Salt Fat Acid Heat by Samin Nosrat. I’m Just Here for the Food: Food + Heat = Cooking by Alton Brown. We measure the salt, fat, acid, and food with spoons and cups and scales, but all too often we just eyeball the heat. An instant-read thermometer is a kitchen essential for the twenty-first century, and your path towards better cooking.
About Instant-Read Thermometers
How They Work
In general, an instant-read thermometer is a one-piece unit with a display section attached to a thin metal probe. You insert the probe, wait a few seconds for the readout to stabilize, then read the temperature from the display section. You may well wonder why it takes a 2‒12 seconds for a so-called instant-read thermometer to give you a temperature reading. As best I can tell, they’re called instant-reads because they replaced dial-type meat thermometers that took 30‒60 seconds to stabilize. Even compared to contemporary technologies, I find it takes our favorite probe thermometer a good 15 seconds to come to a rest.
Internally, instant-read thermometers convert a change in temperature into a change in some electrical parameter, measure that parameter, and display it to you as a temperature. The main technologies used to do this are thermocouples, thermistors, and resistance temperature detectors (RTDs). A lot of people will harp on the importance of which method a thermometer uses, but I’ve come to believe that that debate is overblown. Unless you need to measure temperatures colder than a freezer or hotter than a home oven, the thermometry method is less important than the overall implementation. The best manufacturers figure out how to build a great device that emphasizes the strengths and minimizes the weaknesses of its chosen technology.
|1. Thermapen MK4
|4. Javelin Pro Duo
|5. Maverick PT-55
|7. Maverick PT-100
|9. Taylor Turbo
|11. OXO Perfection
What to Do with an Instant-Read Thermometer
Cooking is, fundamentally, using heat to turn ingredients into food. In the same way you use spoons and cups to measure your ingredients, you use a thermometer to measure your heat. A good thermometer is much more accurate than classic methods of determining temperature by eye or by touch. An instant-read thermometer gives you that information promptly so you can use it effectively when it matters.
The number one use for an instant-read thermometer is cooking meat. Nobody wants an undercooked steak, and nobody cares about the people who want an overcooked steak. If you’re a restaurant chef who’s cranked out a hundred steaks a day for years, you can eyeball a perfect medium-rare. Everyone else should use a thermometer, and an instant read lets you make your decision before it’s too late.
Meat isn’t the only thing to measure with an instant-read thermometer. I use mine to maintain the right temperature on poaching liquid and frying oil. It also comes out for figuring out when the bread is done, or when a stirred custard is ready to come off the heat. For candymaking, an instant-read makes an old-fashioned candy thermometer look like a caveman’s tool. Y’know, for cave candy.
Selecting an Instant Read Thermometer
A good thermometer should be accurate. For most applications, ±1° is an acceptable level of accuracy. That’s enough to feel certain of hitting a 5° wide level of steak doneness. If you’re looking for a thermometer for candymaking, cut that figure in half to ±0.5°. The difference between firm ball and hard ball is very small, so you need to be able to spot it correctly and quickly.
Almost all of these thermometers display temperature down to the tenth of a degree even though they’re not that accurate. To a certain extent, this level of precision is useful since it lets you see which way the temperature is moving and how quickly it’s changing. It’s useful to tell the difference between a pot of poaching liquid at 160° and pretty steady and the same pot at 160° and gradually falling.
A thermometer that reads quickly is better than one that reads slowly. All of the thermometers we tested would make a big jump in the right direction quickly, then gradually move towards the final temperature with smaller and smaller steps until they’d start noodling around within a couple of tenths.
The speed that the temperature of your food changes during cooking determines how fast your thermometer needs to read. A roast that’s already cooked for two hours doesn’t care if it takes ten seconds to read the temperature, but a thin steak can cross from medium-rare to medium in a shockingly short amount of time. Candymakers definitely need a fast thermometer to catch a small change in temperature before it overshoots.
As a matter of comfort, if you’re trying to read a temperature in a hot environment, like right over a grill or a pot of boiling water, a fast thermometer is very much appreciated.
You need to be able to read what the thermometer says for it to do any good. A big display helps with that. So does a backlit display, especially in a questionably lit environment like inside an oven or on a grill in the evening. The ability to rotate the display is also useful.
How does the thermometer feel in your hand? Do you have to practically jam your hand into a pot of cooking water to measure its temperature, or can you stay a few critical inches farther away? Are the buttons in sensible places so you can use them when you want without accidentally bumping them when you don’t want them? If your thermometer is a pain to use, that’s no good.
Your instant-read thermometer needs to hold up to the rigors of the kitchen environment. You want something waterproof enough to survive the occasional accidental dunk and strong enough to take the occasional fall without breaking down.
Our Testing Methods
We used two devices for our testing. For testing accuracy, we needed a provably accurate thermometer that was more precise than the ones we’d be testing. For that, we used a Thermoworks Reference Thermapen, which is accurate to ±0.1° with a display precision of 0.01°. You might wonder if you might want one of these for your kitchen. Bad news — it takes 30‒90 seconds to lock in a temperature, which is not exactly an instant-read.
For our test, we set up a high-tech support gantry over our test vessel (A.K.A. a board with a hole in it set over a pot) so it could come to a temperature without movement or contact.
We also used an immersion circulator to hold a steady 140° water temperature, specifically the Anova Precision Cooker Nano. This review isn’t about the Anova, but we’ve got to appreciate its performance here. According to the reference thermometer, it held the pot of water between 140.00° and 140.05° the entire time. Of course, that’s easier without adding food to suck up that heat, but that’s more precise than I ever expected.
First, we tested low-temperature performance. We were not trying to make a perfect ice bath here, just a pretty cold one. After letting the reference thermometer stabilize at about 34°, I put each thermometer in the bath as close as possible to the reference probe while running a timer. I stopped the timer when the test thermometer stopped moving, then noted the final temperature and time. I repeated this test three times and averaged the results, adding an extra test if there was a significant outlier.
For testing in the usual meat range, we repeated the ice water test in a 140° water bath. Overall, thermometers (or the tests) were more accurate here than with the ice water. Read time was extremely consistent at both temperatures.
For testing readability and comfort, we measured the temperature of a pot of boiling water. This told us a lot about how the different products worked in a difficult environment.
Finally, we ran each thermometer under the running faucet for about 20 seconds each, rotating to get every side wet. We then checked to see which had water inside the casing and which ones still worked.
Accuracy was excellent: 0.7° off in the cold test and dead on in the warm test.
The real selling point is the read time of 2‒3 seconds.This was the best in the test by a good margin. Practically, it means that by the time you’ve got the thermometer positioned and focus on the screen, it’s already showing its final reading. None of the other thermometers in the test came close to this level of performance.
It held up to the water test without any issues. In fact, it’s rated to hold up for 30 minutes under 1 meter of water. There’s exactly one problem with the Thermapen Mk4, and that’s the price tag: $99. That’s not cheap, but you’re getting the best.
If the Thermapen’s price tag is a bridge too far, the ThermoPop is really good and you’re going to like the price: $34.
Readability is solid with medium-sized backlit digits. A button rotates the display in 90° intervals so you can point it where it needs to go. It’s less comfortable to hold than the fold-out style probes, but adequate, especially since it reads so quickly.
The ThermoPop is especially good for left-handers since it’s equally readable in either hand, unlike most of the other models. The water resistance test showed no problems. Overall, it’s a great thermometer at a great price, even if it’s not as good as the Thermapen.
The ThermoPro TP03 is a pretty good thermometer with the extremely good feature of costing just $14.
Accuracy is acceptable, with our cold test reading 1.2° off and the hot test just 0.2° off. Read time was less good, with the ThermoPro needing 5‒8 seconds to stabilize.
The handle is comfortable, with enough length to keep your hand away from the hot stuff while still reading easy.
I especially like its storage features: magnets to keep it on the fridge or vent hood and a loop for hanging on a hook.
The display is medium size with a reasonable backlight.
It survived our water resistance test, though there was a little water in the battery compartment afterward.
On the whole, the ThermoPro is an average thermometer at an extraordinary price. Grab one to take to a friend’s place or on a camping trip. Incidentally, you should take a thermometer for camp cooking. Spending the morning on the toilet because of food poisoning is bad, but spending the morning squatting in the woods with no toilet is much, much worse.
The Javelin Pro Duo is a few steps away from an upset victory in a couple of these superlatives.
It’s not as good as the Thermapen, but the price is a lot better. It’s better than the ThermoPop, but at $55, it’s hard to say that it’s $20 better.
Overall, it’s really good. You get a thermometer with about double the read time of the Thermapen for about half the price. This is a good middle-of-the-road pick.
There’s a lot to like about the Maverick PT-55, starting with its accuracy.
When it comes to accuracy it was off by 0.7° in the cold water test and 0.5° in the hot water test for a very respectable performance.
In both tests it read in about 5 seconds, which is pretty solid.
The display has nice big digits, but the obvious hand position covers them up, leading to a slightly awkward grip. It survived the water-resistance test with no issues.
So after all those good things, including a price of $40, why is it not up with the winners? The controls are awful. There are five buttons and they all suck. The power button needs to get mashed a few times to work. There is a backlight, but it’s on the same button as the temperature hold button. So if you want the backlight, you press once to hold and again for the backlight. Maverick couldn’t spring for a sixth button? Worse, one of the buttons is a Reset button. I never want to have to reboot my thermometer, and I’m automatically leery of the suggestion that I’m going to have to do it so much it needs a dedicated face button. Do better, Maverick.
The Javelin PT12 is the base version of the Lavatools line, and is seriously small. It’ll fit in any pocket, so you can have it on you when you’re moving between the kitchen and the grill.
Accuracy is right there with the Pro Duo, reading 0.5° off in the cold test and right on the money in the hot test.
Read time was a little slower at 6‒7 seconds.
It’s got a hanging loop and magnet for easy storage.
The problem on the Javelin PT12 is comfort and readability. The digits are reasonably sized, but there’s no backlight. The probe is super-short compared to the others in the test, so I had to get my hand uncomfortably close to the boiling water. There’s not enough room for a good grip and a good reading. Ultimately, this is a medium-good thermometer for a reasonable price of $27. It’s not bad, but I’d rather have the ThermoPop in pretty much every category.
The Maverick PT-100 comes in with a heavyweight $79 price tag, so it’s got a lot to live up to. Spoiler: it didn’t.
Accuracy was not that great — 1.2° off in the cold test and 0.2° off in the hot test. That’s good enough, but not amazing.
The grip is long and comfortable, which is good since you’re going to be holding it for a while. The digits are large and backlit, but I find my natural grip obscures them a little. It survived our water resistance test, but there was a little bit of water on the outside edges of the battery compartment. The inner sealer strips did their job, but we still opened it up to dry completely.
Overall, this isn’t a bad thermometer, but the price is a dealbreaker.
The Habor 022 is a cheap no-frills thermometer.
In both the cold and warm accuracy tests it was off by 0.7° for a reasonable level of accuracy.
It took 8‒9 seconds to get a reading, which puts it near the bottom of that category.
It’s small and doesn’t offer a great grip, putting my hand entirely to close to the boiling water for comfort.
The digits are small and not backlit, so it’s tough to read.
It survived the water-resistance test without any troubles.
There’s not that much to say about the Habor. It’s cheap and adequate, but the best reason I could give to buy it is that the store is out of the ThermoPro, a better thermometer for the same price.
The Taylor Precision changes up the design of the folding probe-style thermometer a bit.
Where the display is usually on the side, the Taylor Precision puts it on the top for a more compact design.
It’s almost a good choice for lefties, but the display doesn’t flip, so the digits would be upside-down in the left hand. Accuracy is solid, with a difference of 0.8° on the cold test and 0.1° on the hot.
Read time was 6‒7 seconds, good for a mid-tier performance. For reading it, the display is brightly backlit and easy to read, but the grip I needed to not obstruct the display was awkward.
The Taylor failed out water-resistance test miserably, with the backlight conked out and water pouring from the interior. I appreciate that they tried to change up the instant-read thermometer, but this isn’t good enough for $34.
Kicking off this lowly category, the Powlaken Amagarm’s best feature was the spare battery included in the good packaging. We really appreciated that during testing.
Everything else, though, wasn’t that good. The Amagarm was off by 1° on the cold test and 1.2° on the hot test, by far the worst performance. To get those wrong readings took 5‒12 seconds; not good either. The grip was comfortable and the display was readable when the backlight was on, but it shut itself off fast. This one got destroyed in the water-resistance test with a screen full of water and broken buttons. The $20 price is nice, but there’s not much to like here.
I’ve found a general pattern where OXO’s hand tools are usually pretty good to great, but their electronics aren’t.
This is another data point on that sad plot. The OXO Chef’s Precision was adequately accurate, with readings 1.2° off on the cold test and 0.2° on the hot test. Readings consistently took 9 seconds to come in — entirely too long. It was uncomfortable to hold, especially for 9 seconds right over the boiling water. The display is big and readable, and if it worked better I’d recommend it for lefties. The water-resistance test completely ruined this one. I don’t even think you could wash this. Sorry, OXO, but this one is no good.
The thing that surprised me the most about this test was how good even cheap, bad instant-read thermometers have gotten. It used to be that you had to pay through the nose for any instant-read, let alone a good one, but now you can grab something adequate in the checkout aisle. There’s not an obvious winner here for if you want something more than just adequate.
In any case, pick up an instant-read if you don’t have one. It’ll make you a better cook. If you’ve got an old one, maybe look at the new stuff and see where the technology is going. It might surprise you.
Calibrating Your Thermometer
If you’re concerned that your thermometer is reading incorrectly, you can test it in an ice bath. Some models can be recalibrated, some need to be sent back to the manufacturer for recalibration, and others just get tossed.
Testing a thermometer in an ice bath is a tricky proposition. The basic idea is that a mix of ice and water will be exactly 32°, so you make such a mix and see how far off your thermometer reads. In practice, it’s harder. You have to make the ice bath just right to get to exactly 32°, so you’re simultaneously testing the quality of your thermometer and the quality of your ice bath. To make a good ice bath, fill an insulated cup to the top with crushed ice, then fill most of the way with water. Let the temperature settle for a minute, then insert your thermometer and stir with the probe. If you got this right, you should be reading 32° plus or minus the manufacturer’s stated tolerance.
If you don’t trust that result, you might check your thermometer against a friend’s thermometer that you trust more. It’s important that your friend trust you too, especially if they own an expensive thermometer. Make sure the probe tips are extremely close together so they’re measuring water the same temperature, but not touching so they’re not measuring each other. This is only accurate to within the sum of the two manufacturers’ tolerances, which is probably not going to be overwhelming.
Now, why use freezing water instead of boiling water?
It’s the same reason people at high altitudes despair of getting good coffee: water boils at a lower temperature when the ambient air pressure is lower. Elevation is the big contributor, with the boiling point dropping about 1° per 500 feet over sea level, but the local barometric pressure affects boiling point too. For example, as I write this, an online calculator has my boiling point at 209.61°, about half a degree up from testing last weekend. Long story short — you can’t trust the temperature of boiling water. Freezing water is not affected by pressure (on human-survivable scales, anyway), so you can trust it with reasonable effort.