Bottom line, 1 psi per 10F is not that accurate, but 1 psi per 10C would be pretty close.
I decided to check and adjust the air pressure in the tires on my truck this morning, its been awhile so its time. One side of the truck was facing the sun so I got out my IR thermometer to check the tire temperature. Now this only measures the surface temp and not the air temp inside the tire, so I moved the truck into the shade.
The tires facing the sun were about 85-90F where the shaded side were 65F. Within a minute the sunny side had dropped to 70-72F and then held there for the next half hour, so I’m guessing that that is pretty close to the air temp inside. I was wondering if the rule of thumb of 1 psi per 10F was accurate. While waiting, I decided to do some internet search on temp vs pressure. Looking at the formula for a fixed volume, the pressure change is directly proportional to temperature changes. The temperature change is based on absolute zero though. I know the volume is not absolutely fixed, but in the normal pressure range of say 28-36 psi, the volume doesn’t change very much as compared to say a flat tire to 36 psi.
It turns out that a difference of 65F to 75F is only about 2% when compared to absolute zero, so that means that temp change of 65 to 75F would increase the tire pressure about 2%. Going from 65F to 125F results in an 11.5% increase so using the 2% per 10F seems to be about right for this range. That works out to about 0.6 psi per 10F at 30 psi, 0.7 if you are starting at 36 psi.
My tire gauge is digital and reads in increments of 0.1 psi. It was certified to a NIST standard when new but the certificate is outdated now, but I am using it for comparison more than absolute value. I want the tires on each axle to be the same. Since the tires had about 5F difference side to side, I added 0.3 psi to the hotter tires.