Basic Thermometry


When mounting the sample or a thermometer onto the sample platform, you need to do your best to promote thermal conduction between the sample and the flex-link via the sample platform. This can be achieved through the use of either thermal varnish, which has a thermal conductivity an order of magnitude better than thermal grease, or with the use of silver paste which has a thermal conductivity an order of magnitude better than thermal varnish.

With the use of either the varnish or the paste you will need to clean the surface of the sample and the sample platform well with acetone to clear all organic contamination, only apply a very thin layer of the material between the surfaces, and finally while the substance cures try and apply a downward force to the sample.

A good tip when using the varnish: if you thin out the varnish a bit with some acetone, the layer of varnish can be good and thin for the sample attachment.


The thermometer is chip mounted on sapphire base with alumina body and lid, Mo/Mn with nickel and gold plating on base and lid, Gold-tin solder as hermetic lid seal, 60/40 SnPb solder used to attach leads. The main thermometer sensor is based on a resistance measurement of a metallic element as a function of temperature, usually with two resistors in parallel, one for the low temperature range and one for the higher range.

Note that because the thermometer is based on the measurement of a parallel configuration of resistors, if just one resistor somehow breaks the chip will still give a reading! The only way to check to make sure the chip is still fully functional is to test over the full range of temperatures from 300K down to 4K. If one resistor is broken, the reading will be accurate for one range and not for the other.

The active thermometer is trapped between layers of material that are poor conductors of heat, so once radiation heats the device it will take time to dissipate. And to make things worse, the ceramic actually absorbs radiation and gives a false reading if you just place the thermometer on the sample platform without taking proper precautions. 

There are two main sources of thermal energy, one is the radiant energy from the 30K radiation shield and the other is the thermal energy from the electrical connections to the thermometer. This is not an Ohmic effect, instead it is due to the wires lagged at a warmer point than the sample. This is why good lagging is needed at the 4K stage for accurate readings.

The main methods of getting a proper thermal reading is to protect the thermometer from radiation and to lag the lead of the thermometer to keep energy coming onto the thermometer from the metallic wiring.

To protect the thermometer from radiation, a simple shield can be made from a layer of aluminum foil over the thermometer once the thermometer has been placed on the sample platform. The aluminum foil has a high coefficient of reflectance, thus a very low percentage of the incident radiation is absorbed.  To lag the thermometer leads, more varnish can be used to “glue” the lead to either the platform surface or the top of the sample platform, or preferably both! This is of course in addition to the 30K lag at the thermal clamp.


To address some of the issues one can see with poor thermal lagging, we have run a few simple tests. First, we just placed the thermometer on the sample platform with NO lagging in a Cryostation with the low working distance window. We saw that the main platform got to base, which was ~ 3.8K, and the sample thermometer only got down to ~ 12K. The thermometer was covered with a layer of aluminum for radiation shielding. This is obviously too high and will need additional thermal lagging.

Next we simply lagged the sample thermometer wire to the sample platform with thermal varnish. The lagging was only a few centimeters long. With this configuration, we were able to get the sample platform reading down to 4.4K while the main platform was reading 3.8K. So, with some simple steps, the thermometer can be better shielded from the incoming heating from the wires and give a more accurate reading of the sample platform.

Also, we had our production engineers do a similar measurement, and have successfully cooled the sample to below 4K with the low working distance. The results were: platform 2.9K and the sample 3.39K, and 2.9K on the platform and 3.25K on sample. This was a case where lagging and radiation shielding was used for the sample thermometer. See image below.


After placing new sample platform on the flex-link with a thin layer of varnish, the wiring will still need to be lagged. To achieve good lagging a few things can be easily done.

  1. Make sure the wiring is run the through the 30K thermal clamp at the stage 1 point under the radiation shield.
  2. Varnish the wire to the platform or connect the wire to platform via a simple block with one of the platform screws. Make sure not to pinch the wire so much to short lead.
  3. Varnish the wiring also to top of the sample platform at least a few centimeters long.
  4. Make sure all varnish is cured before trying to cool system, 30 minutes under a heat lamp, or 2 hours without heat.