[SOLVED] Temperature cut off?

I have a change of design.

Our safety system outputs 24V to a door switch or a flow switch, when the switch is broken the system cuts off.

I want a LM35 to run off a 24V supply and when the LM35 reaches 60C i want the system to shut down.

Now i know the LM35 works by 1c = 10mv, so max is 1.5v (150c)

So when the LM35 gets to 60C, there is a voltage of 0.6v on the output.

How can i keep a 24v output below 60C and at 60C or above the output is 0v and cuts the system

Is this possible?

Analog Comparator gives logic level out to drive logic level High side switch. But if thermal sensor is just for regulator, does it have have thermal shutdown?

I would recommend a smart high-side switch with thermal shutdown used in automotive. Tons avail from D-K.

https://www.digikey.com/product-sea...tches-load-drivers/2556406?k=High side switch

eg. https://www.digikey.com/product-detail/en/VN7050ASTR-E/497-14300-6-ND/4552533

Hi Sunny,

So the high side switch will shut down at a certain temperature?

My electronics is quite basic as you can proberbly tell.

Looking at the data sheet it says operating temperatures between -25 to +85c.

Yes , many protection circuits such as short circuit, over temp, over voltage, depending on your needs.

Yes it is possible, there are lots of ways, more than I know to say the least, I'd listen to the SUnnySkyguy.

Another approach is to use an inverter. The 7404 is a ready-made low voltage device (not good for what you now plan to do). It makes a high signal low and a low signal high. You can achieve the same effect with a TTL or a CMOS inverter using discrete components, or even easier is one power transistor (NPN or NMOS) configured as an inverter. Here is a schematic:

https://en.wikipedia.org/wiki/Inverter_(logic_gate)

And this is a simple explanation with schematics too:

https://www.cl.cam.ac.uk/teaching/1112/Hardware/Workshop5_2011.pdf

If you can, get a breadboard, a trimpot, a transistor, the LM35, an LED, a DVM or a DMM which can measure temperature, put them together (I'm happy to do a simple schematic for you if it helps, but I'd need up to a day or two), and with a hairdryer, cigarette lighter or other heating device (if you really need it, which you would to get 60ºC), actually test in the real world when the signal goes low - the LED would go off. Better to test starting off with a lower voltage, and then increasing the voltage to the 24V when the basic version works.
MOSFETs don't seem to do well with a gate voltage over 20V max, they fry I think, they can pass much higher voltages but the gate is like a transistor base, so you would need a device to boost the input signal to the gate/base from the LM35, which could just be a smaller BJT or NMOS/NFET, or an OpAmp set up as a comparator.

I have tested a LM35 on a breadboard with a LED with a 9V input all work fine.

Yes that could be another approach using the inverter.

With the LM35 i would need to make the output 24v below 60C and 0V above 60C.
So increasing -15v - 0.6v (-15C - 60C) to 24v, then 0.61 - 1.5v (61C - 150C) to OV seems complicated.

Logic 1 (24V) Logic 0 (0V)

Myabe sunny is right and the high side switch is the way forward.

Whatever is easiest and least parts count to implement is best, and a device with thermal shutdown sounds a good way to go.