Hello everyone, I am currently setting up a measurement platform to evaluate the pyroelectric properties of thin-film materials using a dynamic AC method and need some advice on selecting the right Thermoelectric Cooler (Peltier module). I recently purchased a few standard generic modules from a local electronics store, specifically the TEC1-12708, TEC1-12706, and TEC1-7103, which you can see in the attached photo of the packaging. I am driving these modules with a 0.1 Hz sine wave at a 10V amplitude using a signal generator paired with a power amplifier. However, under this specific AC driving condition, the observed temperature amplitude only fluctuates by about 1 to 3 degrees Celsius, which I suspect is due to the large thermal mass and steady-state cooling design of these standard modules causing a severe lag in thermal response. My goal is to find a TEC module that can achieve a rapid temperature amplitude of 10 degrees Celsius or more under a 0.1 Hz AC sine wave drive, exactly like the ideal temperature fluctuation shown in the top half of my second attached reference image. I would like to know if DigiKey carries specific lines of TECs designed for rapid thermal cycling, such as micro-TECs or PCR-style modules with very low thermal mass, and if there are any specific brands or part numbers you would recommend for this type of dynamic temperature modulation application. Thank you in advance for your time and help.
We do have modules designed to be thermally cycled, but I don’t know how rapid that would be. They can be found here.
Also, I’m not sure you’re going to get the desired effect driving these with a 0.1Hz sine wave.
Thank you for providing the link to the thermal cycling modules! I will definitely look into those options.
Regarding your concern about driving the TEC with a 0.1 Hz sine wave, the reason we use a sine wave is specific to the Garn-Sharp dynamic method for characterizing thin-film pyroelectric materials. In this method, the pyroelectric current is proportional to the rate of temperature change. By inducing a sinusoidal temperature variation, the resulting material current becomes a cosine wave. This exact 90-degree phase shift allows us to use phase-sensitive detection to perfectly isolate the true pyroelectric signal from other non-pyroelectric background noises.
As shown in the reference graph I attached earlier, driving a TEC with a 0.1 Hz sine wave can produce a matching temperature sine wave (top red line) and current response (bottom blue line), provided the TEC has a low enough thermal mass.
Since a 0.1 Hz frequency means one full cycle takes 10 seconds, it’s actually quite slow, but the challenge is purely overcoming the module’s thermal inertia. Given this 10-second cycle requirement, when looking at your Cycling series, would you recommend prioritizing a smaller physical footprint (e.g., 20x20mm instead of 40x40mm) or a thinner module profile to get the best dynamic temperature tracking?
(Note: In our experimental setup, the thin-film pyroelectric sample is directly attached to one surface of the TEC so it experiences the time-dependent temperature variations, while the opposite surface is firmly mounted to a robust heat sink to prevent overall heat accumulation.)
Thanks again for your valuable expertise!
Selecting a unit with a minimum ratio of inactive mass to rated cooling power would seem like a more important consideration than either of the above. While information about the thickness (mass) of the face plates often isn’t available, selecting a device without a silicone sealant would likely be beneficial for the stated purposes.
Thinner modules generally “leak” thermally through their thickness at a higher rate, which might possibly be either useful or bothersome here depending on the design of the apparatus.
Other things being equal (which they often are not) module footprint size seems like a less important consideration. If anything, one might imagine a larger area to be beneficial, for reasons such as reducing the relative size of perimeter effects or increasing the size of the signal; if the measured quantity is at a scale of microamps per square meter and the sample size is a fraction of a square meter, the signal to be measured is becoming small enough to present measurement complications of its own.
One might also consider using temperature offsets in the system so that the bottom of the target temperature waveform stays above the heat sink temperature throughout. This lets the 2nd law of thermodynamics to work in one’s favor consistently throughout the cycle from a rate perspective, and take advantage of the TEC’s much greater efficacy at heating compared to cooling.
Finally, one may want to consider conducting the experiments under some degree of vacuum, to lessen the influence of convection heat transfer.
Thank you so much for the insightful advice! Your points regarding the thermal dynamics and measurement complications are exactly what we needed to hear.
The suggestion to use a temperature offset to keep the bottom of the waveform above the heat sink temperature is brilliant. We will definitely implement this DC offset to leverage the TEC’s heating efficiency and the second law of thermodynamics. We will also take your advice on minimizing convection into account for our chamber design.
Given your refined criteria, I completely agree that minimizing the inactive mass-to-cooling power ratio is the priority. Since a larger footprint is actually beneficial for maximizing our microamp-scale signal, we are perfectly fine with standard or larger sizes.
Based on these specific requirements : no silicone sealant, a low inactive mass, and a strong cooling power/thermal cycling capability, could you possibly recommend 1 or 2 specific part numbers or series from your catalog that best fit this profile?
Non-sealed options seem comparatively rare, which limits the possibilities. 387005662 might be the best candidate at a glance, though 387005685 would also appear to offer a rather higher Qmax/area ratio than suggested for the items referenced by the photo above.
I cannot guarantee that either will meet your needs, but these seem like the options that would be most likely to do so.