Changing the resistance is a hack, and is in my opinion not a very good idea.
There are two ways to implement TC: the naive one, using a single coefficient, and the more accurate using a curve.
Ideally, and this is rarely done although it's really important, you need to zero the mod internal resistance out. For this you use a special kind of shunt 510 connector made of pure copper, but alternatively any very large piece of copper connecting your 510 positive and negative should do.
When the 510 is shorted, the resistance value displayed on screen is equal to the mod internal resistance. In eScribe and ArticFox, you can enter this resistance in the advanced settings. This way the mod will be able to subtract it from the attachment resistance reading, leaving you obviously with just the resistance of the attachment.
This is important because the "mod part" of the resistance is basically constant. But to make TC work we need to measure resistance changes. When both resistance are compounded together, only a part of it (the atty part) will be variable and this screw things up. When the "mod part" is properly zeroed out, we're left only with the variable part.
Then, the naive way to do TC is to use a simple coefficient (usually advertised as TCR when it is customizable, but also often used under the hood to implement the wire presets like NI200, SS, Ti etc) That coefficient is just the slope of the linear temperature curve. Figure a graph, with temperature on the Y axis and resistance on the X axis and draw a line. The coefficient is the angle (slope) of that line.
The more accurate way to perform TC is to use a curve instead of a line, as this is closer to how the resistance of these metals change with temperature in the real world.
Now you should really disregard the displayed temperature, it doesn't really matter and will never match the temperature of your load. At best it will closely track the mean temperature at the heating element, but again it's not that simple, coils and meshes don't necessarily heat evenly, there are gradients and hot spots etc. But let's assume that it can match the coil temperature overall.
I think the error most people do in this thread and the few other convection attachment ones, is to try to have the displayed temperature match the load temperature. You can do that by tweaking the TCR coefficient and/or the measured temperature offset, but when you do that you only match for a single point. The slope of your curve will be completely wrong. This means that as soon as you change the temperature the match will be lost: you thought you increased by 5°C but you increased by 20°C for instance (extreme example but you get the idea) or inversely you have to increase by 50°C to get only a 10°C change...
Some combinations of TCR+offset value will work but only in a very narrow range. Or for instance you start at 220°C on display and the mod can only go to 250°C so you end up with only 30°C of adjustment range, and it's not good. It's best for instance to have your low temperature point fall in the middle of the mod adjustment range, say 150°C for instance, this way you have ample granularity to adjust up or down, even if the displayed temperature looks a bit meaningless.
On the iHeat for instance, when using the SS316L curve, my displayed temperature is between 140°C and 165°C (and it's not optimal, I'll need to tweak it, do as I say not as I do!!!
![Tongue :p :p](/i/s/tongue.gif)
) But 140°C feels like a 195°C in my other vapes, and 165°C would be close to the max 220°C.