Maker Pro
Maker Pro

Complete temp. detecting circuit-Need help tuning to desired range

greatpanda5

Feb 26, 2015
7
Joined
Feb 26, 2015
Messages
7
We completed a circuit for an engineering capstone project. The temperature sensing ic is an LM355z. The circuit is supposed to output to a 9 volt buzzer once a certain threshold is crossed in temperature. The output chances the voltage with a change in temperatures. The voltage amount can be changed by the 3 potentiometers (1k, 10k, and 100k). We currently have the output going to a 3V LED. The idea is the led will turn on when a certain temperature is reached. Unfortunately, we don't know how to tune the circuit so that it operates at the wanted temperature range. Can anyone tell us how to approach this? We are thinking it could be with tuning the potentiometer or changing to LED. Thanks for the help!




0
 

Arouse1973

Adam
Dec 18, 2013
5,178
Joined
Dec 18, 2013
Messages
5,178
Hello

The POT VR1 is used to calibrate the sensor. At 25 deg C the breakdown voltage of the Zener is approx. 3V (2.98 V actually). VR3 is used to adjust out the offset of the OP-amp this is usually used to trim for a zero output when both inputs are the same.

The trip point is controlled by VR2 so if you set the trip point to 2.98 V and calibrate the temp sensor for 2.98 V at 25 deg C then the opamp should change state when the temp is just below and just above 25 deg C.

Looking at the data sheet for the TL071 it appears the output can only swing to within 2 Volts of either supply, I can't see how this original circuit will work, the transistor will always be on.

You will also have another problem and that is you have no Hysteresis which means for only a slight variation on the inputs the output will switch and then switch back and then switch on etc. You need to really use a comparator that has Hysteresis control or add it yourself as positive feedback.

Adam
 

greatpanda5

Feb 26, 2015
7
Joined
Feb 26, 2015
Messages
7
That's unfortunate, we really just found the circuit online and the description seemed like it made sense. Thanks for the help. We've heard about themocouplet circuits, can you shed any light on this? We really appreciate it.
 

Arouse1973

Adam
Dec 18, 2013
5,178
Joined
Dec 18, 2013
Messages
5,178
You mean Thermocouple. It another form of temperature measuring device. They can be more accurate and have a much larger temperature range. They work by producing a very small voltage difference between two different metals joined at a hot junction (Seebeck Effect).

The circuitry needed is normally an amplifier which has what is called cold junction compensation you need to do this because when the thermocouple is connected to another junction like the measuring equipment this forms another junction (cold junction) which needs to be compensated for.

Because they use two different metals to form the thermocouple they produce a difference voltage that is proportional to the temperature at the hot junction. They don't measure absolute temperature, to do this you need to measure the absolute temperature of the cold junction also and include this in your processing.

Normally an isothermal block is used. You would then need some sort of processing to convert this to a digital readout for instance.

The other issue that you have to contend with is they are not very linear over wide temperature ranges. If you want accuracy and wide operating ranges you will need to factor this in also.

Adam
 

greatpanda5

Feb 26, 2015
7
Joined
Feb 26, 2015
Messages
7
Okay, thanks. That makes sense. The bi-metal arm, we found out,would be a little too clunky for the circuit. Your information makes it seem kind of impractical. Our third and final option was thyristor controlled circuit. More specifically, a thermistor, which is more varying to temperature. The circuit looks simple, can you confirm that? what is your opinion on it? Thanks again, your aid has been invaluable.
 

Arouse1973

Adam
Dec 18, 2013
5,178
Joined
Dec 18, 2013
Messages
5,178
This will sort of work but if the temperature hovers about the turn on voltage of the transistor it will not switch fully and the LED might start off dim and then suddenly go bright as the temperature increases. The same when turning off or cooling down. If the temperature rises quickly and then cool quickly then this will work. It will be better with a Darlington like you have but you may notice the LED behaves differently than you expect for slowly changing temperatures. Why don't you try it and see how it works.
Adam
 

Arouse1973

Adam
Dec 18, 2013
5,178
Joined
Dec 18, 2013
Messages
5,178
Hey Adam, what are your thoughts on an arduino circuit?
Yeah go for it. I used to be dead against beginners jumping straight in to using micro's, but it seems the way they want to go nowadays.
Adam
 

Arouse1973

Adam
Dec 18, 2013
5,178
Joined
Dec 18, 2013
Messages
5,178
Yes that would be nice, thanks for keeping me in the loop. We so often never get to see things finished because people don't come back and let us know.
Adam
 
Top