I doubt the method does give a good approximation even at about 60° north in July, although I did not verify by how much. I live near 49° north and recently tra

I doubt the method does give a good approximation even at about 60° north in July, although I did not verify by how much. I live near 49° north and recently travelled to Scandinavia. While my intuition at home is pretty accurate, my idea of how long it takes for the sun to set so far north in the evening was completely off the mark at first. To calibrate it, I changed my mental model of the sunset to a double-flat angle and estimated how long it might take the sun to (virtually) cover the longer resulting distance.