I think this happened last year. Whenever it happened, I know it's happened before. It only happens when it's cold outside. Yesterday, it was 18ºF which, Priapus was happy to show me, equates to -6ºC. I flipped over to the Climate screen and when I flipped back the display was Fahrenheit again.
I'm not a real fan of metric (you wouldn't be either if you were as close to France as I am ) but I really don't get fahrenheit. Metric makes so much more sense in day to day life compared to fahrenheit which is just weird or kelvin which is mainly for scientists. Water freezes at zero, water is of interest to humans - end of
I wish I know how you did it so I could switch mine to metric. I would rather have celcius or kelvin than fahrenheit.
Hmmm, that would have been fahrenheit then. Think the tv weather forecasts changed metric in the late 1970's when I was still a wee lad.
This is due to bad connections (solder joints) developing in your MFD unit. It can be repaired by someone competent with SMD rework. This problem will also likely get worse over time and eventually you will lose the display alltogether.
A bad connection causes it to go metric? That would mean there is an external component to make it so, like a pulldown/pullup as a select. When this happens does your MFD change to litres/100km? If not, then I doubt this is the cause. It would make no sense to have an external component to select units of just the temperature display and still require a firmware change to convert everything else.
I believe that the MFD can display at least three fuel economy units: 1. miles per US gallon 2. miles per Imperial gallon 3. litres/100km
Intermittent connections in the MFD can cause all kinds of bizarre results. For instance it may not be the simple pull-up line you describe, but rather the I2C data bus used for internal communications. That bus contains the serial EEPROM where the values are stored for the unit selection, and it runs to many of the IC's in there, so a momentary glitch can definitely cause all kinds of undesired operation. Since I have not had the MFD in question apart on the bench, I can't say with 100% confidence this is the problem, but it would be hard to explain otherwise. They are widely known to have electromechanical problems, I saw one displaying colorful "confetti" patterns and rebooting intermittently.
So, you didn't grow up since the conversion? 30+ years to get used to it is growing up with it Now, if you were much older and trying to convert, it would be much more difficult to deal with how hot is 15C is? Do I need a jacket or not.
It would surprise me if Toyota didnt have some sort of CRC or ECC methods in place. If the bus glitches, I would expect the data to either be reassembled or discarded, not just used blindly.
Race conditions are not unusual at all in systems like this when you start going out of bounds on the design parameters. Take any modern electronic system and start artificially introducing power glitches, noise, etc, and you will get all kinds of crazy unexpected results. Engineers often attempt to "predict the future" and build in safeguards, especially when they are critical systems. (such as watchdog timers, checksums, error-correction, etc.) The MFD is not a critical system however, and clearly has some weak electromechanical design elements. A lot of things like this popped up during the international RoHS (Reduction of Hazardous Substances) effort where we eliminated lead from solder. It made otherwise well-engineered electromechanical techniques now more vulnerable to physical stresses, especially in the harsh automotive environment. BTW, I think the MFD is made by Fujitsu-Ten not Toyota. Historically they make non-critical audio gear, so may not have the kind of rigorous fault-tolerant engineering methodologies in place that someone such as Toyota might.
