Subaru Outback Forums banner
  • Our 4th of July contest is live! Enter HERE for your chance to win a $400 Walmart gift card.
  • Hey everyone! Enter your ride HERE to be a part of this month's Outback of the Month Challenge!

21 - 40 of 42 Posts

·
Registered
'06 B9, '10 FXT Prem, '11 OB 3.6R Ltd, '13 FXT Ltd, '14 OB 3.6Rs Ltd, '15 WRX Ltd
Joined
·
38 Posts
Having owned 12 Subaru's since 99, I can affirmatively confirm that the CVTs "don't coast." I hate it. We've got rid of one of them with the '19 Legacy 3.6R remaining. None of the ATs behaved anything like the CVTs.
 

·
Registered
2012 OB , 2017 Impreza
Joined
·
3,562 Posts
Well, the engineers did not consider that coasting to a red light saves fuel over cruising with applied power until the last minute. The engine braking makes it impossible to coast to red lights at the point I would like to...
Lets not forget that those engineers are being paid to create vehicle-dynamics to meet several criteria (MPG, pleasing driving experience, emissions requirements....etc) Perhaps what you "would like" does not meet the same requirements they were tasked to achieve.
 

·
Registered
2020 Touring XT, Crystal White Pearl
Joined
·
251 Posts
I've had 2 Subaru Outbacks. A 2013 Limited and a 2020 Touring XT. The 2013 would do the following... If I let my foot off the accelerator, and was going down, if the car's speed picked up 5 mph, it would go into a lower ratio, to hold the speed.

My 2021 XT does not.

I prefer the 2013 logic.
 

·
Registered
Joined
·
91 Posts
As an aside to all of this, there's a reason theres no gate between drive and neutral in the shift lever if one really wants to freewheel.

I'm not a fan of the trans holding me back in our large rural rolling hills but it's relatively easy to tell it not to.

I've also played with the extremely silly paddle shifters and once in a while you can convince the fake gears to go UP one in a downhill coast. Without published specs it seems somewhat inconsistent but who knows what criteria it uses for fake gears.

Knowing how a CVT actually works, I'd rather the paddles just allowed you to slew RPM smoothly as desired, within a safe RPM range and CVT range, but that would confuse the average former auto trans drivers.

I don't put much stock in the computer learning mentioned. If it stored that information for an extremely long period of time or had a major impact on driving parameters, the dealers would need a way to clear it to have used cars perform equally on a test drive pre-sale or a lead-footed former owner could turn a decent performing vehicle into a complete dog (from it trying to be extra safe) the sales staff couldn't sell. The hysteresis can't be very long on that stuff.

Basically it's just learning the quirks of the engineer. If I want to roll the first button pressed is killing Eyesight/cruise, the second is a shift to neutral. And yeah I know that can be an issue if it stalls. Save the lecture. I rarely feel the need but it's always an option.

Would I prefer it decouple a bit more or even make it driver controlled (like many trucks in towing mode, lightly touch the throttle in a downhill and they quit holding the truck and trailer back when desired...) Yeah. But it's a perfectly livable system. Just have to know what it's going to do and drive it if that's not what you want.

Bruce's comment about the 3.6 actually lowering RPM as you get back into the throttle, I've seen that rarely on the smaller engine if you let it coast off of a really long hill on to flat and just barely feather the throttle back in. To get it to happen you have to time it to just before the car loses any speed.

The poor '00 is so underpowered it grabs two gears on the uphills as it keeps losing speed hahaha. If you don't help it anyway... Poor thing can barely get out of its own way. Haha.

All sorts of different interpretations over the years.

The new to us 01 Toyota Avalon is oooold skool and unlocks everything in the downhills. Wheeeeee! It'll easily gain 10MPH on our particular hills if using its old style cruise, they didn't want to waste a single bit of gravity assistance to keep their fuel econ numbers up with that 1MZ and the venerable four speed with overdrive. Punching the OD off doesn't have much effect just an RPM bump and more noise but it's not a big enough gear ratio chance to accomplish much.

It's a rolling couch. Ha. The squishiest roller coaster ever made. Ha.

The one that really takes off downhill is the 07 Ford truck. Weight and everything unlocked. Whoosh. And you can't downshift that one smoothly, the gear change is too drastic. It'll toss everything around unless you've got a few thousand pounds of stuff in it or behind it.
 

·
Registered
'17 OB 3.6R Touring [ex-'09 OB Ltd. (2009-16); ex-'01 Audi A6 Avant (2001-2009)]; '14 Impreza Sport Premium
Joined
·
924 Posts
I think my '17 Touring w/83k mi coasts really well. I kill my cruise hundreds of feet before stopping or turning and only shed 20-30% of my speed while coasting.

I was thinking the CVT has to play a major part in this, sensing no throttle, no braking, and knows the wheel spin, then "finding" the perfect coasting gear ratio.
yes, engine braking down to about 18 mph then it coasts easily and freely, you can learn this behavior and, eg, approach stop lights smoothly with zero throttle input, when you hit the coasting threshold speed it coasts very freely ime.

Freewheeling is just dangerous and a little stupid imo.
 

·
Registered
2010 2.5 CVT Limited
Joined
·
1,810 Posts
This is one of the first things I noticed about my OB is that it can't coast on flats.
I have to wonder why this is. It would be simple to unlock the TCC and enhance the coasting even if the CVT doesn't have a sprag to allow freewheeling.
 

·
Registered
2019 3.6 Touring & 2018 H6 Limited. Previously also owned a number of Outbacks.
Joined
·
194 Posts
I'm new to both Subaru and AWD so please tell me if this is normal for a 2018 Outback Limited 3.6 we just started driving.
When driving and I take my foot off the accelerator pedal, the car doesn't feel like it is coasting freely (still in drive gear). It feels like the transmission/engine is holding the car back and gently slowing it down. It seems most noticeable in the 20-40 mph range.
Compared to other CVT cars I've owned, it feels as though it would coast more freely if I slipped it into neutral (which I'm not doing and don't plan to do.)
Everything else seems great, good acceleration and gas mileage in the mid-20s for mixed driving.
Is this just the normal feel of a Subaru CVT?
That’s why as my 8th subaru I never had to change brake pads! Especially noticeable on the 6 cyclinder.
 

·
Registered
Joined
·
91 Posts
Another thought cane to mind.

Without knowing the parameters of the test, one wonders if it's all tuned for the EPA gas mileage test specifically whatever it wants, to get the best official fleet mileage numbers they can.

Because we all know EPA mileage never matched real world mileage.

And not coasting down hills ain't ever going to be The Way... to best real world mileage.

But I see this all day every day in modern engineering. Engineer to the test. Regulatory stuff can cost more than doing the right thing for the real world.

Just a suspicion. Could also just be an overemphasis on "safety". Or not wanting the driver to do a fake CVT "downshift" very often.

Lowest common denominators always win in engineerimg for the masses.
 

·
Registered
2018 OB 2.5 Ltd, No Eyesight, No Navigation
Joined
·
3,461 Posts
I don't put much stock in the computer learning mentioned. If it stored that information for an extremely long period of time or had a major impact on driving parameters, the dealers would need a way to clear it to have used cars perform equally on a test drive pre-sale or a lead-footed former owner could turn a decent performing vehicle into a complete dog (from it trying to be extra safe) the sales staff couldn't sell. The hysteresis can't be very long on that stuff.

Basically it's just learning the quirks of the engineer. If I want to roll the first button pressed is killing Eyesight/cruise, the second is a shift to neutral. And yeah I know that can be an issue if it stalls. Save the lecture. I rarely feel the need but it's always an option.
The only reason I mentioned learning was because some Outbacks are apparently not behaving like the original post. For example, my Outback is not. I just checked again yesterday to be sure. The behavior in the original post never even crossed my mind for the past 3.5 years of ownership. If everybody here had the same experience as the original post, then I would not say there is learning happening. (Note I do not have Eyesight, and there may be something to that as well.)
 

·
Registered
2020 Legacy Premium
Joined
·
48 Posts
One thing I heard soon after getting the Subaru that surprised me, but appears to be true, is that if you're coasting in gear, fuel to the engine gets shut off - the engine isn't even running, it's being turned by the wheels, but not firing.

Once your speed drops to about 15 MPH, the fuel is turned back on and it starts up on its own again. Go to some deserted section of straight road where you can safely drive at low speed and try this experiment: set the multi-function display on the panel to show instantaneous MPG and, on a long straight section, from about 40 or 50 MPH, take your foot off the gas and watch the fuel economy jump to 99.9 MPG (that's the highest it can display - it's actually infinity) as you decelerate. Once your speed drops below 20, at some point you may (or may not) feel a slight shudder, and the economy drops from 99.9 to 50 or so (in my 2.5, anyway; YMMV) because you are starting to consume fuel again. I never noticed that shudder until I started looking for it - it's pretty subtle.

If, instead, you coast in neutral, you don't see that 99.9 MPG, because it has to feed the engine gas to keep it turning.

The upshot is that your car really is "dragging" because engine braking is greater than you may be used to. The engine is not generating motive power at all - instead, it's absorbing it.
Every car you’ll drive with a fuel economy gauge behaves this way. When you coast above a certain speed/rpm Your mpg will be more than 99.9. Upon slowing below the point where idle fuel consumption (coasting in gear, engine turning, though not being fed any fuel) results in less than 99 mpg you will again see the mpg drop, indicating fuel actually IS being fed into the motor, albeit very little, reaching 0 when you’ve stopped. Every car does this.
Just as the CVT gear ratio gradually changes during acceleration, it changes during deceleration, though not exactly in the same manner.
 

·
Registered
2020 Legacy Premium
Joined
·
48 Posts
This is one of the first things I noticed about my OB is that it can't coast on flats.
I have to wonder why this is. It would be simple to unlock the TCC and enhance the coasting even if the CVT doesn't have a sprag to allow freewheeling.
Coming off cruise control is where I notice it the most. During normal driving, no cruise, I don’t notice it as much except when coming to a gentle stop. I can feel either the CVT gear ratio change at about 10 mph, or at least releasing a bit, when the car seems to suddenly want to coast a little more freely.
 

·
Premium Member
2020 Onyx
Joined
·
10,336 Posts
Coming off cruise control is where I notice it the most. During normal driving, no cruise, I don’t notice it as much except when coming to a gentle stop. I can feel either the CVT gear ratio change at about 10 mph, or at least releasing a bit, when the car seems to suddenly want to coast a little more freely.
That sudden coast freely when coming to a gentle stop is the torque converter clutch releasing.
 

·
Registered
'15 Outback 2.5i Premium
Joined
·
2,590 Posts
Every car you’ll drive with a fuel economy gauge behaves this way. When you coast above a certain speed/rpm Your mpg will be more than 99.9. Upon slowing below the point where idle fuel consumption (coasting in gear, engine turning, though not being fed any fuel) results in less than 99 mpg you will again see the mpg drop, indicating fuel actually IS being fed into the motor, albeit very little, reaching 0 when you’ve stopped. Every car does this.
Just as the CVT gear ratio gradually changes during acceleration, it changes during deceleration, though not exactly in the same manner.
I'm not sure I entirely follow what you're saying here: "Upon slowing below the point where idle fuel consumption (coasting in gear, engine turning, though not being fed any fuel) results in less than 99 mpg..."

If the engine is not being fed any fuel, then fuel consumed is zero and the distance traveled per gallon consumed is infinite* as long as you're moving at all, no matter how slowly. If the engine is fed enough fuel to idle, then it is being fed a non-zero amount of fuel, and there is a speed where miles per gallon falls below 99.9.

In my case, when I was looking into this, the reported instantaneous MPG would stay at 99.9 until it dropped abruptly (not gradually) to the 60s or so (along with a very subtle shudder), and then declined gradually from there as the car slowed. This suggests that the car is being fed no fuel until a certain point, after which the fuel flow is switched on and fuel consumption became a finite number (and well under 100 MPG).

* Some mathematicians may quibble about this statement because it involves the result of division by zero, and a non-zero number divided by zero is undefined in formal math usage - they are happier saying something like "for non-zero X, the limit of X / Y approaches infinity as Y approaches zero". As a practical matter, however, scientists and engineers are perfectly happy to avoid having to handle special cases separately, so they are willing to say "... so we will treat X / Y for non-zero X and Y = 0 as infinity", because doing so is actually meaningful to them.

For a concrete example, look at how the IEEE Standard 794 for Floating-Point arithmetic handles this - the default option for non-zero X divided by zero is to return the floating-point representation for infinity, and keep going. There is an option to return NaN (Not a Number), and either keep going and let the meaningless result propagate, producing NaN for any result subsequently involving the NaN ("Quiet NaN"), or force the program to either stop with an error or "trap" and handle the special case ("Signaling NaN"). As far as I know, the default is typically used, along with the converse, "any finite number divided by infinity is zero".

The result of zero divided by zero is universally agreed to be undefined. In IEEE 794 math it results in NaN, I think "Signaling" by default, which forces the program to explicitly deal with it (or abort). Infinity divided by infinity is also undefined.
 

·
Premium Member
2016 Outback Premium 2.5 CVT w/EyeSight+SRVD
Joined
·
7,542 Posts
As a practical matter, however, scientists and engineers are perfectly happy to avoid having to handle special cases separately, so they are willing to say "... so we will treat X / Y for non-zero X and Y = 0 as infinity", because doing so is actually meaningful to them.
I recently worked on a similar problem, rather common, of displaying varying data values as a 2-digit percentage (properly rounded) in the range 0 to 100. Very low but non-zero data values were being displayed as 0%, which was clearly incorrect and could be misinterpreted. The simple solution was to introduce a special case: non-zero data values that round to 0% (i.e. values below 0.5% before rounding) are now displayed as "<1%." The same logic can be applied at the opposite end of the display range.

In the case of the Outback instantaneous-MPG display, a display of ">99" might be preferable to "99.9" or "infinity."

There is an option to return NaN (Not a Number) ...
There is also the sometimes-important distinction between zero and NULL. At least these days we seldom have to deal with the difference between positive and negative zeros!
 
  • Like
Reactions: SilverOnyx

·
Registered
2012 OB , 2017 Impreza
Joined
·
3,562 Posts
Because we all know EPA mileage never matched real world mileage.
Speak for yourself.... I have managed to beat the EPA numbers on most vehicles. Driving habits are a HUGE factor and most folks do not have the patence to learn and practice efficient driving. (also extends brake life)
 

·
Registered
2020 Legacy Premium
Joined
·
48 Posts
I'm not sure I entirely follow what you're saying here: "Upon slowing below the point where idle fuel consumption (coasting in gear, engine turning, though not being fed any fuel) results in less than 99 mpg..."

If the engine is not being fed any fuel, then fuel consumed is zero and the distance traveled per gallon consumed is infinite* as long as you're moving at all, no matter how slowly. If the engine is fed enough fuel to idle, then it is being fed a non-zero amount of fuel, and there is a speed where miles per gallon falls below 99.9.

In my case, when I was looking into this, the reported instantaneous MPG would stay at 99.9 until it dropped abruptly (not gradually) to the 60s or so (along with a very subtle shudder), and then declined gradually from there as the car slowed. This suggests that the car is being fed no fuel until a certain point, after which the fuel flow is switched on and fuel consumption became a finite number (and well under 100 MPG).

* Some mathematicians may quibble about this statement because it involves the result of division by zero, and a non-zero number divided by zero is undefined in formal math usage - they are happier saying something like "for non-zero X, the limit of X / Y approaches infinity as Y approaches zero". As a practical matter, however, scientists and engineers are perfectly happy to avoid having to handle special cases separately, so they are willing to say "... so we will treat X / Y for non-zero X and Y = 0 as infinity", because doing so is actually meaningful to them.

For a concrete example, look at how the IEEE Standard 794 for Floating-Point arithmetic handles this - the default option for non-zero X divided by zero is to return the floating-point representation for infinity, and keep going. There is an option to return NaN (Not a Number), and either keep going and let the meaningless result propagate, producing NaN for any result subsequently involving the NaN ("Quiet NaN"), or force the program to either stop with an error or "trap" and handle the special case ("Signaling NaN"). As far as I know, the default is typically used, along with the converse, "any finite number divided by infinity is zero".

The result of zero divided by zero is universally agreed to be undefined. In IEEE 794 math it results in NaN, I think "Signaling" by default, which forces the program to explicitly deal with it (or abort). Infinity divided by infinity is also undefined.
hen not being fed any fuel,
I'm not sure I entirely follow what you're saying here: "Upon slowing below the point where idle fuel consumption (coasting in gear, engine turning, though not being fed any fuel) results in less than 99 mpg..."

If the engine is not being fed any fuel, then fuel consumed is zero and the distance traveled per gallon consumed is infinite* as long as you're moving at all, no matter how slowly. If the engine is fed enough fuel to idle, then it is being fed a non-zero amount of fuel, and there is a speed where miles per gallon falls below 99.9.

In my case, when I was looking into this, the reported instantaneous MPG would stay at 99.9 until it dropped abruptly (not gradually) to the 60s or so (along with a very subtle shudder), and then declined gradually from there as the car slowed. This suggests that the car is being fed no fuel until a certain point, after which the fuel flow is switched on and fuel consumption became a finite number (and well under 100 MPG).

* Some mathematicians may quibble about this statement because it involves the result of division by zero, and a non-zero number divided by zero is undefined in formal math usage - they are happier saying something like "for non-zero X, the limit of X / Y approaches infinity as Y approaches zero". As a practical matter, however, scientists and engineers are perfectly happy to avoid having to handle special cases separately, so they are willing to say "... so we will treat X / Y for non-zero X and Y = 0 as infinity", because doing so is actually meaningful to them.

For a concrete example, look at how the IEEE Standard 794 for Floating-Point arithmetic handles this - the default option for non-zero X divided by zero is to return the floating-point representation for infinity, and keep going. There is an option to return NaN (Not a Number), and either keep going and let the meaningless result propagate, producing NaN for any result subsequently involving the NaN ("Quiet NaN"), or force the program to either stop with an error or "trap" and handle the special case ("Signaling NaN"). As far as I know, the default is typically used, along with the converse, "any finite number divided by infinity is zero".

The result of zero divided by zero is universally agreed to be undefined. In IEEE 794 math it results in NaN, I think "Signaling" by default, which forces the program to explicitly deal with it (or abort). Infinity divided by infinity is also undefined.
With fuel injection the system may cut off the flow, but will still be at the ready even at the slightest throttle signal, if you engage the CC, or if the rpm drops below certain parameters set by the engine designers.

That "99" mpg number you see is nothing more than a displayed number based on fuel flow and vehicle speed. MPG may be calculated at 100 or 1500, but you will only see 99, as that is all the computer can tell you. It is not telling you that fuel has been cut off to the engine. I have driven many cars with an mpg display, and on cruise control, and have many times seen 99 in the display after the car has crested an upgrade or has reached a slight downhill grade from flat. Although it doesn't last long, it happens. And the engine is still being fed a fuel air mixture, as it is still holding speed throughout the process. This is no different than seeing 99 when lifting off the pedal.

Sure, the ECM may at times stop the flow of fuel into the mixture when you lift your foot, but as a driver you don't know exactly when that occurs. As for me, I don't care either. I just try to get as much mpg as I can from my car, whether it's my own or rental or somewhere in between.
 

·
Registered
2018 OB 2.5 Ltd, No Eyesight, No Navigation
Joined
·
3,461 Posts
* Some mathematicians may quibble about this statement because it involves the result of division by zero, and a non-zero number divided by zero is undefined in formal math usage - they are happier saying something like "for non-zero X, the limit of X / Y approaches infinity as Y approaches zero". As a practical matter, however, scientists and engineers are perfectly happy to avoid having to handle special cases separately, so they are willing to say "... so we will treat X / Y for non-zero X and Y = 0 as infinity", because doing so is actually meaningful to them.
Since this theoretical and unnecessary topic of “zero” is still lingering, I might as well throw in my thoughts. In the real world, there is no pure zero and there is no infinity. In other words, even if the fuel in the tank is completely unused, the fuel in the tank would not stay there for infinite years. Fuel is leaving the tank, even if the rate is just 0.0000000000000000000000000000000000001 gallons per day for example. Thus, the mpg never truly has a division by a pure zero problem.

I am surprised I am participating in this topic without first smoking something.
 

·
Registered
2020 Legacy Premium
Joined
·
48 Posts
Since this theoretical and unnecessary topic of “zero” is still lingering, I might as well throw in my thoughts. In the real world, there is no pure zero and there is no infinity. In other words, even if the fuel in the tank is completely unused, the fuel in the tank would not stay there for infinite years. Thus, the mpg never truly has a division by a pure zero problem.
FACTS!
 

·
Registered
Joined
·
91 Posts
Speak for yourself.... I have managed to beat the EPA numbers on most vehicles. Driving habits are a HUGE factor and most folks do not have the patence to learn and practice efficient driving. (also extends brake life)
Never said they couldn't get beat. Said they don't match. Read carefully. Ha.
 
21 - 40 of 42 Posts
Top