What EPUB and PDF actually are
An EPUB to PDF converter looks like a file-format swap. It isn't. The two formats disagree about something fundamental, and understanding the disagreement explains every surprise you'll hit on the way out.
EPUB: a book with no pages
An .epub file is a ZIP archive full of HTML. Open one up and you'll find chapter documents, a stylesheet or two, a manifest listing every file, and a spine that says what order to read them in. What you will not find is a page. EPUB is reflowable: the text pours into whatever screen is holding it. That's why you can bump the font size on your e-reader mid-chapter and the book simply rearranges itself. There is no page 143 in an EPUB, because there is no page anything.
PDF: a book that is only pages
A PDF is the mirror image. Every line of text has already been placed at a specific coordinate on a specific page of a specific size. That rigidity is the whole point — a PDF prints the same from any machine, and page 143 is page 143 forever. Nothing reflows, because nothing is allowed to move.
So converting an EPUB to PDF means somebody has to decide where the lines break and where each page ends. That's typesetting, not file conversion. The EPUB to PDF tool does that decision work in your browser, on your machine, using the settings you pick.
How EPUB to PDF conversion works
The pipeline runs start to finish in the browser tab you already have open.
.epub (a ZIP) → unzip → META-INF/container.xml → the OPF manifest → spine order → each chapter's XHTML → paragraphs → measure & wrap → paginate → draw the PDF.
The unzip uses your browser's built-in DecompressionStream, so there's no ZIP library to download and nothing to install. The container file points at the OPF, the OPF lists every chapter and the spine gives their reading order — front and back matter marked linear="no" gets skipped, because that's exactly what an e-reader does with it.
Then comes the part that decides whether the PDF is any good: measurement. Each character is sized using the standard Core-14 Helvetica advance widths — the published metric tables that every PDF reader uses to place the glyphs. A line the tool says fits is a line that renders inside your margin, not a line that probably does. Words too long for the column get hard-broken rather than allowed to run off the edge, which is what saves you when a footnote contains a 200-character URL.
A worked example
Take A4, Medium text (11pt) and Normal margins (20mm). That leaves a text column 170mm wide and 257mm tall, which holds 45 body lines per page. Feed it one paragraph:
"It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity."
Under a chapter heading reading "Chapter One", the output is exactly two lines, breaking after "the age of":
| Line | Content | Size | Baseline from top |
|---|---|---|---|
| Heading | Chapter One | 15.95pt bold | 26.9mm |
| Body 1 | It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of | 11pt | 38.0mm |
| Body 2 | foolishness, it was the epoch of belief, it was the epoch of incredulity. | 11pt | 43.6mm |
Chapter headings render bold at 1.45× the body size — that's where 15.95pt comes from — and every chapter opens on a fresh page, which is how a book expects to be read. The paragraph above is 36 words and 170 characters, and the tool reports both, along with an estimated reading time calculated at 238 words per minute.
What page size, text size and margins cost you
Three knobs, and each one trades readability against length. The table below is the actual capacity of a page at every combination — body lines per page, measured by the same code that lays out your book.
| Page & margin | Column width | Small (10pt) | Medium (11pt) | Large (13pt) |
|---|---|---|---|---|
| A4, Narrow (12mm) | 186mm | 53 lines | 48 lines | 41 lines |
| A4, Normal (20mm) | 170mm | 50 lines | 45 lines | 38 lines |
| A4, Wide (30mm) | 150mm | 46 lines | 42 lines | 35 lines |
| Letter, Narrow (12mm) | 191.9mm | 49 lines | 45 lines | 38 lines |
| Letter, Normal (20mm) | 175.9mm | 46 lines | 42 lines | 36 lines |
| Letter, Wide (30mm) | 155.9mm | 42 lines | 38 lines | 32 lines |
A4 is 210×297mm and Letter is 215.9×279.4mm — Letter is wider but shorter, so it holds fewer lines at every single setting. If you're printing in the US, pick Letter anyway; a shorter page beats a page your printer crops.
Here's what those line counts do to a real book. A 90,000-word novel — a normal length for literary fiction — converts like this:
| Setting | Pages | Words per page |
|---|---|---|
| A4, Small, Normal margins | 172 | 523 |
| A4, Medium, Narrow margins | 179 | 503 |
| A4, Medium, Normal margins | 217 | 415 |
| Letter, Medium, Normal margins | 218 | 413 |
| A4, Medium, Wide margins | 241 | 373 |
| A4, Large, Normal margins | 264 | 341 |
Same book, same words, and the page count swings from 172 to 264 — a 53% difference decided entirely by three dropdowns. Go Wide on the margins if you plan to annotate or bind the pages. Go Large if you're printing for someone who'd rather not squint. Go Small if you're paying for the paper.
When you need a PDF of an ebook
You landed here because something in your life only accepts PDF. That's usually one of a handful of things.
A device that won't read EPUB. Plenty of e-ink tablets, older readers and corporate document viewers take PDF and nothing else. Converting the book is faster than converting the device.
A print shop. Hand a printer an EPUB and you'll get a phone call. Hand them a PDF with the trim size and margins already set and you'll get pages.
Annotation and study. Course readings, book-club notes, a manuscript a friend sent you for comments — fixed pages mean you and everyone else are looking at the same layout, so "the paragraph near the bottom of page 41" actually means something.
Filing and archiving. Courts, universities and records systems want PDF because it doesn't move. An EPUB that reflows differently on every machine is a bad exhibit.
This works the same whether you're a student turning a public-domain novel into something you can mark up, a self-publisher checking a proof, or someone at a large company who just needs the thing to open on a locked-down laptop.
What doesn't survive the conversion
Being straight about this matters more than the feature list.
Images and cover art are left out. This is deliberate. Images in an EPUB are positioned by CSS written for a reflowing screen — floats, percentage widths, rules that shift with the reader's font size. Dropping them onto a fixed page means guessing, and a guessed layout looks worse than clean text. For an illustrated book — a cookbook, a graphic novel, a photography title — this is the wrong tool and a desktop converter like Calibre is the right one.
Custom fonts, footnote links, tables and drop caps don't come through. You get body text, headings and reading order in Helvetica. Clean, readable, not the publisher's design.
Page numbers won't match the print edition. No converter can make them. The printed book was set in a specific typeface at a specific trim size with the publisher's own spacing. Cite the chapter, not the page.
DRM-protected books can't be opened at all. Store-bought books usually encrypt their chapter files. Stripping that encryption is a separate thing with its own legal questions, so the tool says so plainly instead of handing you a broken PDF. DRM-free EPUBs work fine — Project Gutenberg, Standard Ebooks, most academic and indie publishers, and anything you exported yourself.
Non-Latin scripts: the text comes through and the layout treats CJK characters as full-width so pages don't overflow. But built-in Helvetica doesn't carry every glyph, so Japanese, Arabic or Hebrew may render as boxes in some viewers, and right-to-left text isn't re-ordered. Accented Latin — French, Spanish, German, Polish — is fine.
The file cap is 30 MB, which covers essentially any text ebook; a full-length novel is usually under 2 MB. Files larger than that are almost always image-heavy, which isn't what this tool is for. .mobi, .azw3, .fb2 and .lit aren't supported either — convert them to EPUB first and they'll work.
Related PDF tools
Once the book is a PDF, the rest of the catalog picks up where this leaves off. Use the PDF merger to bind several converted books or a book plus its appendices into one file, and split PDF to pull a single chapter out for a reading group. If the result is heading for email, compress PDF will get it under an attachment limit.
Going the other direction or working from other sources: HTML to PDF handles web pages and raw markup, PDF to Word turns a finished PDF back into something editable, and extract text from PDF pulls every word out as plain .txt. For a scanned book with no real text layer, OCR PDF runs Tesseract in your browser to find the words. And if you're checking a manuscript's length before you convert it, the word counter does that in one paste.
Frequently asked questions
Does my ebook get uploaded anywhere?
No. The .epub is read as bytes inside the page you're on. The unzip uses your browser's built-in decompression, the layout math runs in JavaScript on your machine, and the PDF is assembled locally and handed to you as a download. There are zero outbound requests during conversion — open your browser's Network tab and watch. That matters more than usual for books. A library loan, an unpublished manuscript, or a legal exhibit is not something you want sitting in a stranger's processing queue.
Why won't my book open? It says it's DRM-protected.
Books from most major stores carry DRM, which encrypts the chapter files inside the EPUB. The tool can't read those and won't pretend to — silently producing a broken PDF would be worse than saying so. DRM-free EPUBs convert normally, and that includes everything from Project Gutenberg and Standard Ebooks plus most indie and academic titles.
Will the page count match the print edition?
Almost certainly not. As the table above shows, the same 90,000-word novel lands anywhere between 172 and 264 pages depending on your settings alone — and the publisher's edition was set in a different typeface at a different trim size. Quote the chapter rather than the page number; that's the part both editions agree on.
Can it handle .mobi or .azw3 files?
Not directly — those belong to a different family of ebook formats with a different internal structure, so an EPUB to PDF converter has nothing to unzip. Calibre is free, desktop and offline, and converts them to EPUB in a few clicks. The result works here.
How long does a long book take?
A typical novel converts in a couple of seconds. A very long book — a thousand pages of collected works — takes longer, and the button shows which page it's drawing so you can tell it's working rather than stuck. The finished PDF is named after the book's own title from its metadata, and the title and author are written into the PDF's properties.
Why do converters like this usually want my email address?
Because the common model is to upload your file to a server, convert it there, and then use the download link as the reason to collect an address, hold you in a queue, or offer to remove a watermark for a monthly fee. A free trial gets you the first three books; the fourth needs a card. None of that is required by the actual problem — unzipping a file and laying out text is work a browser has done natively for years. This one runs locally, so there's no file to hold hostage, no account to make and no plan to cancel. It's free to use, there are no paid features behind it, and 10% of every dollar Microapp earns goes to charity off the top, audited quarterly.