Subaru Forester Owners Forum banner

1 - 20 of 22 Posts

·
Registered
2015 Impreza 5-Door Sport Auto (CVT)
Joined
·
2,351 Posts
Discussion Starter · #1 ·
The outside temp readout is about 1 to 2 degrees lower than it should be (compared to an accurate mercury thermometer someone loaned me).

Is there a way to recalibrate the outside temp sensor?
 

·
Registered
2010 2.5X Limited 4-speed Auto
Joined
·
1,979 Posts
I really doubt that Subaru expects anyone to think it is a highly calibrated instrument. In reality +/- 2 degrees is pretty accurate. FWIW, unless your borrowed mercury thermometer is a calibrated lab instrument, you have no idea how accurate it really is. it's probably not much more accurate than the '14's sensor.
 

·
Premium Member
2019 Crosstrek 2018 XT
Joined
·
14,713 Posts
I seem to recall the answer is "no" I am gonna guess that the difference is in the location and the inaccuracy of the mercury instrument.
 

·
Registered
2015 Impreza 5-Door Sport Auto (CVT)
Joined
·
2,351 Posts
Discussion Starter · #5 ·
I seem to recall the answer is "no" I am gonna guess that the difference is in the location and the inaccuracy of the mercury instrument.
Huh?

I stated that the mercury thermometer was accurate, not your Dollar Store thermometer. I would trust the accuracy of a mercury thermometer more than an electonic digital thermometer.

I guess the accuracy is like what Majorlk stated, +/- a few degrees either way.

If it can't be recalibrated I will just add a few degrees when reading the temp.

Just thought there may be a way to adjust it.

It reads colder than what the actual temperature is after driving the vehicle.

At starting the readout is always higher than the actual temperature.
 

·
Registered
2010 2.5X Limited 4-speed Auto
Joined
·
1,979 Posts
Not to be argumentative, but I have four different "accurate" mercury thermometers and no two of them read exactly the same, and one of them IS a calibrated lab-grade instrument. :)
 

·
Registered
2014 Forester 2.0XT CVT
Joined
·
366 Posts
If the Fozzy was an oven and I was baking a cake, a few degree temperature variance would be a concern. But for a general outside air temp, a few degrees is no big deal. I don't think any vehicle brand would build a calabration process into a basic outside air temp thermometer. How do you justify the cost?
 

·
Registered
2015 Impreza 5-Door Sport Auto (CVT)
Joined
·
2,351 Posts
Discussion Starter · #8 ·
If the Fozzy was an oven and I was baking a cake, a few degree temperature variance would be a concern. But for a general outside air temp, a few degrees is no big deal. I don't think any vehicle brand would build a calabration process into a basic outside air temp thermometer. How do you justify the cost?
You are basically correct.

But.

If the readout stated 34 or 35 degrees and the actual temp was 32 or 31 degrees and the roads were icing over, I would not be looking out for icy spots because I would think it is too warm for the roads to freeze up, but after a spin out or near collision from sliding on ice I would know that that readout was incorrect.

Just saying...
 

·
Registered
2014 Touring/Eyesight CVT
Joined
·
78 Posts
I understand what Tony is trying to do. I have a digital wall thermostat at home which has an adjustment to raise or lower the current reading temperature by 1, 2, or 3 degrees. This is so that you can synchronize it worth another thermostat in the house....so it is not a difficult/expensive to include in a thermometer.
 

·
Registered
2014 Forester 2.5 Touring CVT
Joined
·
111 Posts
If the Fozzy was an oven and I was baking a cake, a few degree temperature variance would be a concern.
You have it quite backwards my friend...

I worked a few years at a company that designed and manufactured both mechanical and electronic temperature measuring and control devices. They truly thought that +/- 10 degrees was fantastic for an oven controller and reserved those for high end models. Even with the electronic controllers which were inherently capable of very tight tolerances they truly did not care if parts swapped during assembly created wide variances in operation.

The prevalent feeling was that "Suzie Housewife" would get used to her specific oven and do just fine until the controller eventually failed and had to be replaced. Sure, she may very well have to relearn cooking in her oven after that (imagine a possible +/- 20F or greater swing). But, by then the warranty would be expired and she would have no recourse. Anyway, the thinking went, she would probably blame the repairman sooner than "Us". I lost a lot of battles there I'll say. I am personally surprised and a little ashamed I stayed on as long as I did.

But for a general outside air temp, a few degrees is no big deal. I don't think any vehicle brand would build a calabration [sic] process into a basic outside air temp thermometer.
^TonyFord above is correct- The big deal with external thermometers in automobiles is at or around the 32 F point where black Ice may be forming. I sincerely hope ALL manufacturers are calibrating their temp displays to be +/- 1 degree F at the universal constant of freezing H2O.

How do you justify the cost?
The way to justify the cost is to have your risk analysis folks to weigh it against a few personal injury and/or wrongful death lawsuits.
 

·
Registered
2014 Forester 2.5i CVT
Joined
·
320 Posts
You are basically correct.

But.

If the readout stated 34 or 35 degrees and the actual temp was 32 or 31 degrees and the roads were icing over, I would not be looking out for icy spots because I would think it is too warm for the roads to freeze up, but after a spin out or near collision from sliding on ice I would know that that readout was incorrect.

Just saying...
Sound like you want temperature in one spot to predict icy conditions elsewhere? Windchill factor alone makes that impossible.
 

·
Registered
Joined
·
323 Posts
It's a cheap plastic sensor, that changes slowly. They can be inside a warm shop for 5 hours and still register -25. Even if you could calibrate it you wouldn't be happy with it because it changes so slowly it wouldn't register accurately.

I can't believe how many complaints over the MFD are on here lol. It's just a digital readout of a bunch of cheap sensors sending voltage. I take it with a grain of salt.
 

·
Registered
2008 Forester XT Sport 4EAT
Joined
·
1,663 Posts
I'm pretty sure you can tweak it in how you want it.

It's very likely that the sensor is a thermistor. Thermistors behave such that if the temperature goes up the resistance goes down, so here's what you should do.

1) Disconnect the temp sensor and attach it to a multimeter.
2) Make some measurements of the sensor behavior when you put it in ice water, warm it to body temperature, etc.
3) If you need to tweak it so that your display reads higher than it currently does then add a 10-turn potentiometer (likely a 10 kohm or 5 kohm) in parallel with the temp sensor.
4) If you need to tweak it so that your display reads lower than it currently does then add the 10-turn potentiometer (likely a 10 kohm or 5 kohm) in series with the temp sensor.
5) Plug the temp sensor back into the car harness.
6) Put the temp sensor back in the ice bath.
7) Adjust your potentiometer until your readout shows 32 degrees Fahrenheit.
8) Remove the potentiometer from the circuit and measure it's resistance with a multimeter.
9) Buy a resistor with the value you measured above and install where you had your potentiometer.

Done.


Of course if it's a thermocouple instead of a thermistor then things might get a bit more complicated.

Previous Foresters were thermistors, so it's very likely the 2014 is as well. If you have a factory service manual you might be able to poke around it and find out for sure.
 

·
Registered
2008 Forester XT Sport 4EAT
Joined
·
1,663 Posts
Actually if you put the potentiometer in parallel you'll probably need a really high resistance one, even higher than the 10 kohm I mentioned above. The initial measurements you make will allow you to make some calculations to get a rough idea of what resistance is required for the change.

I'll gladly help run the numbers for anyone if they tell me how much change they need and what they are getting for a reading in an ice bath.

You can leave out using a potentiometer entirely and just do it by calculation as well, which should get you close if not spot on, but the potentiometer is fun, educational and guaranteed to get you dialed in.
 

·
Premium Member
2008 Forester X Premium 5MT
Joined
·
8,064 Posts
You are basically correct.

But.

If the readout stated 34 or 35 degrees and the actual temp was 32 or 31 degrees and the roads were icing over, I would not be looking out for icy spots because I would think it is too warm for the roads to freeze up, but after a spin out or near collision from sliding on ice I would know that that readout was incorrect.

Just saying...
Air temp doesn't necessarily correlate to ground temp either. Air has a far lower thermal mass then a slab of light colored concrete that is 10+" thick or dark asphalt 3" thick. That slab of concrete may keep the ice from melting well into the day after a cold night even after the air temps go above freezing and the sun has been on it for a few hours. On the flip side the dark asphalt might melt the ice a hour into the day once the sun hits it even if the air temps are below the freezing point.

I'm of the opinion that any gauge on a vehivle is nothing more then a refferance gauge and not a calibrated instrument.
 

·
Premium Member
2015 XC diesel CVT
Joined
·
4,053 Posts
Nepbug's got it. Series or parallel add-ons can shift it a bit, but non-linearity may creep in.
 

·
Registered
2015 Impreza 5-Door Sport Auto (CVT)
Joined
·
2,351 Posts
Discussion Starter · #17 ·
You have it quite backwards my friend...

I worked a few years at a company that designed and manufactured both mechanical and electronic temperature measuring and control devices. They truly thought that +/- 10 degrees was fantastic for an oven controller and reserved those for high end models. Even with the electronic controllers which were inherently capable of very tight tolerances they truly did not care if parts swapped during assembly created wide variances in operation.

The prevalent feeling was that "Suzie Housewife" would get used to her specific oven and do just fine until the controller eventually failed and had to be replaced. Sure, she may very well have to relearn cooking in her oven after that (imagine a possible +/- 20F or greater swing). But, by then the warranty would be expired and she would have no recourse. Anyway, the thinking went, she would probably blame the repairman sooner than "Us". I lost a lot of battles there I'll say. I am personally surprised and a little ashamed I stayed on as long as I did.



^TonyFord above is correct- The big deal with external thermometers in automobiles is at or around the 32 F point where black Ice may be forming. I sincerely hope ALL manufacturers are calibrating their temp displays to be +/- 1 degree F at the universal constant of freezing H2O.



The way to justify the cost is to have your risk analysis folks to weigh it against a few personal injury and/or wrongful death lawsuits.

I like the way jc2 thinks....
 

·
Registered
2014 Forester 2.0XT CVT
Joined
·
366 Posts
You have it quite backwards my friend...

I worked a few years at a company that designed and manufactured both mechanical and electronic temperature measuring and control devices. They truly thought that +/- 10 degrees was fantastic for an oven controller and reserved those for high end models. Even with the electronic controllers which were inherently capable of very tight tolerances they truly did not care if parts swapped during assembly created wide variances in operation.

The prevalent feeling was that "Suzie Housewife" would get used to her specific oven and do just fine until the controller eventually failed and had to be replaced. Sure, she may very well have to relearn cooking in her oven after that (imagine a possible +/- 20F or greater swing). But, by then the warranty would be expired and she would have no recourse. Anyway, the thinking went, she would probably blame the repairman sooner than "Us". I lost a lot of battles there I'll say. I am personally surprised and a little ashamed I stayed on as long as I did.
You make a good point. I'm in the restaurant industry and all of our cooling, freezing, cooking and warming equipment have a have a large temp variation. As long as we are within the range we are fine.

If the cheap outside thermometer on the Fozzy is accurate with a +/-3 degrees, that's pretty darn good and no effort to make it any more accurate is worth the effort. As has been mentioned road icing conditions have many other factors and relying on an auto outside temp thermometer to predict ice on roads can be dangerous.

Surface Temperature and Air Temperature Observations: How Are They Related?

Temperature measurements are generally taken with thermometers in a sheltered enclosure at about 5 feet above the ground, usually above a vegetated surface. It is absolutely crucial to understand that the "official" temperatures are reported by thermometers located at around 5 feet whose readings can be very different from road surface temperatures. On clear nights when winds are light, the surface radiates heat to space much more effectively than the air above. On such nights, temperature at ground level can be 2-5F cooler than air temperature only a few feet above. Thus, frost can be occurring at the surface even when official temperature observations are reporting temperatures of 35 to 37F. During the day, the opposite situation can occur, with the road surface several degrees warmer than the air temperature at 5 feet. Air temperature readings on trucks and cars are similarly problematic...they often are colder (night) or warmer (day) that the road surface. The moral of this story is that motorists and maintenance personnel must be wary of icing as air temperatures drop below around 37F.
 

·
Registered
2015 Impreza 5-Door Sport Auto (CVT)
Joined
·
2,351 Posts
Discussion Starter · #19 ·
The moral of this story is that motorists and maintenance personnel must be wary of icing as air temperatures drop below around 37F.

Now I know why the temp readout blinks at 37 degrees!
 

·
Premium Member
2014 Forester XT Touring CVT
Joined
·
12,780 Posts
There is a reason that the MFD (I know the base does not have it, Tony) shows an alert at 37 degrees stating the roads may be icy and to drive with caution.

Edit: Just saw your above post, apparently the readout blinks on models without the MFD.

Just like the fuel economy display, it is not going to precise all of the time. Take it with a grain of salt and drive with caution when the temp gets below 37.
 
1 - 20 of 22 Posts
Top