City ecodriving comparison: 2014 Mirage CVT vs. Mirage 5-speed vs. 1998 Metro 5-speed

[MetroMPG]

(Top row: the actual two Mirages I tried out. Yes, I drove around Ottawa with that lettering plastered all over the car.)

The new Mirage is yet another in the crop of small cars that gets a better EPA rating for the automatic than the manual transmission.

The rating for the CVT is actually higher for both city and highway:

CVT automatic, city: 37 mpg US (6.4 L/100 km = 44.4 mpg Imp.)
CVT automatic, highway: 44 mpg US (5.4 L/100 km = 52.8 mpg (Imp)

5-speed manual, city: 34 mpg US (6.9 L/100 km = 40.8 mpg Imp.)
5-speed manual, highway: 42 mpg US (5.6 L/100 km = 50.4 mpg Imp)

Despite the manual's apparent handicap, it's been my experience that a motivated, efficiency-minded cog swapper can beat the city rating by a bigger percentage than in the automatic.

I suspected the same would be true of the wee Mirage, and this weekend I finally got a chance to compare them head to head, in Ottawa (thanks to the friendly staff at Donnelly Mitsubishi -- plug! plug!):

Route:

- 10.6 km (6.x mi.)
- residential and arterial Ottawa roads
- moderate traffic (but no stop & crawl)
- cool, windy conditions (5C / 41F), damp roads
- speed zones from 40 through 70 km/h (~25 - 44 mph)
- 7 stop signs & 18 traffic lights
- Google map of the route here

Driving techniques:

I intentionally kept it very simple: plain Jane vanilla eco-driving here, with anticipation and minimization of braking being the main tactics.

No pulse & glide, and no engine-off coasting. I even left the engine running at all stops, regardless of length. I wanted non-hypermilers to look at this and think... "hey, even I could do that."

With the manual: moderate acceleration; relatively early upshifts (no more than ~2500-2700 RPM); going right to top gear at low engine loads & low speeds where possible; a little bit of downshifting for fuel cut-off engine braking mode when stronger sustained slowing was needed.

With the CVT: acceleration at the lowest practical RPM, plus I made sure to lift slightly once at cruise to ensure the Mirage's fancy 2-stage sub-gearbox (this is in addition to the variable pulley/belt system) shifted into high gear for lowest engine RPM whenever possible. A couple of downshifts into "B" mode for sustained deceleration fuel cut-off. (PS: downshifting a CVT is kind of fun -- completely unlike other automatics or manuals!)

I kept up with traffic (though of course I let them pull away when I saw a light change to red ahead, for example, and I got off the gas well before they did). No rolling roadblocks here. (Despite that, the Mitsubishi sales guy who rode with me said I drove like a grandma! )

Results:

OK, let's be clear this isn't really a "test". There are piles and piles of uncontrolled variables -- way too loosey-goosey. So a "comparison" it is...

CVT automatic, city: 42 mpg US (5.6 L/100 km = 50 mpg Imp)
= 13.5% over EPA 37 mpg city rating

5MT city: 48 mpg US (4.9 L/100 km = 58 mpg Imp)
= 41.2% over EPA 34 mpg city rating


Bring on the Flea!

Then for fun, after I finished trying out the Mirages, I took the Firefly for several laps over the exact same route to compare it as well.



Three flavours of driving were employed in the Flea:

1) Plain Jane ecodriving (exactly the same style as in the Mirages)
2) Plain Jane ecodriving, but with engine off approaching stops longer than 10 sec.
3) Using advanced P&G / EOC (plus foundation elements of ecodriving, as above)

Results...

1) 55 mpg (US) = 4.3 L/100 km = 66 mpg (Imp)
2) 59.4 mpg (US) = 4.0 L/100 km = 71 mpg (Imp)
3) 73 mpg (US) = 3.2 L/100 km = 88 mpg (Imp)

Considering the Flea is quite modified, and had a ~350 lb weight advantage (lighter car to begin with, with one less person in it), I was suitably impressed with the little Mitsu. Also, it had just ~33 PSI in its tires. Plus, it was pretty cold (5C / 41F) and windy (24 kmh / 14 mph). Considering all that, the Mirage is definitely in "stock" Metro mileage territory right out of the box.

It's just too bad about the manual's gear ratios that limit its highway fuel economy. Though there will be final drive swap options in the future (from the UK/Europe Mirage parts bin)...


Links for more info:

- Full, gory details of the CVT vs. manual comparison:
Gas mileage/MPG test: 2014 Mirage CVT vs. 5-speed (sub/urban Ottawa route)

- General impressions after my first drive of the Mirage:
Brief test drive: Mirage CVT and 5-speed (Donnelly M itsubishi, Kanata/Ottawa, Canada)

- Why the heck does that white Mirage say "64 MPG" on its flank??
Mirage rated 64 MPG highway in Canada, 44 MPG in the US? WTF?!

.

[sheepdog 44]

The Ford Fiesta 1 litre ecoboost "supposedly" gets 45mpg highway but 32mpg city with the manual. Which is a 1 mpg highway improvement (who knows with ford these days). But it is a 1 litre and that Mirage is 1.2 litre. You should give it a test drive and see how it compares.

[MetroMPG]

Neat! 45 EPA hwy is plausible in the Fiesta 1.0. But top gear will have to be very tall to deliver that.

I've been noodling over the Fiesta vs. Mirage showdown already...
2014 Mirag e 1.2 vs. Ford Fiesta 1.0 ecoboost - which has better mileage/fuel economy?

And, agreed on the question of Ford's credibility (though you'd think they would be very cautious about the Fiesta's rating given what happened with their other models recently). Countdown to the first customer lawsuit over misleading ratings starts now?

It'll be fun to see what happens when someone who cares about fuel economy gets their hands on one.

Most Mirage drivers (reporting results on Fuelly or MirageForum) are beating the EPA.

Competition's grand, ain't it?

[MetroMPG]

Sheepdog: I started a new thread for the news about the Fiesta 1.0 turbo. Thanks for spotting that!

Reports: 2014 Fiesta 1.0T EPA rated 45 mpg hwy, 32 mpg city... starts @ $17,240

.

[cbaber]

It would be interesting to see the EPA tests and how they are done. Obviously the M/T is not being driven to it's full potential. I've always wondered why CVT transmissions never get better MPG than manuals. They were suppose to be the answer to auto vs manual problem for MPG, for Honda at least. Infinite gear ratios, always accelerating at the best RPM, low cruising speed, etc. I guess it all comes down to how the power is transferred to the tires, and the M/T just makes better use of it (and with much more control).

With such an obvious error in the EPA tests it makes you wonder if they do it on purpose to sell more CVT's? Honda has never had a good reputation with them, maybe it has turned people off to the name "CVT" for all brands.

*Posted on EM Facebook page*

[MetroMPG]

I've been wondering the same thing -- how the EPA tests handle a manual.

I suspect the "drivers" have to follow guidelines about minimum engine RPM (meaning having to hold a lower gear than we might), and possibly also minimum RPM when upshifting.

Meanwhile, I short-shifted while accelerating, and then went for top gear at every reasonable opportunity, once at "cruising" speed.

As for the CVT, I've driven a few, but none as "aggressive" as the Mirage's in terms of low RPM operation. I was quite surprised, to be honest: it's the first automatic I've driven (outside of hybrid CVTs) that felt designed to work with me, not against me when trying for good economy. I was prepared to dislike it, but it earned my respect.

--

Thanks for the Facebook post - good idea.

[wickydude]

I think the difference is in the transmission control. With cvt, manufacturer's decide what the transmission does. With manual, it's up to the driver.
Cvt is good, but people did not like the way they revved on aceleration. That's why they changed the control to simulate fixed gear ratios.

As to what the epa tests involve, that is well documented. Look here for example:
Detailed Test Information

The thing is that nobody drives exactly according to the epa test schedule. It is just a way to compare different vehicles under controlled conditions. Vehicle A is less thirsty than vehicle B.

[sully06]

Hope they bring that to America

[MetroMPG]

Cvt is good, but people did not like the way they revved on aceleration. That's why they changed the control to simulate fixed gear ratios.

FYI, there are no "simulated" shifts in the Mirage CVT. Engine RPM changes with load - no fakery.

Originally Posted by wickydude

As to what the epa tests involve, that is well documented. Look here for example:
Detailed Test Information

That resource is good, but it doesn't describe any rules on how a manual transmission is to be operated during testing, and I suspect there are rules in order to achieve some kind of consistency.

[Daox]

Originally Posted by sully06

Hope they bring that to America

The Mirage is already for sale in the US. It has been for a month now.