Your Scanned-Look PDF Downloaded as a ZIP. Here Is Why

You fed in a 96-page tender document, waited through the progress bar, and the download button did not say what you expected. It said Download 3 Parts (ZIP). Inside the ZIP: tender-scanned_part-1.pdf, tender-scanned_part-2.pdf, tender-scanned_part-3.pdf.

Nothing failed. This is the tool working correctly on a large file, and there is a specific number behind it.

The forty-page rule

Make Scanned PDF assembles its output in chunks of forty pages. Under that, you get one file. Over it, you get a ZIP of part files.

The part numbers are zero-padded to the width of the total, so a three-part job produces part-1, part-2, part-3, while a ten-part job produces part-01 through part-10. That padding is deliberate: it means sorting the folder by filename puts them in the right order instead of the part-1, part-10, part-2 mess you would otherwise get.

The base name comes from your source file. If you uploaded Document (3).pdf, that is what every part is called. Rename before you send anything.

Why it does not just make one file

The honest engineering answer, because it explains what you can and cannot do about it.

Every page goes through the same sequence: rendered to a canvas at 150 DPI, tilted slightly, run through the grain and tone pass pixel by pixel, encoded as a JPEG, and embedded into an output document at the original page dimensions. Each of those page images is a real chunk of memory, and it stays live until the output document is written out.

Doing that for 400 pages means holding 400 page images at once, and then serialising the whole thing in one uninterrupted pass at the end. That single final step is what hangs a browser tab or gets it terminated by the operating system — particularly on a phone, where the memory ceiling is much lower than people assume.

Chunking keeps the working set flat. Forty pages are assembled, written out, and released; the next forty start from a clean slate. Your tab stays responsive throughout and the job actually finishes.

There is no server doing this. The entire process runs on your device, which is why your file is never uploaded — and also why your device's limits are the ones that apply.

One more detail worth knowing: the ZIP is created without compression. The parts are already JPEG-based, and re-compressing JPEG data inside a ZIP buys almost nothing while costing real time on a big job. So expect the ZIP to be roughly the sum of the parts, not meaningfully smaller.

Putting it back into one file

If you genuinely need a single document:

1. Extract the ZIP. On Windows, right-click and Extract All. On Android, most file managers will open it directly; if yours refuses, install one that handles archives. On iOS, tap the ZIP in Files and it expands into a folder.
2. Open PDF merger. It takes multiple PDFs and runs in your browser like the scan tool did.
3. Add every part. Select them all at once — because of the zero padding, they arrive in the correct order.
4. Check the order anyway. Thirty seconds of looking beats reprinting a tender.
5. Merge and download.

The merged file will be large — it is the sum of a few dozen megabytes of page images. If it is going into an email or a portal with a size cap, put it through the PDF compressor afterwards. The pages are already images, so the image-based Screen level is the one that moves the number, and there is no text layer to lose because a scanned-look PDF never had one.

Consider not merging at all

Merging is a reflex, and often the wrong one.

If the destination is an upload form with a per-file cap, three 8 MB parts sail through where one 24 MB file bounces. Portals that reject a large attachment rarely object to three smaller ones.

If someone asked for a specific section, send the part that contains it. Nobody wants a 400-page attachment to read fourteen pages of it.

If you are archiving, parts are easier to open, easier to preview, and faster to load on an old machine.

If you want the splits to land somewhere sensible

The forty-page boundary is arbitrary. It will cut a table in half, land mid-clause, or separate an annexure from its heading.

The fix is to decide the boundaries yourself, before applying the effect. Run the source through PDF splitter first and cut it at chapter, section or annexure boundaries — keeping each section at forty pages or fewer. Then apply the scanned effect to each section separately.

You get single PDFs every time, no ZIP to unpack, and filenames that mean something instead of part-2. On a document you will hand to other people, that is worth the extra few minutes.

Common snags

"The ZIP will not open on my phone." Some stock file managers still cannot expand archives. Any general-purpose file manager from the store will. Or transfer to a computer and do it there.

"Part 2 starts in the middle of a table." Expected. See the section above — split deliberately first if the boundaries matter.

"It stopped halfway through." The browser ran out of memory or the tab was backgrounded and reclaimed. Split the source into smaller sections and process them one at a time. Keep the tab in the foreground on a phone.

"Only page 1 shows in the preview." By design. The preview panel shows the first page of the result so you can judge the tone and intensity before downloading. The rest are in the file.

"I want to change the intensity now." Changing the tone or intensity clears the previous result and you run again from scratch. Judge from the page-1 preview on a small test file first, then commit to the big one — a 400-page job is not something you want to run three times to compare Light against Medium.

Checking the parts add up

Before you delete the source, confirm nothing was dropped. Open each part and note its page count. Two full parts of forty and a third with sixteen should give you back your ninety-six.

If a part is short, or the last part is missing entirely, the run did not finish — usually memory, occasionally a tab that got backgrounded on a phone while the browser reclaimed it. Re-run on a smaller section rather than hoping the second attempt goes better on the same file.

It is also worth opening the last page of one part and the first page of the next, side by side. That is where a botched job shows itself: a duplicated page, or a gap. In a clean run they follow each other exactly as they did in the source.

The short version

Forty pages per part, zero-padded names, uncompressed ZIP, everything on your device. Merge them with the merger if you need one file, or split before you start if you want the seams in sensible places. The rest of the page-level PDF tools are grouped under PDF Tools, and the full list is on All Tools.