I regularly making Panforte which involves heating an equal parts mix of honey and sugar to 115c. Due to the quantity of sugar a jam thermometer doesn't normally reach the mix in the bottom of the pan to work effectively. My technique at the moment is to stir it and every now and again tilt the pan to pool the honey and then use a normal but cheap probe thermometer. However, sometimes the mix visually looks much hotter (small bubbles, more 'activity') but still reads lower than my desired temperature.

In the finished bake I am getting very inconsistent results in terms of consistently; sometimes they are very soft and other times very hard and brittle. Given the baking times are the same I can only assume the inconsistency is due to the inaccuracy of measuring the temp of the sugar/honey mix.

So my question is is there anything wrong with the technique described above? Why might it be inconsistent? Or is there a better way to accurately take the temperature of the sugar honey mix? Would an IR thermometer be more useful than a probe?

5 Answers 5


IR thermometers are not accurate enough for sugar work. They make some assumptions (like reflectivity of the surface) which are not exactly met in real life.

My way of making small amounts of sugar syrup is to use a small pot with a long handle. I have a 12 cm stainless steel one with a long handle that's very comfortable, and similar vessels exist in even smaller sizes. This has the added advantage of not having too thin a layer of syrup, which can overheat quickly.

If your amount is too small to fill even a tiny pot, you might consider simply making more than you need. Sugar is cheap (OK, I know honey isn't always, but I doubt that you need the best honey if you are heating it) and if you don't feel good at the thought of throwing away food, simply making enough additional syrup to have for two glasses of lemonade will probably be enough to work with a 8-cm pot.

  • 1
    I agree to make more than you need, larger batches are easier to control for one thing. Just freeze the rest for another time.
    – GdD
    Apr 7, 2020 at 11:22
  • As far as small pots go - they make ones for sugar but for really small amounts I use a heavy Turkish coffee pot.
    – kitukwfyer
    Apr 8, 2020 at 15:59

I use IR thermometers all the time , the are very accurate. I use them for fudge, aquariums, ponds, inside and outside temperatures ( you need an appropriate solid target) skin temperature, etc.. I have checked them against electronic immersion thermometers, all good. There are different temperatures in boiling sugar- foam, edge of pan , etc, use some judgement.Maybe because I got experience using them as a boy ,measuring furnace temperatures to 2000 F+ ,I learned everything is not always the same temperature in a given location.

  • So how does this answer the question? [I]s is there anything wrong with the technique described above? Why might it be inconsistent? Or is there a better way to accurately take the temperature of the sugar honey mix? Would an IR thermometer be more useful than a probe? You did answer why it might be inconsistent (varying temperature by location) but the rest ..? Apr 7, 2020 at 21:37

You can use the drop/"ball" method. Drop some of the mixture into a small dish of cold water which will cause it to cool quickly, then feel how soft/firm the drop is. Continue cooking until you get a drop that's firm enough.

If you're familiar with the recipe you can use pretty small drops and usually only have to do 3-4 drops so there's not much waste.

(I do agree with another answer that using a narrow enough pot so that your mixture is deep enough produces the best cooking results for cooked sugar in general, especially in terms of temperature control.)

  • +1 for mentioning the traditional method. I am a tiny bit skeptical that it will be very helpful here - if we have a layer of heated sugar that's too thin for a probe to measure well, it might be that, by the time the ball has cooled well enough to recognize the stage, the candy in the pot will have already heated to the next stage, it needs only a couple of seconds. But it is still a solution that people should be aware of, and sometimes the only one available.
    – rumtscho
    Apr 9, 2020 at 7:00
  • @rumtscho It's the method I always use, since I am too lazy to get out the thermometer. But of course everything is easier when you have a pot proportionate to your recipe. Apr 9, 2020 at 14:01

A probe thermometer should work fine, but you really want to use an electronic probe, not a mechanical one. A mechanical thermometer will sense the average temperature along part of its length, and so it needs to be immersed deeply in the liquid. Even with an electronic probe, use the tilting method to get it as deep as possible. If the probe isn't remaining in the syrup the whole time, wait several extra seconds for the temperature reading to stabilize.

  • 1
    What do you mean by "electronic" probe and "mechanic" probe? The electronics in electronic thermometers is mostly about displaying the data and a few controller functions (like a warning light for failing battery). The probe in the tip tends to be implemented with a thermistor which requires about 2-3 cm to be stuck into the thing for best reading. Are you maybe contrasting a thermistor-based thermometer to thermometers which work based on the expansion of a metal strip?
    – rumtscho
    Apr 7, 2020 at 15:24
  • 2
    @rumtscho Exactly (though I believe thermocouples are more common than thermistors).
    – Sneftel
    Apr 7, 2020 at 15:32
  • 1
    @rumtscho I've got a thermapen mk4 that claims minimum immersion for an accurate reading is 3mm at the tip. Its a lot higher end than most kitchen thermometers though.
    – mbrig
    Apr 8, 2020 at 5:27

I have not tried what you do. But there are a number of multi-cookers on the market. You put whatever you want in the pot and you can set target temperature + time too cook on it.

Just make sure to take a model that allows you to manually set the time and temperature. As well see what temperatures are supported. Whether you can only move with 10℃ increments, min/max temp, etc.

  • Multicookers would definitely not have the needed precision. Let alone the UI to pick the needed temperature. +- 10℃ for syrup is way too coarse
    – Jeffrey
    Apr 8, 2020 at 16:00
  • @JeffreysupportsMonica, I assume you can find one that allows finer grained control. I would be very surprised if you can control temperature more precisely by hand though. Apr 8, 2020 at 16:48
  • @akostadinov For cooked sugar, the cooking temperature (heat of the stove top) isn't controlled, it's the temperature that the sugar gets to that's important... Remember that boiling water is always 100C (at sea level). The mixture contains some water (from the honey or liquids added) and as the water boils off the temperature rises, since sugar can get hotter than 100C. When the thermometer reads a specific temperature, you take the pot off of the stove and the temperature stops rising. Since the temperature may only rise by a couple of degrees a minute, it's easy(ish) to be precise. Apr 8, 2020 at 20:20
  • @user3067860, good to know. I'm not deleting my answer so your nice comment can be visible. Apr 8, 2020 at 20:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.