Temperature tables are emitted by selecting arbitrary sample values
to be used for the linear lookup table. This is fine in the range
where the thermistor produces linear output, but it is markedly wrong
near the extremes where the thermister output begins to curve.
Introduce a new sample selector which chooses samples based on the "most
incorrect estimate" and improves from there to ensure we get a cleaner
approximation across the selected range.
Traumflug: this topic is tracked here:
https://github.com/Traumflug/Teacup_Firmware/pull/208
Previously they were evenly spaced across the temperature range,
now they're evenly distributed across the ADC range ... which
is not entirely the same.
This is related to issue #176.
This is, the parameter entry fields in the GUI, not in the
generated temperature table. Allowing floats is a tiny bit more
precise and also, more importantly, less confusing for the user.
--Traumflug
This is related to issue #176.
The Steinhart-Hart algorithm allows more precise thermistor
tables, but also requires more parameters, which aren't
available for all thermistors. Accordingly, add support for both,
the traditional logic using the beta parameter as well as the new
one.
This also adds thermistor-presets, so users can simply choose
from a pulldown-menu to set their thermistor.
Also, identical thermistors get merged into one table, saving
binary size.
Last not least, a few bugs in this area got fixed.
Usually, all these things go into separate commits, but they were
contributed all in one and separating them is a bit error-prone
for little gain.
This should address issue #130, #134 and #135.