We are looking into setting a wider temperature tolerance range for the QC of our liquid nitrogen temperature measuring devices. We use a couple of different types of devices, some of our models utilize probes that connect to a thermometer or a datalogging device, and in these types, I understand the tolerance of the device and of the attached probes compound; one has to be added to the other to get the true acceptable tolerance. We use the liquid phase of Liquid Nitrogen to create a stable environment at -196 C. We frequently struggle to meet +/- 2 degrees Celsius, much less the +/- 1 degree tolerance used for the higher temps. Is anyone out there using a temperature tolerance range wider than +/- 2 degrees Celsius for ultra cold temperatures? Are there any literature references that substantiate the establishment of the wider range?