[MPC] Add variable power support for PTC heaters

[kageurufu]

PTC heaters need variable power to properly be supported by MPC.

This PR adds three new config options heater_voltage, heater_power_ambient, and heater_temperature_coefficient.

Using the following measurements and formulas, your coefficient can easily be calculated. To simplify this, I've also documented and added a calculator to the MPC documentation.

• $V$ - Voltage
• $T_{a}$ - Ambient temperature
• $R_{a}$ - Resistance at ambient temperature
• $T_{2}$ - Second temperature
• $R_{2}$ - Resistance at second temperature

The formulas then provide your configuration values:

heater_voltage = $V$
heater_power_ambient = $T_{a}$
heater_power = $\frac{V^2}{R_{a}}$
heater_temperature_coefficient = $\frac{(R_{2} - R_{a}) - 1}{T_{2} - T_{a}}$

As an example, I am using TZ-V6 with 80W heaters from TriangleLab, which are 88.6W @ 24C, with a coefficient of 0.003131

Checklist

• pr title makes sense
• squashed to 1 commit
• added a test case if possible
• if new feature, added to the readme
• ci is happy and green

[liampwll]

This only works for heaters with a linear resistance/temperature curve, but with most heaters that have such a curve the change in resistance is small enough to not matter within the region of temperatures that you can print at. The PTC heaters that don't work well with MPC have curves that look more like the attached image:

[liampwll]

I think a better solution may be adding a LUT which can be automatically found as part of the automatic calibration by slowly raising the temperature at a constant duty cycle. The heater resistance would then just be a function of the current temperature and the temperature delta over time as well as the various heater constants.

[Nathan22211]

@liampwll is it how it's done in your firmware? Also, as much as I want to test this PR I don't know if it'll work for a Rapido V2

[liampwll]

It's not done at all yet. We have current sensing so we can do things more directly when it is implemented, but the idea is the same.