Is this some weird CVT behavior?

[Patty Winter]

I drove into my driveway the other day (that being the logical thing
to do in a driveway :-)), and forgot to put my Outback into park
before picking up pen and paper to log my mileage. I must have taken
my foot off the brake, because the next thing I noticed was the car
slowly rolling backwards down the driveway.

The next day I tried just the opposite: I backed in the driveway,
left the car in reverse, and took my foot off the brake. It started
rolling forward.

Call me old-fashioned, but putting a car into a forward gear used
to mean that it couldn't go backwards, and vice versa. Is this
"fluidity" in the Outback the result of the no-gears design of a
CVT transmission? If so, it's a bit unnerving.


Patty

[Frank]

Don't know but interested in knowing about how CVT behaves as it might
be option for my next Subaru.

Saw another one the other day where my brothers Outback's automatic
transmission had a manual mode where you could shift with buttons on the
steering column. Don't know what advantages are.

I live in a hilly area and wonder what best option might be.

[VanguardLH]

So is your driveway inclined even if just a little? Let's assume the
driveway is inclined which is typical to allow runoff, and it is
inclined upward from road to your garage.

You drive forward into the driveway and presumably leave the
transmission engaged (foward). The car is tilted up on the front. You
take your foot off the brake. The car starts to reverse down the
incline. This is what happens when a car is in neutral: it starts to
roll in the direction on the incline.

You drive backward into the driveway and presumably leave the
transmission engaged (backward). The car is tilted up on the rear. You
take your foot off the brake. Thar starts to forward down the same
incline of the driveway. This is what happened when a card is in
neutral: it starts to roll in the direction of the incline.

So it looks like the CVT is going into neutral when there is no load on
the engine. Isn't that how it's supposed to work? If it were still
engaged when it should be in neutral then more fuel is being consumed.

You're used to an engine that remains engaged through the transmission
so there is always some torque in the direction for the currently
selected gear. In the old automatic, you putting your foot on the brake
did not disengage the engine so you lifting your foot from the brake
would have your car start moving in whatever direction (forward or
reverse) you had put the transmission. The "at rest" torque you're used
to with a traditional tranny isn't there with CVT. When at rest (no
load), CVT has put itself in neutral so when you lift your foot off the
brake then the car will start to roll in whichever direction the car is
tilted.

The point of the CVT is to reduce fuel consumption hence raising fuel
efficiency. When at rest, leaving the engine engaged means leaving a
load on it which mean more fuel gets consumed. You aren't used to "idle
neutral" which shifts the tranny into neutral when the car is stopped
and the brakes are applied. This reduces drag (load) on the engine to
increase fuel efficiency. I've read about Mitsubishi with "CVT with
Neutral Logic". Looks like Subaru employs the same scheme.

What do you think would happen if you drove into your driveway (using
either forward or reverse to pull in), used the brakes to stop the car,
manually put the car in neutral gear (whether a traditional automatic or
CVT), and then released the brake? Yep, the car would start rolling
down the hill because, well, it's in neutral.

There are many types of CVT. The manufacturers seem to keep secret how
theirs work. There are what are called "neutral transmissions". When
the load on the engine is minimal, like sitting at a stop or coasting
down a hill, the tranny goes into neutral. With a traditional tranny,
you should rev the engine to get its effective RPM to match your speed
to reduce engine braking effect when engaging the tranny out of neutral.
With a CVT, it can alter the tranny gearing while staying in neutral (to
maintain the smooth driving experience). When coasting, and because the
car may accelerate to a speed above what you want, you may have to apply
the brakes to slow you down hence more brake wear. So they give you the
option to use manual shifting to let you control the engine braking.

I haven't yet had a car with CVT but what you describe sure sounds like
"idle neutral" in an "idle transmission". Your brakes (foot or parking)
are supposed to keep you at a stop so why bothering loading the engine
to eat up more fuel when you're at a stop?

I'm waiting until CVT has more history to decide if I want one. My old
'92 Subaru Legacy is just now having a tranny problem (but suspect the
shop that did the tranny flush put in the wrong fluid). I have a couple
later Subies but still no CVT in them. I'd like a CVT-equipped Subie to
last as long as my others so longevitiy is an issue. Ford's Freestyle
CVT made by Jatco (http://en.wikipedia.org/wiki/Jatco) had sudden
unexpected breakdowns while driving. Nissan's have been critized as
less reliable that standard automatic transmissions. There is also a
difference in driving feel and control, especially for car enthusiasts
(but the same is almost always expressed between automatic and manual
transmissions). CVTs are more expensive to buy and also to maintain or
repair than a standard automatic. CVTs make more noise on startup and
when accelerating. While CVTs can rev the engine to match belt/chain
speed in the CVT for supposedly smooth shifting in or out of neutral,
users have complained of noises similar to clutch slipping in a manual
tranny. Some CVTs are still a bit jerky on acceleration so they're no
better than an automatic. Some users have complained about an abrupt
jerk when coming to a stop (perhaps due to the "idle neutral" feaure).
Because users are still accustomed to change in engine speed when
changing gears but which is absent for a [supposedly] smooth shifting
CVT, paddle shifters are added on the steering wheel to simulate or
override gear changes. To me, for now, CVTs are relegated to the gizmo
category: nice if you're willing to pay for it but not necessary.

[TheSeeker]

Hi,

On Wednesday, June 25, 2014 2:50:59 PM UTC-5, VanguardLH wrote:
> Patty Winter wrote:
>
<snip>

>
> > Call me old-fashioned, but putting a car into a forward gear used
> > to mean that it couldn't go backwards, and vice versa. Is this
> > "fluidity" in the Outback the result of the no-gears design of a
> > CVT transmission? If so, it's a bit unnerving.
>

<discussion of idle-neutral behavior snipped>

>
> Neutral Logic". Looks like Subaru employs the same scheme.

I own a 2014 2.5i Subaru Forester with CVT.
I have not noticed it going into a neutral condition when stopped on an incline while in gear.

<snip>

>
> I'm waiting until CVT has more history to decide if I want one. My old
> '92 Subaru Legacy is just now having a tranny problem (but suspect the
> shop that did the tranny flush put in the wrong fluid). I have a couple
> later Subies but still no CVT in them. I'd like a CVT-equipped Subie to
> last as long as my others so longevitiy is an issue. Ford's Freestyle
> CVT made by Jatco (http://en.wikipedia.org/wiki/Jatco) had sudden
> unexpected breakdowns while driving. Nissan's have been critized as
> less reliable that standard automatic transmissions. There is also a
> difference in driving feel and control, especially for car enthusiasts
> (but the same is almost always expressed between automatic and manual
> transmissions). CVTs are more expensive to buy and also to maintain or
> repair than a standard automatic. CVTs make more noise on startup and
> when accelerating. While CVTs can rev the engine to match belt/chain
> speed in the CVT for supposedly smooth shifting in or out of neutral,
> users have complained of noises similar to clutch slipping in a manual
> tranny. Some CVTs are still a bit jerky on acceleration so they're no
> better than an automatic. Some users have complained about an abrupt
> jerk when coming to a stop (perhaps due to the "idle neutral" feaure).
> Because users are still accustomed to change in engine speed when
> changing gears but which is absent for a [supposedly] smooth shifting
> CVT, paddle shifters are added on the steering wheel to simulate or
> override gear changes. To me, for now, CVTs are relegated to the gizmo
> category: nice if you're willing to pay for it but not necessary.

I have over 20k miles on my unit, too few to comment on longevity or reliability at this point. Compared to my 2002 Forester with automatic though, the CVT is smoother, and quieter than the automatic, and can wring far more out of the 2.5L boxer performance-wise.

There is supposedly a fuel economy boost going with the CVT. The 2014 Forester with CVT is rated at 32 MPG highway (which I do get), and the manual is rated at 29 MPG. I put on about 25k miles/year, so with the cost of gas around here, I can save nearly $300/year in gas compared to the auto.

Sincerely,
Duane

[Patty Winter]

In article <lof972$3l6$1...@news.albasani.net>, VanguardLH <V...@nguard.LH> wrote:

>Patty Winter wrote:
>
>So it looks like the CVT is going into neutral when there is no load on
>the engine. Isn't that how it's supposed to work?

Apparently so. This is the first CVT, so I wasn't aware that CVTs had
this behavior.

>If it were still
>engaged when it should be in neutral then more fuel is being consumed.

I don't think there's a lot of fuel being consumed when the engine
is off, but I could be wrong...

> To me, for now, CVTs are relegated to the gizmo
>category: nice if you're willing to pay for it but not necessary.

Well, they're the only type of automatic transmission that Subaru
offers on Outbacks, so I didn't have any choice.


Patty

[Patty Winter]

In article <lof47e$lfc$1...@dont-email.me>,
Frank <frankdo...@comcast.net> wrote:

[unneeded quotage deleted]

>Saw another one the other day where my brothers Outback's automatic
>transmission had a manual mode where you could shift with buttons on the
>steering column. Don't know what advantages are.

You don't even have to put it in manual mode. You can shift the car
even when it's in drive.


Patty

[cl...@snyder.on.ca]

On Wed, 25 Jun 2014 14:50:59 -0500, VanguardLH <V...@nguard.LH> wrote:

>Patty Winter wrote:
>
>> I drove into my driveway the other day (that being the logical thing
>> to do in a driveway :-)), and forgot to put my Outback into park
>> before picking up pen and paper to log my mileage. I must have taken
>> my foot off the brake, because the next thing I noticed was the car
>> slowly rolling backwards down the driveway.
>>
>> The next day I tried just the opposite: I backed in the driveway,
>> left the car in reverse, and took my foot off the brake. It started
>> rolling forward.
>>
>> Call me old-fashioned, but putting a car into a forward gear used
>> to mean that it couldn't go backwards, and vice versa. Is this
>> "fluidity" in the Outback the result of the no-gears design of a
>> CVT transmission? If so, it's a bit unnerving.
>

A CVT, unlike an automatic, does not use a torque converter. When you
step on the gas it shifts the sheaves to transmit power. And unlike a
standard trans, it does not "engage" the gears with the engine shut
off - so engine running at idle or shut off, with the gear shidt in
drive or reverse, there is nothing stopping the car from rolling away.

[Patty Winter]

In article <8ipmq9h4ae210sl24...@4ax.com>,

<cl...@snyder.on.ca> wrote:
>>
>A CVT, unlike an automatic, does not use a torque converter. When you
>step on the gas it shifts the sheaves to transmit power. And unlike a
>standard trans, it does not "engage" the gears with the engine shut
>off - so engine running at idle or shut off, with the gear shidt in
>drive or reverse, there is nothing stopping the car from rolling away.

Thanks, that's what I figured was happening. Well, good to know before
it happens on a steeper slope!


Patty

[VanguardLH]

Patty Winter wrote:

> I don't think there's a lot of fuel being consumed when the engine
> is off, but I could be wrong...

Reread your starting post. Where did you say anything about the running
state of the engine?

Is your driveway inclined? If so, what happens when you drive the car
to a stop by applying brakes, stop the engine, put into neutral, and
then release the brakes?

Even with automatic or manual transmissions, it is bad practice to use
only the engine to keep a parked car in its place. You brake to a stop,
turn the wheels in the correct direction (straight ahead if flat, away
from curb for uphill, or towards curb for downhill), put the tranny in
first or reverse for a manual or in Park for automatic (the later
usually required using a lockout on the key), and then apply the parking
brake. It's called a *parking* brake for a reason. I don't recall any
method of parking that doesn't end with using the parking brake.

In fact, if you get in the habit of not using the parking brake and just
the compression in the engine's cylinders, you may one day need that
parking brake but pulling on it breaks the cable. If not moved, the
cable can rust in place. There's no continued lubrication of the cable
after manufacture. So keep using the parking brake to move the cable to
make sure it doesn't rust in place.

I know a lot of folks that become accustomed to using the automatic
transmission's parking pawl to keep their car in place when parking; see
http://en.wikipedia.org/wiki/Parking_pawl. That it works on shallow
inclines doesn't obviate it's a dumb way to park. A manual tranny
doesn't have a parking pawl so the user would be relying on the clutch
and engine compression to keep the car in place when parked. Clutches
wear and slip. Parking pawls slip out or break and they rely on engine
braking, too. I've seen many times when users have gotten into the
habit of using the parking brake but apply it at the wrong time. You
stop using the brakes, pull the parking brake handle, and then put the
automatic into park. That reduces strain on the pawl and tranny. It
doesn't take a big smack of someone hitting your car in the parking lot
for the pawl to brake, you not using the parking brake, and finding your
car somewhere else than where you originally parked it. Probably the
biggest reasons drivers don't properly use the parking brake is they are
too lazy, the pawl has been working so far, or they drove the car one
day after forgetting to release the parking brake and smelled the burnt
brakes and didn't want to do that again (they don't bother looking at
the dash to notice the Parking brake warning light).

That your car started rolling after you supposedly "parked it" shows you
aren't parking properly.

[VanguardLH]

In standard automatic transmissions, there was a parking pawl in the
tranny that would keep the car in place while parked. Putting it in
"Park" engaged the pawl. It is very bad practice to use the pawl to
keep the car in place. It is not designed to prevent movement of the
weight of the vehicle yet many drivers get lazy in relying on it to
"park" their car even on an incline. You said you put the tranny in
Park (but didn't mention stopping the engine). I haven't researched to
see if CVT transmissions include the use of a park pawl. Your scenario
indicates there is no such pawl.

The pawl, even when present, is not designed to keep the entire weight
of the car locked into a parking position. It is a ignorant and lazy
practice by drivers of automatic transmissions. Parking on an incline
or getting bumped or pushed by another car can break the pawl inside the
tranny causing more severe problems (besides the car smashing into
something when the pawl wasn't there to block movement).

You got caught using you old bad habit with an automatic: using the
parking pawl (Park), tranny, and engine to keep the car from moving
while parked. Every recommended parking procedure ends with or includes
use of the parking brake. Lots of drivers don't use the parking brake.
Well, you can't teach everyone how to properly park their car. Those
that do use the parking brake and have automatics usually apply the
parking brake too early (i.e., after putting the tranny in Park and
therafter leaving stress on the pawl when engaging the parking brake
afterward).

Now that you have a CVT, perhaps you'll start employing the parking
brake like you were supposed to before when you had an automatic. When
parking, you should get accustomed to hearing that zip noise when
pulling up on the parking brake. In some locales, you can get ticketed
if it can be seen the parking brake lever is down in a parked car
because you didn't properly park your vehicle. It's a cheaper wake-up
call then having to get your transmission replaced. If the pawl breaks
and your car rolls down a hill to smash into another car and it is noted
that your parking brake was not engaged, well, you didn't properly park
so the fault for the accident is all yours.

[dsi1]

Cars with automatics work the same way as your Subaru with CVT. Both
have a torque converter between the engine and transmission. This allows
the slippage you describe and how your car can be at a stop even when
the engine is running and in D or R.

Your observation about the an automatic transmission car not being able
to slide backwards while in drive is not correct. Given enough of an
incline, it will slide alright - even if it's not inclined to. (-:

[dsi1]

On 6/25/2014 10:47 AM, TheSeeker wrote:
>
> I have over 20k miles on my unit, too few to comment on longevity or reliability at this point. Compared to my 2002 Forester with automatic though, the CVT is smoother, and quieter than the automatic, and can wring far more out of the 2.5L boxer performance-wise.
>
> There is supposedly a fuel economy boost going with the CVT. The 2014 Forester with CVT is rated at 32 MPG highway (which I do get), and the manual is rated at 29 MPG. I put on about 25k miles/year, so with the cost of gas around here, I can save nearly $300/year in gas compared to the auto.
>
> Sincerely,
> Duane
>

The CVT would seem to be the ideal transmission. What's not to like
about no gears? I used to have a variable speed drive on my printing
press. It used the same kind of setup as on these modern transmission.
They worked very well although driving the pulleys on the sides of that
V-belt must have created a lot of friction. Anyway, it worked great and
I could slowly crank up the speed and print as fast as I dared. Good
luck with your transmission.

[Patty Winter]

In article <lofq5v$5as$1...@news.albasani.net>, VanguardLH <V...@nguard.LH> wrote:
>
>Reread your starting post. Where did you say anything about the running
>state of the engine?

If I didn't say that it was off, my apologies.

>Even with automatic or manual transmissions, it is bad practice to use
>only the engine to keep a parked car in its place.

My first car was a stick shift (as is one of my current cars), so I
have always been in the habit of using the parking brake in addition
to putting the car in gear (stick shift) or park (automatic). As I
mentioned, I got distracted and had not completed my usual parking
procedure in this case. My previous Subie would not have started
rolling in this case.


Patty

[cl...@snyder.on.ca]

Read the OPs original post. She did NOT put it in park (which she
usually does) when it rolled. Same would have happened with an
automatic or a standard UNLESS it had hill-holder and she had her
foot on the clutch. Some Soobs have hill-holders (forester, for one) -
including automatic Outbacks and Legacys

[John McGaw]

I think that you will find that _every_ automatic transmission car has that
effect to varying degrees. Old-style torque converter transmissions
'leaked' a good bit of rotational energy -- put the vehicle in Drive on an
absolutely flat surface and the vehicle will creep forward; if the car is
very old and/or maladjusted the creep might be more of a lurch. Aim the
same car up a slight incline and put it in Drive and it may creep forward
more slowly and if the incline exactly balances the tendency to creep then
the car will not move but if the incline is more steep then the vehicle
will invariably roll backwards. What you are experiencing is as normal as
normal can be and is the reason that more vehicles are available with
hill-hold.

http://blog.truecar.com/2012/01/31/spotlight-on-technology-hill-hold-or-hill-start-assist-technology-makes-steep-road-starts-safer-for-drivers/

[John Varela]

On Thu, 26 Jun 2014 00:40:32 UTC, VanguardLH <V...@nguard.LH> wrote:

> I don't recall any
> method of parking that doesn't end with using the parking brake.

It used to be that in a car with an exposed cable-operated parking
brake, water could freeze on the mechanism and prevent the brake
from releasing. So it was inadvisable to set the parking brake after
driving on a wet winter's day.

I don't know if any modern cars have exposed brake cables.

--
John Varela

[cl...@snyder.on.ca]

On 27 Jun 2014 00:48:04 GMT, "John Varela" <newl...@verizon.net>
wrote:

Not only that, but wet brake shoes/drums would freeze together and
you were going NOWHERE untill a thaw. Same can happen with disk
brakes.(caliper parking brake)

[Yousuf Khan]

Yeah, I remember that used to be a major issue in the 70's and 80's,
even before I could drive, I remember my father experiencing it in the
70's. And then later I experienced it myself in the 80's.

Those types of brakes also had the problem of rusting out. So I find
that if I use the parking brakes all of the time, then it keeps the
cable operational, and breaks-up any rust in the system.

Yousuf Khan

[Yousuf Khan]

On 25/06/2014 10:25 PM, cl...@snyder.on.ca wrote:
> Read the OPs original post. She did NOT put it in park (which she
> usually does) when it rolled. Same would have happened with an
> automatic or a standard UNLESS it had hill-holder and she had her
> foot on the clutch. Some Soobs have hill-holders (forester, for one) -
> including automatic Outbacks and Legacys

I think Patty is a guy, in this case. :)

Yousuf Khan

[Yousuf Khan]

On 25/06/2014 9:15 PM, dsi1 wrote:
> Cars with automatics work the same way as your Subaru with CVT. Both
> have a torque converter between the engine and transmission. This allows
> the slippage you describe and how your car can be at a stop even when
> the engine is running and in D or R.
>
> Your observation about the an automatic transmission car not being able
> to slide backwards while in drive is not correct. Given enough of an
> incline, it will slide alright - even if it's not inclined to. (-:

A CVT has a torque converter? This is the first I'm hearing of that. I
always assumed it just had an automatic clutch pack to disengage the
drive during idling.

Yousuf Khan

[Patty Winter]

In article <53ad8743$1...@news.bnb-lp.com>,

Huh???

[dsi1]

My understanding is that these cars use a standard torque converter. I'll let someone else validate this or prove it false. The Justy used a very neat electromagnetic clutch that used ferrofluid to make a connection between the motor and transmission. My guess is that clutches and transmissions will go the way of the dinosaurs very soon.

[dsi1]

Welcome to the club. :-)

[cl...@snyder.on.ca]

On Fri, 27 Jun 2014 13:25:58 -0700 (PDT), dsi1 <dsi...@yahoo.com>
wrote:

It appears the Subaru version DOES use a torque converter - a
locking type same as most conventional automatics. There are CVTs
that do not need or use a torque converter.

[Yousuf Khan]

On 28/06/2014 9:15 PM, cl...@snyder.on.ca wrote:
> It appears the Subaru version DOES use a torque converter - a
> locking type same as most conventional automatics. There are CVTs
> that do not need or use a torque converter.

So we're both right.

Yousuf Khan

[dsi1]

How can a car have both a torque converter and no torque converter at
the same time? I'm assuming that the OP has a car with the Lineartronic
transmission. It's true that not all Subaru have torque converters but
all Outback models with CVTs do. Well, that's what I'd bet my $5 on.

[Yousuf Khan]

Some cars use torque converters, some don't.

Yousuf Khan

[dsi1]

On 7/1/2014 7:49 AM, Yousuf Khan wrote:
>
> Some cars use torque converters, some don't.
>
> Yousuf Khan

You certainly got that 100% right.