Featur: Negative Temps for Thermistor

It would be great, when the maximum, miniumum and current values of the thermistor temperature could also be negative.
When I try to insert negative values on WIN10 1.0.0 the values get reset to 0.

You are right...

My first thought was... that's easy to solve.
But... I have to watch it carefully, we need to avoid resistances <= 0.

I will have a look

Hmm.. shouldn't the temperature in Kelvin be in the exponent?
Or is there a simplification for faster calculation?

https://www.tdk-electronics.tdk.com/download/531116/19643b7ea798d7c4670141a88cd993f9/%20pdf-general-technical-information.pdf#page=3

Right now I don't remember the details of this, but, isn't it in the exponent?
// t = temperature in Kelvin
double k = t*t0/(t-t0);
double res = m_r25/pow( e, m_bVal/k );

I'm a bit busy implementing some other stuff right now, so not thinking in this, but..
This is an NTC, so with negative temperatures resistance should rise.. so no risk to get resistance <= 0.

EDIT: Maybe that is was you just said?

Yep. Since the exponent does not have a complex value, the resulting value must be positive. A positive number to the power of something will always be positive for non-complex numbers: a^x>0 for a element of real numbers.  

If you want to get rid of two divisions, the following sould show equal results:

// t = temperature in Kelvin
double k = (t0-t)/(t*t0);
double res = m_r25 * pow( e, m_bVal * k );

The math is "just"

• 1/e^x = e^(-x)
  
• t0-t = - (t-t0)
  
• 1 / (t*t0/(t-t0)) = (t-t0)/(t*t0)
  

I'm not sure whether divisions make much more load on processors...

I'm not sure whether divisions make much more load on processors...

I don't think there is much difference if any at all, but it looks better.

The math is "just"
...

I started to see "fuzzy" from: 1/..

That was the Professors "just" ... like "...as can easily be seen...", or "...simply solvable in three lines..."

That was the Professors "just" ... like "...as can easily be seen...", or "...simply solvable in three lines..."

And I'm not specially good at mathematics... (to say the least).

Issue solved at Rev 813.