Exploring the use of Peltier plates for a microscope heating stage. A cautionary tale.
by David Walker, UK
In the August 2015, April 2018 and May 2018 issues of Micscape I shared my trials of using Peltier plates, both solid and with a central hole to make cooling stages for hobbyist microscopy. A particular goal was to be able to study water freezing under a coverslip with indoor comfort which was successful with later designs. Studying frost formation on solid plates was also fascinating.
Peltier plates can also be used for heating, either by having the hot side uppermost or using a cooling stage design and reversing the polarity of the power supply. Early trials of using both a solid and annular plate in this mode are summarised below.
Peltier plate TEC1 04903, 5.8V, 3.3A solid square plate
The existing setup shown below had been used to study frost formation. A solid plate was tried first as they are cheap, had some spares and did not want to risk the single and more expensive examples of annular plates possessed. The polarity was reversed so the top plate was now the hot side. The copper plate and heatsink now assumed the role of a large thermal mass to heat the hot side. The data sheets warn of not having a bare plate on the hot side so a piece of slide with a well ring filled with immersion oil was used. A type K thermocouple dipped into the oil and attached via a USB interface to PC software was used to monitor the results in real time. The volts were slowly increased in steps and equilibrium was quickly reached at each setting in a few minutes.
Although in data sheets the ΔT temp was stated between the hot and cold side, to date had not found any figures for the maximum permissible temperature of the hot side. A commercially produced Peltier based heating /cooling stages by Linkam had a temperature limit of 120°C so this temperature was the maximum tried and for this plate at least it seemed stable. The temperature also dropped rapidly when the current was switched off as heat is rapidly conducted through the plate, a feature that had noticed for cooling studies as well. These were encouraging results for good control of temps to 100°C so prompted me to try an annular plate which would be much more practical for transmitted studies. The maximum current of 1.8A used was also barely half the plate's max. rating of 3.3A.
TES1 04903, 5.8V, 3A annular plate
The setup shown below was used and was the design finally adopted for earlier transmitted studies with a hole drilled in the copper plate below the plate hole. A fan was not needed for either this setup or the one above to keep the heatsink at room temp. as there was little if any cooling detected. Good control of the temperature was achieved with rapid equilibrium. Until the current was increased a small step to go just beyond 100°C; this time the plate irreparably failed as shown by the rapid heat loss. This was the only example possessed so have had to order another from China, the only source have found to date.
Comments on trials to date
It is not clear from the limited trials to date whether the TEC1 plate is stable at 120°C or was just lucky that it did not fail. On the EVERREDtronics website the TES1 range are stated to have a higher density of smaller elements so maybe this is a factor. Have also noticed a footnote on this site which states that 100°C is the maximum recommended plate temp. for most models which if the case suggest little room for any leeway on the TES1 model used.
Although commercial Peltier cooling / heating stages are presumably reversing the polarity to select the mode, it may be prudent for homebrew trials to retain the plate polarity but place the hot side uppermost for a heating setup. I was hoping to have a dual cooling / heating setup by just polarity reversal to avoid cleaning off the thermal paste and reversing. When the above two plates were used in cooling mode and correct polarity, Vmax and Imax correlated well. When these plates were used as above for heating, although nearing their Vmax, the current was barely half Imax, suggesting with reversed polarity the junctions were not behaving in the same way (possibly stressed?).
The aim of these trials was to have a setup to gently heat aqueous solutions to encourage crystallisation and study in real time under crossed polars. The same speeding up may also be achieved by adopting slightly concentrated solutions and then chilling. For me, the latter seems the safer route for plates owned and the one will adopt in future. If pursuing heating trials, unless have firm data for a given model for their max. operating temp would advise to keep the max. temp quite modest say 75°C although this should be more than enough to speed crystallisation.
There is an argument that cooling is a better approach to force crystallisation as the crystals will be surrounded by solution. Forced evaporation may form crystals surrounded by an aqueous/air interface which can give unwanted artifacts under the microscope.
The EVERREDtronics site list Peltier modules designed to go to 200°C but no annular models were listed which are the best for transmitted studies.
Comments to the author David Walker are welcomed.
Thank you to Michael Wolfson for stimulating email discussions on cooling and heating microscope stages.
Published in the January 2021 edition of Micscape.
Please report any Web problems or offer general comments to the Micscape Editor .
Micscape is the on-line monthly magazine of the Microscopy UK web site at Microscopy-UK
Ltd, Microscopy-UK, and all contributors 1995 onwards. All rights
Main site is at www.microscopy-uk.org.uk with full mirror at www.microscopy-uk.net .