Free Scientific Calculator

Worked example

The angle mode starts on DEG. Type 2 + 3 × sin(30) into the expression box (its placeholder reads "Type an expression, e.g. 2 + 3 × sin(30)"). Before you press anything, the line under the box already reads "= 3.5", because the sine of 30 degrees is 0.5. Press "=" or Enter: the sum is added to "History" as "2 + 3 × sin(30)" with "= 3.5", and the box now holds 3.5. Press "C", type the same sum again, then press "RAD" before you press Enter. The preview changes to "= -0.964094872279", because 30 is now read as 30 radians. Press "DEG" to go back. Next, type 1/3. "Auto" shows "= 0.333333333333". Choose "Fixed" in the "Display mode" menu with "Decimal places" at 6 and it reads "= 0.333333". Press "a/b" and it reads "= 1/3".

Which settings to use, and the limits

• Check DEG, RAD or GRAD before using any trig key. The calculator opens in DEG every time; the choice is not saved. A full turn is 360 in DEG, 400 in GRAD and 2π in RAD, so sin(30) is 0.5 in DEG but -0.988031624093 in RAD. The inverse keys answer in the same unit: asin(0.5) gives 30 in DEG. Exact quarter turns are handled cleanly: sin(180) is 0 in DEG, not a tiny leftover, and tan(90) gives "Undefined". The sinh, cosh and tanh keys do not use the angle mode at all.
• Press "2nd" to reach the inverse and exponential keys. With "2nd" on, sin, cos and tan become sin⁻¹, cos⁻¹ and tan⁻¹, "log" becomes 10^, "ln" becomes eˣ, "√" becomes x², and sinh, cosh and tanh become their inverses. "log" is base 10 and "ln" is the natural log. For another base, type log(8, 2), which gives 3. The keypad also has n!, nCr, nPr, |x|, mod, 1/x, ʸ√x and Rnd. For an odd root of a negative number use nthRoot(-8, 3), which gives -2.
• Type percentages and units the way you would say them. A percent after a plus or minus is a share of the first number, so 50 + 10% is 55. A bare 10% is 0.1, and 200 × 15% is 30. Units go with "to": 5 cm to inch gives 1.96850393701 inch, and 3 kg to lb and 90 km/h to m/s work too. Unit answers are shown to 12 significant digits, so "Display mode", "Decimal places" and "a/b" do not change them, and "M+" ignores them because they are not plain numbers.
• Chain answers with Enter, "Ans" and memory. After Enter the answer stays in the box, and typing a digit adds it to the end (3.5 followed by 2 becomes 3.52). Press "C" or Escape to start clean, or type an operator to carry on from the answer. "Ans" inserts the last numeric answer. "M+" and "M−" add or subtract the value in the preview, "MR" inserts memory and "MC" clears it; an M shows at the top while memory is not zero, and memory is lost on reload. "History" keeps the last 30 results in this browser, with "Use result", "Copy all", "Save .txt" and "Clear".
• Pick a display mode for the job. "Display mode" has Auto, Fixed, Sci and Eng, and "Decimal places" runs from 0 to 12 and starts at 6. Auto drops trailing zeros, keeps up to 12 decimals, and switches to exponent form at 1e15 and above or below 1e-6. Fixed always shows the chosen number of places. Sci writes 47000 as 4.70e+4 with 2 places, and Eng uses exponents in steps of three, so 47000 becomes 47.00e+3. Numbers from 1,000,000 up get thousands separators in Auto and Fixed.
• Use the keyboard, brackets and the phone keypad. Enter works as "=" and Escape clears the box. Use ^ for powers and ! for factorials. The symbols ×, ÷, −, π and √ are understood when typed or pasted, so √9 gives 3. A missing closing bracket is added for you: sin(30 previews as "= 0.5 …", and the dots show it was closed. On a phone or tablet the on-screen keyboard stays closed while you use the keypad, and the ⌨ button brings it up when you want to type. In languages that use a decimal comma, 2,5 is read as 2.5.

Common problems and how to fix them

It says "Could not evaluate this expression — check parentheses and operators."

This appears under the box when you press "=" or Enter and the engine cannot read what is there. Typical causes are a trailing operator (5+), an extra closing bracket (2+3)), two numbers with nothing between them (2 3), or a word it does not know: typing x for times is read as a name, so use ×, * or the × key. In English, separate function arguments with a comma, as in nCr(5, 2). A missing closing bracket is not the problem, because it is added. An empty box does nothing and shows no message.

It says "Error: division by zero or a result too large to show"

The answer is infinite. 10/0 and ln(0) do this, and so do factorials that overflow: 170! works but 171! does not. The same text shows in amber under the box before you press "=", so you can fix the sum early. Look for a zero in a divisor, or a power or factorial that is far larger than you meant.

It says "Error: no real-number result"

The answer would be a complex number, and the calculator only shows real numbers. sqrt(-1) and (-8)^(1/3) both do this. For an odd root of a negative number, use nthRoot, so nthRoot(-8, 3) gives -2. If you did not expect a negative number under the root, recheck your signs and brackets.

It says "Undefined"

The result is not a number. 0/0, log(-1) and asin(2) all give this (asin needs an input between -1 and 1), and so does tan(90) in DEG or tan(100) in GRAD, where the tangent does not exist. If you meant 90 radians, press "RAD" first: tan(90) then gives -1.995200412208.

It says "Couldn't load the math engine. Check your connection and refresh the page."

The maths library is fetched after the page opens, and the line under the box reads "Loading math engine…" until it arrives. This message means that download failed, for example because the connection dropped. Refresh the page and try again. Your calculations are done in the browser and are not sent anywhere.

How it compares

Think of it as a handheld scientific calculator with a longer display: you type the whole expression, see the answer as you type, and edit anything before pressing "=". It has DEG, RAD and GRAD, nCr and nPr, hyperbolic functions, unit conversion, fractions with "a/b", one memory slot and a 30-line history you can save as a text file. It has no graphing, no equation solver and no keys for statistics or matrices, and a complex result gives an error instead of an answer. Numbers are ordinary 64-bit floating point: 0.1+0.2 shows as 0.3 because the display rounds, but "Copy plain" gives 0.30000000000000004. If an exam or employer requires an approved calculator model, check that a browser tool is allowed.