Forums

Join The Community
RegisterLogin

Vampire thirst for electricity at night

I'm expecting delivery this month of Tesla S 85kWh.
I'm concerned about the reported "Vampire thirst for electricity at night"

http://www.greencarreports.com/news/1082949_life-with-2013-tesla-model-s...

Allegedly Tesla S consumes 4.5 kWh energy per day when sitting in the garage in off position. This wasted energy is enough to drive Tesla S 5,000 miles on annual basis (which is about 30% of typical annual use of a car). In California this wasted energy will cost about $300 per year (at my rate of $0.20 average per kWh in San Diego).

Allegedly, Tesla reps confirm this fact although do not necessarily agree with the numbers. The reason for it is the fact that putting all computers to "sleep" caused problems for the Tesla engineers and they decided to keep all computers awake all the time thus causing drain of energy when the car is not used.

Anyone willing to share a real life experience in this regard?

Any word from Tesla when this issue will be resolved.

If you leave your computers plugged in over night, they also use energy. This is a big computer called Model S. What you have to do is find a less expensive way to charge your car. Here in Las Vegas we are charged .05 after 10pm at night. See if your local utilities has a savings option like that.
Enjoy your new 85!
Don

Elon has stated that the vampire drain will be addressed with a software update sometime this summer.

The 4.5 KWh number sounds a bit high, but I don't have any way to refute it either. I can tell you that once charged, the current drops to zero (maybe 10ma, the limit of my meter), so there is no drain when charged up. That said, yes the computer is running in a lower power mode all the time, which drains the battery and requires a topping up the charge at some point.

If it does take 4.5 KW a day (187W/hr), at a more reasonable rate of $0.05 for late night charging (PG&E in Northern CA, E9 Tariff), the cost is $75/year. The power usage may increase at the temperatures extremes to keep the battery temperature at a reasonable level.

Version 4.1 (or maybe 4.0?) software had a power-down mode that turned off several of the computers. This option was removed, as it had a few quirks, but has been promised to return in a future update. The downside of using this mode is the computers take about 30 seconds to come back on-line, so you don't get any displays until that occurs.

While not close to the MS, most newer ICE cars also use the battery while turned off to keep low-power computers running (alarm, FOB detection, On-star, etc.). When you start up and drive in the ICE car, the engine has to work a bit harder (i.e. poorer MPG) to charge up the battery.

Anyway, there are far more things to worry about in life than this minor vampire load. Enjoy your new car, you'll likely forget all about this once you start driving!

Thanks to all who replied.

Since this topic is up my alley (computer engineer by trade) I will dare to comment a little bit more on the subject.

Measuring power consumption when battery is full is futile since the car does not start charging unless the batteries are at least 3% down. Lack of consumption from the power outlet doesn't mean that "the vampire" sleeps.

I agree that off peak rates, where available, are one way to mitigate the expense. However, from pure engineering standpoint it is not an elegant solution to keep charging and discharging the battery for no reason.

I do see some contradiction in the reports. On one hand the original reporter seems like a solid guy and he even bothered to setup a special dedicated electric meter to measure the wasted energy. For this reason I tend to believe him about his estimate of 4.5 KWh per day or 187 Watts of constant power consumption.

On the other hand when I was designing and running data centers not that long ago we always had a rule of thumb 100 Watts per server. An average server is fairly powerful computer with fast CPU, 1 or 2 hard drives spinning motors for fans, hard drives.

While Tesla S is advertised to have many computers on board, most of those are most likely to be what we call "embedded controllers" that have many times less power consumption. Obviously there is a main one that controls the large flat display. But even this one is a modern computer (I assume) without fans and spinning hard drives. With all this said it comes as a surprise to me that the on board computers consume 187 Watts of power.

If I'm correct than either the original reporter was wrong with his 4.5KWh estimate or TeslaS is feeding something else in addition to the computers/controllers when off. Call for comments on this last statement !

Last but not least the 4.5KWh if true may not be that expensive if you manage to get lower rates, however the impact in effective lost miles (13.5 miles per day, according to the same source) is not to be ignored with ease. Imagine scenario when you live 30 miles from the airport and go on a business trip for 8 days leaving your TeslaS at the airport. If you leave home at a capacity of 160 miles, you'll be at 130 miles range by the time you get to the airport. You will lose another 8x13.5=108 miles for 8 days due to the "vampire" effect. Upon your return, your Tesla will have only 22 miles range left not enough to get home.

Let's hope that Mr. Musk will deliver on his promise and this will be something from the past soon.

If you need to exit the car and keep it in neutral, you power off the car. Do we know if that eliminates the vampire drain (powering off without releasing the parking brake)? That could solve the airport problem.

Here are some real numbers if someone wants to do the math. I left my Tesla unplugged in my garage for two weeks while I was on vacation.

When I left, the Rated Range shown on the dashboard was about 210 miles. (I had charged it the night before and driven it to work that day.)

When I came back from vacation I went to take the car out for a spin. The Rated Range was down to 8 miles! That's right - single digits.

So in 14 days, the car "used up" 200 miles of range. I don't want to think what would have happened if I left it for 3 weeks.

The last time I checked my loss was about .53 rated miles per hour which is more like 3.9 kWh per day, still a large number.

@bhs1

That really is bad. I certainly hope the TM software folks have that as their highest priority.

And once they get the vampire issue worked out, I think they also need to get the 12V battery trickle charging from the main battery rather than just from wall current while plugged in. AND provide for auxiliaries like AC to run directly from wall power when plugged in, instead of draining the battery to a certain point then recharging it.

I'm surprised at the unimpressive power management system they have currently, but I guess not everything about the car can be perfect out of the box.

@bhs1 "need to get the 12V battery trickle charging from the main battery"

I'm almost positive that the 12v battery is ONLY charged from the main battery. I can't possibly see the small 12V battery running for hours when it is powering lights, computers, displays, fans and just about everything but the main motor.

Actually the power management seems quite impressive. They are dealing with a lot of issues - instant-on desirability, battery pack charge and temperature management (for best performance, reasonable charge times and longevity). Like any complex system there are going to be tradeoffs.

For example, perhaps you don't care about longevity - so you can make a full charge in 10 minutes which might work for 5 or 10 charges before the battery fails to hold a full charge. Maybe you want the pack to last 20 years - so it takes 20 hours for a full charge every time and you can only charge it to 50% and discharge it to 30%. These are somewhat made up, but there are lots of complex interactions that had to be carefully decided on.

When you "power off" the car, how long does it take to power back on? Can you charge it when it is in the "power off" state?

The current "Power off" button doesn't really power much off. Everything comes back on immediately if you touch the brake or screen. There was an option to really power down in older software, and others on the forum indicated it took about 30 seconds to come on-line. I think you can actually drive the car immediately - the 30 seconds is for the two display processors to boot up.

Don't know about the charging, but I suspect you can charge with the old power-down option set. Hard to tell now that the full off option is gone.

@bhs1 - what was the temperature when you came back?

My real world test at an airport with temps 50-70 deg showed 0.6 miles/hour loss for the first 24 hours, then 0.4 from then on. So c. 14 miles day 1, and just under 10 miles each day after that.

milesperhourlost
10 miles is just over 3KWh per day.

The software update should reduce it to less than 0.5 KWh/day (based on the previously released sleep mode)

What's the best way to monitor vampire loss/re-chargiging iterations? I usually plug in when I get home for the evening (no off-peak rates in San Antonio) and when I get up in the morning, it appears to have only lost 2 miles. Could be topping off in the middle of the night, though.

Thought about setting up an iDevice with screenlock tuned off and then time-lapse record the charging screen of the app. How much extra "drain" do you think I would be causing by having the car respond to the app every 5 or so seconds?

It's been fairly cold since I took delivery of my car in Denver about a month ago. I have consistently been loosing .5 miles of rated range an hour when the car is sitting.

cgi;
If you still have 4.1, that's how it's supposed to work.

I took delivery of my 60kWh Model S on 17 Jan, live in So Cal, start charging @ 11pm each evening. I find I also lose about 2 miles range by the time I check the car at 7am. If I don't use the car until later in the day I will continually lose more - it seems to be about .5mi range/hr.

Ivan - I'm in San Diego as well and, while we are not as fortunate as those who can get $0.05 kWh rates, SDG&E will give you $0.15 per kWh in Super Off-Peak (12M-5am). You need to fill out an application on their website for the EV-TOU (time of use) rates. Signing up for and using this program also means the car charging doesn't push up your total household usage into the Tier 4 rates, which are $0.30 per. It's saved me a significant amount every month.

I left my fully charged 60 kwh model s parked in my garage, unplugged, from Feb 18th through March 6th. Imagine my surprise to have only a 37 mile range left when I got in it once home from the airport. Glad I did not leave the car at the airport - may not have made it home. I did take the car in to the Fremont Tesla service center, and they found no "problems". So be warned - I went from 187 mile range to 37 mile range in 16 days this year. I left it plugged in when I took my next two week trip. Chris

I never leave anything that is 240V/40A+ plugged in when I'm away long term. Sauna, Steamer, Dryer, EV - everything gets disconnected from the breakers. 240V/40A or more is pretty powerful source and can cause a lot of damage in unlikely but possible event of earthquake, rodents, flood etc.

The vampire loss is a weak point for the Model S and (slightly) detracts from the awesomeness of this car. My typical daily mileage is about 10-25, so I'll often charge up only every 2-3 days. I noticed today, for example, that I had driven about 20 miles since the last charge and had 200 miles rated left of the original 240 full charge. I typically average around 300 wh/mile driving in town, so the vampire loss essentially doubled my energy usage.

Return of the sleep mode will be a big plus. I'm assuming this will somewhat delay the responsiveness of the car when you first hop in, so I'd like to be able to schedule "nap time" so my baby is wide awake when I'm ready to drive!

Gentlemen, The power consumption of onboard h/w (cpu etc) is minimal (in single digit watts). Ideally
it should be < 1 watt.

However, the main consumption of the so called "vampire current" is : Tesla maintains the
Lithium ion Battery at a particular temperature. In winter nights the battery is heated up
and in hot garages, the battery is cooled. The main reason is to prolong the life of the battery.
If you monitor, the garage temperature at night and miles lost, you will get a pattern!

Just got back from down under--left the car for 10 days in the garage and my car behaved like nickjhowe--10 miles of range lost per day. You should be able to track it on the Tesla app, so no surprises anymore. Could have plugged it in, but I figured if I left with 220 miles rated, I would have 120 left when I got home.

Ivan--listen to Dennis M. You can save yourself a lot of money. Actually, the .15/kWh is fully loaded. The actual pure electricity charge is only .08/kWh

+1 on the mileage losses.

When I walk thru the garage, you can frequently hear whirring and clicking sounds coming from the car. May be thermal management, but the temperature here in northern AZ has been pretty temperate recently, so not sure that is the main cause of the vampire losses. I think the general rule of thumb for now is don't leave it unplugged for multiple days. Personally, when I need to leave it, I plan to plug in to 110 instead of my HPWC to ensure the 12v battery also has a chance to get more charging time.

I have a separate electricity meter for charging the car and here is a comparison between the electricity consumed and the electricity used to drive the car. I normally drive about 50 miles per typical day and plug in every night.

Electricity meter say I consumed 792 KWh
Trip meter says the car used 571 KWh in driving 1740 miles.

So, the trip meter indicates the car used 328 Wh/mile of battery energy.
The electricity meter says I used 792/1740 = 455 Wh/mile of power off the grid.

(I would post pictures of the electricity and trip meters, but don't really know how to upload them into the forum).

Hopefully, this data indirectly answers your original question, @ivan. Notice that this doesn't deal with the question of vampire load when the car is not plugged in.

@virant, do you have a correlation vs firmware rev?

Last month I was in Washington, with my car outside. Nights near freezing. Now im in san francisco, and the average temp im my garage is closer to 68 to 70. No noticible change in vamp load, I dont think.

Leave it plugged in to a 120V line if 240 worries you. Should be enough to keep the charge level.

viranjit, thanks for sharing your valuable information.
No one here seem to doubt the "vampire" effect, and good metrics like yours help evaluate the scope of it. The very first source I referred to had an estimate of about 179 Wh whereas your data shows about 130 Wh loss. If converted into "miles lost per hour" the above numbers lead to 0.39 to 0.54 miles, which is in-line with other's more empiric observation.
It is significant and I'm sure Tesla will resolve it. Until than do not leave your cars unplugged for more than a week.

Vampire loss is not some inexorable result of keeping the CPU's ready for a fast wake.

TM's chips are similar to smartphones. Your IPhone can maintain hot standby sleep on a much smaller battery with very low drain. As already suggested, the loss is driven by battery management considerations.

When you hear clicks and whirs, that's the sound of relays and pumps working.

Right now, TM is aggressively pampering the battery to protect it from any kind of thermal degradation.

It's overkill, but at production startup, they biased it toward wasting some grid kWhs to overprotect the battery.

As noticed, this also has the effect of spending the battery when you're not plugged in. This will get fixed.

TM is tuning their battery management routines to optimize sleep power rather than simply battery protection.

Over the summer, it should get released after sufficient characterization, and the daily loss will be far smaller.

In extreme climates, the drain will be still be high, but for most people, heavy vampire loss will go away.

bhs1

"So in 14 days, the car "used up" 200 miles of range. I don't want to think what would have happened if I left it for 3 weeks."

Don't worry, Tesla thought about that. The Model S will go in a deep sleep when the battery reaches a very low level and the car can sit like that for 6 months.

Had you come back a few days later, you probably would have found your car in such a deep sleep with the battery perfectly fine.


X Deutschland Site Besuchen