Temperature controlled current source?

On a sunny day (Sun, 15 Jun 2008 12:24:18 GMT) it happened Hammy wrote in :

I wonder if you could use a LM135, 10 mV per degree C, and 2.982 V at 25 C.

So 70-20=50C delta T, makes 500mV delta U, and 2.932V at 20 C. If you set the opamp gain and startpoint correct that should work. You could also use a LM135, PIC 2 channel ADC, soft, PWM out to drive the LED.

Works backwards; higher temp, less current.

John

Your right John.I corrected it I switched the +/- opamp terminals around.

It works good on simulation anyways. :)

?????

That will latch up!

John

That doesn't fix it ! You need to flip the divider.

Also needs a resistor from op-amp to MOSFET gate, else the LM324 will oscillate. An integrating cap from LM324 output to (-) input would be a good idea too...

James Arthur

--- Here, try this:

Version 4 SHEET 1 880 680 WIRE 32 -48 -160 -48 WIRE 192 -48 32 -48 WIRE 352 -48 192 -48 WIRE 352 0 352 -48 WIRE 32 16 32 -48 WIRE 352 96 352 64 WIRE 192 112 192 -48 WIRE 160 128 128 128 WIRE 288 144 224 144 WIRE 32 160 32 96 WIRE 160 160 32 160 WIRE 128 224 128 128 WIRE 352 224 352 192 WIRE 352 224 128 224 WIRE -160 256 -160 -48 WIRE 32 256 32 160 WIRE 352 272 352 224 WIRE -160 384 -160 336 WIRE 32 384 32 336 WIRE 32 384 -160 384 WIRE 192 384 192 176 WIRE 192 384 32 384 WIRE 352 384 352 352 WIRE 352 384 192 384 WIRE -160 448 -160 384 FLAG -160 448 0 SYMBOL Opamps\\\\LT1001 192 80 R0 SYMATTR InstName U1 SYMBOL res 336 256 R0 SYMATTR InstName R1 SYMATTR Value 82.5 SYMBOL voltage -160 240 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 5.6 SYMBOL res 16 240 R0 SYMATTR InstName R2 SYMATTR Value 1K SYMBOL res 16 0 R0 SYMATTR InstName R3 SYMATTR Value 1751 SYMBOL LED 336 0 R0 SYMATTR InstName D1 SYMATTR Value NSPW500BS SYMBOL npn 288 96 R0 SYMATTR InstName Q1 SYMATTR Value 2N3904 TEXT -128 416 Left 0 !.tran 1

You'll need to diddle with R1 and R2 to get zero and span right, and the thermistor is a Betatherm 10K3A1B. RT data is at:

JF

Thanks John!

I tested out your circuit. It works well. Thanks for pointing me in the right direction.

I just used a pot to simulate the thermistor on the bread board and an lf353 as the EA. I'm going to use a TLV2372IDR in the actual circuit the other op can be used for a thermal shut down. No point wasting it.

The only change I made was to parallel a resistor with the thermistor and feed the network with a current source. The thermistor I am getting from Digikey can only tolerate limited power (21mW) before a large error due to self heating is incurred. The current source should limit it to 1 deg C error from self heating. R2 is the thermistor.

Schematic

Simulated Output plot q1 collector current vs. R2 (thermistor), This is pretty close to what I was getting with my LF353 and pot. This is what I was after. If it works this well with the actual thermistor I'll be happy. 30k is 25 deg C and 4.13k is 70 deg C

Thanks