Inspection time. When a food inspector walks in your door, the key question is: Where's your thermometer? And the food inspector has one, too. Indeed, the very act of measuring temperature is Rule #1 for food safety. The word thermometer pops into many standard operating procedures, as well as prep instruction forms, logs, and HACCP plans.
As you wield this essential device, how do you know you are collecting meaningful information?
Accuracy: Scientifically, it means a collection of many factors. Practically speaking, it means whether the reading you write down on a temperature log (or record electronically) is really right.
What can affect it? Construction materials. Response time. How the thermometer is calibrated. Interference. Resolution. How you read the number. Even where you put the probe to begin with. For starters.
Let's look more closely....
Construction materials: The sensor in a thermocouple is generally an alloy. The exact choice of materials influences accuracy. Probes, too, can vary greatly in construction. For electronic thermometers, there is a broad range of circuitry and electronic components available today, each with variations in quality, reliability - and impact on your actual readings.
Response time: How long does it take to get a reading? And do you, as a user, wait long enough? There are actually NIST standards to address this. If you are using interchangeable probes, it's good to know that response time depends on the probe, too.
Calibration: You need to know that when a food is 155°F, that's what the thermometer displays. And then you need to know that that calibration is constant—or know when it's time to re-calibrate. For example, a bimetallic thermometer is very easily knocked out of calibration through dropping or jarring it. Even the process of calibrating is a tricky issue.
You need to know that if a thermometer is well calibrated for the low end of the temperature range, it's calibrated at the high end, too. Experts suggest that to calibrate a thermometer, you need to check it at two or three points within its temperature range (preferably with specialized equipment!). Unfortunately, for a bimetallic, calibrating it at the low end in an ice slush means making it less accurate for critical endpoint cooking temperatures. To get a handle on tricky calibration issues, you can select high quality thermometers that are traceable to a NIST temperature standard.
With units that have attached probes, another calibration factor is probe versus instrument. If your probe is out of whack by 3°, and your instrument is out of whack another 4°, all of a sudden it's possible to be out of calibration as much as 7°! So, it's important to calibrate the two together.
Interference: For electronic thermometers, electromagnetic interference from kitchen equipment such as microwaves, drive-thru radio systems, and high voltage ovens can cause thermometers to display inaccurate readings. To avoid this, you can look for products bearing the CE mark of compliance, which indicates they have been designed to protect internal components from interference.
Resolution: How fine is the availability of readings? Can the thermometer provide resolution to every 2°? Every degree? Or to every tenth of a degree? This relates to the thermometer's design and certainly influences the real numbers you get.
How you read the number: A related issue is the human factor. Can real people see what the number is? On a dial-type thermometer, it's often up to users to interpolate—or estimate the reading as a needle falls between two hash marks. And, if a user is not looking straight at the dial, there's also the problem of parallax (visual distortion). Digital display solves this dilemma. Some thermometers provide backlighting to make reading easy, even in dark locations.
Where you put the probe: limitation of the traditional bimetallic thermometer is the size of its sensing area. Typically, about a 2" segment of the probe has to be in contact with the food for you to obtain an accurate reading. So, what happens if you stick the tip of this thermometer into thin slices of beef for stir fry? Or into a beef patty? Sanitariums agree that for thin foods, a different technology is preferable because you really can't get an accurate reading in this situation.
In fact, the temperature-sensing portion of the probe has to make good contact with whatever you are measuring in order to give you meaningful information. This is why many operators are adopting thermocouple units that accommodate multiple probes. A specialized probe for a griddle, for example, has a large flat area that makes solid contact with the equipment. A thin needle probe works in a pizza or a burger.
Food for thought: The truth is, you can buy a food thermometer for $10. So, why buy a thermocouple for more?
Some operators regard this as an insurance issue. If, for example, your $10 thermometer makes it easy for you to serve undercooked burgers or undercooked chicken, it is supporting you in violating your own food safety or HACCP standards. What's it worth to you to know you are doing what you think you are doing?