GPS Position Stuck??!

[ccdisce]

You did not say which firmware you are on. I am on 2017.38 (see my earlier post) which was installed on 9/28 (same day I got the car). I saw my GPS issue while on 38. Not sure you noticed but someone else reported that the reboots do not work but allowing the car to SLEEP will reset the GPS issue. It did for me while reboots did not. Good luck on your trip.

Oops sorry about not mentioning the firmware rev.. my bad. also the car is a New Inventory with supposedly 70mi which actually came with 114mi on the odo.

The FW Rev that is currently installed is 8.1 2017.38 I do not know what the FW Rev is when I took delivery at 4pm on the 28th Sep.

Before I took off for home after taking delivery I topped off the charge to 90% at a Supercharger at the Service Center that had a clear shot at the satellites.
I also set it to navigate from from the mothership to my home in Stone Mountain GA ( showing in excess of 2000mi ) whereupon it said OK we have to get to the smelly Harris Ranch to charge.. sorry I do not think that we are near Harris Ranch plus I do not have Bio-weapons deployed. It did seem to lock on to some satellites somewhere as it said ''turn here at xxx' ...'make a U turn here (never mind the concrete barrier or the guard rails)
The problem I think is that it cannot find the correct map to slip under the pointer layer or it did find itself at the Service Center in Decatur GA and cannot believe that it has been punted 2000mi from the mothership while it slept on the transporter.

I do not think that the lack of sleep is the issue here , but could be a part of the 'anti-theft process' waiting on the Telematics from the mothership to tell it that its 'OK we have been paid and you have been shipped to GA so go and learn the roads of GA remember to upload the data on command to us'

Shipped and Sold add it to the 3rd qtr tally for Wall St.

 

[ccdisce]

This has happened to me twice now. The first time was last Tuesday after visiting the service center. I took my car back to work and when I left a few hours later the GPS was working again.

The second time was today. I rebooted the MCU using the scroll wheels and that fixed it for about 8 miles and the GPS hung again. I haven't tried rebooting the MCU again.

I'm on 17.36.1b27c6d and have AP 2.5 hardware.

I am on 17.38 AP2.0 car built JUL2017 you may want to check with your SC to see if .38 will fix your problem in an AP2.5 system

I wonder if the GPS signal in your area is:
1. Weak....failing satellites...where is Rocket Man we may need him to launch a few sats You may want to check for satellites with a free app from Google Play it gives an accuracy reading of the incoming
2. Some nefarious person jamming the GPS signal possibly near cell towers for reasons known only to them for us to discover
3. Military testing out their shifting ability of the GPS signals to throw off GPS munitions.

The GPS antenna in the Tesla cars is supposedly in the hatch glass is it bad or weak other(s) have reported so, defective GPS connector on the back of the MCU as low lead RoHS soldering can be a pain sometimes

Yes I am paranoid and I was paid well for being so for the past 30 years, running down all the rat holes to product engineer electronic packages, and when I did not find the root cause I would have to do a thought experiment and then dig a new rat hole to get to the treasure as the old rat hole could have been covered over by the departing party.

Good Luck.

 

[rnortman]

I am on 17.38 AP2.0 car built JUL2017 you may want to check with your SC to see if .38 will fix your problem in an AP2.5 system

2017.38 firmware does not have the GPS fix. I've been bitten by it once since upgrading to 2017.38. Keep waiting...

As others have said, the key is to induce a sleep state. Turn on Energy Saver, turn off Always Connected, and then select Power Off from the controls menu. Wait a couple of minutes (at least) and then power up. That fixed it for me last time. (No guarantees that it will work again...)

 

[ccdisce]

I believe that there is still other problems with the GPS as when I was driving down a straight road yesterday, between the lane markings which were clearly marked, the lane markings on the dashboard center portion where the car is displayed was sliding around left to right over the lane markings and running parallel for a 100 ft say then swinging back over to the other side, I checked through the wind screen and the car was going straight down the middle of the lane.

Probably because the GPS is loosing lock on a few satellites and the placement accuracy went down the tubes, simple interpolation may not be enough and more sophisticated techniques are required in such noisy readings.

Sorta reminds me of driving on a frozen lake in Canada years ago.

 

[rnortman]

So I just had an email exchange with my local service manager about the thing that happened to me a couple of days ago where both the GPS and the cruise were not working. I had assumed that the cruise would not engage because the GPS was telling it it wasn't on a road. It seems to me that there is at least a link between GPS and TACC in that TACC knows the speed limits, which in an AP2 I believe still comes from the navigation database rather than reading signs.

But the service manager said: "The Cruise Control and GPS are not functionally related."

I find that questionable. Anybody here care to comment? Also, anybody following this thread: If your GPS gets stuck, take note of whether cruise control is working or not? Does it know the speed limit? If the GPS is stuck in a location that's off-road (in my case, it was my driveway/garage), does it even allow you to engage cruise? Thanks for adding whatever information you can...

 

[d21mike]

So I just had an email exchange with my local service manager about the thing that happened to me a couple of days ago where both the GPS and the cruise were not working. I had assumed that the cruise would not engage because the GPS was telling it it wasn't on a road. It seems to me that there is at least a link between GPS and TACC in that TACC knows the speed limits, which in an AP2 I believe still comes from the navigation database rather than reading signs.

But the service manager said: "The Cruise Control and GPS are not functionally related."

I find that questionable. Anybody here care to comment? Also, anybody following this thread: If your GPS gets stuck, take note of whether cruise control is working or not? Does it know the speed limit? If the GPS is stuck in a location that's off-road (in my case, it was my driveway/garage), does it even allow you to engage cruise? Thanks for adding whatever information you can...

When it happened to me the speed limit for the road I was on was wrong. My GPS Lock was at my Mother-in-laws house which had a much lower speed limit. So when I put on TACC the car slowed rapidly to 5 miles over the speed limit. However, I am not sure if I can override that on non-freeway driving. I may try that today while testing the car. The default is 5 miles over the speed limit when you invoke TACC.

 

[rnortman]

When it happened to me the speed limit for the road I was on was wrong. My GPS Lock was at my Mother-in-laws house which had a much lower speed limit. So when I put on TACC the car slowed rapidly to 5 miles over the speed limit. However, I am not sure if I can override that on non-freeway driving. I may try that today while testing the car. The default is 5 miles over the speed limit when you invoke TACC.

Thanks for the confirmation that I am not crazy. (At least not because of this...) It is really troubling to me that a Tesla service manager can make the claim that "Cruise Control and GPS are not functionally related" (that's a direct quote). He apparently has not realized yet that he's basically tech support for an enormously complex software system where everything is functionally related to everything else. To somebody who's spent a career building and troubleshooting complex systems, having somebody make a blanket claim that "subsystem X is functionally unrelated to subsystem Y" sends off glaring alarm bells in my head, and makes this guy sound dangerously naive. (Plus even on the surface of it they are clearly related via the speed limit database -- you don't even need to search for hidden connections here, it's right there in front of you.) People who think like that aren't qualified to troubleshoot systems like this.

Sorry, I'm probably reacting a bit more emotionally than warranted, since the guy talked to me like I'm a child.

 

[ishareit]

Same thing happened to me. GPS stuck. Tesla techs couldn’t fix over the phone. Screen reboot and power off and power on the car didn’t work either. Earliest appointment is on Tuesday at my local SC. Can’t use AP over 30mph because the speed limit is stuck at 25mph.

The car is 8 days old now and it’s already been at the service center overnight because of truck hatch closing problem.

Sad

 

[ishareit]

2017.38 firmware does not have the GPS fix. I've been bitten by it once since upgrading to 2017.38. Keep waiting...

As others have said, the key is to induce a sleep state. Turn on Energy Saver, turn off Always Connected, and then select Power Off from the controls menu. Wait a couple of minutes (at least) and then power up. That fixed it for me last time. (No guarantees that it will work again...)

Nope. Didn’t work for me.

 

[rnortman]

Nope. Didn’t work for me.

That's disappointing but not surprising... have you tried doing that and then leaving it off overnight? No loading the app to check on your car either... that will wake it up. Shhh... don't wake the 4500-lb baby!

 

[ccdisce]

I am disappointed and also surprised as it seems as if the trend is to release new hardware and software to the paying public and let them deal with it ' let the software 'learn' its environment under the AI guidelines that we as the Borg have imbued it with'.

If there is a problem that it can't handle let it bail out and go scurrying back to the Service Center for help where it and the owner may or may not get a less than stellar reception.

While I wait for a fix, your post of << Shhh... don't wake the 4500-lb baby!>> brought this image to mind.

I may have to change the name of my car to Baby Huey while I wait for a fix.

 

[James*]

@d21mike @rnortman TACC is not dependent on GPS, except in determining the offset speed from the speed limit. For example, my offset is +8 mph. Even if I'm driving on a residential road, if I just enable TACC, it will go +8 mph over the speed limit. The case where you get limited to +5 mph over is when you engage Autosteer. Autosteer's speed DOES depend on GPS location. If your car's GPS is stuck on a residential road, you will be limited to +5 mph to whatever speed limit was on that residential road, even if you are on a highway when using Autosteer. You can still use TACC to whatever speed limit you want, you just have to manually set the speed limit.

 

[d21mike]

@d21mike @rnortman TACC is not dependent on GPS, except in determining the offset speed from the speed limit. For example, my offset is +8 mph. Even if I'm driving on a residential road, if I just enable TACC, it will go +8 mph over the speed limit. The case where you get limited to +5 mph over is when you engage Autosteer. Autosteer's speed DOES depend on GPS location. If your car's GPS is stuck on a residential road, you will be limited to +5 mph to whatever speed limit was on that residential road, even if you are on a highway when using Autosteer. You can still use TACC to whatever speed limit you want, you just have to manually set the speed limit.

I agree. I was only testing with AUTOSTEER which includes TACC. When I tested with TACC later I can override the speed as you say.

 

[rnortman]

@d21mike @rnortman TACC is not dependent on GPS, except in determining the offset speed from the speed limit.

So first of all, the "except" above basically proves my point... they're not related, except in this way that they're related... so, they're related. That's the tldr version. If you really want to understand the point I tried to make, read on...

As for TACC being "dependent" on GPS, are you sure it's not? I'm not sure. I offer anecdotal incident in which GPS and TACC on my car stopped working at the same time, and both then started working again at the same time. Support could find absolutely no TACC-related errors in the logs; TACC was operating "normally" and simply refusing to activate. And completely coincidentally the GPS was stuck and reporting that I was still in my garage; i.e., not even on a road? Clearly a coincidence, right?

There is a clear coupling between GPS and TACC, as you pointed out: TACC uses speed limits. It does not enforce speed limits, but it uses them. GPS is an input to TACC. The service manager told me, and I quote, "The Cruise Control and GPS are not functionally related." Given that GPS location is in some way an input to TACC, that is clearly false, even if TACC does not "depend on" GPS, they are clearly "functionally related".

If you have worked much with software systems, you know that even the slightest, weakest connection between two pieces of software introduces the possibility that when one of them does something strange, the other will do something strange. Even if it's not supposed to. So TACC may not need the GPS signal, but they may not have tested operating it without the GPS, or with the GPS "stuck" in this way that just started happening in HW2.5 cars. Maybe a -1 is coming in for the speed limit field, and nobody ever tested that, and it breaks the TACC code. This is how these bugs happen. (Or maybe they did test it, and the TACC is doing exactly what it's supposed to, which is refusing to activate when given invalid input... that's how I'd write it.)

On top of this, it strains credulity that both GPS and TACC failing at the same time and resolving at the same time would be entirely coincidental. There is clearly a link. I'm pretty sure I've had instances where TACC would not engage based on GPS location. I think it really wants to be on a road, and it may care what road type designation the current road has on the map. (I am not very sure about the latter, but it seems entirely possible. It also seems entirely possible that the service manager would be clueless about this.)

 

[DriveMe]

Had a stuck GPS this afternoon. Did a two-button reboot while driving. It didn’t resolve the issue, though there was a brief message, saying that the Navigation system needed service and something about maps not loading. Streaming music worked fine, so it wasn’t the Internet issue.
We then stopped for dinner. When I started the car afterwards, the GPS was still stuck, bun then recovered a minute later (and switched to the night mode).

Judging by all the reports here, I tend to think it’s a software bug. Hopefully it will be fixed soon.

Brand new 75D with AP2.5 and 2017.38.

 

[rnortman]

Judging by all the reports here, I tend to think it’s a software bug. Hopefully it will be fixed soon.

Tesla has said that it's a software bug (or at least a software/firmware fix), but I'm curious what about the reports here confirms to you that this is software? Do you just me it's not something wrong with your car specifically? That is probably true... but that alone doesn't mean it's not a hardware problem. It could still be a hardware problem that affects all recent builds (design flaw) or perhaps many/most recent builds (manufacturing/QC glitch).

 

[marksvend]

I just installed update 2017.40.1 last night, and what happened when I drove out of my garage this morning? The GPS was stuck in my garage. I had to park to shut down the car for 3 minutes so I could use Autopilot on my commute.

My Model S has already been in for a service appointment for this issue, which resulting in the resetting/update of some firmware.

 

[Darmie]

I just installed update 2017.40.1 last night, and what happened when I drove out of my garage this morning? The GPS was stuck in my garage. I had to park to shut down the car for 3 minutes so I could use Autopilot on my commute.

My Model S has already been in for a service appointment for this issue, which resulting in the resetting/update of some firmware.

Darn, I was hoping they had this fixed.

 

[Darmie]

shifting to park, turning on Energy Saver and turning off "Always Connected", and then telling the car to power off (these settings are all available in the Controls menu). I waited about 2 minutes and then pressed the brake to turn the car back on.

I know a little off topic. I attempted this myself on one freeze with no help. Maybe I did not leave it off long enough, which by the way fixed it self the next day. Anyway, if you do power down your car, can you get out and lock it and are you able to re enter with it powered down? I was concerned not to know the answer the wrong way by getting out of the car when It was powered down.

 

[yargok]

I've had this happen to me twice over the last week on the .38 firmware. There is a "bandaid" fix to this problem if you call Tesla. 1/2 reps knew the answer right away, for the second rep I had to push him to try the fix. Both times I got my problem resolved. There is now a bulletin about the problem (known bug) and will be fixed in a future firmware.

You need to ask the Tesla Rep to "disable and reenable driver assistance features". To do this, the car has to be in park and completely powered off. Sounds like a couple of more weeks before we put this one to bed.