I am posting a copy of charlieb's post from the other thread. I thought it addressed it rather well.
charlieb:
Let?s start with some basics. I apologize if some of this is already ?old news? to you, but I need to start somewhere.
The temperature that a wire will reach depends mostly on two things. One is the resistance of the wire. That is a function of the size of the wire. If you put the same current through two wires of different sizes, the larger wire (with its lower resistance) will not get as hot as the smaller wire. The other key factor is the conditions that surround the wire. The ampacity tables are based on one value. If the surrounding temperature is higher, then the same wire with the same current will reach an overall higher temperature. That is why there are derating factors for ambient temperatures above 30C.
With that as a start, I will now say that I think you are looking at the tables from the wrong perspective. The three columns are not about how hot a wire will get at the posted current values. They are saying that a wire with one type of insulation (e.g., TW) can sit in a 30C ambient area, can take the current that is shown for that size wire (e.g., 55 amps for #6 copper), and it won?t get hotter than 60C. The name of the insulation (again, we are using TW for now), and the fact that that name appears at the top of the 60C column, tells us that that wire can handle an overall temperature of 60C, without suffering damage to its insulation. Similarly, sending 65 amps through a #6 copper wire will result in an overall temperature no higher than 75 C, and if the wire has type THW insulation it will be able to handle that current and that temperature, without suffering damage to its insulation. You can say something similar about a #6 copper wire with a current of 75 amps and with THHN insulation: it won?t get hotter than 90C, and the wire can handle that temperature without suffering damage to its insulation. That is what the columns are telling us.
However, if you do put 75 amps through a #6 copper wire with THHN insulation and in an ambient temperature of 30C, the wire?s insulation will be able to handle it, but the point at which you connect that wire to a breaker, a relay contactor, a terminal block, or a switch might not be able to take the full 75 amps without it getting too hot. Not many manufacturers have yet listed their terminations for more than 75C. That is why we generally don?t use ampacities higher than those listed in the 75C column, even if we use wire with 90C insulation.
Now I can (finally) get around to answering your question.
We can use the 90C column for derating (if, of course, we have a wire with one of the insulation systems shown at the top of that column) because the wire itself can handle the current value shown in that column. A 250 MCM wire, aluminum, with XHHW-2 insulation can handle every bit of the 230 amps shown in the table. But with an ambient of 112F, we need to reduce the current going through the wire, to prevent it from getting above the 90C that the insulation can handle. So we derate it by a factor of 0.87. (By the way, your example used 0.82, and that is not right. Look under the 90C column for an ambient of 112, and you will see 0.87.) Multiplying 0.87 times 230 gives you 200 amps. So under this set of ?conditions of use? (please look up the article 100 definition of ampacity), the ampacity of this wire is 200 amps. That is higher than the 175 you need for the load. So you can use this wire size and type. Please note that you are putting 175 amps through a wire that (using the 75C column) would have had an ampacity of 205, but in our case has a more limited ampacity of 200, you will not exceed the 75C rating of the terminations, and you will come nowhere close to the 90C rating of the wire itself.
