I recently discovered that the traditional Taylor analog candy thermometer that I’ve been using for the past few months is reading temperatures 8 degrees higher than they actually are. I thought I was boiling my caramels to 248F when, in fact, they were only reaching 240. I’ve been really concerned by how soft my caramels have been turning out (and in hindsight, their suspiciously short boiling times), but since I’ve starting making them in bigger batches on a powerful stovetop (I switched commercial kitchens a month ago), I thought that the problem had more to do with scaling up the recipes and different boiling times.
But finally, I boiled up a pot of water, and instead of reading 212F, my thermometer read 220F.
I ran out to some stores, and was pretty disappointed by the options — either another analog version or a digital probe on a rope. I’ve used a probe in the past and I didn’t really like its odd clumsiness, loose rope, or slow reaction time w/ candy, but I couldn’t bring myself to buy my third Taylor analog, as if I was some kind of groupie rather than a desperate chocolatier. I got the probe.
So, I made a test batch of caramel, cooked to 248F, and lo, it set up like real chewy caramels should. I liked how much more comfortable it was to read the temp on the digital display on the counter, rather than lowering my head to eye level to keep track like with the analog, but I didn’t like how the digital jumped in temp, and wouldn’t let me set an alert for two other numbers that I care about — either 247 or 236 ( I swear, the interface just skips over those numbers; it’s not like they’re both odd or even). And it didn’t even beep when it reached numbers that I set it to.
I also wanted to test something with my marshmallows recipe, so I made a small test batch of my passion fruit. The marshmallow fluff came out very, very thick. I knew that there would be some difference, given that I’ve been boiling to 228F instead of 236F (which is a difference of about 7% water), but this was thicker than any batch I’d ever made before.
It seemed like a good time to try out the boiling water test on this new thermometer. It read 208F — 4F lower than the actual temp. So. Both thermometers are jacked, and read 12F apart.
This lead to two concerns: what to do with my thermometers and what to do with my recipes.
I just want a trusty thermometer that helps me make consistent candy. Theoretically, I could still use either thermometer, and mentally adjust for the differences in temp while testing them weekly to keep track of how much I should adjust. But that seems prone to mistakes (and making me annoyed). I looked at the reviews on Amazon for candy thermometers, and few get higher than 3.5 stars, and those that get higher tend to have 1-2 reviews only. It seems candymakers are not high on thermometer technologists’ priority list.
I finally decided to splurge and get the only thermometer that I feel trust for: the thermapen (in BonBonBar blue!). I’ve used it before at a job, and it was accurate. Its sensor is a thermocouple instead of a thermistor, so it takes only 3-4 seconds to get a reading (an accurate one, no less) and its sensor is in the last 1/8″ of its tip so it doesn’t need to be submerged by an inch or more like other thermometers.
The catch is that it’s handheld, so it can’t be clipped to the side of the pot. And it’s $89. Logically, my worst fear is losing my grip and dropping my $89 thermometer into a pot of bubbling sugar.
But it should last for years, and the company seems to be great about customer care — recalibrating if necessary (and incidentally, responded well to my emailed question and my questions over the phone; and it didn’t hurt that the rep solemnly told me that he had checked out my website, and my chocolates were the most beautiful he’d ever seen :)).
I think it’s worth the money in prevented failed batches and the number of doomed thermometers that I would have bought in the next few years… as long as I don’t lose my grip.
I hope that my only problem is that I won’t know which I love more, my thermapen or my laser thermometer.
As for the thermometers I have now, I’ll return the probe since it arrived faulty. I never tested my analog thermometer when I first got it, so I really don’t know if it was off to begin with or rate of decline. I wish I did because then I’d have a better idea of how things turned out at a variety of temps; I think it’s been declining for about a month.
And it’s true that I can’t be completely disappointed by them — they may have improved two of my lines. Perhaps my caramels should be boiled to 252 instead of 248 (but definitely not 240), and perhaps my marshmallows should be boiled to 228 instead of 236 (but definitely not 240). I wouldn’t have tried such variations on my own. I’ve noticed that my marshmallows have been getting lighter and fluffier lately, and I rather like that. I may adjust a little more, but it’s a good new avenue.
Of course, I’m going to have to test this all over again when my thermapen arrives, just to make sure. First, I’ll boil water…