what a shame, thanks for the in-depth analysis though @DogP . Need a job at Honda, they could use your expertise

 

Heaven knows, they've proven to have very little (if any) expertise at all.
.

 

Holy cow DogP, thank you for all the work you've put in. If you are ever in FL and wanna grab a beer send me a message!

 

they were aware for many years now...
the 1999 owners elsewhere on Acurazine brought it up when it hit them at the 15 to 16 yr mark, I joined in on the frey when it hit my '04 at the 16 yr mark, and the new flood of 16 yr old failures with the exact same problems hitting Acuras/Hondas are coming up on the thread... again.
Same probs, same time spans, same 'response' from Acura.
FFS if it takes a company several decades to 'look into it', they are only exposing themselves as failures at this point.

Dont expect any better treatment than the owners who were ignored years ago...

 

lol exactly.
Us '04 owners got the same answer about it being 'looked into' and they were 'aware of it' years ago. nothing was done about it
Now Acura is resorting to 'put up with it to such and such a date and it will fix itself'. AKA 'we arebt going to do aything about it'

And if it doesnt fix itself, Im sure there will be more years of the 'we are looking into it' response.

 

Except this IS NOT user wear and tear
Its NOT normal nor an acceptable post-warranty event.

This is a massively repeated known and recorded blatant failure on Acura's part, year after year after year. The same new delusional owners give the same "I just got off the phone with Acura and they promise they are looking into it" nonsense, not having their bubble burst yet.

I dont gafk HOW OLD something is, if the manufacturer provided proven defective products IT IS up to them to fix or replace the product no matter how old the thing is. There were recalls all over the place for non life threatening Acura product failures (windshield washer pump), sprinkled in with a few serious ones... came long after the warranty was up. What makes this one any diff.? Its a prob THEY CAUSED by providing inferior function that didnt reveal itself until a shockingly simillar 'end date', and its not due to anything of environment or user cause. Its a sourced chip or programmer or electronic cause which IS their responsibility for.
'
I dont gafk what fanboys of the forums say about how 'unreasonable' Im being in this regard. A FLAW is a flaw, THEY DO SO know about this as thousands of acura/honda inquiries and complaints have been phoned or written into dealerships over the decades, and we have the exact same 16 yr or so flaw hitting the cars in the last few months that the 1999 owners (who ALSO got the run around and were ignored) faced all those yrs ago. Hopefully they are still on the forum here somewhere and seeing that 'they finally came for me' syndrome has hit the same way again, and they are not alone or singled out...

 

""" All in all ppl on Acuzine have saved me 'dealership costs' amounting to probably an entire used Toyota Corroda for kick around in salty winters, like I used to have (This TL is immaculate, underbody et al, corrosion non existant, was winter store...."""

annnnnnnd mere days after posting this, a gdmn deer committed suicide all across the hood so neatly after sunset. Im inclined never to drive after dark again unless urgently needed.
Dashcam shows how instantaneous it all happened. I didnt stand a chance, ABS braking or not.

Looks like I will be using all that Acurazine help/saved cash, on parts and repairs. FFS life is cruel 'sometimes'.

 

I used to own a 05 and yes the clock annoy as hell, some member here message me right after the new year and I responded to them "welcome to the 04-06 navi clock club" and we has been deal with it since the plandemic.

 

Alternatively, you can just push some buttons and turn a key at 1:00 local time for the same result.

 

"fix"

 

It seems like this has become too widespread for Honda to ignore. We'll see if it actually gets fixed.

 

'IF' a fix becomes available? Thank you for being a loyal Acura customer?

Just how many of us so-called loyal customers do they really think are likely to stay loyal after this?

This kind of thing is something Lexus would swoop right down on and have a fix post-haste, or better yet, would properly design their systems to never have this problem in the first place!

 

I just did this on my 08 TL-S and worked fine, nice to have the correct time back. Guess this is a better "fix" than Acura is offering which is nothing.
Unplugging the battery twice a year is not too bad as long as you have navi and radio codes.

 

Every state needs to get rid of day light savings time

 

Heh, if I had a job at Honda, they wouldn't let me work on this problem anyway, because their current plan is to wait until August when it sorta fixes itself, which is 100% free and takes no effort. :-P Though if they released a 2022 map update with the GPS fix, they'd probably sell a record number of those discs! Right now, I doubt they sell many $150 disc updates each year for ~15 year old cars, when everyone has maps on their phones.

Update on my fix attempts though - I did find my notes on extracting files and rebuilding the navi disc from the last time I messed with this, and it's not as tedious as I remember. So, I extracted the files and dug through some disassemblies of the binaries this afternoon, and have a few ideas on solving this (and the nag screen). Haven't gotten a chance to actually test anything out yet... maybe this week?

Oh, something interesting... I completely disconnected the GPS receiver after the last testing, since I didn't want it loose and banging around the trunk. I've been driving the past few days without it (~20 miles each way to/from work), and the navigation unit still knows where I'm at! So, I guess between the yaw and car speed sensors (and probably assuming that I'm always driving on a road), it really only needs the GPS receiver for coarse positioning.

DogP

 

Not sure if anyone's following along, but I might as well dump some more info here just in case it helps someone in the future.

I figured it'd be easier to find the code where it's pulling the data if I knew more about the Furuno format... so I briefly went back to that last night. The data that I was receiving looked a bit strange, so I checked it by hand on an oscilloscope and noticed that it actually uses odd parity (i.e. 9600 8O1). Once I saw that, the data looked more recognizable, and it turns out that they basically ripped off the Trimble TSIP format, but used their own packets (maybe Furuno requires an NDA so Trimble doesn't find out? ).

So, every message starts with 0x10 <Packet ID>, then the data, and ends with 0x10 0x03. Since 0x10 is used as a special character... if any data has 0x10 in it, it is repeated (e.g. 0x10 data outputs as 0x10 0x10). Inside most of the packets there looks to be a check byte, which is an XOR of every byte between the preamble and postamble, not including any repeated 0x10 bytes.

I see at least one place in the code that seems to be parsing this data, and the data is fairly straightforward. For example, in the packets I received hex bytes: 02 06 05 08 42 11, which is clearly BCD corresponding to 08:42:11 (UTC) on 06/05/2002 (1024 weeks ago).

Also, there's a Log VP Data function where it logs a bunch of stuff a PC card... I popped a card in the unit (used a PCMCIA to SD adapter) and let it log for a while. It does look like the raw GPS packets get dumped in there, though I haven't dug through it in detail.

I should have some code to test soon... I was hoping I had a CD-RW laying around so I could iterate without burning through a whole spindle of CDs, but I didn't find one. Maybe I'll dig one up (or DVD-RW) somewhere tomorrow.

Oh, and without the GPS receiver, my navigation finally went off in the weeds a couple days ago. Maybe it was the snow/ice causing the speed sensor to be wrong, or maybe the errors just finally accumulated enough to jump to the wrong road... but after coming home it thought I was a couple miles away.

DogP

 

I think most people now use their phones for maps so it would probably be best to install an aftermarket head unit that can connect to their phone. Also, the latest head units probably have better resolution to use with a backup camera.

 

I don't know how it sort of knew how you were to start with after you disconnected the GPS units - it's not like the car has an Inertial navigation system built into it - need gyros for that, eh? It does have the accelerometers needed though. Also saw a TV show last week related to this - "Trafficked with Mariana van Zeller" did an episode where they tracked how cars are stolen in the US and delivered to overseas locations, West Africa in this case (they break into cars so fast it's not funny, 5-10 seconds - they jiggle the door handle and do something they wouldn't show on TV***) - she traced everyone involved from the car thief to the guy who sells them to end buyers. One of the first thing the thieves did was rip out the OnStar transponders so the cops couldn't track them - TTBOMK OnStar is disabled until/unless you hit the button inside the car to use it - if it was constantly sending out a signal it would kill the battery of a car that sat for a while. Older OnStar systems broadcast to GPS satellites - that takes a lot of power - newer ones can use cellular towers, but like the old Bluetooth problem 2000 era Acuras had I'd think OnStar pinging cell towers constantly would also drain the battery over time if the car's not being driven. So were these thieves just being overcautious, or can you really track a stolen car with OnStar that hasn't had it activated by the driver?


*** As Acura owners know, it's impossible to jimmy the locks, they're protected by a metal sleeve that blocks the coat hanger or jimmy tool - how the hell these guys were breaking into BMW's and Mercedes I have no clue - didn't set off any alarms either. Once inside they program a blank key fob by plugging into the car's diagnostic port under the dash. Another method is to use RF amplifiers to intercept the signal constantly being exchanged between the key fob and the car, record it, and program it onto a blank key fob - but that only works if the fob and car are close enough to communicate - they suggested if you park your car on the street right outside your house or apt., or in your driveway, you store the fob in a poor man's Faraday cage while it's inside your house to prevent this kind of theft. The thieves also change the dash VIN plate in some manner that's apparently undetectable - thought that was damn near impossible?

 

It doesn't have an "Inertial Navigation System" in the strict sense, but it does receive a speed pulse from the car, and has a "Yaw Sensor", which I assume is a gyro (though I guess it could also be a magnetometer). So... it knows roughly which direction you're turning, and how fast you're going, and probably assumes that you're on a road (correcting absolute position when you turn for example), then as long as it doesn't accumulate enough error between turns, it's pretty easy to see how it could know position without GPS.

But if for example I'm driving through a neighborhood with a lot of side streets, and turn on street B, but it thinks I turned on street A... then it'd correct my position to the wrong road. Then errors from there get exaggerated if roads aren't completely parallel, since again it's thinking I'm staying on that road, when I'm actually going a slightly different direction on a different road.

DogP

 

I ended up getting slightly sidetracked last night. Since I already had the GPS receiver accessible, and figured out enough about the message format... it looked really easy to just throw in an MCU to intercept/modify the data. So I gave it a try, and it works!

Of course the goal is to apply a fix to the native navigation unit software, but since I can confirm that this works, I'm confident that it'll work if I can get the navigation unit to simply modify that data when it arrives.

Not that I expect anyone else to do this (nor do I even recommend it), but I'll throw the code up on Github in case anyone's interested in checking it out. It's running on a PIC16 MCU... the code just BARELY fit (including the <time.h> functions ate up ~90% of the flash, but it made converting the date only a few lines of code! :-) ). It uses a signed 32-bit time_t though, so you're on your own in 2038. ;-) These MCUs do come in really small packages, so it'd be easy to make a small PCB that sits between the GPS receiver and navigation unit motherboard... but that's just a backup plan if I can't figure out how to do it in the navigation unit software.

Few pics below for those interested.

DogP

 

^^^^ Now you just need to share with Honda

 

Boss move @DogP

 

Holy crap, the mad scientist is making it happen ..

it’s giving me PTSD from my previous CompEng classes in college that I eventually flunked out of but still

 

Seems there will be a fix in August

 

so sounds like they're not going to do anything?

i guess i wont be considering a honda/acura for my next car then!

 

Hi @DogP , do you have a link to github project?

 

I just uploaded it this evening: https://github.com/pdaderko/PIC-Proj...vi_Clock_Fix.X .

DogP

 

Thanks @DogP !
I can see that GPS module shield's 4 legs are soldered into the main PCB. What about the 10-pin connector? Is there a connector that unplugs, or are all 10 pins from the GPS module soldered into the main board? How did you desoldered the GPS module?

 

Unfortunately, the 10-pin connector is also soldered to the PCB. I used a vacuum desoldering gun, though you could also desolder it with a soldering iron and soldapullt.

DogP

 

Nothing really to share here, I just wanted to commend DogP on the information he's shared and his obvious vast knowledge...don't schrug it off - very impressive sir.

 

That's all we can do for the time being. Set mine today at exactly 1:00. Refreshing to see the real time on the dashboard again.
The calendar thing never bothered me as much as always having to look at the wrong time, especially when my 1st job out of college was working for the company that made the very first electronic digital clock for the consumer market, circa 1977
.

 

Originally Posted by jeffstlnote View Post
Perhaps a "temporary" fix if anyone wants to try this at 1pm/am without doing all the battery or fuse stuff...

Yes - that dash button reset method does not work with an RL.

 

Sorry to hear that! Surely there is a way to do an RL, how though I don't know.

 

I'm not quite sure disconnecting the battery at 1:00 pm will work to reset the clock on an RL either - I don't recall ever having to reset the clock after changing a battery, and doesn't it get the time from the GPS NAV system anyway?

 

https://acurazine.com/forums/third-g.../#post16791159

So based off the latest bulletin there is no fix from Honda/Acura other than to disconnect the battery or wait until August?

 

As expected.. they'll just sit on their hands and tell us to "wait until August". what a joke

 

Also is it only an 07-08 TL issue? What about 04-06 TLs with Navi?

 

Thanks for the link to that service bulletin Mizouse - odd that you have to disconnect both battery terminals at exactly 12:40 pm and reconnect at 1:00 pm to get this procedure to work. I have a sticky note in my car manual labeled "bad things that happen when you disconnect the battery", will have to look thru it to see if it's worth having to change all those settings back to where I want them just to correct the clock. Also saw this is the SB - DST automatic adjustment will be broken forever?

What can we do to avoid the time changing to an incorrect value again after August 17, 2022?
The automatic DST function will not work properly causing the incorrect time at the perceived DST date. We recommend
that after August 17, 2022, the client sets the “Auto Daylight Saving” function to “OFF” and adjust the DST manually.

 

The '04/5/6 TL's with Nav have been suffering this malady since January of last year. I can only assume come August, they'll correct themselves too! I have no faith Honda Motor Company is going to fix this anytime soon.
.

 

Yep, I have been pulling the fuse for the clock for over a year now. Considering my clock went out a year before the 07/08 models I have little faith mine is going to fix itself in August.......

I was actually hoping mine would fix itself in January of 2022 (a year later) but instead everyone else's broke lol