NTC lookup tables versus Steinhart-Hart law

[Daren Schwenke]

Is there a compelling reason to use the thermistor lookup tables in the beaglebone code as it is or can I implement it via the Steinhart-Hart law equation?

I'm using an unknown thermistor and it seems approximating the curve would be easier with the latter.

[andy pugh]

On 5 January 2015 at 12:13, Daren Schwenke <darens...@gmail.com> wrote:
> Steinhart-Hart


Seems computationally intensive and is only an approximation. The
lookup-table can be arbitrarily accurate and has lower CPU overhead.

--
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

[Charles Steinkuehler]

I used the lookup table because it's more accurate than the equations
for the specific thermistor I typically use (I pulled the table values
out of the data-sheet). The AVR firmware folks use the lookup table
approach because it saves CPU cycles, but I think most of their tables
are created using just the beta value of the thermistor.

If you've got the Steinhart-Hart coefficients for your thermistor and
want to code up the equation, I see no reason why it shouldn't work well.

--
Charles Steinkuehler
cha...@steinkuehler.net

[andy pugh]

On 5 January 2015 at 12:43, Charles Steinkuehler

<cha...@steinkuehler.net> wrote:
> The AVR firmware folks use the lookup table
> approach because it saves CPU cycles,

In my day job I program car ECUs. Every sensor uses a lookup table
rather than a polynomial.

[Daren Schwenke]

I don't have the values, but I also have no data sheet.  Seems to be around 60k for the 25 value, so an ice bath and boiling water should give me the three data points I need to calculate the Steinhart-Hart (from what I read).

[andy pugh]

On 5 January 2015 at 15:40, Daren Schwenke <darens...@gmail.com> wrote:
> I don't have the values, but I also have no data sheet.

Bear in mind the possibility that a temperature scale calibrated in
"Darens" might be perfectly acceptable.

[Daren Schwenke]

Given I also need to do a resistance to frequency conversion for my 555 based input, a single "Darens" scale is starting to sound attractive.  

I'm using the fan speed input on my mobo to get this feedback loop.  Reads 360rpm at 25C and 12000rpm at ~280C (or whatever the real lowest temp my Weller goes to) 

I really need an accurate thermometer and the candy thermometer I have is way off.

My code is just getting less re-usable by the moment. :)  Thank you.