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

GPS is off position

This site may earn commission on affiliate links.
Yes it’s off. Today it’s to the left of the road. I reached out to service. Too bad because the MCU2 upgrade has been great otherwise.

8C3F97DC-C494-4C3A-B45D-1CABF88E9B6F.jpeg
BE5D1DB2-E381-40CB-8AF0-D0BF9AB227A2.jpeg
 
That is indicative of a GPS antenna issue. Given the fact your MCU was just replaced, odds are that the cable was not tightly connected to the MCU or that it was crimped/damaged during the MCU replacement. If not that, the cable is crimped/damaged somewhere on the run to the antenna or the antenna (located at the top of the rear glass) is damaged. Those are much less likely. My bed is on the connection to MCU.

Take pictures/videos to document and get the service center to troubleshoot and fix it.
 
The FM radio antenna and the GPS antenna are both there...but they are under where the glass is bonded to the hatch. They are not accessible without removing the glass from the hatch (stupid design from a maintainability standpoint but that's another issue). None of that should have been touched in your MCU installation. Problem is likely at the MCU end of the cable. Think of it like failing to fully screw in an antenna cable. Some signal will get through, but it will be degraded. That's most likely why you have GPS position but errors.
 
I have had the same thing on my 2020 Model S: onplace most of the time but occasionally off by 50m or so.

I live in an area where the mapping is six years old - i fact I apparently have my home in bushland with no buildings anywhere near! The GPS gives me directions down non-existent roads and ignores onesthat I'm actually on, so the fact that it's occasionally off elsewhere doesn't bother me too much.

Using the Garmin on my previous car made me realise that the GPS position on that unit was actually NOT GPS at all - only "monitored" by the GPS. If I had a route in and decided to for example take an off ramp instead of keeping on the route, the GPS indicated position would continue along the planned route until the car's position was off by about 100m, and then it would jump to the actual position. That indicated to me that the mapping was what Garmin was using as a display, not the GPS fix - the car was always (normally) showing dead centre of the road yet the GPS fix could not be that accurate. I wonder if the Tesla GPS does the same thing?
 
@Wol747 You are describing something else. All GPSs show you the most recent position received from the satellites which (under normal conditions) is continuously updated. However when driving in places where the signal is lost (most common in tunnels and in cities with large buildings), you can go into a GPS dead zone. At that point there are three options:
  1. Show no position
  2. Show the last position
  3. Update the last position with your direction of travel and last speed until you get a new fix to update the position.
Most GPS units do number 3. What you are describing (an intermittent lost of position data and transition to Ded(uced) Reckoning is very common. What the OP is describing is a different problem where the GPS position is being updated, it is just off by a certain direction and distance.
 
What you are describing (an intermittent lost of position data and transition to Ded(uced) Reckoning is very common.

No, that's not what I am describing. (Relating to Garmin and Navman in my experience.)

Increase the map scale to maximum. The car icon is normally dead centre of the road, which is probably more accurate than the GPS is capable of. (I'm talking about the Garmin and Navman, but I can't see any fundamental difference in the Tesla.)

You are driving along a motorway with a route in the computer. If you decide to take an exit not programmed, as you exit and drive along the parallel exit lane the car symbol keeps to the motorway onscreen as if you have not exited, until you get several tens of metres away from the planned route, and then suddenly jumps to the correct position. Nothing to do with signal, that's OK all the time.

The issue with the symbol being well off to the side of the actual road is often something different - incorrect mapping usually. There are places round here that have the car apparently hundreds of metres in forest away from your position consistently - the mapping is wrong probably caused by the canopy obscuring the actual road when the human or computer does the mapping, and a guess is made.
 
@Wol747 You are describing something else. All GPSs show you the most recent position received from the satellites which (under normal conditions) is continuously updated. However when driving in places where the signal is lost (most common in tunnels and in cities with large buildings), you can go into a GPS dead zone. At that point there are three options:
  1. Show no position
  2. Show the last position
  3. Update the last position with your direction of travel and last speed until you get a new fix to update the position.
Most GPS units do number 3. What you are describing (an intermittent lost of position data and transition to Ded(uced) Reckoning is very common. What the OP is describing is a different problem where the GPS position is being updated, it is just off by a certain direction and distance.
Really? And now tell me that gps of tesla does that by car readings ans not gps? As this starts for me when I went into after market 20 inch wheels on my model s....
 
There are several different scenarios regarding GPS indication offsets - extended loss of signal is the more obvious example but the display offset in the case of your deliberate divergence from the programmed route (driving away from the "magenta line")is not the GPS per se but how the receiver program acts on the GPS position in relation to the programmed route. My impression over many similar "issues" is that the manufacturers want to keep the indicated position on the programmed route as much as possible and only jump it to the accurate GPS position when the offset exceeds a certain distance - which makes sense because any slight GPS anomaly would be distracting to the driver. I think I'm correct because when there's no route programmed in this doesn't happen in the same area.
 
There are several different scenarios regarding GPS indication offsets - extended loss of signal is the more obvious example but the display offset in the case of your deliberate divergence from the programmed route (driving away from the "magenta line")is not the GPS per se but how the receiver program acts on the GPS position in relation to the programmed route. My impression over many similar "issues" is that the manufacturers want to keep the indicated position on the programmed rout as much as possible and only jump it to the accurate GPS position when the offset exceeds a certain distance - which makes sense because any slight GPS anomaly would be distracting to the driver. I think I'm correct because when there's no route programmed in this doesn't happen in the same area.
For me it happends when I have navigation set on or not. It is alsp happening random, sometime is 100% accurate , sometime offset... So even if are my wheels.. thats so stupid... I mean does your smartphone knows what tires yoh are using... no, that mean this feature is so stupid, now two
option go back to ugly OEM wheels, or stay like this...
 
So I did some more testing today... I created a very small JavaScript app, which takes the geo location and display it on the browser. I opened it on my car and when my arrow was a bit off ( into the field) i took a photo, and then check the cordinates from this time, it was indeed there where it was on the Tesla Navigation, I checked it on google maps. So At this time I was indeed receiving a wrong coordinates.
The only thing that I have done since that started was 2 things. After market wheels, and new software 26.9.... Any ideas before I go back to tesla?
I also want to try it with my old wheels, but changing them requires time and money...
I mean I can agree that if the wheels somehow does not have the same dimensions they may interfeer with some internal calculation, but those should be only when there is no GPS. At the end we have GPS navigation... I'm really confused right now...