How to Use a Meat Thermometer: Placement, Calibration and Safe Temps

Every temperature on this page is quoted from USDA FSIS, FoodSafety.gov, the FDA Food Code, or a land-grant extension service, with that source’s own last-updated date.

How do you use a meat thermometer?

Four steps, and only one of them is the actual measurement.

1. Know your thermometer’s sensing area. A thermocouple senses at the tip. A dial bimetal thermometer averages temperature along 2 to 2½ inches of stem. That difference decides everything else.
2. Insert to the correct depth in the thickest part, away from bone, fat and gristle. For thin foods, go in through the side.
3. Wait for the type’s response time — 2 to 5 seconds for a thermocouple, about 10 for a digital instant-read, 1 to 2 minutes for a dial.
4. Check more than one spot on anything large or irregular, and check every piece when you are cooking several.

USDA FSIS puts the core rule this way in its Food Thermometers factsheet (last updated March 21, 2025): “The food thermometer should be placed in the thickest part of the food, away from bone, fat or gristle. For thin foods, the food thermometer should be inserted through the side until it reaches the center of the food.”

One refinement worth borrowing, and it is the single best idea anywhere on this subject: J. Kenji López-Alt at Serious Eats argues you should push the probe all the way through the meat until it emerges out the far side, then slowly withdraw it and take the lowest reading you see. You are not looking for the geometric center — you are looking for the coolest point, and on an irregular cut those are not the same place. It costs three extra seconds and it is more reliable than guessing where the middle is.

Seven free homestead tools.Livestock due dates, raised bed soil, companion planting, canning times and storage life.See the homestead tools →currenthomesteading.com

What temperature should meat reach? The USDA chart

This is the table everything else on this page serves. It comes from FoodSafety.gov (last reviewed November 21, 2024), which is the more complete of the two federal charts, with the FSIS variations noted.

FSIS’s own Safe Temperature Chart (last updated April 14, 2025) adds a standalone “Ground Poultry 165°F” row and drops bison, goat, rabbit, venison and the shellfish detail. Use FoodSafety.gov’s version as the base.

Where the two federal charts disagree

None of these will hurt you if you cook to the higher number. But they are worth knowing about, because a reader following FoodSafety.gov will overcook a USDA-plant ham by 25 degrees, and a reader who finds the stale FSIS tips list will overcook pork chops by 15.

How do I know if my meat thermometer is correct?

You calibrate it. There are two accepted methods, and one of them is a trap.

The ice-water method — use this one

FSIS, verbatim: “fill a large glass with ice cubes. Add drinking water to the glass and stir well. Immerse the food thermometer stem a minimum of 2 inches into the mixture, touching neither the sides nor the bottom of the glass. Wait a minimum of 30 seconds and check the reading without removing the stem from the water. It should read 32 degrees F.”

The University of Maryland Extension’s FS-1092 factsheet (August 2019) sharpens it: use finely crushed ice, not cubes, packed dense with potable water. A slushy, tightly packed slurry is what actually pins 32.0°F; a glass of cubes with water sloshing around them does not. Maryland also gives the recalibration schedule: “Daily is ideal, but at least weekly,” plus after any drop or large temperature change.

The reason to prefer this method is simple: the freezing point of water does not change with elevation. 32°F in Miami is 32°F in Leadville.

The boiling-water method — and the altitude trap

FSIS also describes boiling-point calibration, and then says something remarkable: “Boiling water’s temperature may vary in high altitudes… Check with the local Cooperative Extension System office or health department for the exact temperature of boiling water in your high-altitude area.”

That is a federal agency telling you the number you need is somewhere else. So we built it.

Boiling point of water by elevation

Download the full table as CSV (250-ft increments, 0–10,000 ft)

How these numbers were produced. Station pressure from the NOAA/National Weather Service pressure-altitude relation, inverted for elevation; boiling point from the Antoine equation using the NIST Chemistry WebBook coefficient set for water (Bridgeman & Aldrich 1964, valid 344–373 K): A = 5.08354, B = 1663.125, C = −45.622. Validation: the method returns exactly 212.0°F at sea level and reproduces Colorado State University Extension’s published anchors within 0.7°F at every point (2,000 ft: computed 208.4 vs published 208; 5,000 ft: 202.9 vs 203; 7,500 ft: 198.3 vs 198; 10,000 ft: 193.7 vs 193). It also matches FSIS’s own rule of thumb — “with each 500-feet increase in elevation, the boiling point of water is lowered by just under 1°F” — at a computed 0.91°F per 500 ft near sea level. Assumes U.S. Standard Atmosphere; the actual boiling point on any given day shifts roughly ±1°F with a ±25 mb weather swing, which is another reason to prefer ice water.

Boiling point in specific cities

Elevations from the USGS 3DEP Elevation Point Query Service at each city’s downtown coordinate, queried August 19, 2026, at 1-meter resolution. City elevation is a range, not a single number — USGS separately publishes Denver’s city high point as 5,470 ft, which works out to 202.0°F, a 0.4°F difference from the downtown value. Immaterial for calibration, but worth being transparent about.

How accurate does a thermometer have to be?

Three different answers, and the gap between them is instructive.

Read those together and the point lands: a thermometer that reads 4°F off is normal at retail and illegal in a restaurant. That gap is the entire argument for calibrating yours.

Where exactly do you put the thermometer? Placement by cut

All quotes below are from the FSIS Food Thermometers factsheet, last updated March 21, 2025, unless noted.

How deep does the probe go? It depends on the type

The bimetal trap: when the sensor is longer than the food is thick

This is the failure mode nobody explains, and FSIS states it outright: a dial bimetal thermometer “senses temperature from its tip and up the stem for 2 to 2½ inches. The resulting temperature is an average of the temperatures along the sensing area… it is not appropriate to measure the temperature of any food less than 3 inches thick.”

Stick a dial thermometer straight down into a three-quarter-inch burger and most of that sensing coil is measuring air. The reading is an average of meat and kitchen. You have two fixes:

• Side-insert so the entire 2–2½ inch coil sits inside the food, or
• Use a thermocouple or thermistor, where the sensor lives in the tip.

FSIS also warns about stem conduction: “heat conducts along the stem’s metal surface faster than through the food,” so the spot touching the tip reads hotter than the meat beside it. Its remedy is to take a second and third reading elsewhere, letting the thermometer equilibrate at least one minute each time. That applies to bimetal and liquid-filled types, not to thermocouples.

What about infrared thermometers?

They cannot tell you whether meat is done. An infrared thermometer reads surface temperature only. It is a actually useful tool for checking a griddle, a pizza stone, or a pan — and useless for verifying the inside of a chicken thigh.

Worth being straight about the sourcing here: we searched FSIS, FoodSafety.gov, the full FDA Food Code 2022, and the Penn State, UConn, Maryland, SDSU and Purdue extension services. None of them address infrared food thermometers at all. The Food Code mentions “infrared” exactly once, and only about heat-lamp shielding. The best available source is trade, not government: Robert W. Powitz, Ph.D., MPH, writing in Food Safety Magazine (October 1, 2006), states plainly that “the IR thermometer will only measure surface temperatures and not internal temperatures,” notes that most foods have an emissivity near 0.95 while shiny surfaces vary widely, and describes typical distance-to-spot ratios around 6:1. The physics is uncontroversial regardless of who publishes it.

And pop-up timers?

Better than their reputation, and the failure mode is not what people think. FSIS: pop-up devices “are reliable to within 1 to 2 degrees F. But proper placement is important. Checking with a conventional food thermometer is always recommended as an added precaution to properly gauging both safety and doneness.”

So the mechanism works — a food-grade nylon housing with a stainless spring and an organic firing material engineered to melt at a set temperature, in use since 1965. What you cannot verify is whether the factory put it in a spot that represents the coolest part of your particular bird. Trust the device, verify the location.

The government table almost nobody has seen: time and temperature

165°F for poultry is not a magic threshold where bacteria die. It is the temperature at which the required kill happens instantly. Hold poultry at a lower temperature for long enough and you achieve exactly the same lethality.

USDA publishes the numbers. They live in the FSIS Cooking Guideline for Meat and Poultry Products (Revised Appendix A), December 2021, Document ID FSIS-GD-2021-14 — a 92-page PDF where the tables are embedded as images, which is why they never show up in text searches and effectively do not exist for consumers.

Table 2 — Meat products

FSIS: “Temperatures stated are the minimum internal temperatures that must be met in all parts of the meat product for the total dwell time listed.”

FSIS footnote 5, verbatim: “The required Log reductions are achieved instantly (0 seconds) when the internal temperature of a cooked meat product reaches 158°F or above.”

Tables 3 and 4 — Chicken and turkey, 7-log Salmonella reduction

Fat protects bacteria, so the dwell time rises with fat content. The 1% and 12% columns bracket every real case. Note that turkey needs longer than chicken at the same temperature — FSIS built two separate tables because Salmonella’s heat tolerance is species-dependent (Juneja et al., 2001a).

Source: FSIS Cooking Guideline for Meat and Poultry Products (Revised Appendix A), December 2021, FSIS-GD-2021-14, relating to 9 CFR 318.17(a)(1), 318.23, 381.150(a)(1) and 417. Because fsis.usda.gov blocks automated retrieval, we read the complete document from the North Carolina Department of Agriculture’s mirror and verified the title, date and document ID match. For products mixing poultry species, FSIS directs you to use the longest dwell time from either table.

Safe is not the same as done — the barbecue problem

Every temperature above answers one question: is this safe to eat? None of them answers the other question a pitmaster cares about: is this tender?

USDA draws the distinction itself. FSIS: “‘Doneness’ refers to when a food is cooked to a desired state and indicates the sensory aspects of foods such as: Texture, Appearance, Juiciness… Most pathogens are destroyed between 140 degrees F and 165 degrees F. Consumers may cook food to higher internal temperatures to achieve a particular ‘doneness.'”

So USDA explicitly sanctions cooking past the safe minimum. It just declines to say how far, or why. Here is the why.

The collagen science

Texas A&M University’s Department of Animal Science is the one institutional source that publishes a number: “Briskets and pork butts are two of the most popular barbecue cuts, and both of them have a high percentage of connective tissue containing collagen. Collagen is converted to gelatin when heated… And it requires a higher internal temperature (ranges from 185 to 195°F) to achieve this conversion.”

The peer-reviewed literature adds the part that actually explains low-and-slow. Purslow (2018) in Meat Science reports that collagen’s thermal denaturation temperature measured at fast heating rates falls in the 62–67°C (143.6–152.6°F) range, but — critically — “collagen denaturation is a multistep, non-equilibrium process that is highly heating-rate dependent and can occur at 55–60°C in slow heating regimes, such as slow roasting or sous-vide cooking.” Tornberg (2005) puts denaturation at 53–63°C and notes that if the collagen fibers are not stabilized by heat-resistant bonds, they dissolve into gelatin on further heating.

That heating-rate dependence is the scientific basis for cooking low and slow, and it is why time at temperature matters more than the number on the display.

The reassuring math

Put the Appendix A table together with a twelve-hour cook and something useful falls out. Per Table 2, meat held at 145°F for four minutes already achieves a 7-log reduction. A brisket passes 145°F somewhere around hour three and then spends the next eight or nine hours climbing to 200°F. It is pasteurized many times over long before it is tender. Everything after hour three is texture, not safety.

Our smoking times and temperatures chart gives the pull temperature and the USDA safe minimum side by side for eighteen cuts, which is the practical version of this section.

Seven things people believe that USDA does not

One number we will not give you: a figure for carryover cooking. The “pull it 5 to 10 degrees early” rule is real craft practice, but no USDA or extension source quantifies carryover in degrees, and it varies with mass, shape and pit temperature. USDA acknowledges that carryover happens — that is the stated rationale for the 3-minute rest — without putting a number on it. Neither will we.

Frequently Asked Questions

Do I leave the meat thermometer in while cooking?

Only if it is designed for it: oven-safe bimetal dial, oven cord probe, wireless/smart probe, or liquid-filled. Instant-read thermometers, thermocouples, thermistors and forks are meant to be used near the end of the estimated cooking time, then removed. Kenji López-Alt makes a further point worth weighing: a leave-in probe gets placed before cooking, when you cannot yet know where the coolest spot will end up, so it is better used as an early-warning system than as the final word.

What temperature is 135°F for meat?

135°F is a doneness preference, not a safety threshold — roughly medium-rare-to-medium on beef. It is 10 degrees below USDA’s 145°F safe minimum for whole cuts. Per FSIS Appendix A Table 2, meat held at 135°F for 37 minutes reaches a 7-log reduction, which is why steakhouse-style cooking is not automatically unsafe — but that holding time is a commercial standard measured throughout the product, not something you can verify on a grill.

Should I calibrate a new thermometer?

Yes, and then at least weekly, plus after any drop. Use ice water, not boiling water — especially above about 1,000 feet.

My thermometer reads 3°F off. Replace it?

If it is consistently off by a known amount and you cannot adjust it, you can simply do the arithmetic every time — though that is one more thing to get wrong at 11 p.m. with a bird resting. FDA’s Food Code requires ±2°F in commercial kitchens, so 3°F would fail a restaurant inspection. UConn’s practical line: a few degrees, adjust; more than a few, replace.

How do you calibrate a meat thermometer at altitude?

Use the ice-water method, not boiling water. Water boils below 212 °F at elevation — 202.4 °F in Denver, meaning a cook who calibrates against 212 °F there builds in a permanent 9.6 °F error (12.6 °F in Flagstaff). Ice water sits at 32 °F at every elevation on earth, so it calibrates correctly everywhere.

Is color a reliable way to check if meat is done?

No. USDA’s own research found ground beef patties that looked well-done at 131 °F, and more than 25% of patties browned before actually reaching the safe 160 °F minimum. A thermometer reading is the only reliable check.

Where should you insert a meat thermometer?

Into the thickest part of the food, away from bone, fat and gristle, and wait for the reading to settle. For thin foods like burgers or chops, insert through the side so the sensing area sits inside the meat. Check whole poultry in three places: the innermost thigh, the innermost wing, and the thickest part of the breast.

Why do USDA sources disagree on reheating temperature for ham?

Because two current USDA pages give different numbers. FSIS says to reheat a USDA-plant fully cooked ham to 140 °F; FoodSafety.gov says 165 °F. Both pages are current and neither has been corrected to match the other.

Sources:

1. USDA FSIS, Food Thermometers, last updated March 21, 2025.
2. FoodSafety.gov, Safe Minimum Internal Temperatures, last reviewed November 21, 2024.
3. USDA FSIS, Safe Temperature Chart, last updated April 14, 2025.
4. USDA FSIS, Cooking Guideline for Meat and Poultry Products (Revised Appendix A), December 2021, FSIS-GD-2021-14. Read via the North Carolina Department of Agriculture mirror; FSIS notice at FSIS Notice 59-21.
5. USDA FSIS, High Altitude Cooking, last updated August 13, 2024.
6. USDA FSIS, Doneness Versus Safety, last updated July 10, 2024.
7. USDA FSIS, Color of Cooked Ground Beef as It Relates to Doneness, last updated May 23, 2024.
8. USDA FSIS, Washing Food: Does it Promote Food Safety?, last updated February 7, 2024.
9. USDA, Cooking Meat? Check the New Recommended Temperatures, Diane Van, USDA Meat and Poultry Hotline, May 25, 2011.
10. FDA, Food Code 2022, §§ 4-203.11 and 4-302.12(B).
11. University of Maryland Extension, FS-1092, How to Calibrate a Thermometer, Jackey & Henley, August 2019.
12. South Dakota State University Extension, How to Calibrate a Meat Thermometer, Amanda Blair, October 23, 2023.
13. Colorado State University Extension, High Altitude Food Preparation, revised 2013.
14. NOAA/National Weather Service, Pressure Altitude.
15. NIST Chemistry WebBook, Water, Antoine equation parameters (Bridgeman & Aldrich, 1964).
16. USGS 3DEP Elevation Point Query Service, queried August 19, 2026.
17. Texas A&M University Department of Animal Science, Barbecue Science.
18. Purslow, P.P. (2018), “Contribution of collagen and connective tissue to cooked meat toughness,” Meat Science 144:127–134, DOI 10.1016/j.meatsci.2018.03.026.
19. Tornberg, E. (2005), “Effects of heat on meat proteins,” Meat Science 70(3):493–508, DOI 10.1016/j.meatsci.2004.11.021.
20. University of Connecticut Food Safety, Using Thermometers in the Kitchen, January 11, 2022.
21. Robert W. Powitz, “Infrared Thermometry,” Food Safety Magazine, October 1, 2006.
22. J. Kenji López-Alt, “How to Take the Temperature of Meat,” Serious Eats, updated January 18, 2024.

Related reading

• Smoking Times and Temperatures Chart — pit temp, time per pound, pull temp and USDA safe minimum for 18 cuts.
• What Is Barbecue, Really? — including the two FSIS pages that give different smoker temperatures.
• Turkey Thawing and Cooking Calculator — back-plans thaw, brine and cook times from your serve time.
• Smoked Brisket: How-To Guide
• Reverse Sear Steak
• Across the Brand Avalanche network: Current Homesteading’s canning times and pressures chart, built from NCHFP data the same way this page is built from FSIS data.