The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

Using thermometers to test for bread doneness

Jeffrey Hamelman's picture
Jeffrey Hamelman

Using thermometers to test for bread doneness

This post will be a little long; sorry, but I want to be thorough.

Several years ago, I was teaching baking classes at the Japan Institute of Baking (JIB) in Tokyo. A big part of the Institute is devoted to research on all manner of bread topics. An engineer there gave me a gift--a small probe that had a USB port on one end and a sheath to cover it. It is used to track the temperature changes within a loaf of bread. I asked him when bread reached its maximum internal temperature, and he said it was attained when the bread was approximate two-thirds through the bake. Hmmm. When I returned to Vermont, I put the probe into my desk at King Arthur and other activities took my attention--I didn't test it.

A year or so later, I was in Nantes, in France, visiting with dear friend Hubert Chiron, one of France's most important bakers, writers, and researchers. We were at INRA (Institut National de la Recherche Agrinomique), where he works. I asked--"Hubert, when does bread reach its maximum internal temperature?" "About two-thirds through the bake," he replied (I had not told him about the Japanese conversation). Hmmmm.

When I returned from that trip, I pulled out the probe and did the experiment for myself. Here's how it works: You take the sheaf off the end, plug the USB port into a computer and start the program. On the computer screen you see a grid that graphs temperature change on one axis and time on the other. Then you remove the probe, put the sheaf on, and wrap the whole thing in the center of a loaf of bread. It stays there throughout the final rise and the entire bake. Once the bread is baked, out comes the probe. On the computer you can see that there was no increase in temperature for a long while, since the bread was initially rising at room temperature. Eventually it curves upward after it has been loaded, and the temperature begins to increase. Sure enough, the temperature rises pretty quickly and then begins to taper off. Eventually--about two-thirds through the bake--it pretty much flatlines and temperature increase is minimal. 

Not too long after that, I was teaching a five-day class at King Arthur, and one of the students wanted to test doneness of a loaf using a thermometer. I told the class about my experiences in Tokyo and France, and my own recent experiment. Two things happened: one was that one of the students was a writer for Cooks Illustrated. His ears perked up, and a month or so later he sent me a one-page article he had written for Cooks; basically he had replicated the experiment I had done, with the same results. The second thing was that another student in the class just happened to have, in his glove compartment, a rather sophisticated temperature probe. He went and got it, and we inserted it into a loaf of ciabatta that was about to be loaded. Being a bunch of dweeby bakers, all of us just stood around, riveted to the display that showed the temperature rise. When the bread hit 210F internally, out it came. It was half done at most. I wouldn't go near it, but I offered a slice to anyone who wanted one. No takers. Believe me, if you were blindfolded and squeezed that loaf and a roll of Charmin, you wouldn't know which one was bread and which was toilet paper. 

I know that there are plenty of people who have their own opinions and practices that are different from mine, and that is totally fine. I started out working with French and German bakers, and squeezing and thumping were the ways that doneness was ascertained (along with the length of time the bread was in the oven and its color). It really is foolproof once the skill is acquired. One might also say it is more respectful to the bread, since there are no small holes in the bottom where a thermometer was plunged. And it sure feels good on the hands. 

Jeffrey

Ilya Flyamer's picture
Ilya Flyamer

That is so interesting Jeffrey, thank you for sharing this info. And we are so lucky to get your comments and advice here!

I never use a thermometer for my hearth loaves, but on occasion I might use it for bread baked in a pan to make sure the bread is done - since there one can't simply look at the colour on the sides, knock on the crust, or use other sensory cues. And I mostly bake in the pans when I make rye bread, which perhaps complicates it even more. Have you compared the temperature curves in that case, to the actual time the bread is baked through?

Jeffrey Hamelman's picture
Jeffrey Hamelman

Hello Ilya,

Your question about temps in loaf pan breads is, like many of the others that have recently been posted, a good one. I have about zero scientific genes in my DNA, and do not know the answer to some of the questions posed, nor to yours about temperature curves in loaf breads. I can say this--taking the bread out of the pan when it looks done and giving a good squeeze is a reliable way to ascertain doneness. For me, I definitely want to feel a definite "wall" of baked dough, even on brioche loaves. I've removed hundreds of 2.2 kg 100% rye loaves from their pullman pans over the years, and there is no hardship in doing that. We should keep in mind that there are as many desired ways to bake bread as there are bakers. For me, I like a bold bake on most all breads, or as I like to put it, "a crust you can trust." But other bakers prefer different degrees of doneness--some want a lighter loaf, and of course there are some well-known bakeries these days that bake to a super dark color and refuse to budge, even if customers ask for something lighter. I personally find that to be a somewhat disrespectful attitude (maybe even an affectation). I spent decades at the oven, and even though as I mentioned I like a full bake, my philosophy has always been to bake through a range of doneness so that as many customers as possible get the kind of loaf they wish for. If bakers are public servants, and we surely are, then we owe it to our customers to offer them the product they want, as long as we do not compromise our personal standards and values. For instance, I personally would never use shortening in place of butter, or caramel coloring, and so on. 

As far as the praise that some people are offering me, I am very grateful and humbled--thank you. I can say with honesty that I have absolutely no secrets when it comes to baking. I have learned an immense amount from other bakers, and it would be arrogant and selfish not to be completely open to other bakers. I love to share what I've learned, if for no other reasons than that I love baking, and I love bakers. We are, after all, in the same great club. Thanks.

Benito's picture
Benito

It doesn't surprise me at all, the fluid components in the bread have to boil before they cook off so the boiling point needs to be reached for that to start happening.  It makes complete sense.  Thanks for sharing that Jeffrey, it is great that the experiments confirmed what is logical.

We always appreciate your contributions to TFL and bread baking in general sir.

Benny

Another Girl's picture
Another Girl

... but it makes complete sense. It was an aha moment and a duh moment all at the same time – so obvious that I can't believe I never realized it before. For the record, I do squeeze and thump, but my error has been in trusting the thermometer more than my senses. Suffice to say that my lazy temperature-taking days are officially behind me. 

While I don't have the mind of a scientist, I do like to understand things, so here's a question for you scientifically-inclined folks: It's been my observation that an instant-read thermometer usually inches up in the final minutes of a bake, which usually coincides reasonably closely with the bake time indicated in the recipe. Why would this be so if the bread reaches maximum internal temperature two-thirds of the way through the bake? (I'm not asking because I doubt what everyone is telling me; to the contrary, I fully accept it. I'm asking because I'm trying to fill the gap in my understanding.) Would it be completely off base to speculate that an instant-read thermometer simply isn't designed to handle all the densities and changing states of matter inside a baking loaf? 

Finally, a gigantic thanks to Mr Hamelman for reaching out with this information. I second MT's Mt Rushmore nomination. Mt Bake More?

Ilya Flyamer's picture
Ilya Flyamer

Does the water actually boil before the loaf is done? I think the normal target for internal temperature I find online is <100°C, depending on the type of bread around 92-95°C. So lower than water boiling temperature.

harum's picture
harum

The crust reaches far higher temps during baking than boiling point, which means the dough right underneath the crust and some distance in boils away. 

Chiabatta dough is full of water and bubbles so it heats up almost instantly compared to, say, high percentage rye doughs.   However, (speculating here) for the physical/chemical process that defines the conversion of raw dough into a fully baked crumb just reaching a high temp might not be enough: it might also need time for starch gelatinization, protein denaturation, moisture redistribution, enzyme deactivation, etc.  Which is demonstrated  by the chiabatta, for which the maximum possible, nearly boiling 210 F obviously wasn't enough.   

For some (even wheat) breads internal 180 F means ready to pull out of the oven -- while its center crumb still keeps heating up and well before the max possible temp is reached, meaning a thermometer is a good friend here.

MTloaf's picture
MTloaf

Taking a temperature is important to bread making but it is not a reliable measure of when it is done. Living at a higher elevation makes it even more so. I was always creeped out about probing a freshly baked loaf for doneness. It does work pretty well for meat though which is what they were originally intended for. Unlike meat I error on the side of overdone and judge by color and heft. I was amazed, in the recent videos, how you could pluck a loaf from the oven with your bare hands and give them a squeeze. I would be tossing it like a hot potato.  

Thanks for posting it is much appreciated. If they ever create a Mount Rushmore for bread bakers. I am sure you will be on it.

harum's picture
harum

Interesting info raising a lot of questions.  One is if this applies to home baking with bake temps much lower compared to industrial ovens.  If it does, there could be other variables including leavening type (yeasted vs. sourdough), dough type (whole rye vs. white wheat), loaf type (brick form vs. hearth), resting time (cut immediately vs. a day later), etc.

There's an old experiment where a fully baked wheat bread was achieved by taking the loaf out of the 450 F oven once the center has reached 165 F and then letting it sit at room temp.

suave's picture
suave

I've been telling people about that CI article for years, in vain.  Myself, I've learned to bake using the method I discovered early on in Maggie Glezer's (I think) book - until done and five more minutes.

SCruz's picture
SCruz

If the max temp is reached at 3/4 baking time, then why is the internal temperature still rising five minutes before I know they'll be ready to take out of the oven?

 

leslieruf's picture
leslieruf

I admit to using a temperature probe to check the temperature but whilst I don’t thump, tap or squeeze, I have done so in the past.  I check the probe and if it doesn’t come out reasonably cleanly I pop the bread back for a few more minutes.  Maybe in future I will check for that hollow sound  more frequently but surely also knowing your oven plus experience are big aids in deciding “doneness” of your bread. 

Leslie

albacore's picture
albacore

A brave admission in this thread Leslie, but I'm with you - I think it's a very useful tool to use, especially for beginners (which I know you're not!) and for odd shaped loaves, like the cottages i've been baking.

it does give you the minimum doneness standard - you can then carry on for a crisper crust/bolder bake as you wish. Of course, once you know your oven and loaf shape/type, you can happily work to times instead.

I remember when I was learning to bake in the early days, before the advent of digital thermometers with thin probes, the advice was to tap the bottom of the loaf and if it sounded hollow, then it was done; I always found this most unreliable - the loaf often sounded hollow long before it was done.

And lastly, what is "done"? I don't think that "done" is a single state of being for a loaf (sorry for the existentialism!).

Surely there is a range of doneness from pale to the bold bakes that some prefer and this range will translate into several minutes of bake time?

Lance

hanseata's picture
hanseata

I follow Peter Reinhart’s advice, using an instant thermometer to check for doneness: 190-195°F for enriched breads and 200°F plus for lean doughs.

It would be highly unpractical to turn out a (possibly under-baked) Vollkornbrot from the pan to knock on the bottom to gauge its doneness.

suave's picture
suave

I've acually see recipees where a rye bread is popped out of the tin and finished on a sheet.

Ilya Flyamer's picture
Ilya Flyamer

Yes, there are recipes for that, and I've done that in the past, which is useful to ensure it's completely baked through. But you also might want to avoid doing that - many rye breads are expected to have a soft crust, for example. Moreover, typically they are much easier to remove from the pan after letting them cool down for a few minutes, so that would complicate matters.

Donkey_hot's picture
Donkey_hot

These are not serious concerns, 5-7 minutes on hearth still produces soft crust in rye breads, and one must develop a plan for easy pan release to avoid complications.

There is another reason not to poke your bread with a probe - in home ovens you lose about 50F every time you open the door.

Donkey_hot's picture
Donkey_hot

Mr. Hamelman,

Thank you so much for taking a time and coming to talk to us here.  Really appreciate your lifetime work and dedication,  and so generously sharing your knowledge and experience. Your legendary book is always on my desk, even on the days when I am not baking.  If I could have only one book, it would be "Bread".  Also, thank you for The Isolation Baking Show, it tremendously helped so many people in so many ways. 

As of sticking a probe into my baked bread, I've never done it because I never had to.  Every method I've adopted from you works just fine.  I know when my bread is ready.

dmsnyder's picture
dmsnyder

If I understand the premise, the internal temperature of a loaf reaches its "maximum temperature" "two-thirds through the bake." 

What confuses me is the following:

1. My experience using an instant read thermometer is different. The internal temperature continues to increase for the last 5-10 minutes of the bake.

2. The assertion implies that the internal temperature does not continue to increase during the final third of the bake. Besides the experience described above, logic suggests that could only be true if the crust was a perfect insulator or their was a physical process cooling the interior of the loaf. Say the "maximum internal  temperature" is 205ºF. The ambient temperature of the oven is a lot higher. One would expect the loaf to eventually equilibrate to the oven temperature, unless water evaporation cools it or the crust insulates it.

I recognize that the cooking process reflects the contribution of both temperature and time, so achievement of a target temperature does not necessarily indicate that the loaf is fully baked.

Maybe enlightenment will come with further reflection, but if some one can help dispel my confusion, that would be appreciated.

David

albacore's picture
albacore

"Say the "maximum internal  temperature" is 205ºF. The ambient temperature of the oven is a lot higher. One would expect the loaf to eventually equilibrate to the oven temperature, unless water evaporation cools it or the crust insulates it."

David, I think what's happening is that as long as there is moisture in the loaf, the energy input to the loaf from the oven is being used to turn that moisture into steam, which then escapes from the loaf. This is because the latent heat of vaporisation of water is so massive. I think you've come to the same conclusion when you say "unless water evaporation cools it".

Only if there was very little moisture left in the loaf would the temperature rise above 212F. I guess this is what happens in the crust area.

Lance

dmsnyder's picture
dmsnyder

The curve of change in temperature versus time would show a rise then a plateau at 2/3 of the bake, then a second rise after enough water has either evaporated or been bound to starch that it cannot cool the crumb any longer. 

It would be interesting to know what the "maximum internal temperature" was in the experiments and whether it corresponded to what we generally regard as the target for a fully baked loaf. 

David

albacore's picture
albacore

I did search for some scholarly articles, not entirely successfully, but these two give some clues:

Source 1

Graph 1:

 

T is the internal distance from the crust

Source 2

Graph 2:

 

Lance

greyspoke's picture
greyspoke

Jeffrey, your story about the temperature probe (the one where the loaf ended up not done) probably needs converting into an experiment.  I imagine the temperature probe was the type like a shortish round skewer with a braid-protected wire on the other end you can run outside the oven to the display yes?  If you stick such a probe into a lump of meat, it will be buried in something dense and good at conducting heat.  Any heat seeping down the probe and its wire from the outside will be dissipated into the meat along the way, and might not affect the temperature at the tip much.  So the probe will give a reliable reading in that scenario.  A loaf is different, much lighter, holding less heat (or coolness if you prefer) and conducting it less well also.  Heat coming down the probe from the outside (the end of the probe and its wire will be in the hot oven) might not get soaked away and might affect (upwards) the temperature at the tip. 

This doesn't mean that you will get a bad reading of you take the loaf out, stick the same probe (all initially at room temperature) in and wait to see if the measured temperature creeps up to your desired level.  If anything, in this scenario the probe will under-read, not over-read, but probably not by much because the probe isn't being left in the loaf very long.

The experiment would be to bake two identical loaves, side-by-side, one with the buried sensor and one with the real-time probe from outside, then compare the results.  Also, measure the internal temperatures with a stick-in probe after baking.

By way of disclosure, I use a temperature probe to detect done-ness, along with feel, sound, smell and of course experience with other loaves.  When they get above the mid 90s C in the middle that seems to work for me.  I would have to bake them longer if I wanted dark crusts, but my oven is a bit rubbish and isn't really up to that anyway.

(I do have a science degree, though I never really practiced it.)

TIM