That's very likely to be the spatial distribution of heating. Thermal conduction slows down quadratically with distance, so if your sensor is twice as far from the transistor as from the transistor, the speed will be four times slower at low power, because most of the heating is occurring in the transistor.
Hence the benefit of having two thermistors.
Cheers
Phil Hobbs
twice as far from the transistor as from the heater, the speed will
Oh crikey, I just figured this out. (I'm a total idiot!) (I'm kinda embarrassed to post but what the heck.) For whatever reason, I was thinking about the power dissipation in FET and R's wrong. Oh course at low current all the voltage drop (and power) is across the fet.
Oh well, at least we put the Fet in the middle and R's around the edge.
On Friday I was playing around with the sensor position. (on my prototype.) I had the FET and 3R's on the four corners of my ~4"x5" hot plate.
Sensor in the middle Gain was ~constant, ~4-5 sec lag. Sensor between 2 R's was as before more gain needed at low power (now all understood.)lag ~1.5 s
Putting the sensor on the side, midway between the Fet and one R was wacky! (The local gain of the fet is negative)
Hmm, given a circular plate, fet in middle, 3R's symmetric around the edge, there should be some points where the "gain" is flatter.
George H.
More exactly, negative above the 1/2 power point. GH
Here is something you might enjoy reading:
Karlquist The Theory Of Zero-Gradient Crystal Ovens
SW
Have something to add? Share your thoughts — no account required.
Ask the community — no account required