About CaffCalc

What CaffCalc is

CaffCalc is a free caffeine safety calculator and a small library of research-compiled articles about how caffeine behaves in the body. It exists to answer a question that most search results answer badly: how much caffeine is too much — for me, specifically?

It has no sponsor, no affiliation with any beverage manufacturer, and at present carries no advertising.

Who runs this site

CaffCalc is built and written by Ivan Ushakovsky, a software engineer. It's an independent project — not a company, not a clinic, not a media brand. One person, one calculator, and a growing pile of reading about caffeine.

Why I built it

I wasn't sleeping well. Not dramatically — just worse than I thought I should be, for weeks at a time. Eventually I did the obvious thing and counted: five or six espressos a day, a couple of them well into the afternoon.

At 63 mg a shot, that's 315 to 378 mg. Comfortably under 400 — the figure the FDA describes as not generally associated with dangerous effects in healthy adults, and the figure every article about caffeine repeats. By that number I was fine. I was not fine.

Two things are wrong with reading it that way, and between them they turned into this site.

The number isn't calculated for you. 400 mg is a population benchmark for healthy adults. The EFSA's assessment, quoted far less often, works in milligrams per kilogram of body weight instead. When I eventually built the calculator and answered its questions honestly — including that caffeine hits me harder than average — it put my own ceiling at 340 mg, not 400. On a six-shot day I wasn't near the limit. I was over it. Nothing I'd read would ever have told me that.

And the number says nothing about when. Caffeine's half-life averages around five hours, so two espressos at four in the afternoon still leave something like a can of cola's worth of caffeine in my blood at midnight. My daily total was defensible. My four o'clock habit was the actual problem, and no calculator I could find would tell me so, because they were all answering a different question.

There's a third thing, which I've never fully resolved. Years ago in Italy I watched people order espresso after dinner — pensioners, unhurried, clearly about to go home and sleep. It bothered me for a long time. The usual explanations are a lifetime of daily tolerance, genetic variation in the CYP1A2 enzyme that clears caffeine, and a smaller pour than most tourists picture. I suspect all three contribute and I can't prove the weighting, so I'll say plainly that I don't know. But it settled something: there is no single number that is true for everybody, and anyone who hands you one is telling you about a population, not about you.

These days it's espresso in the first half of the day only — no more than four shots from the office machine, or two or three from a coffee shop, where they pull them considerably stronger. Same drink, different arithmetic, and I sleep. This site is the calculator I wanted while I was working that out.

What the calculator does differently

Rather than returning a flat figure, the calculator works from your own inputs:

It also runs entirely in your browser. Your weight, age and pregnancy status are never sent anywhere — see the Privacy Policy.

What I'm good at, and what I'm not

What I bring instead is the habit of checking. Professionally I'm a backend engineer — I spend my working life on systems where an unverified assumption surfaces later as a bug, and the reflex carries over. Out of work it shows up in odd places: I'll spend an evening keying out a beetle or a roadside weed against a reference until I'm actually sure which species it is rather than which one it resembles, and I'd rather take a bicycle crank apart to find out why it creaks than replace the part and hope. Chess, for whatever it's worth, rewards the same thing — working out why a move is good, not just that it worked.

Articles on CaffCalc are compiled from published guidelines and peer-reviewed research, and checked against their sources before publishing. Applied to caffeine, that means:

• Every number on this site traces to a named regulator, database or paper, and the link is right there.
• Where studies disagree, the page says they disagree. Caffeine research is full of observational findings that get written up as though they were causal, and I'd rather bore you than overstate.
• Where a caffeine figure is a range, I publish the range instead of averaging it into false precision.
• If a claim is popular but I can't find the origin of it, it doesn't go on the site. A surprising amount of what "everyone knows" about caffeine traces back to nothing.

The references I work from most:

• U.S. Food & Drug Administration — consumer caffeine guidance
• European Food Safety Authority — scientific opinion on the safety of caffeine
• USDA FoodData Central — caffeine content of foods and beverages
• American College of Obstetricians and Gynecologists — caffeine in pregnancy
• PubMed — peer-reviewed literature on caffeine metabolism and effects
• Manufacturer nutrition labels, for the caffeine content of specific branded products

The parts my kids are responsible for

I have two sons. The calculator has separate age bands for under-13s and 13–17s because at some point I wanted to know the honest answer for them — about a cola, about an energy drink a classmate had handed round — and the honest answer turned out to be more interesting than either "it's fine" or "absolutely not". No health authority recommends caffeine for young children; the European safety guidance for children and teens is set per kilogram of body weight. That's why the tool asks the questions it asks.