I'm working on an experimental setup for my thesis. As I have limited experience with electronics, I would appreciate some feedback on the setup before I order the components.
The idea is to create a small, insulated Styrofoam box that can regulate temperature between 0-60 °C with a heating/cooling rate of 1 °C/hour. The setup will include three temperature sensors: one measuring the ambient temperature in the box, and the other two inserted into samples inside the box.
I'm considering using the ESPDUINO-32, as it would make data collection easier. To control the heating and cooling modules, I plan to use MOSFETs, which should help maintain a consistent heating rate and prevent overheating.
Any advice or suggestions would be greatly appreciated!
ESPDUINO-32
3 x DS18B20 temperature sensor
2 x MOSFET 15A
20x4 LCD with I2C adapter module (Which will show ambient temperature, and temperature of two samples)
Peltier module with heatsink
Heating pad (15V)
Power supply adapter (voltage and current?)
Voltage regulator or buck converter?
Breadboard
Jumper wires
4.7k ohm resistor (for DS18B20)
2x 10k ohm resistors (for MOSFETs)
Barrel jack adapter or screw terminals? (for power supply connection)
Heat sink compound (for Peltier module heatsink)
So I have a few questions,
Can I regulate the peltier module with the MOSFET? Which peltier module would you recommend for this application?
A Peltier module needs heat sinks on both sides of the device and a fan to cool the heat sink on the hot side. Are you intending to alternate the hot and cold sides?
Thermal calculations are tricky without some experimentation. You said "small" and you're looking for a slow temperature change so that's good.
If the volume is perfectly insulated (impossible in the real world) and if you can "take forever" it doesn't take much energy to get as hot or cool you want.
A separate heating element (which can simply be a power resistor) might be better than reversing the Peltier.
If you can get your hands on a regular-old light bulb that's an easy way to get a "small" amount of known wattage for heating. Electricity is converted 100% to heat. (Even the light is converted to heat when it hits the walls of the container.) The Peltier well be less efficient so you'll need more energy for it.
Virtually all heating & cooling is done by cycling on & off. You've probably noticed that with your furnace you may be able to hear your refrigerator when it cycles on & off. So you turn-on the heat and then turn it off when you reach the target and then on again when the temperature drops.
In your case you can step the temperature at 1 degree per hour, or 1/10th of a degree per 1/10th of an hour, etc. The temperature can't change instantly and the temperature will vary at different places in the container so it can never be "perfect"...
Usually there is some "swing" or hysteresis so it goes slightly above the target (when heating) and then slightly below before cycling-on again. That can be a fraction of a degree, or you may not need it at all with a solid/state system.
You should know Ohm's Law, (Current = Voltage/Resistance) which describes the relationship between voltage, resistance, and current. Most of the time voltage is constant or "known" and current depends on resistance. (Resistance is "the resistance to current flow".) Your power supply has to be capable of supplying the required current but the actual current depends on the load.
And you should know how to calculate power. The basic formula is Watts - Voltage x Current. And there are variations so if you know the voltage & resistance you don't have to calculate the current, etc.
I only was planning on using it for cooling and the heating pad for heating, as i thought that would be easier.
Ah okay, thank you for your response!!
Do you think a heating pad will also work? So by adding resistance the light can be dimmed/adjusted so that it is very gradual?
Study the documentation for Peltier. You must have a heat sink on both sides. When you turn the Peltier off, all the stored internal heat must be dissipated. or the diodes will open up. 1/2 the internal heat will go to the hot side and 1/2 the internal heat will go to the cold side. So, any cooling will occur ONLY while the device is powered. When turned off, the cold side will get hot for a while.
Cation, Peltier diodes do not like PWM and they require a lot of current. Adding a small fan in the box will help even the temperature gradient in it. You can also use heating elements, running them at a lower voltage will not hurt them but actually extend their life and they will be much cooler.
When using a Peltier to cool a box, the heat must flow through the box insulation, with the Peltier module sandwiched between finned heat sinks on either side of the box insulation.
Run the power to the project thru the Peltier. They are a current device. Lots'o Amps, less so on the Volts. They install at the thermal boundary. Use liquid immersion for Peltier hot side cooling to reduce/eliminate fan depending on coolant reservoir.
If it’s a small well insulated box , you will probably need a permanent cooling load to get any chance of temperature stability .
Old fridges used to have a heater inside !! For a similar purpose . It’s also common for climatic cabinets to have a cooling load when operating around ambient temperature .
Small peltier cooler and a resistor as a heater , PID controlled . Don’t forget to mention us as providing your solution in your write up
Duh! (slaps forehead) You don't need BOTH a Peltier AND a heater. Just reverse the current thru the Peltier and the hot and cold sides reverse. Drive it with a bipolar signal and you can make a nice hot or cold adjustment.
Yes I know how a peltier works . In this application you will need a load for the heater to work against otherwise you won’t get a stable control as the box is small and it’s heat loss low . At the set point you would have the heater and chiller probably on to balance each other .
Commercial climatic cabinets need cooling on when operating around say 10-30C and maintain set point using the heater against the cooling load .
What you propose may work , but conventionally a load to work against is needed to heat or cool a box . Bit like trying to control the speed of something on a frictionless surface , it’s easier with a bit of friction