Tesla Model 3 Down: Won't Power Up, and is Inaccessible

[mongo]

So no. There are no excuses. Vehicle MUST, I emphasize, MUST consume less than 100mA when asleep.

Why must an EV align itself with ICE standards? And which standard? Many ICE cars also have an ignition off draw (IOD) fuse that gets pulled when the car is shipped to reduce draw even more. How low at what cost for what benefit?

One mile of range is 250Wh, at 12V that is 20Ah or so, or a draw of 835 mA. Is losing a mile a day a big problem?

App functionality has nothing to do with parasitic draw.

Sure it does. For the app functionality to work requires an active two way communication cell or Wi-Fi along with a network processor.

Vehicles with limited power resources use low power receivers or polled/ triggered active systems such as unlock on door button press or handle pull. To active instantaneous response via app requires an already existing connection. To do wake on approach requires, at a minimum, a fast poll rate.
Sure there are ways to do it using secondary application specific processors, but that takes extra development time and introduces complexity. Tesla used a system that provided high confidence of working instead of optimizing power draw. Seems reasonable to me.

 

[ℬête Noire]

PS:
I don't know how but latest BMW's have nominal sleep draw 7-21mA. Can be found in that PDF

That must be effectively a complete shutdown of all computing systems? How long does it take to come out of that "sleep" mode?

 

[mongo]

That must be effectively a complete shutdown of all computing systems? How long does it take to come out of that "sleep" mode?

Automotive processors can go to a sleep/ low power mode and then get woken up by network traffic or a wired signal. Some things are on switched ignition so zero draw when off. Both styles do not require a long initialization sequence to start operating. However, things like infotainment systems with a full operating system can take longer to boot up again.

Tesla is more like a computer, instead of using sleep mode, it keeps running in active mode which provides instant response and avoids the issues than can come from entering/exiting sleep mode.

 

[arnis]

That must be effectively a complete shutdown of all computing systems? How long does it take to come out of that "sleep" mode?

Waking starts as soon as door has been opened/vehicle unlocked. And if you push the START button, everything works immediately (with an exception of navigation map). This even includes glow plugs for diesel engines. I researched how is that possible. Why, in frigid cold, owner approaches the vehicle, opens the door and IMMEDIATELY pushes START button, vehicle will crank immediately. Glow plugs take few seconds and in really cold weather up to 5 seconds. Well... glow plugs are activated way before driver pushes start button. No magic. BMW keeps startup time top-notch

They disable modules with bi-stable relays (Except alarm (if equipped) and vehicle access module). This means most modules are not energized when vehicle is asleep, 0mA. Vehicle access module wakes everything up as soon as key is near the vehicle (2-5 steps) by zapping those relays with ON command.
My friend left the vehicle for 3 months in winter. When he arrived, vehicle did not wake up with remote (antenna module was de-energized). Handle had to be pulled. Discharged battery warning was displayed. Clock/date was reset, GPS data was lost (GPS lock took few minutes), Vehicle recommended a longer drive or manual charge.
In summer, the sleep period could be 4-6 months, but lead acid battery capacity drops in cold. 105Ah battery for heavily equipped 5-series.
Battery SOC is kept at 80% when vehicle is being driven. This is required for regen (3kW).
This happened in 2017. Vehicle was from 2010. Factory AGM battery.

 

[nvx1977]

Battery TMS is not cooling the pack when vehicle has been left alone. This includes Tesla. BMS is checking for battery temperature and might switch on heating if temperature is below specified limit (like -15*C). But there is no cooling. Cooling requires AC compressor. Which requires contactors.

That is really good to know. It means that we owners should actively think about parking the car in cooler areas (shade, parking garage) during the summer. Heat destroys Li-ion batteries, particularly ones holding high charge.

 

[arnis]

It doesn't "destroy". It speeds up degradation. Leaf chemistry speeds up significantly at around 30C. Tesla way higher. Likely 40C, or maybe even 45C. If SOC is below 40-50% degradation is slower even at hot temperatures.
No need to worry with Tesla. Chemistry is pretty stable.

So in case of Tesla, good owner in hot climate might lose 10% in 9 years, bad owner loses 10% in 7 years. Doesn't really matter.

 

[nvx1977]

as you say, it's a curve of degradation. I'm anal enough to want to stay as low on that temp+charge curve as possible. I fully appreciate the Tesla battery is pretty darn resilient, another reason I hesitate to choose another EV brand. I'm still going to lower my charge limit in the summer and find shade wherever possible. Hypermiling the battery

 

[mongo]

That is really good to know. It means that we owners should actively think about parking the car in cooler areas (shade, parking garage) during the summer. Heat destroys Li-ion batteries, particularly ones holding high charge.

The manual warns against exposing the pack to temps above 140F/ 60C for more than 24hrs. Where degradation really kicks in is another question.
It will cool the pack when plugged in for charging, possibly when plugged in and not charging, (anecdotal reports) , it will cool the pack once you start driving if it was hot, and could when unplugged and really hot.

 

[nvx1977]

I have to park in an open parking lot for work. In the summer, the car is just gonna bake for 9 hours straight. While Tesla's chemistry is more resilient in heat, and these conditions are nowhere close to what's stated in the manual, degradation will still occur. The heat makes the undesired chemical reaction (lithium plating on the anode) more favorable. It's not all or nothing; it's a smooth, exponential curve, which at the extreme temp end leads to thermal runaway. I intend to keep my battery on the flat part of the curve as much as possible.

I have no idea if that means my battery lasts 11 years instead of 10, or if it lasts 10.1 years instead of 10. I just like to keep the battery in as optimal conditions as possible when it's under my control.

Sorry, getting off-topic. I'll shut up about this now

 

[JRP3]

Try element3batteries.com for specs.

element3batteries
Ouch, $439 for 12V x 30ah= $1219.44 per kWh! Someone really needs to start mass producing 4 LiFePO4 cells and some circuitry for a reasonable price.

 

[mongo]

element3batteries
Ouch, $439 for 12V x 30ah= $1219.44 per kWh! Someone really needs to start mass producing 4 LiFePO4 cells and some circuitry for a reasonable price.

A kWh pack would not have 3x the housing/BMS/R&D/shipping/ packaging costs...

 

[JRP3]

Right, because the plastic housing and circuit boards are where the markup is At half the price it would still be overpriced but at least somewhat reasonable.

 

[Saghost]

Battery TMS is not cooling the pack when vehicle has been left alone. This includes Tesla. BMS is checking for battery temperature and might switch on heating if temperature is below specified limit (like -15*C). But there is no cooling. Cooling requires AC compressor. Which requires contactors.

Do you have data or documentation on a Tesla that demonstrates this definitively?

Tesla has always been somewhat flexible in their definition of "on" - I don't believe that it always opens the contactors when you get out of the car, and I do believe that it will do active cooling of the pack while parked if required.

 

[mongo]

Do you have data or documentation on a Tesla that demonstrates this definitively?

Tesla has always been somewhat flexible in their definition of "on" - I don't believe that it always opens the contactors when you get out of the car, and I do believe that it will do active cooling of the pack while parked if required.

It will open contactors when HV is not needed (may be a short time after exit) for safety and to reduce drain due to the contactors themselves along with inverter idle loads.
It will activate them without a person in the car if needed for:

• 12V charging
• Pack heating
  
• Cabin pre-conditioning / over temp control
• Pack cooling
  
• Charging
• Summon

(Pack heating/cooling when not charging is a unclearly defined operation)

 

[Saghost]

It will open contactors when HV is not needed (may be a short time after exit) for safety and to reduce drain due to the contactors themselves along with inverter idle loads.
It will activate them without a person in the car if needed for:

• 12V charging
• Pack heating
  
• Cabin pre-conditioning / over temp control
• Pack cooling
  
• Charging
• Summon

(Pack heating/cooling when not charging is a unclearly defined operation)

That was my impression from what I'd read - it keeps the computers somewhat awake, and will engage the contactors while "off" to cool the pack or charge the 12V when it feels that's appropriate.

 

[mongo]

That was my impression from what I'd read - it keeps the computers somewhat awake, and will engage the contactors while "off" to cool the pack or charge the 12V when it feels that's appropriate.

Mine too. So annoying that we have to guess about the car's state machine instead of being able to just read the code... Still not annoyed enough to apply and move to CA...

 

[ℬête Noire]

Mine too. So annoying that we have to guess about the car's state machine instead of being able to just read the code... Still not annoyed enough to apply and move to CA...

There have been a few brave souls to successfully dump out and unencryp at least parts of the firmware, right? What state is that scene in?

 

[Evoforce]

If the Model 3 is the same as S-X, I can't believe it wouldn't be, it cools/heats the pack when necessary, no matter the circumstance. Plugged in, or not, occupied, or not. You will also lose a little off from range between each charge because of thermal management when it is unplugged.

 

[scaesare]

Interesting, that would at a minimum require an extra level of charging protection (temp vs current) for a Li type over PbA to prevent Li plating. In the bad idea/ inefficient world, seems like they could also use a different chemistry with some alternator diodes to reduce output voltage (with any appreciable current).

1) I think Tesla current draw was a design decision due to use of full PC style HW vs low power sleep mode microcontrollers. Makes the devepment cycle a lot simplier/ faster if you can rely on the huge battery pack under the floor instead ifof creating a reliable low power instant boot version on Linux.

3) For a long time cars were fixed voltage alternator (and you had to top up the water), then sealed batteries reduced water loss, finally OEM started adjusted voltage to temp. With battery temp monitoring and voltage control heath is maintained. Culomb counting is useful for system health, but a little overkill for PbA (great on though). I think it was a reasonable reduction of complexity.

Edit: battery blankets, like block heaters, are still useful. Check out the options for automatic backup generators.

Now I really want to know what point #2 was!

 

[mongo]

Now I really want to know what point #2 was!

Just for you...
I had nothing to add regarding this (other than an anecdote about the 24V setting being really bad for 12v batteries):

2) Tesla has incorrect charging algorithm for 12V battery. By the way, same story was with 2011-2012 Leaf's. They killed 12V battery as often as once per year. Algorithm was fixed later on and now premature failures are rare.