Refrigerator/Freezer MU - HOW?

[calgal]

Our calibration lab is would like to become ISO 17025 accredited, and since that requires all calibrations to include MU - How do I calculate MU for something like refrigerators/freezers?

I currently have MU budgets for pipettes and balances, and I understand the contributing factors.

I just have no idea what to do with something simple like temperature readings.

[Hershal]

In Reply to Parent Post by calgal

Our calibration lab is would like to become ISO 17025 accredited, and since that requires all calibrations to include MU - How do I calculate MU for something like refrigerators/freezers?

I currently have MU budgets for pipettes and balances, and I understand the contributing factors.

I just have no idea what to do with something simple like temperature readings.

This is very much like an oven survey, but in reverse.

You will need to know the setting that you use, and if like many labs you simply have a commercial or residential model then the settings will be some number or set of numbers, depending on what is inside. The same is true if those are the types you calibrating for other organizations such as a test laboratory that has the freeze/thaw test methods.

Typically in an oven survey - that is, a FULL survey - there will be nine readings. Three readings at each of three levels (top, middle, bottom). That is repeated for each setting necessary. Use a digital with a thermocouple. You can get 10-30 readings for each over the course of a couple of hours, which you need for your Type A.

Type B contributions include the calibration uncertainty of the digital thermometer, uncertainty for the thermocouple itself, and if there are multiple settings then there is a contribution from the time to get to the set point from previous set point.

Now, in truth, the number of readings can also be taken from a single point, and a simple difference taken as a Type B for the other levels or points normally taken. That difference will be a rectangular distribution, but a slightly more conservative approach would be a U shaped distribution. Your uncertainty then would be slightly larger, but if that is not an issue for customers then it is the approach I would take.

Also, if you seeking accreditation, go to www.ilac.org and see what accrediting bodies are in your country (I did not check your profile). If you are in the U.S., there are six, and I strongly suggest detailed discussion with all before signing with one. Discussion should be more than a quote, you need to talk to the Lead Assessor and have a detailed technical discussion before a decision is made.

Hope this helps.

[Hershal]

Oh, I just checked, welcome to the cove.

Please do continue with questions. We love that!