Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Caution - Annual Service - Car power-down - GPS issue

This site may earn commission on affiliate links.
I'm new to this forum but just wanted to drop a note of thanks to everyone. I ran into this issue when my Roadster went in for service. Even though I agreed to pay the fee to fix the GPS issue, the SC was unable to fix it and gave me the car back without a solution. After reading this thread, I created a makeshift harness with some test leads and was able to flash the GPS. Everything is working now. Thanks to everyone - this is a great community. By the way, in my journey to fix this, I ordered a GPS 18X LVC from Amazon and it happened to come with version 3.40 of the firmware already on it. I made sure to buy it as a "Ships and Sold by Amazon" to ensure I was getting recent stock and not from some random seller. It was $61. Might be useful info for others. Turned out I didn't need it for this issue but I can use it for other things.
 
A couple of months ago we noticed the GPS issue with the time & date. It doesn't affect the operation of the car so we ignored it.
Our Roadster is up in Reno for it's annual and they did a log dump after rebooting the VMS. Of course they found the date and time were off. The tech did some research and found that Tesla might provide a Garmin fix as part of a firmware update. Is that what's being mentioned by some others or is this new?
 
A couple of months ago we noticed the GPS issue with the time & date. It doesn't affect the operation of the car so we ignored it.
Our Roadster is up in Reno for it's annual and they did a log dump after rebooting the VMS. Of course they found the date and time were off. The tech did some research and found that Tesla might provide a Garmin fix as part of a firmware update. Is that what's being mentioned by some others or is this new?

sounds like the same thing, though others have said their SC's weren't as amenable to doing an update
 
A couple of months ago we noticed the GPS issue with the time & date. It doesn't affect the operation of the car so we ignored it.
Our Roadster is up in Reno for it's annual and they did a log dump after rebooting the VMS. Of course they found the date and time were off. The tech did some research and found that Tesla might provide a Garmin fix as part of a firmware update. Is that what's being mentioned by some others or is this new?

Wow. Free Garmin from Tesla? I just had @CM_007 replace mine. Not free, but not unreasonable either. I'll be curious to hear what Tesla says if they replace it for free. Are they saying the Garmin's they put in there were defective?
 
Wow. Free Garmin from Tesla? I just had @CM_007 replace mine. Not free, but not unreasonable either. I'll be curious to hear what Tesla says if they replace it for free. Are they saying the Garmin's they put in there were defective?

The tech did not mention any cost so my "assumption" would be there is a cost (especially on a ten or more year old car).
 
Replacing a GPS sensor that is functional but delivering the incorrect date and time is not necessary. Updating the firmware in the GPS sensor will fix the incorrect date and time. The cable I made and loaned serially to several owners facilitates that firmware update process. I mailed that cable to Los Angeles today for another owner.

My guess as to what the Reno SC was talking about is that the incorrect date could also be corrected by modifying the firmware in the Roadster VMS by adding 1024 weeks (7168 days) to the date provided by the sensor. The OVMS firmware faced a similar problem with an incorrect date coming from the GPS in the modem module (not used for Roadster, but on other cars that don't have their own GPS). The OVMS firmware was modified to add the offset. This is the first I've heard that Tesla might do such a modification to the VMS firmware.
 
What Steve said.

Bob, I recommend you get on the list for Steve's cable. The update only took a few minutes, and fixed the log problem going forward (logs already written retain the incorrect date). Having good date information in the logs could help with future troubleshooting, so fixing it is recommended.

Steve: Perhaps coincidentally, I also have not had any "possible flatbed" alerts from the OVMS module since doing the update last December.
 
Bob, I recommend you get on the list for Steve's cable. The update only took a few minutes, and fixed the log problem going forward (logs already written retain the incorrect date). Having good date information in the logs could help with future troubleshooting, so fixing it is recommended.

Steve - PM sent.

From the service invoice:

"Concern: The customer authorized diagnostic for date and time reset/malfunction
Confirmed the concern. Date and time are wrong due to GPS roll over in April 2019. This was a global impact and should be addressed through an update in the future. No time table for the update at this time.
"
 
Replacing a GPS sensor that is functional but delivering the incorrect date and time is not necessary. Updating the firmware in the GPS sensor will fix the incorrect date and time. The cable I made and loaned serially to several owners facilitates that firmware update process. I mailed that cable to Los Angeles today for another owner.

Are you willing to ship the adaptor to Sweden if I pay for the freight?
 
Are you willing to ship the adaptor to Sweden if I pay for the freight?
Yes, but the cable is currently with an owner in Los Angeles, then there are two others waiting in Northern California and Florida, so it might be a while.

Are you skilled with electronics and computers? You might want to try cobbling together a fix like I did before I made the cable. See my post with a picture and description of how I did that. I had a little 5V power supply to use. If you don't have one, you could use an old USB cable and cut off the end that doesn't plug into the computer. Two wires in that cable, probably red and black, would supply 5V power. @Dr No in Austria did this; see his post and the ones following it.
 
I'm also assuming that since it appears that the VMS isn't turned off during an annual service (even if APS is disabled?), the power to the GPS isn't dropped either. And with my GPS flipping out a day or two after the service, it's not a matter of power being the trigger; something else is going on...

I know it's kinda of old and you've move on but I just realized that my GPS reset on Feb 1st so I had to read this thread from beginning again!
Although I grabbed log regularly during winter, I don't always check them up because...winter means garage queen here.
So I started really looking at the logs again when I got the car out beginning of May.

I remembered this thread, started to read it back through, realized I missed my chance to see Henry and get this done when he borrowed the cable (he's my closest roadster neighbour...that I know) although, it looked like winter season so probably wouldn't have worked out.
I went on the hunt to figure out when the reset happen.

I thought this might be of interest to you @gregd ,I found this in the messages.0 log file:
Feb 1 19:47:30 kernel: ttyS1: 1 input overrun(s)
Jun 17 20:24:08 kernel: ttyS1: 1 input overrun(s)


Luckily Jun 17th hasn't passed yet in 2020 :) ...
So I'm pretty sure, the reset happened on Feb 1st between 19h47 & 20h24 since time is not impacted only the week.
Of course in February, the car was definitely in my garage, nowhere near service centers!
But I was working on a radio replacement project this winter and this was a Saturday evening...so I probably shorted power unwillingly for a split second and bang...GPS reset. No fuse were blown so I never noticed it at the time, so it must have been instantaneous.

Position, elevation, orientation & time in the VDS is good. I now have 137 day offset, luckily both 2000 & 2020 have Feb 29th.

So, if you get requests to ship the cable across your northern border, @slcasner , I will also be in need of borrowing it.
Right now, I've seen you already have a backlog and wouldn't recommend cross-border shipment anyway until border crossing returns more to normal.
FedEx lost some roadster part I ordered recently from US.

So once your backlog quiets down and shipping returns to 'more normal' (whatever that will be!), I'd appreciate if you put me on the list then. Thanks.
 
I know it's kinda of old and you've move on but I just realized that my GPS reset on Feb 1st so I had to read this thread from beginning again!
Although I grabbed log regularly during winter, I don't always check them up because...winter means garage queen here.
So I started really looking at the logs again when I got the car out beginning of May.

I remembered this thread, started to read it back through, realized I missed my chance to see Henry and get this done when he borrowed the cable (he's my closest roadster neighbour...that I know) although, it looked like winter season so probably wouldn't have worked out.
I went on the hunt to figure out when the reset happen.

I thought this might be of interest to you @gregd ,I found this in the messages.0 log file:
Feb 1 19:47:30 kernel: ttyS1: 1 input overrun(s)
Jun 17 20:24:08 kernel: ttyS1: 1 input overrun(s)


Luckily Jun 17th hasn't passed yet in 2020 :) ...
So I'm pretty sure, the reset happened on Feb 1st between 19h47 & 20h24 since time is not impacted only the week.
Of course in February, the car was definitely in my garage, nowhere near service centers!
But I was working on a radio replacement project this winter and this was a Saturday evening...so I probably shorted power unwillingly for a split second and bang...GPS reset. No fuse were blown so I never noticed it at the time, so it must have been instantaneous.

Position, elevation, orientation & time in the VDS is good. I now have 137 day offset, luckily both 2000 & 2020 have Feb 29th.

So, if you get requests to ship the cable across your northern border, @slcasner , I will also be in need of borrowing it.
Right now, I've seen you already have a backlog and wouldn't recommend cross-border shipment anyway until border crossing returns more to normal.
FedEx lost some roadster part I ordered recently from US.

So once your backlog quiets down and shipping returns to 'more normal' (whatever that will be!), I'd appreciate if you put me on the list then. Thanks.
My roadster among things is getting the "GPS Firmware Update" at @racevpr Gruber's...maybe you need that update too?
 
I remembered this thread, started to read it back through, realized I missed my chance to see Henry and get this done when he borrowed the cable (he's my closest roadster neighbour...that I know) although, it looked like winter season so probably wouldn't have worked out.
Actually, Henry's attempt at updating the firmware was not successful for reasons that are not fully understood. And that was with my first implementation of the cable; subsequent borrowers have succeeded with the second implementation. Henry said he may try again in warmer weather.
I found this in the messages.0 log file:
Feb 1 19:47:30 kernel: ttyS1: 1 input overrun(s)
Jun 17 20:24:08 kernel: ttyS1: 1 input overrun(s)


So I'm pretty sure, the reset happened on Feb 1st between 19h47 & 20h24 since time is not impacted only the week.
Of course in February, the car was definitely in my garage, nowhere near service centers!
But I was working on a radio replacement project this winter and this was a Saturday evening...so I probably shorted power unwillingly for a split second and bang...GPS reset.
Yes, that calculates out correctly. The loss of 1024 weeks would move the date back to June 17, 2000. I assume the times in the messages.0 file are in the UTC timezone, so that would have been in your afternoon rather than evening.
So, if you get requests to ship the cable across your northern border, @slcasner , I will also be in need of borrowing it.
Right now, I've seen you already have a backlog and wouldn't recommend cross-border shipment anyway until border crossing returns more to normal.
FedEx lost some roadster part I ordered recently from US.

So once your backlog quiets down and shipping returns to 'more normal' (whatever that will be!), I'd appreciate if you put me on the list then. Thanks.
Well, if Henry does decide to give it another try, perhaps you can meet up with him as you've suggested. Otherwise, sure, I'm OK with sending it, but you may need to remind me.
 
  • Like
Reactions: Roadrunner13
Yes, but the cable is currently with an owner in Los Angeles, then there are two others waiting in Northern California and Florida, so it might be a while.

Are you skilled with electronics and computers? You might want to try cobbling together a fix like I did before I made the cable. See my post with a picture and description of how I did that. I had a little 5V power supply to use. If you don't have one, you could use an old USB cable and cut off the end that doesn't plug into the computer. Two wires in that cable, probably red and black, would supply 5V power. @Dr No in Austria did this; see his post and the ones following it.

I am also in LA, but if I read this thread correctly, the early versions of the car do NOT have this GPS issue?
Any way to check?
If I may have the issue, I would otherwise love to try to take advantage of someone in LA having the update cable (do a quick meet & update).

Thank you.
 
I am also in LA, but if I read this thread correctly, the early versions of the car do NOT have this GPS issue?
Any way to check?
You can determine whether you have the GPS 18 or GPS 18x sensor by taking off the left front wheel and and inner fender to expose the sensor so you can read the label. (I haven't done this, so I can't describe it more accurately.) The firmware update applies only to the 18x. But that seems like a lot of work.
If I may have the issue, I would otherwise love to try to take advantage of someone in LA having the update cable (do a quick meet & update).
I sent the cable to Jason Fudenberg for his car, but suggested to him that he find out whether any members of TeslaClubLA also needed the update. If you are a member then you would know him. You can also determine whether the sensor is 18 or 18x using the update cable because the model number is sent as part of the data stream.
 
I went on the hunt to figure out when the reset happen.

I thought this might be of interest to you @gregd ,I found this in the messages.0 log file:
Feb 1 19:47:30 kernel: ttyS1: 1 input overrun(s)
Jun 17 20:24:08 kernel: ttyS1: 1 input overrun(s)


Luckily Jun 17th hasn't passed yet in 2020 :) ...
So I'm pretty sure, the reset happened on Feb 1st between 19h47 & 20h24 since time is not impacted only the week.
Of course in February, the car was definitely in my garage, nowhere near service centers!
Interesting! I do a monthly grab of the logs, for archival purposes. I opened the log from 31-Dec 2019, which is right after I did the firmware update, and scanned back through the logs. Nothing in there resembling your error, nor anything else for that matter, around the time of the GPS event. It's looking pretty normal (messing with SDA a lot, for some reason) on 13-Dec, then all of a sudden it's in May with a USB event (looks like my USB thumb drive, grabbing the logs), then back to 31-Dec, presumably after the GPS firmware update. So, nothing in there by way of a GPS error. My GPS reset occurred between 14 and 16-Dec 2019.

What is ttyS1, anyway? Looking back in the logs (May - August 2015, if the time stamp is correct), there are a bazillion of them recorded, along with troubles related to "setting terminal parameters"... I don't recall there being any sort of service events back then, and I wasn't driving the car much at the time. Odd... Looking through the logs is, perhaps, best left to others. Kind of like not asking the butcher what's in the sausage; some things are better left unknown.
 
  • Funny
Reactions: drewski
Interesting! I do a monthly grab of the logs, for archival purposes. I opened the log from 31-Dec 2019, which is right after I did the firmware update, and scanned back through the logs. Nothing in there resembling your error, nor anything else for that matter, around the time of the GPS event. It's looking pretty normal (messing with SDA a lot, for some reason) on 13-Dec, then all of a sudden it's in May with a USB event (looks like my USB thumb drive, grabbing the logs), then back to 31-Dec, presumably after the GPS firmware update. So, nothing in there by way of a GPS error. My GPS reset occurred between 14 and 16-Dec 2019.

What is ttyS1, anyway? Looking back in the logs (May - August 2015, if the time stamp is correct), there are a bazillion of them recorded, along with troubles related to "setting terminal parameters"... I don't recall there being any sort of service events back then, and I wasn't driving the car much at the time. Odd... Looking through the logs is, perhaps, best left to others. Kind of like not asking the butcher what's in the sausage; some things are better left unknown.

ttyS1 is the standard name for the second serial port's device node in Unix/Linux. ttyS0 would be the first serial port. "tty" is a historical designation meaning teletype/teletypewriter, "S0" means the zeroth Serial device. Anyway, it just means these data are arriving via a serial port somewhere.
 
ttyS1 is the standard name for the second serial port's device node in Unix/Linux. ttyS0 would be the first serial port. "tty" is a historical designation meaning teletype/teletypewriter, "S0" means the zeroth Serial device. Anyway, it just means these data are arriving via a serial port somewhere.
Right, I understand about Linux devices. The curiosity is what particularly is attached to that one. The GPS might be it, or it might be something else. Given my car's logs a few years ago with a bazillion complaints about it, and not being able to set the console parameters, I'm thinking it may not be the GPS but rather something else.