Model 3 SR+ LFP Battery Range, Degradation, etc Discussion

[Baluchi]

This graph (from the Tessia app) shows the temperature vs efficiency in my LFP car. Not surprising to see that my worst average efficiency was at the lowest temps (5F/-15C to 10F/-12C), at 302 Wh/mi. My best average efficiency was from 70F/21C to 75F/24C, at 214 Wh/mi. Those numbers go up a little bit as it gets really hot, presumably due to more A/C usage.

This data shows a 41% increase in power consumption between ideal conditions and really cold conditions. But the blue bars on the graph are the number of miles I've driven at each temperature, so you can see that a vast majority of my miles are driven between 25F/-3C and 85F/29C, when efficiency is much better than those worst-case scenarios.

As a note, the Wh/mi shown here are higher than what I see in the car, I think because Tessie includes non-driving energy in its calculations (i.e. charging losses or preconditioning). But the trip reading in my car indicates I often use under 200 Wh/mi while driving in ideal conditions.

 

[theothertom]

This graph (from the Tessia app) shows the temperature vs efficiency in my LFP car

Good data, thanks. Interesting that efficiency kicks up slightly (gets worse) at the higher temps.

 

[Baluchi]

Good data, thanks. Interesting that efficiency kicks up slightly (gets worse) at the higher temps.

It's also possible I just drove harder, on average, in those temps, resulting in higher consumption. My rule of thumb is to keep it in Chill mode on weekdays and try for max efficiency. But on the weekends I allow myself a bit more fun and use Standard mode. Those weekend miles are a small fraction of my driving, but it's hard to say if they skewed any of the data.

Also worth noting that I got the car last September, so I haven't had it in July or August yet. I suspect the graph will get "smoothed out" as I go through more seasons.

EDIT: By the way, I did a video a couple months ago demonstrating charging my Tesla with a small portable generator. It's long and I would say fairly boring, but may be interesting for anyone into in the kind of dorky stuff we talk about in this thread.

 

[bin31z]

I’m frankly amazed by LFP 2022 with ryzen efficiency. Drove to work and got 193 wh/m. Dropped to 220 when going home with spirited drive through Malibu canyon but still good compared to what I’ve seen in nca model 3 and S. Hopefully degradation is much less as well.

 

[Baluchi]

I’m frankly amazed by LFP 2022 with ryzen efficiency. Drove to work and got 193 wh/m. Dropped to 220 when going home with spirited drive through Malibu canyon but still good compared to what I’ve seen in nca model 3 and S. Hopefully degradation is much less as well.

Gosh, that's yet another variable with Intel Atom vs AMD Ryzen. I doubt we'll even notice the difference in efficiency without the benefit of big data, but still worth noting.

 

[wizziwig]

Also don't forget that the original 2021 LFP used 980 performance-model rear motors to maintain faster acceleration despite the extra weight of LFP. Current 2022 LFP cars with slower acceleration use the 990 motors. There are probably efficiency differences between these motors. The weight difference between 55 and 60 kWh LFP will also affect efficiency - especially in stop/go traffic.

 

[wizziwig]

One more thing regarding parked efficiency differences. The newer cars with the Li-ion 12V batteries may experience different vampire drain when parked for longer periods. These batteries are much smaller capacity (around 1/5 of the old lead acid) so car may need to wake more frequently to keep them charged.

 

[amit_gunner]

One more thing regarding parked efficiency differences. The newer cars with the Li-ion 12V batteries may experience different vampire drain when parked for longer periods. These batteries are much smaller capacity (around 1/5 of the old lead acid) so car may need to wake more frequently to keep them charged.

So, you are saying it would have higher vampire drain?

 

[Labs]

So, you are saying it would have higher vampire drain?

I would think so. Especially in cold temps as it has to keep the 12V Li-ion above -10 to charge…

 

[Dave EV]

One more thing regarding parked efficiency differences. The newer cars with the Li-ion 12V batteries may experience different vampire drain when parked for longer periods. These batteries are much smaller capacity (around 1/5 of the old lead acid) so car may need to wake more frequently to keep them charged.

While the lithium 12V has less capacity than a lead acid, lead-acids don't like to sit around partially charged and they don't like deep cycles (even the so-called deep-cycle batteries like the one used by Tesla), either. Not to mention that lead-acids take a while to really fully charge as well and charging efficiency is lower than lithium, too. I expect overall efficiency of the lithium 12V to be at least as good as the lead-acid, if not better.

 

[Ruffus23]

can't see tesla using a more expensive/less superior battery in their car.

lead is dead!

 

[AlanSubie4Life]

While the lithium 12V has less capacity than a lead acid, lead-acids don't like to sit around partially charged and they don't like deep cycles (even the so-called deep-cycle batteries like the one used by Tesla), either. Not to mention that lead-acids take a while to really fully charge as well and charging efficiency is lower than lithium, too. I expect overall efficiency of the lithium 12V to be at least as good as the lead-acid, if not better.

Someone should take a look at this with the appropriate equipment. I wonder how fast they can push energy into the Lithium as compared to the lead acid; this would matter as well if it were different.

In the end it’s a question of duty cycle spent charging the 12V and the energy spent elsewhere during that charging period (and to a lesser extent the energy going into the 12V itself).

My hope is they actually reduced standby power in more recent vehicles, but for some reason Tesla seems to have decided 6-10W is acceptable. Reducing that of course reduces the amount of energy you have to put into the 12V, and (more importantly) also reduces the extraneous energy use while charging the 12V (that’s another thing they could improve).

Any takers on this analysis with the new Lithium battery? Or has someone already analyzed it and posted around here (I am not spending much time here these days)?

 

[wizziwig]

Took delivery of LFP RWD Model 3 with Li-ion 12V and Ryzen about a week ago. During that time, I have observed 1% daily vampire drain which matches what Tesla wrote in their online owner manual. All car settings are still at factory defaults and I have not used their phone app at all. While grabbing some stuff from the garage, I also caught the car waking once (heard contactors engage). It remained awake for 16 minutes before going back to sleep. I assume it was charging the small Li-ion pack from the main traction battery (car was unplugged).

A quick back-of-the-envelope calculation seems to indicate it would need to charge several times a day if the drain was linear and only used to recharge the battery. In reality not all the drain is going back into the small battery. 60 kWh pack * 0.01 = 0.6 kWh lost per 24 hours. The small Li-Ion pack only holds ~ 0.1 kWh per this article and linked teardown videos.

I do have the equipment to measure the current and voltage but not a lot of spare time to watch the car all day waiting for it to charge. I'll post an update if I get lucky and catch it doing anything interesting. Will have to pop the hood and see if the battery terminals are even accessible. In the teardown it looks like it's all covered by a harness.

As for why Tesla would do this... could be same reason they've done everything else (remove radar, passenger lumbar, center speaker, center console data usb ports, etc.) - to reduce cost or get around supply chain issues. In the teardown, he says the cells in the pack can be found for $24 - maybe much less in bulk + case + bms, etc. They are currently always pairing this battery with the higher drain Ryzen infotainment systems (Model S and 3) so maybe some attribute like higher voltage or charge/discharge rate of this battery is useful there.

 

[AlanSubie4Life]

In the teardown, he says the cells in the pack can be found for $24 - maybe much less in bulk + case + bms, etc.

Yeah I guess I suspect that is the reason. Seems like Tesla should be able to make Lithium-ion packs relatively cheaply. But who knows. I guess we’ll see if replacement costs beat the $85 12V when people order them from the service center!

It doesn’t sound like the sleep power of 6-10W or the ~150-250W idle power has changed much on the new vehicles. I guess for idle power someone should check whether Sentry Mode drains at the same rate (leave it on for a couple days) as the older vehicles to get a rough idea of whether any of the new computers have an impact on that. Hopefully not worse! (I know they had to adjust some efficiency numbers in other countries, but that may be different.)

 

[wizziwig]

Just for some perspective, the Kia Niro EV, Hyundai Kona Electric, and Chevy Bolt EV have essentially zero vampire drain. Myself and others measured them at about 20 mA draw from the 12V pack during sleep. That's only 0.25 watts! Even at such low consumption, the cars still wake and respond to requests from their phone apps. If left undisturbed, the KIA and Hyundai only wake from sleep once every 24 hours for a 20 minute top-up of the 12V battery.

Tesla's 1% daily vampire drain is a bit shocking in comparison. Multiply that by a much larger global fleet and it's a lot of wasted electricity. Is this because they use x86 architecture processors to run their cars? Connected standby seems to be very efficient on the ARM SOCs.

Since I work from home, I tend to take a lot of short trips - mostly local errands, dining, shopping, etc. While still only pennies per day, it may end up costing me more to park the Tesla than to actually drive it.

 

[eevee-fan]

Just for some perspective, the Kia Niro EV, Hyundai Kona Electric, and Chevy Bolt EV have essentially zero vampire drain. Myself and others measured the KIA at about 20 mA draw from the 12V pack during sleep. That's only 0.25 watts! Even at such low consumption, the car still wakes and responds to requests from their phone app. If left undisturbed, it only wakes from sleep once every 24 hours for a 20 minute top-up of the 12V battery.

Tesla's 1% daily vampire drain is a bit shocking in comparison. Multiply that by a much larger global fleet and it's a lot of wasted electricity. Is this because they use x86 architecture processors to run their cars? Connected standby seems to be very efficient on the ARM SOCs.

Since I work from home, I tend to take a lot of short trips - mostly local errands, dining, shopping, etc. While still only pennies per day, it may end up costing me more to park the Tesla than to actually drive it.

Could all Tesla actually be secretly mining dogecoin? Haha

 

[AlanSubie4Life]

That's only 0.25 watts!

Yep. And while the Tesla loses about half of its energy due to the overhead used while recharging the 12v, it’s reducing that standby sleep consumption considerably that would really help the overall drain....

Drop it down to ~1W (or even 0.25W - seems totally possible), and the energy per unit time (so average power consumption) due to the recharge event also goes down by a factor of ~10 (peak power is the same but it happens 10 times less often, so once every ~5 days) .

It's always been mysterious to me that Tesla specified the system with such a high sleep power. Here we are, four years later, and it's apparently exactly the same, for no apparent reason. Obviously, it can take <1W to be able to wake up to requests from an app and have no impact on features or function. Just need the system to be specified differently.

 

[wizziwig]

Could all Tesla actually be secretly mining dogecoin? Haha

I think you may have solved the mystery and why TSLA stock/profit are so high.
Think of how much more they can mine with the Ryzen GPU!

On a serious note, I see 3 possibilities:

1) Could the main traction batteries they use (NCA and LFP) self-discharge at a higher rate than NMC? (typically < 2.5% per month).
2) The car consumes a lot more when awake (such as when charging the small 12V pack) or spends more time awake.
3) The car consumes a lot more when sleeping.

At some point I'll put a current clamp on the 12V pack to see how much it draws while car is sleeping. If it's negligible like other EVs, then it must be some combination of 1) and 2). Disappointing that the "Since Last Charge" screen in the car doesn't actually report what the label says. It only tracks consumption while in D/R gear. That makes tracking the energy consumed while idle much more difficult. The Chevy Bolt has a similar screen titled "Since Last Full Charge" but it tracks every kWh leaving the main pack - even breaking it down into % consumption for Driving/Accessories, Climate, and Battery Conditioning.

Maybe someone has already done this type of research over the years. I'm new to Tesla ownership and didn't do much googling on this topic yet. Still seems worth doing another investigation in case anything changed on these latest models.

 

[AlanSubie4Life]

At some point I'll put a current clamp on the 12V pack to see how much it draws while car is sleeping.

Remember to track the voltage concurrently of course! You should find 6-10W (if nothing has changed) for power draw when the contactors are open and the car is asleep (for voltage, the old lead-acid batteries went from 14.5V when charging to 12.6-12.8V to trigger a recharge, with a float voltage of 13.4V (you see this in idle mode when the 12V is not actively being charged but the contactors are closed and the DC-DC is on)).

Maybe someone has already done this type of research over the years.

Yes, but always good to see solid data posted, especially from a new model year. Definitely measure and post! (You may be able to find an better fitting thread for it if you search for prior data.)

 

[TexaCali]

Since I have one of the fist 2022 LFP cars delivered in North America, I thought I’d give a 90 day update. My car was delivered on Nov 13th with 15 miles on the odometer. Odometer was 2454 this morning.

The short summary is the car reported 271 miles after a full charge last night - 0.4% loss,if that’s what it really is and not just a BMS calibration. For more details, read on…

Environment:
Vehicle sits outside 100% of the time (parked in my driveway, not covered). I’m in San Jose and temps since delivery have ranged from low 30’s to high 70’s. Mid 40s’ to mid 50’s were probably the most common temps during this period.

Use:
Primary use of the vehicle has been for my work - private music instruction in clients’ homes. I drive from my home to the first client (15-20 miles highway), spend about 1 hr at the client’s home, then drive 2-6 miles to the next clients home, repeat multiple times, drive home. Total about 40-50 miles a day, mostly after dark, and most miles are freeway going to/from my home to the client area. I made extensive use of the HVAC in November and December when temps were lower and humidity was high. As temps have warmed, I’ve been using the HVAC less and Wh/mile continues to improve. Lifetime average is 219 Wh/mi, but current use is under 200 Wh/mi.

Charging:
I charge the vehicle to 100% and then drive until there is not enough charge left to get through the next day’s use. Recently that has meant charging once per week - Level 2 at home. Zero Supercharger use so far.

This week is fairly typical of what I see, so lets look at the numbers:
214 miles traveled, 199 Wh/mile, 18% battery remaining. That means I really used 60x(1-.18)=49.2 kWh, and efficiency is really 49,200/214 = 230 Wh/mile when we count the vampire/idle drain. Looking at it another way, 214x199= 42.6 kWh, therefore 49.2-42.6 = 6.6kWh was consumed while the car was parked (11%). It was exactly 1 week between charges, so that’s 943 Wh lost per day (about 1.6% per day).

Though the idle drain is a bit high, overall I’m extremely pleased with the car. It is still amazingly efficient while still somewhat sporty and fun to drive. My use case is allowing the BMS to see the battery at various states of charge and minimizing the average charge state, which should yield a long battery life. Time will tell, but so far so good!