0
\$\begingroup\$

My team is doing a senior project that involves using polymerase chain reaction to analyze DNA. From what we know, most modern thermal cyclers uses the Peltier thermoelectric modules for active heating and cooling. For those who aren't aware, a thermoelectric effect creates a voltage when there is a temperature difference on each side of the involved surface.

Our research indicates that while Peltier thermoelectric modules are simple, accurate and do not involve any moving parts. However, they are very energy inefficient. Most input energy into the thermal cycler is lost as heat.

Our team then tried to propose other alternative methods for active heating and cooling. Below are some of our ideas (heat sink is out of the question).

1.) Cooling the thermal cycler with fan modules, while heating the thermal cycler with wires.
2.) Cooling the thermal cycler with a custom water cooling system (like those used in car engines and computers).
3.) Using a mixture of the fan and water cooling system.

Cost is not the biggest constraint here (although it can be touched upon). Does anyone here have any experience with doing heating and cooling system with electronics? If so, could anyone explain the ups and downs of the methods I've suggested? If there are any better alternatives, please do suggest them so we could do further research. Thanks.

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Do you have a spec that says efficiency matters? Usually, when everybody does something one way, there is a good reason for it. \$\endgroup\$
    – Matt Young
    Sep 28, 2015 at 20:05
  • \$\begingroup\$ Hi Matt. Yes, energy efficiency is the biggest criterion here. I'm sorry for not making it clear in my question. \$\endgroup\$ Sep 28, 2015 at 20:08
  • \$\begingroup\$ Sorry, but efficient cooling and Peltier do not mix. If you're working at relatively low power levels (in the 40W region) you can't beat a compressor based cooler. Take a look at "Danfoss BD35" as a starting point. Vibration and noise may be issues but you can isolate compressor from everytnhing else. \$\endgroup\$
    – user16324
    Sep 28, 2015 at 20:13
  • \$\begingroup\$ For a twist a Stirling engine based cooling and heating might be fun thing, but at that temperature levels probably not very efficient, and hard to get. \$\endgroup\$
    – Arsenal
    Sep 28, 2015 at 20:54
  • 1
    \$\begingroup\$ What are your constraints?? Also, what are your specs? Start with figuring out WHAT you need to achieve. HOW becomes less important, so long as you meet spec. If any or all of those listed options meet spec, they may all be perfectly acceptable. A lot of engineering resources get wasted by trying to achieve better than good enough. \$\endgroup\$ Sep 28, 2015 at 22:48

2 Answers 2

1
\$\begingroup\$

Its not about efficiency, its about choice of application. If I am making an air conditioner of 1000 watts, I will still choose conventional vapor compression ( even though it has thousand times more global warming potential than carbon dioxide itself, and high efficiency thermoelectric materials ZT > 2, are still in development).

But for smaller loads 10 watts, 20 watts, 100 watts etc. ,to actually cool them, or cooling/heating/thermal cycling, thermoelectric systems are better, nor will I get a 20 watts air conditioner in market. The best part about thermoelectric systems is the response time from heating temperature to cooling temperature and vice versa, its very quick in seconds especially in your case, where cooling load is generally small in PCR.

Now, I do not know the response time and what is required in your test, but you should check that whether your above proposed systems will respond quickly to required temperature in the desired time, as you want them.

You can make thermoelectric cooling heating system yourself, its quite simple.

\$\endgroup\$
0
\$\begingroup\$

Just taking a swing at it, I think this may be the most all-around efficient solution.

Reversible Heat Pumps

It gets hot, it gets cold, and it's thoroughly studied. (And weirdly, since it's moving heat as well as creating it, the heating mode produces more heat at the hot side than it consumes!)

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.