Prius Heating/Cooling & Fuel Economy

Yes, thanks for reminding me. There are two 400W heating elements in the air vents under the dash. I noticed them for the first time on Saturday. As I'm sure you know, in a Canadian winter 800W isn't going to heat much! But it was just "cool" on Saturday morning and I had left the A/C on so I got some heat immediately when I started Pearl.

In modern cars with "climate control", such as the Prius, "A/C" means both heating and cooling.

Lets start with the "cooling". The compressor is variable speed and runs off the traction battery (HV). If it runs long enough without the engine running it will cause the engine to start (stopped waiting for someone, for example - don't use "N", use "P" - the battery won't charge when you are in "N"). I haven't noticed it start the engine while waiting for a light.
It is purely a cooling device.
If you are very observant you will notice a slight drop in mileage in the summer from energy lost to cooling the interior, but you want the interior comfortable, as the traction battery also uses that air to cool itself.

On to heating. Prius uses a "normal" in-car engine coolant heater. In winter it will cause the engine to run much more than it would in summer. It's a good idea to block the grills to help keep the engine warm. I recommend getting a Scangauge II so you can accurately monitor the coolant temp. if you do block the grills. If you do not block the grills the Prius becomes a "cold winter car". It just will not warm up. But even with grills blocked, you also will not be able to warm up the interior by letting the car idle. You HAVE to drive a few blocks. Pearl blows nice hot air after about 3 blocks at city speeds, even when it's -30C ambient. The engine is so efficient you will see the coolant temp drop at lights if the ambient is below -20C. So I suppose it could be argued it's not "waste heat". I'm not sure if the two 400W heaters stay on continuously, but they do go off if the cabin temp reaches or gets close to the "set point" (the temp you set the auto climate control to).

 

I'm completely sure that at 50 kilometers per hour open windows is better than air conditioning. I suspect that the crossover speed is quite a lot higher than that, probably higher than "highway" speeds.

I've seen astonishingly little numeric evidence of the windows versus air conditioning comparison, and have never seen any acknowledgment at all of the option of opening the windows less than all the way.

Here in New Mexico, so long as the car has not sat out in the sun first, our famous "dry heat" means that just opening two windows on one side about 2 inches provides a good cabin air circulation at higher speeds. I seriously doubt the drag from that is worse than air conditioning up to the absolute maximum speed of the car.

Windows versus air conditioning is a pet peeve of mine and I have intentions of doing real tests on my Audi A4 before summer is over. My personal opinion is that the great majority of people who were driving in a condition where some degree of open windows makes them actually comfortable will find that option gives them better mileage than using the air-conditioning.

For the Prius, and not other cars, a complicating factor would be the balance between gasoline saving for higher cabin temperature vs. loss of ultimate traction battery life. But I suspect that in highway cruise in non-hilly country the traction battery is not in much need of cooling assistance.

 

I find that the best arrangement for me, alone in the car most of the time I admit, is to keep the driver side window open and the right rear window cracked a couple of inches. I find the airflow cools the cabin as much as I would expect, and I imagine the airflow providing cool air to the battery cooling intake.
I've kept myself reasonably comfortable, and have used the A/C only twice this summer.
I average 3.8 liters per hundred k in the summer.

 

I've done some simple tests to see how much drag open windows causes. It's a lot. I haven't done drag vs speed tests, but I suspect above about 70 km/hr the drag is a greater loss than the power fed to the compressor motor.

My tests show that on a particular very long downhill (mountain decent) the car will coast well in excess of 130 km/hr windows up and compressor on or off. The Prius takes off on downhills so be careful. Use "B" on long ones to control your speed.

Windows all the way down limited the descent speed to about 110 km/hr (still well in excess of the 80 km/hr posted limit - so what's that, a $500 fine?).

Using "B", of course, limited the speed to around 90 km/hr. But I was testing the aerodynamics, honest, officer!

Oh, and if you think this windows up for efficiency thing is Prius specific, it is not. The anti-Prius (2001 Nissan Pathfinder) worked the same only better. I could hold its' speed on that downhill to 95 km/hr with just the windows open! And it weighted a -lot- more than Pearl!

One last thing. You WILL want to block the front grills in the winter. My experience with the Scangauge II as monitor last winter showed it's safe up to about 10C ambient. At that temp and above you should have the Scangauge to monitor the temps. We use pipe insulation to block the grills. Simple, inexpensive, and effective. I use zip ties to hold mine in, but lots here just use the friction of the foam insulation in the grill. Easy to insert and easy to remove. Bonus, it protects the rad and A/C coil from salt corrosion, something we don't have to worry too much about here in Alberta, but you do in the east.
You will also want good winter tires. I use Nokian WR all year round. Your stock tires may be ok, as the Canadian cars are now all "touring models", so you got reasonable tires, not the junk Goodyear Integrites my car came with last year.

 

The Prius MPG Simulator performs modeling using a variety of user-specified criteria, including speed, outside air temperature (OAT), automatic climate control (ACC), and the following combinations of window positions: all up, all down, two halfway down, and two diagonally opposed 1" down.

The chart below shows results with 90F (32C) OAT, with and without ACC set to 75F (24C), and all window position models. The models assume the following:

• Steady speeds
• Level terrain
• No wind
• 190' elevation
• 30.00 barometric pressure
• Smooth dry pavement
• Michelin Hydroedges inflated to 50/48 PSI
• E10 fuel
• 300# load, driver + cargo

(Forgive me for not doing all the metric conversions. I speak US measures. )

The conclusions:

• All windows up and no ACC gives the best results across all speed ranges -- for the brave soul who can tolerate it.
• Two diagonally opposed windows 1" down is next best, and might be tolerated by some at high speeds.
• Two windows halfway down is superior to ACC at speeds below the low 50s in MPH (low 80s in km/hr).
• All windows down is superior to ACC below 40 MPH (64 km/hr).

Specific values of course will change somewhat with different combinations of typical summertime OAT and ACC thermostat settings, but I doubt the overall relationships change much.

 

Nice Jim. Just one correction, as 50 MPH = 80 Km/hr the cross over point is in the 70s km/hr area. About where I guessed.

Oh, and you still want the interior air to be "comfortable" so the battery can cool itself. So "toughing it out" with no A/C is probably not a good idea. It might overheat the traction battery and that would definitely shorten its' life.

 

Even here in Phoenix I had it half blocked last winter and noticed a good improvement. I know people that use block heaters here in Phoenix as well and notice a difference. The Prius is a unique beast since the engine keep shutting off while your driving. Once that happens it cools off pretty fast, and then you have to run the ICE just to get it warmed back up. That eats into your mpgs pretty quick.

Rob

 

That's interesting. Do you have any idea what the simulator used for input information to provide the drag estimates that went into this?

I feel quite sure it's not running a computational fluid dynamics simulation of the car in real time, so rather likely it has a numeric model somebody put together. Unless the model is sound the results are just a picture of a guess.

The post from David Beale is about as close to real data as I've seen, while even for that I would caution that the comparison needs to be fairly closely at the same wind condition and that one can't measure the A/C effect that way. As the test seems to have been on a standard route for him, I have hope that he was sensitive to that concern.

 

Yes, the simulator uses mathematical modeling, but supported with extensive real world test data. From the site I linked previously, the words of the developer:

"I had to fit two different Prius vehicles with anemometers, wind direction equipment & data loggers for more than 30,000 miles (people thought I was a storm chaser) to come up with the formulas & factoring values used in this program."
​

I suggest giving the site a good read. That should give you a pretty good sense of his methodology and thoroughness.

 

I suspect your car -has- a block heater. When I purchased Pearl last year, as I was about to leave, I remembered and asked if she had it installed. The salesman replied "Of course it has a block heater!" It's a 400W device that plugs into a cavity in the aluminum block.
My grill blocking is considered here as uglier than some others. It looks better if you find the 3/4" soft stuff used for air conditioning and cut it in half lengthways, then fold those in half and insert between each grill gap. I just used the ethafoam stuff, 3/4" for the top grill uncut, and I think it was 2 1/2" for the lower grill uncut, both gray from Rona. Two zip ties for each. Pic attached.
radblock2007 small.jpg

 

I try not to run the AC and generally open the driver's window 3 centimeters and the right rear passenger window 3 centimeters. I try to imagine air flow through the vehicle and into the rear vent. All windows closed makes the car too stuffy, even with various configurations on the MFD. Only on extreme cold or hot days are all windows up with the AC or heater with air recirculating. Even in vary hot and sun intense conditions, conditions are tolerable with the cross-window configuration and not running the AC. I have a block heater for winter use (early morning pre-warmup). I ordered the BH from Canada and had to argue to convince the local dealer to install it (who saw no need for the BH whatsoever).

 

Actual tests on my A4

Today I conducted some mpg testing on my 2002 Audi A4 1.8T. I'll describe the methods and results in some detail because I have in mind doing similar testing on the Prius.

I compared mpg in fifth gear at speeds from 30 through 80 mph, at a standardized summer air conditioning condition:


I compared most interesting windows/Aircon configurations at 70 mph, and compared two at 30 and 40.


Regarding the configuration questions so often debated:

1. For this car under these test conditions, 70 mph was well under the crossover point for moderate airconditioning to give better mpg than even full open windows.
2. My preferred two windows opposite side partial opening had so little additional drag at 70 mph that it was below the measurement resolution of my test (I observed an 0.1 mpg improvement--presumably measurement reproducibility error).
3. The air conditioning penalty, by contrast was easily detected, amounting to about 10% even at 70mph, and much higher percentage at lower speeds.

My conclusions and advice:

1. At city speeds, if you find fully open windows make you comfortable, don't think you are wasting fuel by preferring them to aircon--they are much better.
2. At highway speeds, even fully open windows may be competitive (on my car they win), and if you can find comfort with a pair of moderately open windows, you likely win over aircon at all conditions at which you are likely to cruise.

Method Details:

The test site is on U.S. route 60, about 10 miles east of the I-25 intersection at 34d25mN, 106d39W, altitude 5200 feet msl. It is typical New Mexico U.S. route--two medium width lanes with very small paved shoulders, decently paved and graded.

The site was chosen from Google Earth, where I searched for a level location with good sight lines, opportunities to turn around, and low traffic.

It was an hour drive from Albuquerque. I preconditioned the interior (and thoroughly warmed the engine) during the drive down, so first ran the tests at a comfortable automatic aircon setting. The controls were set to 69F, though the actual cabin air temperature was considerably warmer than that. The Audi cycles the aircon compressor unless it is maxed out. I believe I had a stable condition in the range many would find comfortable, though far too warm for a Texan.

I standardized each data point, averaging mpg over 60 seconds using the Audi's excellent trip computer function once in each direction. The West-bound runs all started at an obvious visual marker--the only speed limit sign in miles. At the end of the first West-bound run at a given speed, I created a GPS marker, which I used to match the road used for the East-bound run. I stabilized speed before the initial point and engaged the Audi's magnificent cruise control. Nearly all runs at a given speed (in particular all the 70 mph runs) used the remembered cruise control set point from the first run at that speed. I relied on the displayed value (mentally averaged for the few tenths of mph variation) on a Garmin GPS V running with WAAS engaged for speed. Most runs appeared to be within 0.2 mph of intended speed.

The 2-way method helped compensate out error due to my failure to find perfectly level road (this is not western Ohio, nor even Kansas) and any wind. In the event, wind appeared to be negligible during the test. The tall grass was not waving, which puts it well below 5 mph. The ambient temperature reported by the Audi varied from 90 to 92, but was almost always at 91.

I did a few primitive tests of reproducibility, which suggested that within this couple of hour period my results might be reproducible to about 1 or 2%.

On the trip down, I compared comfort from the left forward, right back window partially down to my own preferred counterpart, which leaves both driver side windows full up, and lowers the two passenger side windows a little. At equal opening I found I prefer my method, which to my taste gives a broader cabin circulation, with less annoying blast right at my left ear. I doubt the air drag differs appreciably. I carried a bit of scrap wood to calibrate the window lowering to 2 inches--perhaps 30% more than my usual actual practice.

I think the Audi (and my former BMW 325i) aircon is on the weak side. Higher output units run at full on other cars might well impose larger mpg penalties. This Audi is pretty slippery (though less so than the Prius or the Insight). Most other cars would probably have a worse degradation of mpg with speed.

I'm going to this painful detail partly because I'd like testing results on the Prius to be creditable and credited, so seek feedback on things which trouble you, or just reasonable improvements.

My own intended changes and improvements:

I won't again do the multi-speed testing with automated aircon--it is clearly the least reproducible of the configurations. Even though my method averaged 120 seconds of operation, the fraction of compressor on time probably varied appreciably. I'll use the 2 inch down for passenger side configuration instead.

I'll drink water during the test next time.

 

All to often, turning off the a/c has as adverse impact on MPG, because as the traction battery heats up, the less electric only operation (stealth and warp stealth) potential you're going to have. (Which of course, is where the staggering fuel economy comes from).
That far outweighs the negligable, indirect decrease in economy by running the A/C. Especially in warmer climates.
Does the Prius Sim take battery temp potential into account?

 

Hi David, where did you get the ethafoam from? I went to Home Depot and all they have is a self-adhesive black stuff.

Thank you.

 

I got it from Rona, but it's the same as the black self-adhesive stuff from Home Depot. Except Home Depot didn't have any in stock when I looked last fall :noidea: so I went to Rona. They had it in gray, and it matched my car. They also had a better selection of sizes. I was shocked.

 

I guess it was just my dealer (Kingsway Toyota). NOT installing an EBH in a new car is just being a cheapskate. And you can quote me! They have a generous profit margin when they sell a Prius here! There is no good reason to cheap out and not install an EBH. It is -needed- in Canada, though I will admit I've never actually used mine. I'm retired, and can't predict exactly when I'll be driving Pearl. She sleeps in the unheated but well insulated garage. The coldest I've ever measured it in there is -10C. That was after a few weeks of -30 to -40C weather. Yuk! Last winter it never went much below -30C. Tropical!