Hot End and Bed Temperature Measurement Question

Hi, I was wondering how the temperature of the heated bed and/or hot end is measured. I am using a 100k thermosister ,but I am not sure how the software knows what temperature surrounds to what resistance. I understand that the thermosister changes its resistance in accordance with temperature, but I don't understand how the software knows which resistance corresponds to what resistance. Is it something that I measure and plug in or dose the software do it through the PID controller? Also I read the section on temperature measurement and controlling in on the repetier host guidehttp://www.repetier.com/documentation/repetier-firmware/rf-installation/]Repetier host guide[/url], but I did not understand any of it. Could someone answer my questions and/or post a link to the answers.
Thank You very much

Edited 1 time(s). Last edit at 01/26/2014 09:31PM by gunnd14.

I am not an firmware expert, but I think the firmware has different thermistor temperature curves logged in it. So in order to get the correct curve you would need to uncomment the correct curve that corresponds to the thermistor you are using.

The PID (Proportional, Integral, Derivative) controller is the feedback controller that the firmware uses to regulate the temperature to the temperature curve. Adjusting the values of the P, I, or D will give you various types of responses. If you are interested in learning more about PID controllers, and there is A LOT, just check out the (Wikipedia page.

Hope that helps.

---

greghoge.com

HUGE 3D PRINTER PARTS SALE!!!

What the firmware does is basically a hack. Because no one knows how to come up with a way to make firmware perform the complicated exponential equation to read the temperature, it uses a "look up table" to compare the current of the thermistor to a long list of temperatures and their expected current.

Quote
ohioplastics
What the firmware does is basically a hack. Because no one knows how to come up with a way to make firmware perform the complicated exponential equation to read the temperature, it uses a "look up table" to compare the current of the thermistor to a long list of temperatures and their expected current.

It's actually measuring voltage with an ADC, but you're right that it's effectively measuring current. In the end, what we're really measuring is the resistance of the thermistor. It does use a lookup table and interpolates, I think with a straight line, between the table entries to convert the reading to C. I think the equation is well known since you can actually generate your own table with a Python script that used to come with the firmwares. But I think it would be too slow for realtime control to try to calculate that in firmware. Anyway, it has been shown a while ago that this method is good enough even with a thermistor.

Edited 1 time(s). Last edit at 01/26/2014 11:33PM by brnrd.

A lookup table is a good solution if what you are measuring is really difficult to describe by an equation, as you can adjust values in the table to match the actual performance of the sensor instead of the theoretical performance. The accuracy is a function of how many data points are in the table, so that interpolations get done over a span that is essentially a straight line. If you need more accuracy add more points. You can even have values that are close together where the curve changes quickly and farther apart as it approaches a straight line. So I think it is a really good solution all in all.

Years ago I used to drive a motor controller with an analog signal. Then the manufacturer changed the resistors in the input circuit from 1% to 10% and suddenly the machines ran at widely varying rates. We had a tachometer so we automatically computed a new lookup table any time the computer got rebooted. It wasn't a PC so that usually only happened rarely.