The Mirai’s air conditioning and fuel consumption

1. 𝐔𝐬𝐞𝐬

As with any car with air conditioning, the Mirai’s ac needs fuel to operate. How much fuel does it consume? This question is of interest to fuel economy: what hit does the MPGe take with the ac on? In the Mirai’s case it is also of interest when keeping the car cool using the ac while you’re away: what does it cost per hour, and will sufficient range remain when you return?

If the latter application of the fuel cell seems like a wasteful perversion, using it to power your house would have to seem even more so. You won’t think so if you aim the leftmost vent at the steering wheel before you leave: on your return you’ll be able to hold the left side of the wheel without burning your fingers—no such luck however with the right side that I can see, sadly.

The ac might even save a life if you accidentally forget and lock your kid or pet in a hot car for even a fraction of an hour, see the gruesome statistics for kids at . But only if accidentally: don’t rely on the ac to save a life, as occasionally ac’s fail, sometimes with fatal consequences.

2. 𝐀𝐧 𝐞𝐱𝐩𝐞𝐫𝐢𝐦𝐞𝐧𝐭

To answer these questions I parked in the sun at 2 pm yesterday with the windows up. The outside air temperature (measured in the shade as usual) was 80 F, a typical mid-July afternoon in Palo Alto. (Temperature of anything exposed to direct sunlight is not well defined as it depends very heavily on the nature of the surface so exposed.)

Now pick whichever one of the following two paragraphs is more to your taste and skip the other. 🙂

1. The car got hot inside. The car is 16′ long: 5′ for the white hood and trunk, 5′ for the black roof, and 6′ for the transparent windshield and rear window. Therefore around 20% more solar heat enters the car through those two windows than heats the black roof, and then there’s also the side windows. The white hood and trunk contribute next to nothing to cabin temperature.

2. With the sun’s rays beating down through the glass onto the dash, rear parcel shelf, steering wheel, seats, and other upholstery, all black, these interior surfaces became painfully hot to the touch. And although the windows are transparent to short wave radiation (SWR) from the 5700 K sun and are therefore not heated by it, they are opaque, i.e. black, to the 20x longer wavelength radiation (LWR) from the car’s black interior surfaces at 300-340 K. In this way the solar radiation that doesn’t heat the car’s exterior by virtue of passing through the windows, instead heats all surfaces in the interior including the inside surface of the windows. The radiation trapped in this way accumulates to make the car literally a hothouse (aka greenhouse).

Opening the windows a crack goes some way to allowing that accumulated heat to escape, namely via convection.

Running the ac does a far better job than cracking the windows. (To lock the doors while the ac is on, use the mechanical key, which as a side effect renders the remote inoperative for all but the trunk until you unlock the doors again with the mechanical key.)

But at what cost?

3. 𝐓𝐡𝐞 𝐤𝐞𝐲 𝐦𝐞𝐚𝐬𝐮𝐫𝐞𝐦𝐞𝐧𝐭

To get some idea of this I set the climate control (both driver and passenger) to 75 F, and ran the ac for two hours with the car stationary, with ECO ac off during 2-3 pm and on during 3-4.

The ac consumed a total of 0.1 kg, evenly split between the two hours, hence 0.05 kg/hour. I found no detectable difference with ECO on, either in consumption or passenger comfort, but perhaps it might make a difference on a more cloudy day.

Using 33.7 kWh = 115,000 BTU (in agreement with each other to within a surprising 0.01%) for the energy of 1 kg of hydrogen, that comes to 0.05*33.7 = 1.685 kWh or 5750 BTU. Room air conditioners need about 30 BTU per square foot so the Mirai’s ac would be adequate for a 200 s.f. bedroom. This might seem like overkill for the Mirai’s cabin area of only 50 s.f. but bear in mind that the average bedroom does not sit out baking in the hot sun on a hot road or parking lot trapping LWR: I know of no for kids’ bedrooms.

4. 𝐂𝐨𝐧𝐜𝐥𝐮𝐬𝐢𝐨𝐧𝐬

𝐶𝑜𝑛𝑐𝑙𝑢𝑠𝑖𝑜𝑛 1. If at 65 mph you get 65 MPGe (miles per kg) without the ac on, that means the car uses 1 kg/h. Turning on the ac therefore raises that to 1.05 kg/h, i.e. 5% more fuel consumed.

My 50 mph (indicated) run that I reported here the other day over 43 miles (therefore taking time 43/50 = 0.86 hours) got 100 MPGe with the ac on. That’s 50/100 = 0.5 kg/h, hence a consumption of 0.5*.86 = 0.43 kg. But in 0.86 hours the ac uses 0.05*.86 = 0.043 kg, so 10% more fuel consumed with the ac on than with it off! (This morning I repeated that run at an indicated 57 mph with the ac off and got 82 MPGe, a big drop from 100 MPGe and showing that turning off the ac doesn’t come near compensating for going 7 mph faster.)

So as a percentage the ac hurts more when you’re trying to be economical by driving more slowly. That’s obvious when you consider that the ac burns fuel at a steady rate whereas slow driving reduces overall fuel consumption while also taking longer to reach your destination and so requiring more ac.

𝐶𝑜𝑛𝑐𝑙𝑢𝑠𝑖𝑜𝑛 2. If you use the ac to keep your car cool while parked, 1 kg of hydrogen will last 1/.05 = 20 hours. A full tank can therefore keep it cool for 5*20 = 100 hours. A quick way to see if you have plenty of hydrogen is to look at the gas gauge at upper left of the dash, which seems to be marked off in 8 divisions of about 0.6 kg each with perhaps a reserve in the neighborhood of 1/3 kg, just guessing. So each of those 8 divisions should give 0.6*20 = 12 hours of cooling. But don’t forget to allow enough extra in order to remain well in range of a hydrogen station.

5. 𝐌𝐞𝐭𝐡𝐨𝐝𝐨𝐥𝐨𝐠𝐲

Burning question: where did I get 0.05 kg/h for ac consumption? Certainly not from watching the range decrease. The way the Mirai calculates range is far too inscrutable to infer any reliable kg numbers!

What I did was to scroll right on the multi-function display to (i) (Information), then scroll down to the Drive Monitor and watch the MPGe going down for one or both of TRIP A and TRIP B (selected on the left of the steering wheel). This works provided there are already some miles traveled on that meter (shown at the lower far left), otherwise MPGe will remain at zero. The more miles the better the precision; fortunately A had 145.6 miles though B only had 28.5. Neither changed during the experiment because I didn’t go anywhere, but their MPGe’s in the Drive Monitor kept decreasing during the 2 hours of ac operation.

Here’s the key bit of magic for obtaining the figure of 0.05 kg/h.

𝐌𝐢𝐥𝐞𝐬 𝐝𝐢𝐯𝐢𝐝𝐞𝐝 𝐛𝐲 𝐌𝐏𝐆𝐞 𝐠𝐢𝐯𝐞𝐬 𝐤𝐠 𝐨𝐟 𝐇2 𝐜𝐨𝐧𝐬𝐮𝐦𝐞𝐝 𝐬𝐢𝐧𝐜𝐞 𝐫𝐞𝐬𝐞𝐭𝐭𝐢𝐧𝐠 𝐭𝐡𝐚𝐭 𝐦𝐞𝐭𝐞𝐫

So as MPGe declines, that quotient steadily rises, reflecting the ongoing consumption of hydrogen. The amount it rises in one hour is the amount of hydrogen consumed per hour. Need I say more? (No, but that never stopped me before. 😉 )

6. 𝐔𝐧𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲 𝐚𝐧𝐚𝐥𝐲𝐬𝐢𝐬

As MPGe is only given to 3 digits of significance you get very little accuracy after just 10 minutes. An hour of ac however is enough to tell that consumption per hour is in the range 0.045 to 0.055 kg/h and that 0.05 kg/h should therefore be good to about 10%. (One way to improve accuracy is to wait until each change of the MPGe and immediately record the time on a stopwatch to the nearest second, but I didn’t think of that until later.)

Following my usual practice I had reset TRIP A after the last fill-up. I noticed that when 2.5 kg had been consumed the gas gauge at upper left showed slightly less than half empty. This suggested that “half-empty” on the gas gauge was not really exactly 2.5 kg down but perhaps only 2.4 kg down. Since the gauge is divided into 8 parts, that suggests 0.6 kg per division rather than 5/8 = 0.625.

7. 𝐂𝐚𝐯𝐞𝐚𝐭𝐬

All the above numbers are only preliminary – my estimates may be inaccurate, and your mileage may vary. More data from other Mirai owners might improve these numbers, or for that matter from Toyota who may know them by now to three decimal places.

Also I’ve assumed a perfect odometer but from what I’ve seen so far the odometer is not calibrated to be spot on until the tires are almost bald, at which time one wheel revolution will take the car forward about 78.5″. It therefore must be reading about 3% low when the tires are new, taking the car forward 81″. (Some day someone will invent a more accurate odometer that’s independent of tire wear; in the meantime I rely on GPS for accuracy.)

And lastly many manufacturers add 2 mph to their speedometer reading.

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s