Send me a note:
darin AT metrompg D-O-T com,
Reflections on side mirrors: testing drag vs. MPG
Posted Thursday, August 31/06 in Mods & Tests
Mirror, mirror on the door -
Sorry. I won't do that again. But that is indeed the topic of this installment of what sadly appears to be becoming a once-monthly update at MetroMPG. (I will try harder in September - it's not for lack of subject matter, believe me.)
My desire to find concrete information about the effect of side mirrors and fuel economy was piqued from seeing the number of high-efficiency concept cars that substitute video cameras for side mirrors. Why? Is it just for the techno bling factor? Or are there real gains to be made by rethinking side & rear vision?
Mirror drag overview ...
High-efficiency production cars go without ...
In addition to concept cars, a couple of iconic high-efficiency production cars also eschewed door mirrors - on the passenger side, at least. But the question is complicated somewhat because the cars in question - the Civic VX and the Metro XFi - were also fairly low-buck vehicles, so the reason they went without could easily have been cost savings. Remember that as recently as the 90's when these cars were available, passenger side mirrors were optional on many cars, whereas today both outside mirrors are equipped on even the cheapest car.
Mirrorless concept cars - splitting hairs?
Looking at any group of slippery concept vehicles, it quickly becomes apparent that the lack of mirrors is no coincidence. Whether it's the Ford Probe V, all three of the American PNGV 80 mpg concept cars, or the more recent Daimler Chrysler Bionic concept, side video cameras are unquestionably the norm.
That would seem to seal the deal, then, wouldn't it? Mirrors must have a measurable effect on Cd. The question then turns to "how much of an effect?" After all, aero engineers love to leave no stone unturned in the pursuit of perfection. So are the side mirrors pebbles, or boulders in the stream?
References to side mirror contribution to Cd ...
With a bit of Google sleuthing, I discovered several credible references to the question. The first, relating to GM's PNGV Precept:
"Instead of outside side mirrors, the car uses tiny cameras — a set-up similar to that seen on the Cadillac Evoq concept car unveiled at the 1999 North American International Auto Show. "We knew we couldn't afford the 30 counts of drag, so we went with camera mirrors," said George Claypole, Vehicle Integration Engineer for GM's Advanced Technology Vehicles. - Automotive Engineering International Online
("30 counts of drag" is Cd .03)
Next, mirror drag is quantified at 3-6% of total vehicle drag for a modern car in a report called Contribution of different devices to the total drag (PDF):
"A very well designed mirror increases the drag of a car of approximately ΔCd=0.012. But this value can reach 0.025 to 0.030 for the worth [worst] designs. Actually it seems reasonable to think that the average value for ΔCd is of around 0.015"
Estimating mirror MPG effect on a Metro ...
According a spreadsheet tool for figuring aerodynamic and rolling resistance, my car theoretically gets 56.7 mpg (US) @ 55 mph (using Cd: 0.34, frontal area 1.858 square meters, @ air temperature 25 degrees C). That's really not far off my actual fuel consumption at that steady state speed.
Removing one mirror: theoretically, using the most conservative figure (the "very well designed" ΔCd=0.012) from the Chalmers report, taking off one mirror reduces the Cd by .006. Frontal area is also reduced by about 0.0375 square meters.
MPG at 55 mph / Cd (delta Cd) / configuration
56.70 / .34 (-0.0) / stock, both mirrors on
= +1.71 mpg US / 3% - theoretical fuel economy gain from stock.
Worth trying? Sure!
Drag-reducing mirror configuration - legal considerations
While reducing drag is a worthy pursuit, I still want to be able to see to the rear and side of the car. Also, I don't want to break any laws by changing my mirror configuration, so the first question: what are the legal issues of mirror use where I live? A quick search revealed that in Ontario:
66. (1) Every motor vehicle other than a motorcycle shall be equipped with [...] a mirror or mirrors securely attached to the vehicle and placed in such a position as to afford the driver a clearly reflected view of the roadway in the rear, or of any vehicle approaching from the rear. R.S.O. 1990, c. H.8, s. 66 (1). - source
So it would appear that legally, a side mirror isn't required (unless the inside mirror is obstructed).
Mirror re-configuration options
While the concept car designers head for the video camera parts bin without hesitation to solve this problem, there is the issue of cost and complexity that makes that route a little less practical for the average tinkerer.
One option is smaller slightly convex racing mirrors, as was retrofitted on this car:
Another option is to install a pair of folding mirrors (to fold for highway trips), and add a smaller convex mirror to the aft end of the folded driver's side for use when folded (and shoulder checks to the passenger side).
To compensate for folded or removed outside mirrors, there are a number of internal mirrors that offer an expanded or panoramic view:
In the end, I decided to completely remove the passenger mirror and install a folding mirror on the driver's side to replace the OEM mirror (which was a non-folding style). In combination with a couple of convex mirrors added inside the car, this retains most of my side mirror functionality, and preserves the option of using the left mirror for rear-view duty on those rare occasions I'm carrying stuff that blocks the inside rear view.
Blackfly mirror swap DIY
I replaced my stock non-folding but otherwise aerodynamically shaped driver's side mirror with a folding but less aerodynamic mirror from a 1993 Suzuki Swift 3-door.
It took a little cutting (tin snips & hack saw blade) and drilling to make it fit, and I also had to add a piece of black plastic to fill the gap since the mirror base is a different shape than the OEM mirror). But all the modifications are hidden & reversible. Should I want to revert to original condition, I'll be able to with no problem.
However, it became clear pretty quickly that while the folded mirror might present less area to the wind, the open hinge was an aerodynamic disaster. So I drilled a hole on the top hinge to receive an hinge "insert" which I made from a piece of sheet metal shaped & contoured with body filler to be somewhat aerodynamic (moreso than a plain flat edge, anyway). To install the insert, I simply drop the "pin" (a bolt) into the hole drilled through the mirror hinge. I can easily install/remove this while sitting in the car.
I have been trying out a few cabin mirror variations to compensate for the removal / highway folding of the outside mirrors. A round convex mirror mounted on the left A pillar at eye level (inside the car) preserves the view to the left side and left rear. An "extension" to the right side of the windshield-mounted center mirror plus an additional round convex mirror inside the glass on the same sight line to where the outside mirror was brings back the right side & right rear view.
Experiment: testing the new configuration ...
Route & conditions
Recently I had near perfect weather conditions for doing an experiment that was likely to uncover only small results, if any: light north winds, which are blocked by a tree-lined ridge that runs along the north side of my usual "test" road, leaving the route in nearly still air.
Air Temperature (ATMP).. 67.6 F / 19.8 C; Aug 23/06; 7:30 - 8:30 PM
I did an A-B-A comparison, where A = both mirrors on, and B = passenger mirror off (with the hole covered) and driver's mirror folded, with the aero hinge insert in place.
Speed was 88 km/h / 54.7 mph, with cruise control set once before entering the test section, cancelled with the brake and resumed for each pass. (Yes, it was an odd speed choice. I aimed to set the cruise control at 90k, but didn't quite make it.) Runs were averaged bi-directional to cancel any effects of wind/grade.
@ 88 km/h / 54.7 mph - in MPG (US)
A/B ... West ... East ... Average MPG (US) / L/100 km / MPG (imperial)
A ..... 55.0 ... 55.3 ... 55.2 / 4.26 / 66.29
55.24 mpg (US) - average of all A runs
1.27 mpg (US) - average difference of between all A and all B runs
Not quite the 3% improvement predicted by my (imperfect) aero/rolling resistance spreadsheet - and I even had a slight "advantage", since those calculations were for removal of only 1 mirror, where I effectively removed one and a half. And it's certainly not as high as the GM numbers would have predicted. Still, it's a pleasant surprise. I'll be keeping my new mirror configuration.
darin AT metrompg D-O-T com, or here