Thermometers

[tbostock]

What does everyone use to calibrate their thermometers in the Blood Bank? Currently, we use a glass certified thermometer. But they are expensive, break easily, and are difficult to send out for calibration due to their fragile nature. I'm looking at digital models; does anyone use these? The company states:

"To assure accuracy, an individually serial-numbered Traceable® certificate is provided from an ISO 17025 calibration laboratory accredited by A2LA. It indicates traceability to NIST standards."

Then, if I sent the digital out every year for calibration, I should be all set? About how much should I expect to pay annually for calibration for this type of thermometer?

Thanks for the advice.

[bxcall1]

I am currently having the same dilema. My plan is to see if our blood cneter will calibrate one or two of my thermometers so I can trace back to their NIST thermometer.

[John C. Staley]

If your facility has a clinical engineering or biomedical engineering department check with them. Ours had an electronic NIST I could borrow when needed and they had it calibrated on an annual basis. This worked great for me.

[Lcsmrz]

There is a difference between a certified thermometer and a calibrated one. The former comes with a certificate of accuracy that states that the thermometer was tested upon manufacture against an NIST-traceable thermometer and has met certain stated limits, typically +1C for BB thermometers. There is usually no expiration date.

The latter comes with a calibration report stating at which points the thermometer was tested (usually 2, 3 or 5 points), what the deviation vs the NIST is at each point (usually to two decimal points), and carries an expiration date of usually one year.

Then, there's the issue of the validity of test conditions for comparing dissimilar thermometers: total immersion, partial immersion, using a partial immersion LIG thermometer at an immersion level different than the line on the thermometer, and so on. I have the most problem with using the long NIST-calibrated thermometer in tight quarters.

We inspect and test our thermometers on receipt and annually thereafter for accuracy against our lab standard, an NIST-calibrated thermometer. Since the cost of recalibrating the thermometer approximates a new one, we buy a new one every two years (use on receipt, then again just before it expires). We do our best at standardizing the test conditions ...

Your quality plan should state the level of accuracy requried for thermometers and what the standard will be. For routine BB work, 1C resolution is sufficient. For enzyme work, 0.1C or less resolution may be required.

I like John's idea of having the facility provide one NIST-certified thermometer for use by all depts. A big cost savings!

And I've been trying to get away from using LIG thermometers for years, without much success.

[Franklyn]

In my experience, the NIST therms break most often during use when you are trying to manuver the long glass tube into or around equipment. It just isn't a good fit and takes a long time to do at great risk to the thermometer. We purchased a NIST calibrated thermometer that we send out for annual re-calibration. It really isn't all that expensive and the thermometer we purchased has a wide range (something like -50 C to 100 C). We then use it to calibrate our own electronic thermometer/meter with multiple test probes. We calibrate each probe at a variety of temps (-30 for FFP, 0 to 6C for blood, 37 for water baths etc.) and then use the electronic thermometer for all our other calibration needs. Breakage of the NIST has been eliminated and it is MUCH easier to calibrate equipment displays when you have a flexible lead with a remote display. The brand we have used is "Fluke" and the unit was less than $300. We have approximately 100 thermometers (including equipment displays) to calibrate annually and anything that sped up and simplified that chore was welcome. We could send out the electronic meter for professional calibration but have declined to do so out of consideration for the comfort level of assessors. The old timers just like to see a liquid in glass thermometer and we just want the get through the assesments with as little drama as possible.

[AMcCord]

Our Bio Med department contracts with a company to come in and check the calibration of their thermometers (and other stuff) once a year. They notify us when the company rep is coming and I deliver the 2-3 thermometers, tachometer and stopwatch I want checked. BioMed returns my stuff when they are done and brings me a printed report for each item from the company. The report includes points checked, the ID of the NIST standard checked against, etc, etc. From there I check all my other thermometers and centrifuges. Works well for us.

[Mabel Adams]

FDA apparently requires monthly calibration of digital thermometers unless the manufacturer specificially states that they can be calibrated annually. I am trying to move to a digital thermometer as my calibrated standard and would just as soon not calibrate other digital thermometers against it monthly. So, where can I buy decent inexpensive digital thermometers (to be used for taking the temp of blood shipments on arrival) that specifically state that they can be calibrated annually?

[dmpollock]

I have been using these thermometers for annually certifying all the other thermometers. They have a 2 year certification from manufacture. We usually have about a year and a half left when they arrive. When the certification is up I order a new one. Much cheaper than recertifying an old thermometer. Also no compaints from my FDA inspector, even though it is electronic.

Thermometer (NIST digital), Fisher Scientific, 14-648-12

Edited April 1, 2009 by dmpollock

[Mabel Adams]

Are the other thermometers you are certifying annually from this one all Liq-in-glass or are any of them digital?

[dmpollock]

I have used the Fisher thermometer (14-648-12) for everything from waterbath to blood warmers, including digital and glass thermometers. They cost less than $30. About 6 years ago when we last sent out the NIST glass thermometer for certification it was $100 every year plus shipping costs.

I like the Fisher thermometer because the probe fit into the blood warmer where the IV tubing would go. The measuring part is at the tip of the probe. Here is a picture (the one at the bottom) is most like the real thing: https://fscimage.fishersci.com/images/F20378~wl.jpg

Edited April 1, 2009 by dmpollock

[Vickibucklin]

Hi Terri,

Try googling "calibration cap". A calibration cap is a fixed precision resistor that simulates a predetermined temperature. It is very inexpensive, maybe $50, and it's a certified instrument that is UKAS traceable, an international standard equivalent to NIST. If all your thermometers fit this tool, you don't need to send anything back annually. Just buy one cap annually to use with all your thermometers for verification of accuracy that is accepted on an international basis. Pardon - I should say you only need to send those back that don't match the cap's reading within the degree of variation stated for accuracy.

What does everyone use to calibrate their thermometers in the Blood Bank? Currently, we use a glass certified thermometer. But they are expensive, break easily, and are difficult to send out for calibration due to their fragile nature. I'm looking at digital models; does anyone use these? The company states:

"To assure accuracy, an individually serial-numbered Traceable® certificate is provided from an ISO 17025 calibration laboratory accredited by A2LA. It indicates traceability to NIST standards."

Then, if I sent the digital out every year for calibration, I should be all set? About how much should I expect to pay annually for calibration for this type of thermometer?

Thanks for the advice.

[JohnD]

Hi Terri,

Try googling "calibration cap". A calibration cap is a fixed precision resistor that simulates a predetermined temperature. It is very inexpensive, maybe $50, and it's a certified instrument that is UKAS traceable, an international standard equivalent to NIST. If all your thermometers fit this tool, you don't need to send anything back annually. Just buy one cap annually to use with all your thermometers for verification of accuracy that is accepted on an international basis. Pardon - I should say you only need to send those back that don't match the cap's reading within the degree of variation stated for accuracy.

When calibrating digital thermometers either they should be calibrated as a system (meter and probe) or calibrate them separately using the 2 different standards, one for probe and the other for the meter. Calibrating just the meter isn’t sufficient and will lead to errors.

When calibrating Liquid in glass thermometers a procedure such as NIST monograph 150 should be observed. The working standard should be 4 times more accurate then the unit under test in order to have a confidence factor of 95%. When calibrating using a liquid bath the standard and the UUT tips should be at the same depth to account for the temperature gradients within the bath fluid. In the case with 305mm full immersion thermometers either you must use a bath deep enough or calculate a stem correction to obtain the correct reading. Most cal labs use a fixed mounted scope with a reticle to read the glass thermometers to prevent any parallax and obtain a higher resolution of the reading.

[TStec]

I have done my homework on this point because it was my job to verify all the liquid in glass thermometers and the alarm checks. We purchased two FLUKE 50 series II digital thermometers as our NIST traceable thermometer. We send one for annual calibration just before the other expires. We get a 4 point calibration because we use this to calibrate the thermocouplers used in our Central Alarm system which monitors LN2, Ultra low freezer, plasma freezers, refrigerators and incubators. The calibration is performed by State Technologies, recommended by central alarm company.

Before that we would purchase a cheaper NIST traceable thermometer (<$50) that had a 2 year expiration date. We had a quality monitor to verify all digital thermometers once a month which got to be many digital thermometers. Before the NIST traceable digital thermometers, we were purchasing a new liquid in glass thermometer annually (~$300).

We have eliminated all the liquid in glass thermometers in bottles from our storage equipment except in the remote location equipment. We found that no one checked these bottled thermometers anymore when an alarm occurred unless the central alarm system was down. Now if the central alarm is down, we use the FLUKE to check storage temps every 4 hours, placing the thin wire probe into the bottles where the alarm system thermocouples reside. Then we call the remote locations to check their liquid in glass thermometers.

Therefore, the only glass thermometers we now have to QC are waterbaths, and testing incubators/ racks and those remote locations. When we do the qc for the central alarm thermocouples we test temp against NIST traceable FLUKE, then we test the audible alarm conditions by slowly raising and lowering the probes. I never could get that stirring slowly to record temp at alarm, too impatient! Our monthly quality monitor is of our storage condition monitoring on a schedule as recommended by the manufacturer (per FDA) not all those bottled thermometers periodically that I used to do.

During our last AABB assessment, we were questioned on the storage of derivatives with the red cells, there was a lower limit of 2 degree C compared to the 1 degree lower limit for red cells. I produced records from our central alarm system, grafics that showed the 2 degree limit was never reached in that refrigerator. We changed the alarm criteria to 2 degree C so now the RBC and other products live happily ever after in same fridge!

Edited April 15, 2009 by TStec