I think what it comes down to is there is really no good place if you want a true air temperature.
Anywhere on/underneath the vehicle will be influenced by heat radiation from the roadway. Fresh blacktop in the summer sun will be much hotter and even in the winter the sun will make the roadway warmer than the air just a bit off the road.
Mike Romain proclaimed:
... or a bridge or other structure that allows chilled air to get under the roadway. Or coming around a corner at the end of a nice sunlit stretch of dark pavement onto an area where water has flowed across the road and frozen on the shady sections--or worse, only the bottom is frozen and there is free water on top of it. Or the road just changes color...
Within reason, the best location is near the front of the vehicle. Paying attention any time the temp gets within a few yards of freezing is a driver issue.
I'd sure use a good filling, non-hardening non-water based adhesive and thoroughly fill any places where the wiring goes thru bodywork, has a connector, etc.
Lee Ayrton proclaimed:
With very few exceptions, the air is always colder than the surface, even beginning as close as a few centimeters above and or below the surface. Over solid ground, in all my years measuring these trivia, even when you think you might, if there was any insolation at all, the surface would be warmer than air just a few CM above it. Frozen ground on occasion would be colder, but generally lightly colored with no insolation. Over areas with canyons and other places where air from well beneath the roadway can get near the roadway, you are more likely to run into colder surfaces. If the minor degree difference in mounting location is taken by the driver as anything significant, that driver is a sliding accident just waiting to happen. There is more temp difference along a single mile of typical roadway than any trivia vertically.
billy ray proclaimed:
"True air temperature" is tricky to get with any sensor. Sufficiently so that the weather bureau has guidelines for constructing an enclosure that protects the sensor from any direct radiation yet allows free air flow but goes on to note that you need to then force a small air flow past a dry sensor before taking a reading. The enclosure is to be mounted at a standard height above the surface which cannot be rock, asphalt, etc. And then you are expected to calibrate your equipment periodically against precision thermometers and even then although the temps are reported to a tenth of a degree they are really accurate to only a degree or two. Once you realize all that, plus the general cheapness, lack of inherent accuracy, and lack of calibration capability at multiple points across the temperature curve of the typical home and/or automobile sensor, the location becomes pretty much more self delusion than worthwile effort. In the air stream but blocked from direct high speed air flow and moisture is about as good as you are gonna get. Calibration is possible on some units but is generally done only at one point--which if it were my sensor would be 32 degrees F. Then realize that if the temp is 40 or lower, it is a good idea to be prepared for icing on all but the most unusual real roads.
The temp over a dark roadway will be hotter, although the vehicle is moving and there may or may not be enough other vehicles to stir the air enough to eliminate the differences except on very wide spans of asphalt like the old runways at Camp Roberts. Anyone who relies on an auto temp sensor for the degree of accuracy to really predict icing is kidding themselves anyway. Anything below 40 or so is dangerous.
I have one of these
The sensor.. well I had a heck of a time with that.
First I tried over one of the doors and discovered that engine heat gets channeled up the door frames.
Next I ran it all the way out and put it just behind the grille. Great when moving but sitting still or parked and the engine heat threw it WAY off.
Finally I got some insulating foam to keep the sensor from directly contacting the bumper and mounted the sensor on the bottom of the plastic bumper guards as far out in front as I can.
Direct sunlight (on hot days) still heats the bumper and gives you a false high temperature reading but the rest of the time I get about as accurate a reading as you can reasonably expect. Drifter "I've been here, I've been there..."
I guess I should have better stated my thoughts.
On the Norwood Lateral (Ohio 562) there is, within a mile of the ramp to I-75, an outdoor time/temperature sign at the rear property line of a local business.
When I worked evenings I would drive past that sign on the way home (after midnight in most cases) and the temperature shown was almost always within a degree or two of the temperature shown on the dash on 3 of my last 4 vehicles (the only ones with "outside" temperature displays)
On days when I worked mornings and got off work at 5 PM I would go past that sign and the temperature shown on my dash would always be higher.
The greatest difference I have noted was 30 degrees higher on my dash than on the sign. (please recall I worked evenings 80+% of the time)
If there was heavy traffic or an accident so traffic was crawling you could watch the dramatic drop when you would edge underneath the overpass and you and the pavement were shaded from sunlight.
I could continue this but Bobbie's "Welcome back to the world" shindig starts in 35 minutes and I have to get back to getting the food ready. needless to say his mother has been a wreck since last December....
That's why a/c iciing indicators basically ignore the temperature and use air flow through a calibrated port to detect it. Your 40 F number is about right - carb or intake ice detection starts about there and it only takes once to get your undevided attention.
I'm with Mike, I think you want it closer to the road so you get an idea of what life is like where the tires are. The tires can see ice when the temp is upwards of 37F at the road surface. If the probe is up behind the spare, then the temp displayed could be upwards of 40F when there can still be ice on the road.
Ice can form on bridges and in shade if the car thinks it is 37, or below. Obviously, this gives a margin for error, but the system doesn't respond quick enough to detect actual conditions for ice, so they set an alarm on my car that chimes at 37, then comes on and stays on at 32. That is, it lights a display that then goes out at 37, then lights the display and remains lit at 32. Both temps sound a chime for a couple of seconds.
Nice discourse folks on a topic that we have not discussed for a while. For me, I pretty much know when to look out for icing and where/how it lays and such, so I am really looking at this as more of a novelty/passenger toy than a life-critical device. Thanks to all the folks for their thoughts on where to put this thing. The link that Drifter posted,
MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.