Critique My Shiny New Circuit

[Yossarian451]

So thanks to those who fixed my little comparator mixup. Here is the final product. Comparator Happiness . Yeah it is basically an extension on the common circuit, but mine allows me to set an exact turn on and turn off temperature. I like it better than the ones that just do a timer to prevent cycling, and it costs significantly less. Once I finish the final setup I will take pics. Ohh and some of the resistor values are somewhat arbitrary because I had them lying around. But those are mostly just the ones that are coupled with the leds, and the grounds for the flip flop pin.

 

[TuxDave]

Although I have no idea what's needed in a refridgerator control unit, good job. Is there any reason why you have two comparators with the inputs swapped on the left side? I'm thinking you could've used a double ended differential amplifier instead of a single ended one.

And what's that thing in the middle. It looks like a J/K Flip Flop but I'm a little confused about what's happening in regards to the clock pin.

 

[Yossarian451]

Originally posted by: TuxDave
Although I have no idea what's needed in a refridgerator control unit, good job. Is there any reason why you have two comparators with the inputs swapped on the left side? I'm thinking you could've used a double ended differential amplifier instead of a single ended one.

And what's that thing in the middle. It looks like a J/K Flip Flop but I'm a little confused about what's happening in regards to the clock pin.

Umm, sorry about the diagram for the j/k, it is a neg-edge trigger so I have the cap there to provide a negative edge to the signal that stays high. The LED allows me to know that I have a signal coming from the Comparators, mostly for just my own reference. The top comparator turns on the fridge when the thermistor goes high, the bottom is switched so that it goes high when the temp goes below the specified temperature. The j/k will stay on anywhere between the range and turn off below the bottom mark. Also since the thermistor is not linear using a pot with a feedback resistance would be hard to accurately mark. This gives me better control on what range I keep the fridge at rather than relying on a simple diferential effect of the air, etc. So far in my limited proto tests it works like I want it to.

Next step I will solder and attach to the housing. The relay will go to an actual outlet that I have the power cord of the fridge attached to, and the input I gutted an old power supply box. Hopefully it will be pretty solid.

 

[Mark R]

Nice.

Personally, I'd have used 1 op-amp to buffer the thermistor, and configure a 2nd as a schmitt-trigger comparator. Then drive the relay from the buffered output from that.

To be honest, you probably wouldn't need to buffer the thermistor, as long as you used highish resistor values on the schmitt trigger. So the entire circuit could be done with a single op-amp and a transistor + a few passives.

 

[Yossarian451]

Originally posted by: Mark R
Nice.

Personally, I'd have used 1 op-amp to buffer the thermistor, and configure a 2nd as a schmitt-trigger comparator. Then drive the relay from the buffered output from that.

To be honest, you probably wouldn't need to buffer the thermistor, as long as you used highish resistor values on the schmitt trigger. So the entire circuit could be done with a single op-amp and a transistor + a few passives.

How would you accurately control the temperature which it is cut out at with the schmitt-trigger. I say this because the thermistor is non-liner so if it was a lower temperature the value of the feedback resistor would be different than at a higher temperature to give say a 5 degree cutoff point. I am interested in alternative ways. I also used the comparators because I had experience with them and since they come 4 in a chip, I used it in this configuration.