Lack of “Always Connected” on Model 3 causing noticeable vampire loss

[MikeBur]

Wanted to share observation that api/ logging tools on Model 3 disables it from sleeping. Some tools are better than others, though anyone polling the apis at a minute or less refresh rate may cause in the order of >5% loss per day.

I believe if/when “always connected” mode is enabled will help significantly based on S observations.

Just a heads up, this is what was causing constant trickle-charge top-up messages and considerable drain when disconnected from power source.

Another interesting charging / battery observation is I’m seeing 6-11 degree (F) cabin temperature rise when charging...

 

[jomo25]

Which tools are you using for logging?

 

[apacheguy]

Any luck getting your CAN logger on the power train bus?

 

[MikeBur]

TeslaFi, TeslaLog, Alexa, etc. taking it down to just TeslaFi with aggressive sleep (ie pause for 15mins) doesn’t help that much, ie appears car fully wakes up to any query.

Changing password to invalidate all tokens, reinforced confidence on this being the issue.

Still some vampire loss, and continues to top up 3-4 times during the day, though significantly less. There’s some coorelatikn between when the top up occurs and phone enters/exits range which is unsurprising.

 

[MikeBur]

Any luck getting your CAN logger on the power train bus?

Um, no... and this was Bill’s Logger, which he’s not updating. Have thought about digging deeper, though honestly wondering how much would be useful considering this is incomplete fw imo.

 

[llngoc]

I am seeing a 1 mile/hour drain which is significantly higher than the Model S and X I have. At this rate, it will probably render the car useless if I decide to park it at long term parking for a 2 week vacation.

Also, does anyone notice that when the charge cable is plugged in, it seems like the car is constantly trying to top itself off? I am getting "charged" notification from the Tesla app a few times a night about 20-30 minutes apart. This never happens with the Model X, once it is done charging, it does not seem to be re-charging anymore.

 

[omgwtfbyobbq]

Are the vampire losses on the 3 higher than the S/X without something polling it, or only when something's polling it? If the baseline is higher, I wonder if Tesla's polling the cars to keep an eye on how the batteries behave.

 

[MikeBur]

I am seeing a 1 mile/hour drain which is significantly higher than the Model S and X I have. At this rate, it will probably render the car useless if I decide to park it at long term parking for a 2 week vacation.

Also, does anyone notice that when the charge cable is plugged in, it seems like the car is constantly trying to top itself off? I am getting "charged" notification from the Tesla app a few times a night about 20-30 minutes apart. This never happens with the Model X, once it is done charging, it does not seem to be re-charging anymore.

I will almost guarantee you that you have a logging app connected. TeslaLog in particular appears to be very chatty.
Logging apps disable ability for the 3 to sleep, and responding to app or api requests appear to have to wake up car significantly.
The continual top off appears to occur because charging restarts when you lose approx 3 to 5% of battery SoC below charge limit set. I has this exact issue. It’s wors3 if you’re using 110v

Easy way to Check. Change your tesla account password and give it a day.
After that, introduce apps as you go. I’ve found I can run TeslaFi in sleep mode and this be significantly better.

 

[MikeBur]

Are the vampire losses on the 3 higher than the S/X without something polling it, or only when something's polling it? If the baseline is higher, I wonder if Tesla's polling the cars to keep an eye on how the batteries behave.

Vampire losses seem to be as good or lower, IF:
A) You don’t keep register key phone within range
B) you don’t have app/service that uses the api, eg teslalog, TeslaFi or connected app running, eg remote S, tesla app itself

I’m doing Some additional tests current on vampire loss, though it appears just us8ng TeslaFi and enabling car to sleep after 15 minutes of inactivity is best combo so far for logging, yet minimal vampire loss

 

[doubleohwhat]

A) You don’t keep register key phone within range

What's the real world range on this? I work from home and my office is just on the other side of a wall from the garage. I'm hoping I won't have to turn the garage into freakin faraday cage.

 

[llngoc]

I have learner to switch in the app to my Model X to avoid the constant mirror folding/unfolding every time I walk arond the house and get close to the car. Let me check and make sure all the Remote S app is not running on any of the iPads.

What's the real world range on this? I work from home and my office is just on the other side of a wall from the garage. I'm hoping I won't have to turn the garage into freakin faraday cage.

 

[MikeBur]

I have learner to switch in the app to my Model X to avoid the constant mirror folding/unfolding every time I walk arond the house and get close to the car. Let me check and make sure all the Remote S app is not running on any of the iPads.

Whilst debugging the vampire loss, I ended up putting cameras in the garage to see and hear what was going on. (You can’t see the contractors engaging though you darn well can hear them )

This is what I found, regarding BT connection and waking up:

There does appear to be a number of false positives within an 8’ radius of the car through one (garage) interior wall, and (wooden) garage doors

I suspect / hope there’s software smarts that locate the user with triangulation and this will get better.
This may be why there are 4 Bluetooth device listed when you add your phone as a key. There is another Bluetooth device for phone/media.

These 4 devices are pretty much always connected to our phones when with 20’ radius of car. Though this doesn’t unlock/unfold mirrors, hence the posit above.

I do not know if the 4 “BT location sensors” (my term, not sure if there’s another) are BT or BLE, though it may not matter as the energy used for either is insignificant compare to the rest of the systems. I have some concerns on the baseline energy use of the car even when it’s asleep just based on trickle charge top up frequency relative to S and X

Regarding other behaviors: the contractors on 3 are a lot more chatty, and whilst unsure if this is the DC>DC or trickle top up, I believe it is actually both.

These are findings on 2017.52 - I’m convinced this can and will improve over time, though Tesla really needs to get the always connected option (as per the manual) rolled out and hoping they can get the car into deeper sleep with less invasive wake up when interrupted.

 

[llngoc]

By "Contractors engaging", do you mean the clunking noise? I can hear it through the drywall in my office which is next to the garage. It seems to have improved a little since the 2017.50 FW but is still coming on once in a while. With the older FW, my wife and daughter were startled numerous times when they come home and heard the car started to make loud noises.

Whilst debugging the vampire loss, I ended up putting cameras in the garage to see and hear what was going on. (You can’t see the contractors engaging though you darn well can hear them )

This is what I found, regarding BT connection and waking up:

There does appear to be a number of false positives within an 8’ radius of the car through one (garage) interior wall, and (wooden) garage doors

I suspect / hope there’s software smarts that locate the user with triangulation and this will get better.
This may be why there are 4 Bluetooth device listed when you add your phone as a key. There is another Bluetooth device for phone/media.

These 4 devices are pretty much always connected to our phones when with 20’ radius of car. Though this doesn’t unlock/unfold mirrors, hence the posit above.

I do not know if the 4 “BT location sensors” (my term, not sure if there’s another) are BT or BLE, though it may not matter as the energy used for either is insignificant compare to the rest of the systems. I have some concerns on the baseline energy use of the car even when it’s asleep just based on trickle charge top up frequency relative to S and X

Regarding other behaviors: the contractors on 3 are a lot more chatty, and whilst unsure if this is the DC>DC or trickle top up, I believe it is actually both.

These are findings on 2017.52 - I’m convinced this can and will improve over time, though Tesla really needs to get the always connected option (as per the manual) rolled out and hoping they can get the car into deeper sleep with less invasive wake up when interrupted.

 

[MikeBur]

By "Contractors engaging", do you mean the clunking noise? I can hear it through the drywall in my office which is next to the garage. It seems to have improved a little since the 2017.50 FW but is still coming on once in a while. With the older FW, my wife and daughter were startled numerous times when they come home and heard the car started to make loud noises.

Yes, exactly. The clunk is when the HV battery pack is being connected/disconnected, like a relay, which requires the 12V low voltage to engage.

This is why if the 12V battery dies there is no ability to “start” the car: the contractors cannot be engaged.

It may seem ironic, though when the 12V starts running low the charging system has to ensure there is enough power to engage the contractors in order to use the 400V to 12V converter to recharge the 12V battery. If you have more 12V loads, eg dash cam, etc running then the 12V battery drains faster.

 

[llngoc]

It almost seems like it is a negative feedback loop here. Do you know if I can use a trickle charger with a cigarette adapter to charge the 12V battery?

Yes, exactly. The clunk is when the HV battery pack is being connected/disconnected, like a relay, which requires the 12V low voltage to engage.

This is why if the 12V battery dies there is no ability to “start” the car: the contractors cannot be engaged.

It may seem ironic, though when the 12V starts running low the charging system has to ensure there is enough power to engage the contractors in order to use the 400V to 12V converter to recharge the 12V battery. If you have more 12V loads, eg dash cam, etc running then the 12V battery drains faster.

 

[MikeBur]

It almost seems like it is a negative feedback loop here. Do you know if I can use a trickle charger with a cigarette adapter to charge the 12V battery?

Shudder. Please don’t do that. It might work, though nowhere near as effective as using UMC on that same 110 for car health...

 

[erthquake]

Is wifi still disabled in the Model 3 software? If so, those API/logging tools have to use the cell antenna, which probably requires much more power than connecting to a nearby wifi access point. Downloading software updates and uploading Autopilot data would also apply.

 

[MikeBur]

Is wifi still disabled in the Model 3 software? If so, those API/logging tools have to use the cell antenna, which probably requires much more power than connecting to a nearby wifi access point. Downloading software updates and uploading Autopilot data would also apply.

WiFi not enabled yet, correct.
LTE takes more energy than WiFi for sure, though even the most inefficient older 3G radio energy use would be inconsequential compared to waking the car up.
No one knows if autopilot data/shadowing is happening on the 3 yet, though one would hope they’d send when car awake, not wake car to send...

 

[llngoc]

Just an update. I quitted all the Remote S across all my iPhones and iPads and have switched the "active" car from the Model 3 ro the Model X on my Tesla App and then quitted the app. It looks like I am not getting much vampire drain anymore. My car was charged to 60% since 11am this morning and it is still sitting at 60%. Previously, this would have been at 57-58% already by this time.

I have learner to switch in the app to my Model X to avoid the constant mirror folding/unfolding every time I walk arond the house and get close to the car. Let me check and make sure all the Remote S app is not running on any of the iPads.

 

[MikeBur]

Just an update. I quitted all the Remote S across all my iPhones and iPads and have switched the "active" car from the Model 3 ro the Model X on my Tesla App and then quitted the app. It looks like I am not getting much vampire drain anymore. My car was charged to 60% since 11am this morning and it is still sitting at 60%. Previously, this would have been at 57-58% already by this time.

Yay! As I said, I believe it’s likely much easier to just change your main account password and then start bringing things back - particularly if you tie in to Alexa, logging tools, etc.

Still, another confirmation that we really really need always connected rolled out? I’m sure Tesla’s working on this.