting will make you less able to react properly in terms of controlling your car in an emergency avoidance or similar.

In short: Please do not drive in neutral. The gains you are assuming are there are not, and it's dangerous in the right circumstances.
posted by Brockles at 7:15 AM on October 13, 2010 [1 favorite]

Crap, I moved a decimal and goofed. These should've read

2. ... you travel another 1/3 of a mile and burn an additional 0.01 gallons of fuel, for a total distance of 0.416 miles while consuming 0.01+infinity gallons of fuel


2b. ...you travel 0.083 miles and burn 0.83 gallons of fuel. You then have 20 seconds to travel at 60mph at 30 mpg, so you travel another 1/3 of a mile and burn an additional 0.01 gallons of fuel, for a total distance of 0.417 miles while consuming 0.84 gallons of fuel, for a real average of 0.49 mpg.
posted by jon1270 at 8:09 AM on October 13, 2010

PopMech covered this recently.
posted by trinity8-director at 9:17 AM on October 13, 2010

Jon1270,

My math is off but my conclusions are still correct. I was trying to figure out how to calculate it using distance but those formulas are a little fuzzy for me since I was in high school and I couldn't remember how to do it. The way you're calculating it is far more accurate but still not as accurate as possible since cars don't accelerate under constant acceleration and I haven't the foggiest idea how the rate of acceleration affects the fuel mileage.

He cites 5.9mpg as the mileage under "spirited" acceleration. What his definition of "spirited" is, I have no idea. Car and Driver lists the 0-60 time as 7.8 seconds but that might be for the automatic. In any case, I think 5.9mpg is a more realistic figure to use in your example.

When I do, I get the same .0321 gals of fuel consumed of the 1/4mile of slow acceleration and .0194 gals of fuel consumed over the same 1/4mile with "spirited" acceleration and constant speed. Please check my math here, if you'd be so kind, as I'm still not sure that I'm doing this right.

Every now and again, someone writes into either Road & Track or Car and Driver magazines asking about this. They ask some automotive engineers and they say the same thing. Accelerating not at all is best for fuel mileage but when you do have to accelerate, use most but not all of the throttle. That is why I'm confident in my conclusions even know my example isn't quite right.
posted by VTX at 9:26 AM on October 13, 2010

Driving in neutral probably has some efficiency gains in an older (non-fuel-injected) car, but in a modern one with "zero throttle coast" -- where the car actually shuts off the fuel supply and lets the rotation of the drivetrain run the alternator and other critical parts of the engine -- it's probably very close to a wash, or even advantageous to stay in gear.

I have a ScanGauge on my car and have played around with shifting into neutral versus zero-throttle coast, and I do not think that it is worthwhile. Even if there is a very small efficiency gain (which I am skeptical of), you are also creating wear on your clutch and transmission that you would want to factor in. Plus it is just one more thing to think about and possibly distract you while driving.

If you are looking to save gas, the single best thing you can probably do is to look ahead while driving, plan ahead, and slow down using the throttle/engine rather than brake. Every time you touch the brake, you should in your mind think "WASTE!" Not because the brake actually consumes gas, but because it means you didn't take your foot off the throttle soon enough. (Obviously if the car in front of you slows down you have no choice; I'm not suggesting that you do anything unsafe.)

A good example scenario is exiting from a highway onto an uphill ramp that terminates in a stop sign. If you know you're going from 65MPH and will have to be at zero for the stop sign, you might as well just take your foot off the throttle once you're in the decel lane ... don't zip down the ramp at 60 and then come to a screeching halt in front of the sign.

Planning ahead, looking ahead, keeping a bigger "buffer" between your car and the car in front of you while in traffic (so you can keep a slow and steady pace rather than having to start-and-brake) will all make bigger contributions to fuel economy than shifting into neutral will, and they're safer. Go for the low-hanging fruit first.
posted by Kadin2048 at 11:00 AM on October 13, 2010

Please check my math here...

Well, lessee... (fiddles with the spreadsheet I made this morning)

If you can accelerate from 0-60 in 7.8 seconds at a fuel economy rate of 5.9mpg, then you'd have to maintain 60mph/30mpg for another 11.1 seconds to finish the quarter-mile you mentioned. For that quarter-mile you'd burn 0.017 gallons of fuel at an average fuel economy of roughly 14.5mpg. If my assumptions about what you were asking are correct, then your .0194 gals is wrong, though not in a way that undermines your case. What does undermine your case is that every one of these 0-60 time / mpg combinations is being pulled virtually out of the air. There simply isn't enough data to make meaningful calculations (though it is a fun exercise). You aren't drawing "conclusions;" you're making assertions and trying to illustrate them, but you haven't given a shred of evidence for them. Which is not to say the basic idea is wrong -- I don't really know.

Which is just a long way of saying, cite please.
posted by jon1270 at 11:13 AM on October 13, 2010

you are also creating wear on your clutch and transmission that you would want to factor in.

There is no additional wear whatsoever from running a transmission in neutral, so nothing to factor in. Neither is there any appreciable wear on your clutch from the move that would matter more than marginally/immeasurably.

However, it is an unnecessary distraction, that's for sure.
posted by Brockles at 11:38 AM on October 13, 2010

I think my math is right. I'm still using 10 seconds 0-60mph at 5.9mpg. I got smart and made myself a little spreadsheet after that post (great minds think alike!) and I get the same .017 gallons at 7.8 sec 0-60mph at 5.9mpg. If my calculations are correct, the mileage can be as low as 2.47mpg when accelerating at full speed before you break even. That, or course, assumes that 7.8mpg is an accurate figure for accelerating from 0-60mph in 30 seconds.

You're right, I'm pulling some of these figures out of the air but even then the numbers look realistic to me and I'm trying to underestimate when I can. When I adjust things so that they break even the numbers start to look a little whacky. My gut is no substitute for science and hard data but my gut thinks its right.

I'm not going to stress out over too much. I can't find anything from either of the car magazine's websites (though I'm certain it has shown up). Specifically, I saw it in an issue of Road & Track in the "Technical Correspondence" section but they don't have anything from those sections online from more than the last few months. Even then, there wasn't any data, simply the assertion by the automotive engineer via the editor. A car's cruise control does seem to back this up though. If you set the cruise at 60mph and hit cancel, slow down to about 40mph (under 30mph and the cruise shuts off), then hit resume. You car will accelerate back up to 60mph at a pretty brisk pace. My guess is that this speed is pretty close to the optimum for that car but that could just be because I want it to be true. The acceleration might be fast for some other reason.

Searching the internet yields advice both ways but with no actual data to back it up. I've wanted to test this out myself for quite a while now but I just don't have the hardware to do it. The mpg figures we're using are pretty accurate, we just don't know what the actual rates of acceleration associated with those numbers are. "Spirited" is a lot different than 0-60mph in 10 seconds.

What we really need it someone with a scangauge to run tests doing something like 1 mile or 1/2 mile runs from a stop to 60mph at full-throttle, as little throttle as possible, and multiple points in between.

This article gets as close as I've been able to find.
posted by VTX at 12:16 PM on October 13, 2010

If my calculations are correct, the mileage can be as low as 2.47mpg when accelerating at full speed before you break even.

I don't know whether I agree, because I don't know which variable you're holding constant between the two cases. Are you comparing 2 cars that travel for 30 seconds, or 2 cars that travel for 1/4 mile?

I haven't read the article just yet and don't have time to at the moment, but I think it's worth noting that BSFC is a totally different animal from MPG. It's quite possible for an engine to produce more horsepower per gallon while accelerating without achieving better mileage.
posted by jon1270 at 1:00 PM on October 13, 2010

Sorry for getting judgmental here, but I can't restrain myself anymore.

As a hypothetical question, this is a fun math puzzle. But if you (and others) actually drive like this, I'm staying off the road. This is seriously dangerous, and also illegal. Whatever tiny amount of fuel or money you might be saving is less than nothing compared to the financial, physical and psychological cost of even a minor accident. Don't think it can't or won't happen to you because you're a good driver, or a nice person. It absolutely can. Please just don't do it.
posted by Corvid at 1:08 PM on October 13, 2010

Jon1270,

I'm holding the distance traveled (1/4 mile) constant.

Corvid,

1. There is nothing illegal happening with driving like this. That is part of the reason we're talking about accelerating up 60mph and then maintaining that speed.

2. There are very few instances where you would accelerate up to 60mph from a dead stop but that isn't the point. The point is that this scenario makes the math easier. You can then apply the lessons learned to other scenarios such as accelerating down on on-ramp from 5mph to enter the freeway in which case accelerating faster is actually safer. How fast do think we're talking about here? You're supposed to be going the same speed as traffic when you're getting on the freeway. In a small car like the Mazda3 we're talking about in these examples, you have to accelerate moderately hard just to get up to speed in time.
posted by VTX at 3:08 PM on October 13, 2010

This is seriously dangerous, and also illegal. Whatever tiny amount of fuel or money you might be saving is less than nothing compared to the financial, physical and psychological cost of even a minor accident.

I initially assumed you were referring to the coasting in neutral, but if you are referring to the 'full out acceleration versus constant speed driving' discussion there is neither anything dangerous or illegal being discussed. I think you are over-reacting.
posted by Brockles at 3:56 PM on October 13, 2010

I did indeed have coasting in neutral in mind when I said "this" is dangerous and illegal. Sorry for the confusing reference, and for hyperventilating a bit. Coasting in neutral is illegal in NY, and in various other jurisdictions too I'm sure. Actually, if you can figure out how to coast uphill, NY is OK with that:

§ 1216. Coasting prohibited. The driver of any motor vehicle when
traveling upon a down grade shall not coast with the gears of such
vehicle in neutral, nor with the clutch disengaged.
posted by Corvid at 5:30 PM on October 14, 2010