I have a Dec 2016 S60 RWD that I've been experiencing loss of range with.
For reference, we have just over 17.5k miles. Lifetime Wh/mi is 321. All power saving modes are enabled (Always connected turned OFF, Energy Savings is turned ON). Note: I NEVER use the range prediction to look at battery State of Charge (SoC). It's an algorithm based on both battery capacity (which is in and of itself an algorithm) plus highly variable driving parameters). Battery capacity as a percentage should be a more reliable and stable method of tracking long term changes to the battery health itself. All data presented below is a mix of data reported by the vehicle itself via the displays and reported by TeslaFi. Data sources marked where applicable.
The time I noticed loss of range coincided with the onset of summer in northern LA, so I attributed some of the loss to cabin overheat protection/battery cooling/etc. However, now that cooler "normal" weather has returned, I have not seen any improvement.
As part of the Southern California wildfires, Tesla remotely and temporarily increased by S60 to an S75. I have data for both battery capacities.
Example from November (while as an S75):
Charge set to 90% limit - charge start 11/14 1:01AM @ 8% SoC. Charge complete 11/14 6:16AM @ 90% SoC. Charge added 55.06kWh. (From displays)
Driving stats: 45 drive cycles, total mileage 130.5 miles. (TeslaFi) User interface reports 43.5kWh used for an efficiency of 333Wh/mi. (Car displays)
Charge set to 90% limit: charge start 11/22 1:01am @ 11% SoC. (Car displays)
The first discrepancy is in the amount of charge added versus the amount of charge used. For a 8%->90% charge, the battery added 55.06kWh. For a 90%->11% discharge, the battery used 43.5kWh. Delta: 11.56kWh for 3% battery delta, over 8 days. Rough average (not including 3% delta) of 1.445kWh per day of vampire drain. Seems high to me.
Per wk057's analysis of the BMS, the starting usable capacity of an S60 is around 62.4kWh and an S75 is around 72.6kWh.
Based on the numbers above for my drive plus the starting usable capacity (S75):
Personal current battery capacity (charge) = 55.06kWh/(90% SoC - 8% SoC) = 67.15kWh
This would equate to a roughly 7.5% battery degradation in 2 years/17.5k miles. Not terrible, but not great either.
Personal current battery capacity (driven) = 43.5kWh/(90% SoC - 11% SoC) = 55.06kWh
This would equate to a roughly 24.15% battery degradation in 2 years/17.5k miles, which is atrocious.
Example from October with S60 (with fewer driving cycles):
Charge set to 90% limit: charge start 10/7 1:13AM @ 58% SoC. Charge complete 10/7 2:54AM @ 90% SoC. Charge added 20.3kWh. (From displays). Interestingly, TeslaFi reports 18.8kWh used to add 20.3kWh, for an efficiency of 108.2%. I'll chalk this up to a TeslaFi error, but interesting nonetheless.
Driving stats: 3 drive cycles, total mileage 105.44mi. (TeslaFi) User interface reports 31.9kWh used for an efficiency of 320Wh/mi. (Car displays).
Charge set to 90% limit: charge start 10/8am @ 28%. (Car display).
Same math as above: 58%->90% charge, the battery added 20.3kWh. For a 90%->28% discharge, the battery used 31.9kWh. 30% delta (58%-20%) equals 11.6kWh (abs(20.3-31.9)), scale to 100%, gives me 38.66kWh. Obviously something is amiss here.
If we just look at the driven battery capacity: 31.9kWh/(90% SoC - 28% SoC) = 51.45kWh. Using the 62.4kWh from wk057, we get an expected degradation/vampire loss of 17.55%, which again, seems very high for 2 year battery degradation + 1 day of vampire.
Now that the data is out of the way, I scheduled an appointment with my local service center for Monday, December 10th. I laid out the case above (but with less detail) and received a call from a service advisor that their techs were going to do remote battery diagnostics. I provided the dates above for the S75 case, and this is the response I received:
"I am remotely diagnosing your concern at this time and wanted to follow up with my finding’s.
I have run a remote battery health check of your pack, and found there to be no hardware issues currently flagged. I have also confirmed your pack has been successfully upgraded to 75kWh.
What I have found, is your pack has a range calculation inaccuracy, which means the range displayed is not entirely accurate. This would explain why your consumption of 43.5kWh is equating to 79% of the packs available capacity. In fact this is a miscalculation of the packs consumed energy. Please note this inaccuracy does not affect the actual capacity of your pack nor the actual range your pack is capable of reaching.
I have performed a battery degradation calculation using logs stored for your vehicle and have found the true degradation to be around 6.6% since the inception of your vehicle (Approximately 2yrs).
We are constantly evolving the algorithm for range calculation to avoid inaccuracies like this and it should get better with future firmware updates.
If you have any further questions, please don’t hesitate to reach out."
Correct me if I am reading this wrong, but what they are saying is that there is an range calculation inaccuracy (despite me never having said that I use the range calculation, only battery percentage), but then they mention that there is a miscalculation of the packs consumed energy, which seems like a major flaw to me. They note that this does not impact actual capacity or actual range, yet in my experience, it does impact one or both of those parameters. A true battery degradation of 6.6% sounds reasonable, so that part is fine. And then they say that this should get better in a future firmware update, which (me being an engineer) is what we say when we have no intention of actually fixing it, but customer service needs something to tell a customer.
Expanding on my note above about the actual impact on either the actual capacity or range: we are limited to charging based off what the battery pack is reporting. If the battery pack reports 11%, it believes it only has 11% capacity left, and the firmware begins to provide notifications that charging is required. On the flip side, the charge circuitry also controls how much charge can be added to the battery, especially when the charge limit is set to 90%, which is a function of the built in fuel gauge. So if there is a fault in the fuel gauge, or in the calculation of the "packs consumed energy" as noted by Tesla, then there is a very real impact to at least the actual range the pack is capable of reaching. The pack itself may be fine, but the fuel gauge may not be.
I responded back with the above concerns and they responded that they have escalated it for a 2nd look.
TMC hive mind: Is there something I am missing here? Do I have an error in my analysis calculations? Or is vampire drain greatly increased with recent software updates?
Happy to answer any questions! Thanks!
For reference, we have just over 17.5k miles. Lifetime Wh/mi is 321. All power saving modes are enabled (Always connected turned OFF, Energy Savings is turned ON). Note: I NEVER use the range prediction to look at battery State of Charge (SoC). It's an algorithm based on both battery capacity (which is in and of itself an algorithm) plus highly variable driving parameters). Battery capacity as a percentage should be a more reliable and stable method of tracking long term changes to the battery health itself. All data presented below is a mix of data reported by the vehicle itself via the displays and reported by TeslaFi. Data sources marked where applicable.
The time I noticed loss of range coincided with the onset of summer in northern LA, so I attributed some of the loss to cabin overheat protection/battery cooling/etc. However, now that cooler "normal" weather has returned, I have not seen any improvement.
As part of the Southern California wildfires, Tesla remotely and temporarily increased by S60 to an S75. I have data for both battery capacities.
Example from November (while as an S75):
Charge set to 90% limit - charge start 11/14 1:01AM @ 8% SoC. Charge complete 11/14 6:16AM @ 90% SoC. Charge added 55.06kWh. (From displays)
Driving stats: 45 drive cycles, total mileage 130.5 miles. (TeslaFi) User interface reports 43.5kWh used for an efficiency of 333Wh/mi. (Car displays)
Charge set to 90% limit: charge start 11/22 1:01am @ 11% SoC. (Car displays)
The first discrepancy is in the amount of charge added versus the amount of charge used. For a 8%->90% charge, the battery added 55.06kWh. For a 90%->11% discharge, the battery used 43.5kWh. Delta: 11.56kWh for 3% battery delta, over 8 days. Rough average (not including 3% delta) of 1.445kWh per day of vampire drain. Seems high to me.
Per wk057's analysis of the BMS, the starting usable capacity of an S60 is around 62.4kWh and an S75 is around 72.6kWh.
Based on the numbers above for my drive plus the starting usable capacity (S75):
Personal current battery capacity (charge) = 55.06kWh/(90% SoC - 8% SoC) = 67.15kWh
This would equate to a roughly 7.5% battery degradation in 2 years/17.5k miles. Not terrible, but not great either.
Personal current battery capacity (driven) = 43.5kWh/(90% SoC - 11% SoC) = 55.06kWh
This would equate to a roughly 24.15% battery degradation in 2 years/17.5k miles, which is atrocious.
Example from October with S60 (with fewer driving cycles):
Charge set to 90% limit: charge start 10/7 1:13AM @ 58% SoC. Charge complete 10/7 2:54AM @ 90% SoC. Charge added 20.3kWh. (From displays). Interestingly, TeslaFi reports 18.8kWh used to add 20.3kWh, for an efficiency of 108.2%. I'll chalk this up to a TeslaFi error, but interesting nonetheless.
Driving stats: 3 drive cycles, total mileage 105.44mi. (TeslaFi) User interface reports 31.9kWh used for an efficiency of 320Wh/mi. (Car displays).
Charge set to 90% limit: charge start 10/8am @ 28%. (Car display).
Same math as above: 58%->90% charge, the battery added 20.3kWh. For a 90%->28% discharge, the battery used 31.9kWh. 30% delta (58%-20%) equals 11.6kWh (abs(20.3-31.9)), scale to 100%, gives me 38.66kWh. Obviously something is amiss here.
If we just look at the driven battery capacity: 31.9kWh/(90% SoC - 28% SoC) = 51.45kWh. Using the 62.4kWh from wk057, we get an expected degradation/vampire loss of 17.55%, which again, seems very high for 2 year battery degradation + 1 day of vampire.
Now that the data is out of the way, I scheduled an appointment with my local service center for Monday, December 10th. I laid out the case above (but with less detail) and received a call from a service advisor that their techs were going to do remote battery diagnostics. I provided the dates above for the S75 case, and this is the response I received:
"I am remotely diagnosing your concern at this time and wanted to follow up with my finding’s.
I have run a remote battery health check of your pack, and found there to be no hardware issues currently flagged. I have also confirmed your pack has been successfully upgraded to 75kWh.
What I have found, is your pack has a range calculation inaccuracy, which means the range displayed is not entirely accurate. This would explain why your consumption of 43.5kWh is equating to 79% of the packs available capacity. In fact this is a miscalculation of the packs consumed energy. Please note this inaccuracy does not affect the actual capacity of your pack nor the actual range your pack is capable of reaching.
I have performed a battery degradation calculation using logs stored for your vehicle and have found the true degradation to be around 6.6% since the inception of your vehicle (Approximately 2yrs).
We are constantly evolving the algorithm for range calculation to avoid inaccuracies like this and it should get better with future firmware updates.
If you have any further questions, please don’t hesitate to reach out."
Correct me if I am reading this wrong, but what they are saying is that there is an range calculation inaccuracy (despite me never having said that I use the range calculation, only battery percentage), but then they mention that there is a miscalculation of the packs consumed energy, which seems like a major flaw to me. They note that this does not impact actual capacity or actual range, yet in my experience, it does impact one or both of those parameters. A true battery degradation of 6.6% sounds reasonable, so that part is fine. And then they say that this should get better in a future firmware update, which (me being an engineer) is what we say when we have no intention of actually fixing it, but customer service needs something to tell a customer.
Expanding on my note above about the actual impact on either the actual capacity or range: we are limited to charging based off what the battery pack is reporting. If the battery pack reports 11%, it believes it only has 11% capacity left, and the firmware begins to provide notifications that charging is required. On the flip side, the charge circuitry also controls how much charge can be added to the battery, especially when the charge limit is set to 90%, which is a function of the built in fuel gauge. So if there is a fault in the fuel gauge, or in the calculation of the "packs consumed energy" as noted by Tesla, then there is a very real impact to at least the actual range the pack is capable of reaching. The pack itself may be fine, but the fuel gauge may not be.
I responded back with the above concerns and they responded that they have escalated it for a 2nd look.
TMC hive mind: Is there something I am missing here? Do I have an error in my analysis calculations? Or is vampire drain greatly increased with recent software updates?
Happy to answer any questions! Thanks!
