[billnowicki]

I have a 2018 Outback 3.6.

I've noticed that if I park in a parking garage that the time displayed on the clock will show an incorrect time when I start the car until after I drive it out of the garage and therefore receiving an update via satellite.

If this is true, then this is the FIRST car I've ever owned that does this. I've had many previous cars that are updated via satellite yet ALL of them retained the time when not in sight of a satellite.

So ...is this how everyone's cars work?

Thanks!
Bill

 

[Yoda_One]

It is GPS controlled

Laughing at oneself and with others is good for the Soul!

 

[Mad Mallard]

there is also the possibility your garage door folds space and time...

 

[billnowicki]

I'm thinking it's a 'battery' issue ... somehow it's not saving the time when the car is turned off and out-of-sight of the satellite/GPS

In other words, plenty of cars have their clocks updated via satellite/GPS ... but they don't lose the information when parked in a garage

 

[Ktiemann]

Maybe the car stops pinging the satellite when it realizes its not receiving a signal and just stops updating the time?

You could set it manually, too. That's what I did when I gave up trying to get mine to update via bluetooth.

 

[johnre]

It is GPS controlled

Another example of putting technology to work where it really isn't needed, and suffering from unintended consequences.

Ancient technology (before 1980's) consisted of mechanical clocks and electric wind mechanisms. They worked somewhat accurately (if adjusted correctly) until they no longer worked. Which was usually a couple of years, IIRC.

Starting in 1980's - digital clocks with crystal oscillators, CMOS digital microprocessors, and very low power drain to the battery worked flawlessly to an accuracy of a few seconds a year.

Starting with 2000's and GPS - same digital clock technology, but now with a GPS and more computer power to occasionally check the GPS satellites' time signals and correct your time. This produced some advantage when crossing time zones, but little advantage in having the correct time displayed - unless a few seconds accuracy wasn't good enough for you. I think it would be for most people.

But the side effect of this is that if there's a marginal GPS signal environment such as that caused by a parking garage, apparently the GPS clock update process receives a corrupted signal, and doesn't know to leave "good enough" alone on the displayed time. And the power drain to run the computer far exceeds that for the clock alone, and of course we well know the outcome of that part of it.

 

[skipw]

there is also the possibility your garage door folds space and time...

Lol! Returning to the OP's problem...

I have a 2018 Outback 3.6.

I've noticed that if I park in a parking garage that the time displayed on the clock will show an incorrect time when I start the car until after I drive it out of the garage and therefore receiving an update via satellite.

So ...is this how everyone's cars work?

Not mine. But I don't have an '18, or the built-in nav system, which I presume you do. Those automagically set the clock from GPS time.

That said, your situation doesn't seem right. I wouldn't expect the clock to drift enough to notice even if not reset for a month or longer if it's working right.

Has it lost the time of day completely (like "waking up" showing 12:00), or is it losing time (i.e. running slow) by a few minutes or few hours while it's turned off? If the latter, is the amount of time lost equal to the length of time it was off?

First SWAG: whatever source is supposed to power the clock when the car is off isn't working, so when the car is turned off, the clock either resets, or it stops measuring time until the car is turned back on. After that, the time displayed will be wrong until a valid GPS time becomes available (after leaving the garage), at the earliest, so the clock can be reset. [Unless it's off by exactly 12 hours, in which case you wouldn't notice! ]

Probably easiest would be to check for a blown fuse. According to the FSB for the '15, I might suspect main box (under the hood) fuse #29 (circuit MB-3, Audio & Nav, direct battery connection, 30 A). I'm not sure if that's the same on an '18, though.

 

[skipw]

Maybe the car stops pinging the satellite when it realizes its not receiving a signal and just stops updating the time?

I seriously doubt the clock display is updated directly from GPS time. It's likely driven from a clock chip (or time-of-day logic built into the BIU) that's synchronized to GPS time occasionally.

Another example of putting technology to work where it really isn't needed, and suffering from unintended consequences.

Ancient technology (before 1980's) consisted of mechanical clocks and electric wind mechanisms. They worked somewhat accurately (if adjusted correctly) until they no longer worked. Which was usually a couple of years, IIRC.

That was my experience, too. After a year or two, you could count on the thing being exactly right twice a day.

The horrible environment in a car - broiling hot to frigid, dust and humidity, vibration and shock - had to wreak havoc, and took its toll on the relatively delicate mechanism in a mechanical clock.

Starting in 1980's - digital clocks with crystal oscillators, CMOS digital microprocessors, and very low power drain to the battery worked flawlessly to an accuracy of a few seconds a year.

Starting with 2000's and GPS - same digital clock technology, but now with a GPS and more computer power to occasionally check the GPS satellites' time signals and correct your time. This produced some advantage when crossing time zones, but little advantage in having the correct time displayed - unless a few seconds accuracy wasn't good enough for you. I think it would be for most people.

But the side effect of this is that if there's a marginal GPS signal environment such as that caused by a parking garage, apparently the GPS clock update process receives a corrupted signal, and doesn't know to leave "good enough" alone on the displayed time. And the power drain to run the computer far exceeds that for the clock alone, and of course we well know the outcome of that part of it.

I'm betting the first part of that last paragraph isn't the case. I just can't fathom a rational reason not to just let a CMOS clock chip (or whatever) free run in the absence of valid GPS time. As I speculated in the prior post, I'm guessing it's a problem with the CMOS (or...) clock - most likely its continuous power source. If it were a system design error, lord knows we'd have been hearing about it a lot here. As far as I recollect, we haven't been.

 

[snowbelter]

I also have an '18 with Navigation. I've noticed that the time on my clock is off when I park in a parking garage, not my garage at home. It corrects pretty quickly when I drive out of the parking garage. I believe the minutes are correct when I leave the garage, but the hour is off. I had a '15 with Navigation where the time was also set automatically, and don't recall that I ever had this happen.

 

[billnowicki]

Perhaps related ... for some strange reason, the phone contacts, favorites, are not saved for more than a day. I will save some favorites, confirm they are there ... but then the next day, they are GONE.

Again ... thinking some kind of memory loss due to battery loss, or something

 

[skipw]

I also have an '18 with Navigation. I've noticed that the time on my clock is off when I park in a parking garage, not my garage at home. It corrects pretty quickly when I drive out of the parking garage. I believe the minutes are correct when I leave the garage, but the hour is off. I had a '15 with Navigation where the time was also set automatically, and don't recall that I ever had this happen.

Just out of curiosity, could it be showing Greenwich Mean Time? That's 4 hours ahead of EDT, 5 ahead of CDT, etc. Some GPS units report 0 latitude, 0 longitude (equator at Greenwich meridian, in the Gulf of Guinea off the west coast of equatorial Africa) when they can't get a reliable location.

 

[CNY_Dave]

I would guess it's not a separate clock at all, just a computer display, and they were too cheap to put in a small real-time-clock chip to keep track of the time.
So the software probably churns through an interrupt when in sleep mode, the clock timekeeping then only as accurate as the clocking circuit for the processor.

 

[skipw]

I would guess it's not a separate clock at all, just a computer display, and they were too cheap to put in a small real-time-clock chip to keep track of the time.
So the software probably churns through an interrupt when in sleep mode, the clock timekeeping then only as accurate as the clocking circuit for the processor.

The clock in my car (no nav) keeps time just fine. It gets reset when changing to and from DST and when I drive to a different time zone for more than a couple of days. About every few months on average. It's never off by more than a minute or two (it doesn't show seconds), and that's mostly dependent on how accurately it was set.

Commodity clock chips cost a few dimes apiece when purchased in quantities of a few thousand, and are probably pennies each when ordered in hundreds of thousands. I can't imagine it being cheaper to design completely different clock systems from the ground up for nav and non-nav cars, like with a standalone CMOS clock for the non-nav cars and software clock, as you describe, in cars with nav. But what do I know? Maybe they're all done in software now; if so, mine is accurate enough that accuracy is not an issue.

I'd be interested in hearing more details about the OP's and snowbelter's problems.

 

[patchelect]

When my car was garaged for it's first 5 years I never noticed it being an issue, but then again I was more interested in backing out than the time at the moment. Of course I have an '11 and they may be different.

 

[snowbelter]

Happened again this afternoon. Parked in a garage. When I started my car up, my clock read 11:15. Problem was that it was actually 3:15. My clock changed to 3:15 once I was outside the garage.

 

[skipw]

Happened again this afternoon. Parked in a garage. When I started my car up, my clock read 11:15. Problem was that it was actually 3:15. My clock changed to 3:15 once I was outside the garage.

Are you 4 hours east, or 8 hours west of Greenwich?

 

[snowbelter]

Google tells me I'm -4 hours GMT during Daylight Savings Time.


I'm sure this wasn't happening (changing time) several weeks ago when I first got my car. I haven't lost my Sirius signal in this garage. In my garage at home, I lose Sirius when I drive in, but my clock is fine.

 

[Tishers]

Hours jumping while the minutes are accurate have nothing to do with oscillator drift. When it jumps that is when the entire software system wakes up and aligns the time zone settings from whatever it went to when the brain was asleep. Also, there may be an RTC but it may not be in the module that displays time on the head unit or a dash display.

There are multiple CPU's in the car, one of them may have the duty of maintaining time while the others slave off of the master timekeeper. If you were to put a protocol analyzer on the CAN buses you might see the clock synch request messages from the slaves to the master or the clock set command from the master to the slaves. I bet that they aren't doing that until the other bus activity involved in.starting things up has settled down some.

Professionally I work with distributed control systems (for the electric utility industry) with an emphasis on communication protocols. Some devices do a time synch right away, others postpone it for a while. It still amazes me to see some of the time-stamped data from slave devices transiting the bus with a time stamp relative to 1 January, 1970 from the slave but the master redoes the time stamp to an NEMA referenced time before transmitting back to the host.

 

[billnowicki]

So ... is the consensus that it's a battery/fuse issue ... or an overall design flaw/oversight with Subaru ...?

 

[Danver]

A couple of months after I first got my Outback I noticed the time being off by several hours (minutes still appeared to be correct). Car was parked outside in the driveway in an open area. In my case the nav system had some sort of error reading the sd card (I can't remember what the actual error message said). I took the sd card out and it read fine on my computer, but when I put it back in I still had the same error. I took it out and put it back in a couple of times and it started working on its own. In my case I assume that kept the nav from detecting what time zone the car was in so it used a default setting. So I guess it makes sense that it might do the same thing when parked in a garage, but I'm surprised this hasn't been a common complaint. This seems like one of those things we Outback owners would love to complain endlessly about. :grin2: