I remember a year ago, Martin posting on his now defunct TeslaFounders blog about the fridge. I was so impressed that I cut and pasted it into a document that I kept. I thought we had posted it on the TM site too, but there was a time where blogs were being taken down so that might have been a casualty. I noticed that on this blog it was just referenced through a now defunct link. I will repost it here because of the controversy noted in the "How are charging kWh measured?" post under Performance and Technology in the Tesla Motors Owners site in case you are interesting in visiting that. -Ian O
Filed under: Energy, General, The Founder's Car — mfeberhard - Oct, 2008
Details matter. You can get a lot of things right, but one seemingly little thing wrong can make a huge difference. Such is true with the energy consumption of my Tesla Roadster. Let me explain.
Soon after I got my car, I noticed a funny thing: the ESS coolant pump seems to run all the time. Even when the car is off. Even if it has been off for a long time. Even when the car is plenty cool. You can hear it run, and many people have commented about the noise of the pump and the noise of coolant gurgling into the overflow reservoir.
The Tesla people tell me that when the battery is half-way discharged, and the car is off, and the ESS is cool, then the pump will shut off. Fine. But my average daily drive is less than 60 miles, and I have only driven far enough to drain half the battery 4 times since I got my car three months ago. So, except a few hours on these four occasions, this poor pump has been running 24 hours per day, 7 days a week for three months solid.
I noticed another funny thing: if I charged the car up and let it sit for a few days without driving, I found that the battery was no longer anywhere near full. I guess that the constantly-running coolant pump, along with its support electronics, is draining the battery.
Curious, I installed a real electric meter ahead of my car’s charging station, as noted in a previous blog. This meter simply accumulates a measure of the energy passing through it on the way to charging my car. It is exactly like the power meters that electric companies use to measure what we use for our electric bills.
I’ve had this meter in place for a month now, and the data it has so far collected is revealing.
How much power does this coolant pump draw? To find out, I charged the car up and read the meter. Then I unplugged the car and let it sit for 4 days. Then I charged the car back up again, and read the meter once more to see how much energy the car had consumed those four days of sitting in my garage.
Whoa! The car consumed a whopping 14 kilowatt-hours in four days, just sitting there! Doing the math, the pump draws about 146 watts all day long, all night long, every day. This works out to 1,278 kWh per year. To put this in perspective, a really nice 26-cubic foot side-by-side refrigerator with an ice maker uses only 618 kWh per year, so this pump is the energy-equivalent of two huge refrigerators!
Okay, so with this pump running all the time, how much energy is my car really using? Since I installed the meter, I have driven 999 miles, and the meter registered 439 kWh consumed. That works out to 439 Watt-hours per mile, way higher than I ever anticipated.
By subtracting out the time that I was actually driving the car, I can calculate how much energy the pump consumed while the car was parked. This works out to 96 kWh of wasted energy in the month since I installed my meter.
Subtracting this wasted energy, the car’s energy consumption at the meter is only 343 watt-hours per mile. A little high, but in the right ballpark. Here is the kicker: 22 percent of the energy consumed by my car happens while my car is parked! Twenty two percent. Imagine that.
(How do these numbers compare to the energy consumption reported by my Tesla’s onboard computer? At the end of each day’s drive, the computer reports energy consumption between 260 watt-hours per mile and 310 watt hours per mile, with an average somewhere around 280 watt-hours per mile. This is a little higher than other drivers might experience largely because I live at the top of a 2,400-foot hill, so every drive includes a decent hill climb. If I subtract the energy wasted by the pump from the meter readings, and compare that to the energy consumption reported by the car’s computer, I conclude that the charger is about 87% efficient. Not great, but not bad either.)
Aside from the shame of wasting 1.3 megawatt hours every year to gurgle fluid through my car, this brings two questions to mind: The first is battery life. 22% of the energy cycled through my Tesla’s battery is not used to propel the car, and the cycle life of the battery is 22% shorter than if this pump did not run when the car is off. As much as I have promoted lithium ion batteries for cars, cycle life is still a difficult problem for batteries.
Let’s do the math. Let’s say these cells are rated for 500 full cycles, and let’s say the car goes 220 miles on a charge. That means that the cells will reach the end of their life in about 110,000 miles. Not perfect, but not bad either. Now let’s load on an additional 22% of energy to run the coolant pump. Now the batteries will only make it 90,000 miles. That little pump will cost me 20,000 miles of precious lifetime driving range for my Tesla! The expensive Tesla ESS will expire a year or two sooner just because of that pump. Ouch!
The second question is the life expectancy of the pump. I expect that Tesla used an automotive-grade pump from a good supplier. I am also sure that no other car leaves a pump running 24/7. Consider a typical car designed to run for 200,000 miles at an average speed of 30 mph. Such a car is designed to run for 200,000 / 30 = 6,666 hours. Let’s say the designers want some room for error, and design the water pump for that car to operate for 10,000 hours without failure. 10,000 hours life expectancy would be a good-quality automotive pump.
Now, let’s run that same pump 24/7 instead of the couple of hours per day it would run in our typical car. Running 24/7, that pump will pass 10,000 hours in only 13 months. That’s all – end of life. Just to make it through Tesla’s 3-year warranty, that pump would need to last 26,280 hours without failure. To last just 5 years, the pump would need to run 43,800 hours. Hopefully, Tesla installed a pump rated for at least 50,000 hours of operation without failure, implying an MTBF of at least 70,000 hours, assuming an exponential failure distribution. Does any automotive parts manufacturer even make such a pump?
Maybe this constantly-running pump is an artifact of the temporary drive train in my car. Maybe it will shut off when not driving after Tesla finally installs my 1.5 drive train. I sure hope so, but the Tesla people have told me that it will run the same way with the 1.5 drive train. What a waste that would be – a waste of energy, a waste of batteries, a waste of reliability.
Still Fun to Drive! --ME
