Hi, I need to measure the temperature of a sample with high accuracy (accuracy better than 50 mK as per specification) and long-term stability in the range 77-180K. My application has 2 set-up: one immersed in LN, one in a vacuum chamber. I would use the Pt-103, calibrated version for high accuracy. I also need to be able to easily remove the sensor, so I would chose the Pt-103 AM packaging. I cannot bond or tape sensors on the sample.
I recently heard that the PT-103 AM reading is not stable when used in vacuum. This was unexpected according to my experience. As a matter of fact, this model is 'not reccommended' to be used in vacuum in the last datasheet. The suggested model is the PT-111, but is it not provided with the AM packaging.
My questions are:
- have you any further information on this issue on the Pt-103? What are the contraints to use these sensor in vacuum (pressure interval, for instance) and/or immersed in LN? Are special calibration curves available for this applications?
- Is there any packaging for the Pt-111 similar to the AM one?
Post by Lake Shore Jeff M on Sept 12, 2019 7:46:27 GMT -5
Yes, based on an investigation of an issue reported by another customer, we have determined that the PT-102 and PT-103 sensors are not suitable for use in any vacuum. We have found that most of these sensor exhibit a change in their resistance as vacuum is applied causing the interpolated temperature to increase. Once vacuum is removed, they return to their normal state. As a result, we do not recommend them for use in a vacuum environment.
Is there a specific reason you choose the Platinum sensor? If you are not using the sensor in a magnetic field, you could use a silicon diode in the CU package. If you are in a magnetic field, the Cernox sensor in the CU package would also work. In all cases, to obtain a 50mK accuracy, you will need a calibrated sensor.
Thank you Jeff for your reply. I had selected Platinum Sensors for their long-term stability. I cannot easily remove sensor to re-calibrate them. Platinum is reported a +-10mK long-term stability, while Silicon diodes +-40 mK (at 77K). Can you provide more data on lon-term stability? In this case, silicon diodes can be an option. Another question: do you reccommend PT-103 AM for using immersed in LN (atmospheric pressure)? As an alternative, can DT-670-CU used immersed in LN? Thank you again for your support.
Post by Lake Shore Jeff M on Sept 12, 2019 15:26:06 GMT -5
Basically, our long term stability was determined by thermal cycling (gradual transition in temperature, not a shock) several calibrated sensors of each type between room temperature and 77K 200 times. We then recalibrated the sensors and computed the average change which is what is defined as the long term stability.
Our recent testing of the Platinum sensors revealed that the PT-102 & PT-103 sensors are not hermetically sealed. However, we know that this is a common use so as long as it is a gradual transition it should be OK. Our experience with shocking the sensor by immersing quickly into the LN, can cause the case to crack.