Near annual replacement of 12V battery is typical according to Tesla Service Tech

[Johan]

What's important in the Model S application is cycle count -- the capacity, size and weight are also factors, but the application is demanding on cycles. The C&D is a deep-cycle battery, and suitable for the application. The issue is really the vampire load -- it's way too high IMHO, and is stressing the 12V battery, even a very good battery like the C&D.

Good points. I never put my car on energy saving mode and have the box "always connected" ticked... I wonder if that accelerates wear?

 

[SteveS0353]

Good points. I never put my car on energy saving mode and have the box "always connected" ticked... I wonder if that accelerates wear?

Tesla has systematically reduced the vampire load. They've added visible things, like the energy saving (duty-cycling) modes, always connected modes, etc, and probably done other things in the software "under the covers" to duty-cycle the electronics to reduce the vampire load. Keeping the car plugged in doesn't help either, since the power architecture is wall connector -> charger -> HV battery -> DC-DC converter -> 12V battery. I suspect there are more opportunities in the software to reduce the vampire load even further.

- - - Updated - - -

Size, weight, amp hours are satisfactory. Most of the thousands out there cause no trouble. I've had mine nearly 2 1/2 years, 62000 miles, never a whimper.

Cost, small batteries (if high volume stuff) are cheap. It's probably a standard Toyota or Honda size! Who knows. They may switch to some LiIon stuff when they gey the Gfactory going. Lead acid is a poor choice.

they have the warning system in place, it's not deep discharge. The car goes to sleep so the vampire load is cut down on long off times.
They are going to be servicing it every now and then, so they check it when you're in, replace it, probably don't even tell you, big deal. I thought it was a pretty small "automotive" battery: I was thinking more "motorcycle" type. I've never seen it. I asked yesterday, and they had not changed it; it didn't need it.

There is probably a very wide bell curve, and you clearly have an outlier. There are numerous posts of 12V replacements shorter than 18 months too -- some even 6-12 months. There was a reported batch of batteries from another manufacturer which failed early, and these are the subject of a preemptive replacement program.

 

[Cosmacelf]

I haven't had the luck of some here having an ICE battery last 5 years. My wife's Infiniti QX56 battery lasts about 2 years before needing replacement. And with all the junk they attach to the positive terminal, and all the electronics that need resetting after a battery replacement, it isn't a painless job either. Oh, and my dealer invariably says there is no need for replacement after two years, this is after I've had the car not start a couple of times. "Checks out fine, don't replace it!". Tesla's proactive approach is much appreciated.

Also, again, my SUV uses more oil, and I do use synthetic in it, but my oil changes cost north of $100. Twice a year (synthetic oil).

I'm just saying that ICEs are not anywhere near better than the Model S as far as cost and convenience of maintenance goes.

 

[rpo]

My last three ICE cars had batteries that lasted 6-8 years. I never changed a single one and sold the cars with the originals still in place and functioning fine. Hopefully Tesla is (or has already) reduced the failure rate for newer cars.

 

[invisik]

Maybe after one is out of warranty a better battery could be installed as you'd be paying for it.

When my car is plugged in, why doesn't it draw from shorepower to keep the 12v charged to help reduce vampire drain? I'd think that to be part of the plugged-in Tesla is a happy Tesla concept. The car doesn't seem to do much with shorepower except for charging and runing pre-heating.

-m

 

[Cottonwood]

The 12V lead-acid battery does get recharged by a DC-DC converter from the high-voltage main battery. The issue with the 12V battery longevity is related to the so-called "vampire load", power to run the on-board electronics. Typically, the vampire load is 1.2kWh per day. There are various sources on the forum for this number, and represents ~4miles of range loss per day from the main HV pack.

All rechargeable batteries have a cycle limit. In other words, you can only discharge and recharge batteries a specified number of times before they "wear out" and can't support the load. The above battery had a capacity of 33 ampere-hours (33AH on the label). At 12V this is 0.396kWh of capacity. The 1.2kWh vampire load represents 1.2 / 0.396 = 3 charge / discharge cycles per day. The data sheet for the Tesla battery (thanks, by the way to the OP for providing the picture) has a capacity limit graph. Assuming Tesla allow the 12V battery to discharge 50% before recharging it from the DC-DC converter, the battery has a cycle limit of 1,800 before wear out. At 3 cycles per day, the 12V battery in the Model S would have a typical lifetime of 600 days, or ~18 months. Some batteries will last longer, some will fail sooner, but if one plots the bell curve, I would expect the mean will be about 18 months to 12V battery replacement.

This is exactly the issue, but your calculations need a couple of corrections. The largest of which is that when you do 50% discharges, you need to do twice as many cycles. That would mean 6 cycles per day in your example at 1.2 kW-hr per day vampire loss. On top of that, this kind of draw is an 8-hour draw for 100%. Looking at the chart at the bottom of https://www.cdtechno.com/pdf/lit/12_1061_0412.pdf, it gives a capacity of 30 A-hr rather than 33 A-hr at a 24 hour draw. That increases the cycles per day to 6*33/30 or 6.6 cycles per day.

If you assume 1800 cycles per battery that the spec sheet claims, the net result is 1800/6.6 or 272 days battery life. Because the vampire is sometimes less thirsty than 1.2 kW-hr/day, the vampire stops drinking when the car is on, and the batteries probably do better than spec'd, we usually see more than a year from these batteries. If you want your lead-acid battery to last longer, then use all the power savings modes to make the vampire less thirsty.

The reason that these batteries fail so often is simply because the vampire is sucking the life from them. A larger, heavier lead-acid battery could last longer, but the real solution is for Tesla to improve sleep efficiency in the Model S. The Roadster has a MUCH less thirsty vampire. My guess is that it will take a major hardware rev of the processors to tame the vampire. With all of the other development going on at Tesla, I bet this is just low on the priority list.

 

[ecarfan]

I thought it was a pretty small "automotive" battery: I was thinking more "motorcycle" type. I've never seen it. I asked yesterday, and they had not changed it; it didn't need it.

The S 12V battery is smaller than a standard car battery, but I think it is slightly larger than a motorcycle battery.

Note: the 12V that was removed from my S yesterday was original to the car and it was replaced with an identical battery, as shown in my photo above.

I am amazed to read in one post in this topic that one owner has had a 12V battery last 2 1/2 years!

 

[Ohm Alone]

Is there any other battery type that should last longer given the existing parameters than lead acid deep cycle? Li-ion perhaps? But, why not NiMH? After all, Prius NiMH batteries seem to last upwards of 400,000-600,000 km (250,000-375,000 miles) and years of use in taxi fleets in Vancouver BC.

 

[Cattledog]

The 12V lead-acid battery does get recharged by a DC-DC converter from the high-voltage main battery. The issue with the 12V battery longevity is related to the so-called "vampire load", power to run the on-board electronics. Typically, the vampire load is 1.2kWh per day. There are various sources on the forum for this number, and represents ~4miles of range loss per day from the main HV pack.

All rechargeable batteries have a cycle limit. In other words, you can only discharge and recharge batteries a specified number of times before they "wear out" and can't support the load. The above battery had a capacity of 33 ampere-hours (33AH on the label). At 12V this is 0.396kWh of capacity. The 1.2kWh vampire load represents 1.2 / 0.396 = 3 charge / discharge cycles per day. The data sheet for the Tesla battery (thanks, by the way to the OP for providing the picture) has a capacity limit graph. Assuming Tesla allow the 12V battery to discharge 50% before recharging it from the DC-DC converter, the battery has a cycle limit of 1,800 before wear out. At 3 cycles per day, the 12V battery in the Model S would have a typical lifetime of 600 days, or ~18 months. Some batteries will last longer, some will fail sooner, but if one plots the bell curve, I would expect the mean will be about 18 months to 12V battery replacement.

Excellent description, thanks.

 

[m1ker]

Does anyone have experience with this in an are where there is no Tesla service centre? In Calgary AB the closest SC is in Vancouver (12h away driving). Do the fly someone out?

 

[AnOutsider]

Had my third battery installed last week as well (25 months in).

Hoping this doesn't become a yearly thing, but maybe now I've got the latest version.

 

[SteveS0353]

This is exactly the issue, but your calculations need a couple of corrections. The largest of which is that when you do 50% discharges, you need to do twice as many cycles. That would mean 6 cycles per day in your example at 1.2 kW-hr per day vampire loss. On top of that, this kind of draw is an 8-hour draw for 100%. Looking at the chart at the bottom of https://www.cdtechno.com/pdf/lit/12_1061_0412.pdf, it gives a capacity of 30 A-hr rather than 33 A-hr at a 24 hour draw. That increases the cycles per day to 6*33/30 or 6.6 cycles per day.

If you assume 1800 cycles per battery that the spec sheet claims, the net result is 1800/6.6 or 272 days battery life. Because the vampire is sometimes less thirsty than 1.2 kW-hr/day, the vampire stops drinking when the car is on, and the batteries probably do better than spec'd, we usually see more than a year from these batteries. If you want your lead-acid battery to last longer, then use all the power savings modes to make the vampire less thirsty.

The reason that these batteries fail so often is simply because the vampire is sucking the life from them. A larger, heavier lead-acid battery could last longer, but the real solution is for Tesla to improve sleep efficiency in the Model S. The Roadster has a MUCH less thirsty vampire. My guess is that it will take a major hardware rev of the processors to tame the vampire. With all of the other development going on at Tesla, I bet this is just low on the priority list.

Thanks Cottonwood, and you are correct. I neglected to compensate the number of cycles as they are 50% depth (my crude assumption), and that means the car would need 6 x 50% cycles per day to feed the 1.2 kWh vampire load. That shortens the battery life to ~300 days. The small correction you applied for the available capacity at this level of discharge is appropriate, so I concur on 6.6 x 50% cycles per day to feed 1.2kWh, nominal lifetime of 272 days (~9 months), if the discharge / charge cycles are really 50% capacity.

I read somewhere else on TMC that someone had tracked the duty cycle of the 12V, and it seemed that when the car is idle, the recharge comes on every 3 hours -- about 8 times per day. Assuming 8 recharge cycles per day is correct, then the depth of discharge is ~38%. The curve of cycles to wear out versus depth of discharge is strongly dependent on the depth of discharge and not linear. Eyeballing the cycle curve in the data sheet, I estimate the wear out at 38% depth of discharge is ~3,200 cycles. Using that, would mean 3,200 / 8 = ~400 days, or ~13-14 months.

In any event, we agree that the vampire is sucking the life out of the 12V battery, and I'm not surprised to see 12V battery replacements typically around 18 months. As you say, the vampire load could be less at times, and with tolerances on every other thing, these are at best educated WAGs (wild-ass guesses) of theoretical battery lifetime.

 

[santana338]

My guess is that it will take a major hardware rev of the processors to tame the vampire. With all of the other development going on at Tesla, I bet this is just low on the priority list.

Thanks for the analysis guys. It seems like this issue is settled. Is it time to move on to speculate what Tesla is going to do with the processors to reduce vampire drain? I would assume all accessory processors for windows, etc are in a deep sleep state or even off. I would think just the basics are on: things like the key fob radio, 3G radio (oh, there is a power hog), and charge monitoring/charger. Anyone care to guess what they have now and how it could be improved?

 

[mknox]

My last three ICE cars had batteries that lasted 6-8 years. I never changed a single one and sold the cars with the originals still in place and functioning fine. Hopefully Tesla is (or has already) reduced the failure rate for newer cars.

Just thinking the same thing. My wife's 07 Pontiac is on the original battery that came with the car. It sits outside and starts like a champ at below 0 F these last few weeks.

 

[AoneOne]

A larger or better battery will certainly last longer, but I'm hoping that Tesla's next electronic design will have greatly reduce vampire drain. It's almost as if Tesla, at first, didn't appreciate the effects of the vampire draw. The obvious cost was the extra electricity consumption, but we've now learned that it wears out the 12V battery and, presumably, requires periodic reconnection of the main battery to the DC-to-DC converter, increasing the number of battery switch cycles, as well as DC-to-DC power cycles. That cycling, too, reduces reliability, though not as dramatically as the effect on the 12V battery.

So, perhaps there will be a replacement for the primary electronics package that reduces the vampire drain ten-fold, and includes LTE, a faster processor, more memory, etc. We can hope.

 

[Cottonwood]

Thanks Cottonwood, and you are correct. I neglected to compensate the number of cycles as they are 50% depth (my crude assumption), and that means the car would need 6 x 50% cycles per day to feed the 1.2 kWh vampire load. That shortens the battery life to ~300 days. The small correction you applied for the available capacity at this level of discharge is appropriate, so I concur on 6.6 x 50% cycles per day to feed 1.2kWh, nominal lifetime of 272 days (~9 months), if the discharge / charge cycles are really 50% capacity.

I read somewhere else on TMC that someone had tracked the duty cycle of the 12V, and it seemed that when the car is idle, the recharge comes on every 3 hours -- about 8 times per day. Assuming 8 recharge cycles per day is correct, then the depth of discharge is ~38%. The curve of cycles to wear out versus depth of discharge is strongly dependent on the depth of discharge and not linear. Eyeballing the cycle curve in the data sheet, I estimate the wear out at 38% depth of discharge is ~3,200 cycles. Using that, would mean 3,200 / 8 = ~400 days, or ~13-14 months.

In any event, we agree that the vampire is sucking the life out of the 12V battery, and I'm not surprised to see 12V battery replacements typically around 18 months. As you say, the vampire load could be less at times, and with tolerances on every other thing, these are at best educated WAGs (wild-ass guesses) of theoretical battery lifetime.

We are in complete agreement!

Your WAG is right on and way better than a SWAG (Stupid Wild Ass Guess)!

 

[ecarfan]

So, perhaps there will be a replacement for the primary electronics package that reduces the vampire drain ten-fold, and includes LTE, a faster processor, more memory, etc. We can hope.

No need to hope. Look at the recent history of microprocessors and memory. I can guarantee those things will happen. We are talking about "consumer electronics" here. Okay, this item has wheels and a big electric motor. But otherwise it's basically the same thing.

 

[miimura]

A larger or better battery will certainly last longer, but I'm hoping that Tesla's next electronic design will have greatly reduce vampire drain. It's almost as if Tesla, at first, didn't appreciate the effects of the vampire draw. The obvious cost was the extra electricity consumption, but we've now learned that it wears out the 12V battery and, presumably, requires periodic reconnection of the main battery to the DC-to-DC converter, increasing the number of battery switch cycles, as well as DC-to-DC power cycles. That cycling, too, reduces reliability, though not as dramatically as the effect on the 12V battery.

So, perhaps there will be a replacement for the primary electronics package that reduces the vampire drain ten-fold, and includes LTE, a faster processor, more memory, etc. We can hope.

If you want to talk about cycles, don't forget that in order for the DC/DC to recharge the 12V, the main contactor has to close too. Those cycles aren't infinite either.

 

[SteveS0353]

No need to hope. Look at the recent history of microprocessors and memory. I can guarantee those things will happen. We are talking about "consumer electronics" here. Okay, this item has wheels and a big electric motor. But otherwise it's basically the same thing.

Short of new lower power hardware, there are likely a few things that software can do to reduce the vampire power drain by aggressively duty-cycling various hardware subsystems. Tesla has already done some of that with the energy saver mode. To put this into context, the vampire drain when the car is off is ~1.2kWh per day on average. That's a average current from the 12V battery of a little over 4 amps. A significant power drain is likely the various radios. Bluetooth, WiFi and 3G radios consume a lot of power when running. Turn off the various radios when they are not being used, and aggressively duty cycle them when the car is off and they are still required for whatever reason. High frequency CPUs and memory consume a lot of power too.

I have no idea if any of these things are implemented, but here are a few ideas...

• Humans perceive 0.8 of a second latency as pretty much instantaneous. There is no need to power the key-fob radio 100% when the car is off and locked, poll for 0.2 second every second (1:5 duty cycle), and save basically 5x the key-fob radio power. It might be possible to reduce the frequency of key-fob polling when the car is off too.
• There is likely no requirement to power the Bluetooth radio when the car is off.
• It may be possible to duty cycle the WiFi radio looking for a WiFi network when one is not active.
• It may be possible to reduce the beacon frequency of an active WiFi network when the car is off to keep the network alive and responsive but save power.
• There is likely no requirement to power the 3G radio when there is a valid WiFi network while the car is off.
• Even with "stay connected" checked in power save mode, when there is no active WiFi network, duty cycle the 3G radio when the car is off.
• Reduce the clock frequency of the CPUs that do need to stay awake to support some function.
• Stop the clock or remove the power on CPUs which are idle. There could be a delay to start up in some cases, so power up needs to be carefully designed.

It's all about attention to detail -- looking for the last µA of current and being creative about power down and power up sequencing.

- - - Updated - - -

Maybe after one is out of warranty a better battery could be installed as you'd be paying for it.

When my car is plugged in, why doesn't it draw from shorepower to keep the 12v charged to help reduce vampire drain? I'd think that to be part of the plugged-in Tesla is a happy Tesla concept. The car doesn't seem to do much with shorepower except for charging and runing pre-heating.

-m

The power architecture is shore power -> charger -> HV battery -> DC-DC converter -> 12V battery. Shore power does not feed the vampire directly. That would be an improved architecture, but that's not how it's implemented AFAIK.

 

[Johan]

Great insights Steve!

One follow up question with regards to cycling of a regular 12V battery: If there is a constant power draw from the terminals of a 12V battery but at the same time a steady charge applied that balances out the draw would that even result in any cycling? Basically what I mean is that it's all about voltage delta right, so if shore power was somehow directly charging the 12V battery or if the contactor between the HV battery to DC-DC converter to 12V battery were left open all the time while the car was plugged in, it would be like the short power powering all the 12V electronics directly, without causing wear and cycling on the 12V, right?