John said:
---
Check the same subject on seb.
What the **** is it with some of you clowns, anyway?
I do a guy a favor with a little tutorial to which he refuses to
respond and then I get flak from some fucking idiot about wasting
bandwidth, from you with your snide goddam "paper" bullshit, and from
Givens with his "I was gonna write the same thing myself, but"...
To hell with all of you niggardly sons of bitches.
It is possible to achieve some semblance of precision with the Radio
Shack parts. The OP's relay driver is garbage. An example of a 3oC
hysteresis t-stat shown below- make Rh adjustable for other -5% carbon
film is good enough:
View in a fixed-width font such as Courier.
RADIO SHACK THERMOSTAT 3oC HYSTERESIS
Rb
V+ >-/\/\--------+-----+------+
| | |
/ D --- /
R \ / Rthermistor
/ --- /
\ | \ LM741 V+
| | | |\ |
+---+ | +-----------|-\
| | | | | >-----+->Vout
| / Vz | / +---|+/ |
| R --- / Rpot | |/ | |
| / // \ / | --- |
| \ --- \ | /// |
Rb | | | | | |
+---/\/\-------+-----+------+ | |
| | | |
| | | Rh |
--- +------------------------+----/\/\----+
///
V+=12V Rb=220 Vz=1N4733A 5.1V D=1N914 R=10K
Temperature Hysteresis=3oC==>
for Ro=10K thermistor and R(35o)=6.8K and R(15o)=14.9K
BETA=4088 where R(T)=Ro*EXP(BETA*(1/T-1/To))
TCR R(T)= -0.39K/oC @25oC ( TCR=Ro * -BETA/T^2 )
-2.925V
TCR Vin(-)= ------- x TCR R(T), assuming zener +diode=5.85V
2 x R(T)
==> TCR Vin(-)= +57mV/oC or 170mV for 3oC hysteresis
6V
so that Rh= -------- = 180K for required hysteresis
170mV/5k
Actual Vhyst=160mV or 2.8oC at 25oC
At T=15oC TCR R(T)= -0.62K/oC so that Hysteresis= 2.6oC
At T=35oC TCR R(T)= -0.25K/oC so that Hysteresis= 3.0oC
T-Hyst=2.8oC +/-0.2oC 15oC < T < 35oC