Peltier device used as a heater, trying to build temp controller. Any Suggestions?

wayneh

Joined Sep 9, 2010
17,152
The thermistor can be part of your resistor divider. It takes the place of the 100k pot. You'll need to work out the fixed resistors to use, to achieve the range of PWM control that you need, and you'll need to trim it so that hit holds the right temp in your application.

Personally, I'd prefer a comparator based thermostat circuit.
 

Dodgydave

Joined Jun 22, 2012
9,956
You can put in series with D2, probably with a preset resistor in parallel with it,so that when it gets hot the resistance will decrease and pull pin 2 up faster, and keep the fet off longer, what thermistor value is it?
 
Last edited:

wayneh

Joined Sep 9, 2010
17,152
One thing I'd do before using the thermistor is determine - by experimentation - the PWM duty cycle that gives you the desired result. The arrange your resistors and thermistor to produce that duty cycle.

I think small changes to the duty cycle are likely to produce wide changes in temperature. You might find out you need a duty cycle below 10%. You may need to put a current linter in series with the Pelletier to achieve smoother results.
 

Thread Starter

Redline870

Joined Nov 13, 2015
22
So this is what the engineer I am working with came up with so far:



The 555 timer is working correctly. We powered that by itself and it is fine. There is a problem with the circuit for the op amp however. As you can see, the simulation software he used says there should be 2.42V at pin 6 on the op amp. It is actually at 4.24V and does not change when the potentiometer is adjusted.

This is how I have it all connected:



If anyone has any ideas of what may be wrong, or sees something I hooked up incorrectly, feel free to chime in. Thanks
 

wayneh

Joined Sep 9, 2010
17,152
I don't think your op amp is powered correctly.

Never mind. I checked the pinout and it looks OK.

The 741 usually cannot get anywhere near the V+ power rail, so I'm not sure how you're getting such a high voltage output. Likewise it cannot sense very near either rail, so in this application your biasing needs to be very careful to stay in the operating range. You might choose a more modern op-amp. LM358, maybe. (different pinout - beware)
 
Last edited:

Thread Starter

Redline870

Joined Nov 13, 2015
22
Is the LM358 your top recommendation? I will buy some now if you think that will work better. I'm getting near the deadline and just want to get this thing done.

I thought maybe I had a defective op amp so I tried 3 others with the same result.
 

Dodgydave

Joined Jun 22, 2012
9,956
Use a Lm358 or Tlc271, if its working on your simulator, either you have a wrong connected circuit on the breadboard, or software issue.
 
Last edited:

ronv

Joined Nov 12, 2008
3,770
Is the LM358 your top recommendation? I will buy some now if you think that will work better. I'm getting near the deadline and just want to get this thing done.

I thought maybe I had a defective op amp so I tried 3 others with the same result.
The 741 won't work. It only responds to inputs 3 volts above ground and 3 below Vcc.
The 358 would be ok.
 

wayneh

Joined Sep 9, 2010
17,152
Is the LM358 your top recommendation?
It should be OK, and it's just one of the ones I have on hand and use from time to time. So maybe not a "top" recommendation, just a more-or-less generic op-amp that should work. I also have TLV271 on hand, and these are rail-to-rail. Nice, but possibly overkill for this application.
 

Thread Starter

Redline870

Joined Nov 13, 2015
22
That's so weird because it worked in the Multisim simulator software.

I don't currently have access to a LM358. The electrical engineering lab doesn't have any, nor do they have a TLC272. They gave me 2 different options they thought might work, a LT1635 and a TL071CN.

Would either of these work?
 

Thread Starter

Redline870

Joined Nov 13, 2015
22
This dude is dead set on the fact that this 741 op amp is going to work. He said he guarantees it will work. This is the new diagram he came up with:



I am not knowledgeable enough to tell him it's not going to work. How do I tell him the 741 is not going to work? What's the details on why it won't work? I'm sick of wasting time with failed designs! Very frustrating.
 

Dodgydave

Joined Jun 22, 2012
9,956
Make the circuit that he suggested, and see what happens, stop wasting time with software programmes,

its a differential amplifier with a gain of 50, the output of the op amp will swing as the temperature increases, the voltage will drop altering the pwm output.
 
Last edited:

Thread Starter

Redline870

Joined Nov 13, 2015
22
I got it working! I used the LT 1635, checked the pinouts, talked with the electrical engineering professor who recommended the part, plugged it in with the same resistor values and voila! Maintained 33.1 degrees Celsius all night, and super fast response from the thermistor if I spray it with key board air can or hit it with a heat gun. Super happy!

One last question and comment on this. It seems to have a fair amount of noise as you can see from the next picture. It doesn't seem to really effect the performance but if it's an easy fix, I would like to fix it. But I'm also of the mindset that if it aint broke, don't fix it. It seems to change when I move the power leads to the peltier and also when I move the leads for the thermistor. Should I wrap wires with something and ground them? Or should I not worry about it since it does work.



Lastly I just wanted to thank everyone that has responded to my questions and made suggestions. This was my first time posting in this forum and it was nice to not get condescending, belittling comments. I have seen that a lot in other forums with people who are not well educated or informed on what they are asking about, and it was nice to just get knowledgeable well-informed answers.
 

wayneh

Joined Sep 9, 2010
17,152
So the green trace is temperature? And the yellow is the signal to the peltier? Looks a lot like it's basically operating as a thermostat, which is fine.
 
Top