Peltier module use improvement

Question

I am trying to setup a box, where I can keep the temperature to constant/near constant number of degrees. I have chosen to use a Peltier module for cooling but the result isn't satisfactory.

A similar problem is faced here Peltier module not cooling

and I am trying to understand how to improve my setup.

The unit I use is rated 12 V 6 A with a cool side small fan and a double fan for the hot side. When the unit is connected to power it warms up on the hot side and is cold on the cool side.

https://www.banggood.com/Geekcreit-12V-6A-DIY-Electronic-Semiconductor-Refrigerator-Radiator-Cooling-Equipment-p-1074404.html?cur_warehouse=CN

After a while ice is built up on the aluminium radiator so the Peltier works. I can also control the fans via PWM from an MCU.

But although the cool side works, I don't get much of a temperature drop in the box. I have noticed also that the fan sucks the air and does not blow it to the box area.

My questions are:

1. Should the cool side small fan drive the air on the Peltier or away from it? What should I do, maybe place a larger fan on top to drive air away from the cool radiator and into the box?
2. Can I control the Peltier temperature with PWM, as I do with the fans, or will i end up burning the unit. I should mention that I control the PWM with two MOSFETs (TIP220) to drive the max 12 V to the fans.

Answer 1

Should the cool side small fan drive the air on the peltier or away from it. What should I do, maybe place a larger fan on top to drive air away from the cool radiator and into the box?

What you need to do is circulate air between the Peltier cool side and the volume of the box. It doesn't matter whether the fan is moving air from cooler to box, or the other direction, as long as you have a return air path to circulate the air around without it leaving the interior of the box.

Can I control the peltier temperature with pwm, as I do with the fans, or will i end up burning the unit. I should mention that I control the pwm with two mosfets (tip220) to drive the max 12v to the fans.

Fast PWM (many Hz or quicker) is inefficient, but doesn't damage the Peltiers. Steady current of the right level is far better. Slow PWM (or on/off control) is both inefficient and will damage the Peltier module.

The difference between fast and slow is whether the temperature of the Peltier follows the on/off cycle of the PWM. Peltiers have a wear-out mechanism whereby repeated temperature changes can de-bond the elements from the top and bottom plates.

The difference between PWM and steady current is the fact that the heat generated in the Peltier is proportional to current squared, whereas the heat shifted into the cold side is only proportional to current. Turning the current on and off (ie PWM) generates more heat for the heat shifted than does a steady current of the right magnitude.

Answer 2

Should the cool side small fan drive the air on the peltier or away from it. What should I do, maybe place a larger fan on top to drive air away from the cool radiator and into the box?

It doesn't matter as long as the air is circulating in the box. What does matter is if you have any kind of control sensor (thermistor, RTD ect), then you want that to be in the downstream air of the peltier heatsink.

Also, by radiator I assume you mean heatsink and it needs to be large enough to transfer heat to the air effectively. Also you need to have a high enough R value on the box to effectively cool it. There are ways to calculate how much heat will leave the box through the insulation, I won't go into that because it's a detailed thermal problem. But you have to have enough insulation on the box to cool it.

Can I control the peltier temperature with pwm, as I do with the fans, or will i end up burning the unit. I should mention that I control the pwm with two mosfets (tip220) to drive the max 12v to the fans.

You can do pwm, I usually would do a 5Hz or 10Hz cycle and then control the power with a duty cycle. PWM with a switch like a mosfet that has a low voltage drop is best for power purposes. You can use a high power amplifier to do variable voltage control, at the expense of power dissipation. The other problem with PWM is it gives off noise, and it can be hard on power supplies.