Merge PDFs in the right order: select, arrange and check pages locally
Admin ·
To merge PDFs, choose at least two files, put them in the required order and select any page ranges before exporting. ToolMellow creates one merged.pdf on your device. Blank range fields include every source page; optional page thumbnails let you rearrange, duplicate, remove or rotate the selected pages.
Start by deciding what the combined document must contain. File order, printed page labels and physical page positions can differ. Keep your originals and inspect the download, especially when you need bookmarks, forms, accessible tags or links as well as visible page content.
Decide what belongs in the combined PDF
A merge can join complete documents, select a few pages from each, or build a custom page sequence. For a report and its appendix, ordinary file order may be enough. For a packet assembled from extracts, write down the desired sequence before changing any thumbnails.
This workspace copies PDF pages into a new document. Convert Word documents or images to PDF first using the appropriate tool. Merge PDF does not accept those formats directly, add a cover automatically or create a new table of contents.
| Task | Starting choice | Check before export |
|---|---|---|
| Report followed by appendix | Blank ranges; report file first | Every source page belongs in the packet |
| Selected pages from two reports | Per-file page ranges | Use physical positions, not printed labels |
| Interleaved or repeated pages | Ranges, then Arrange selected pages | Final card sequence includes every intended occurrence |
A practical local merge workflow
Choose PDF files and wait for their page counts. Additional picks append to the queue. Move or sort files, enter ranges if needed, then choose Merge PDFs. Arrangement is optional: the range-derived selection can be exported directly.
After a successful job, Download saves one merged.pdf. Inspect it in the PDF reader or application that will receive the packet. A successful export establishes that a file was created; it does not prove that every interactive or accessibility feature survived.
- Keep the original PDFs and confirm that each file is an appropriate unsigned, unencrypted source.
- Set file order and ranges before spending time on manual page arrangement.
- Check the resulting page count, order and repeated pages before replacing or sharing a working copy.
File order determines the initial page sequence
With all pages selected, a three-page file A followed by a two-page file B produces A1, A2, A3, B1, B2. Move files up or down, use Sort by filename, or choose Reverse order to change that initial sequence. Filename sorting uses numeric comparison, so report2 can precede report10.
Ranges stay associated with their files when the queue moves. Changing files, file order or ranges clears previous results and resets manual page arrangement. Complete these source choices first; changing them after a careful interleaving rebuilds the selection from the file ranges.
Select physical page positions and preserve repetitions
A blank Pages field includes all pages from that source. Enter one-based physical positions with ASCII digits, commas and ascending hyphen ranges. The first page is 1 even when its printed label is i, cover or 12. For a five-page source, 1-3, 5 selects four pages.
The order of comma-separated entries is preserved, and repeated entries create repeated output pages. The range 3-1 is rejected; enter 3,2,1 to request that reverse sequence. Each selected occurrence counts toward the combined 300-page limit.
| Entry for a five-page source | Selected order | Meaning |
|---|---|---|
| Blank | 1,2,3,4,5 | Include every page |
| 1-3, 5 | 1,2,3,5 | Ascending range plus one page |
| 3, 1-2, 3 | 3,1,2,3 | Preserve order and repeat page 3 |
| 3-1 | Rejected | A hyphen range must ascend |
Arrange pages after choosing files and ranges
Arrange selected pages opens the page editor. Use move buttons or drag cards to change order, duplicate a page, remove a page or add a clockwise 90° turn. Reverse pages reverses the current page plan, which differs from reversing the source-file queue.
The editor displays twelve cards per window. Next pages and Previous pages change the displayed window; all pages remaining in the full plan are exported. Restore pages from file ranges discards manual edits and rebuilds the original range-derived selection.
Marked pages select bulk actions, not export inclusion
An individual checkbox or Mark all selects cards for Rotate marked 90° and Remove marked. Merely marking one card does not restrict the merge to that page. Unmarked cards still export when they remain in the plan.
To exclude a page, remove its card or change the source range. You can remove all cards from one queued source, but the merge still requires at least two queued PDFs and at least one output page. If every card is removed, Restore pages from file ranges provides a way back.
Mixed page sizes and rotations remain meaningful
A merged document can contain portrait, landscape and differently sized pages. Copying retains each page’s geometry, including its crop boundary and an explicit UserUnit when present. Merge PDF does not normalize all pages to A4, add margins or offer a page-size setting.
A card’s added quarter turn is combined with the source page’s existing rotation. A page already at 90° with another 90° becomes 180°. Duplicated occurrences can receive different rotations. Review both the visible crop and orientation in the downloaded PDF; a thumbnail is only a review image.
Page copying does not rasterize the document
The merge copies selected page content and resources instead of rendering each page into a replacement image. Existing text and vector artwork are not deliberately converted to pixels, and the merge has no image-quality, PPI or compression setting.
A scanned page remains a scan. Merging does not run OCR or create selectable text where none existed. It also does not guarantee a smaller file: repeated content, resource copying and document rewriting can affect size. Use Compress PDF deliberately when reducing bytes is a separate task.
Visible pages and document-level features are different
Document-level bookmarks, interactive forms, accessibility tags, attachments and cross-page links are not guaranteed to survive this page-copy workflow. A visible page can look right while navigation, reading order or a form behaves differently. Keep the originals when those features matter.
Merging is not redaction, sanitization or flattening. Page content and resources may remain, and no general metadata-removal promise is made. Use a suitable document editor for feature-preserving assembly, then test the final bookmarks, links, form fields and accessibility in the receiving workflow.
Source limits and selected-page limits both apply
The queue supports at most twenty nonempty PDFs with 50000000 combined input bytes. Every full source must independently contain 1–300 pages before ranges are evaluated. A 301-page source is rejected even when you only request page 1; prepare a shorter source separately.
The final selection can contain at most 300 occurrences across all sources, including repeated ranges and duplicated cards. Two sources with 200 pages each therefore need shorter ranges to fit. These limits describe the application’s bounds; they do not guarantee that every permitted document will finish on every device.
| Boundary | Value | What counts |
|---|---|---|
| Queued files | 2–20 for Merge PDF | At least two required when processing |
| Combined inputs | 50000000 bytes | Decimal 50 MB across the queue |
| Complete source PDF | 1–300 pages each | Checked before selecting ranges |
| Selected output | 1–300 occurrences total | Includes repeated ranges and duplicates |
| Serialized output | 100000000 bytes | Decimal 100 MB; separate from input bound |
| Displayed page cards | 12 per window | All remaining plan pages still export |
| Worker deadline | 60 seconds per inspection/export job | A deadline, not a success or speed guarantee |
Preview settings affect inspection, not merged quality
The source’s Preview button opens a panel; Show PDF preview starts its rendering. The result has its own preview. Detail choices of 800, 1600 or 2400 pixels change review images, not the copied PDF pages. Closing a preview releases its local renderer.
Preview has separate limits. A detected embedded image above 24000000 pixels or 12000 pixels on one side can prevent rendering while page copying remains available. Result preview also re-inspects under a 50000000-byte input limit, whereas serialized output allows 100000000 bytes. If an output exceeds the preview limit, inspect the downloaded file in a suitable reader. A failed preview and a failed merge are different outcomes.
Download one PDF or continue explicitly to another tool
The merge produces one result named merged.pdf. Download saves that result; this route does not offer a multiple-result ZIP action. Check where your browser saved it and avoid confusing an earlier file with a newly generated version that uses the same filename.
Compress this PDF, Review text for Word and Export page images send the completed file to the corresponding tool in the same tab. The handoff is one-time memory, with no file payload stored in a URL or persistent storage. Reloading clears it. Review the next tool’s own limits before processing.
Local processing still needs application resources
The selected PDF bytes are processed on your device by local workers. Visiting ToolMellow still loads the page, application code and, when needed, local-site PDF rendering resources over the network. Local file processing does not mean the site is installed for offline use.
Keep a separate copy of valuable sources. Cancel terminates the active export worker and retains the original queued files without a partial successful result. A same-tab handoff is temporary memory; use the downloaded file when you need a durable copy.
Recover from invalid inputs or an interrupted job
Encrypted or damaged PDFs are unsupported. Detected signature fields or signature markers are rejected because rewriting can invalidate signatures; detection is a conservative editing guard, not cryptographic verification. Obtain an appropriate unsigned, unencrypted source through your normal document workflow.
Inspection or export errors do not provide a partial completed packet. Remove or replace a rejected file, correct an invalid range, or reduce the job after a timeout, then try again. Clear files empties the queue and results. Changes to sources, ranges or arrangement clear stale results so they are not mistaken for the revised plan.
| Symptom | Decision to check | Next action |
|---|---|---|
| Page range error | Physical positions, ascending hyphens and ASCII syntax | Correct the range or leave it blank for every page |
| 301-page source rejected | Complete source cap applies before selection | Prepare a source with at most 300 pages |
| Download succeeds but preview fails | Independent byte/image/rendering guards | Inspect the downloaded PDF in an appropriate reader |
| Timeout or local-engine failure | Document or device resource constraints | Cancel if active, reduce the job and retry |
Check the combined document before sharing it
Open the downloaded PDF and inspect its first and last pages, transitions between files, repeated pages and changed orientations. Test required document features rather than assuming that a readable thumbnail proves they survived.
- Confirm every remaining plan page is intended, including unmarked and off-window pages.
- Check mixed sizes, crop boundaries, text readability and scan quality in the receiving reader.
- Test needed bookmarks, links, forms and accessibility, and keep original sources separately.
Common questions
In what order will my merged PDF files appear?
Files initially follow the queue order, with each file’s range entries in their entered order. Move, sort or reverse the file queue before arranging pages. The optional page editor then determines the final plan order.
Can I merge only selected pages from each PDF?
Yes. Enter one-based physical positions in each source’s Pages field, such as 1-3, 5. Blank includes all pages. Use ASCII digits, commas and ascending hyphen ranges; repeated entries remain repeated.
Why is my 301-page PDF rejected?
Every complete source must have 1–300 pages before range selection. Requesting just page 1 cannot bypass a 301-page source rejection. Prepare a shorter source separately, then keep the combined selection within 300 occurrences.
Do marked thumbnails control which pages are merged?
No. Marking targets bulk rotate/remove actions. Every card remaining in the full plan exports, including unmarked pages and pages outside the current twelve-card window. Remove a card or change a range to exclude it.
Does merging PDFs reduce image quality?
The merge copies page content and resources without rasterizing pages or applying an image-quality setting. Scans remain scans, and there is no OCR or compression step. Inspect the download because this does not guarantee preservation of every document feature.
Will bookmarks and interactive forms survive a merge?
They are not guaranteed, nor are document-level tags, attachments or cross-page links. Keep originals and test the final features in the receiving reader. Choose a document editor suited to preservation when these features are required.
Can I merge signed or encrypted PDFs?
Encrypted or damaged inputs are unsupported, and detected signature fields/markers are refused to avoid silently invalidating signatures. Use an appropriate unsigned, unencrypted source. This guard is not cryptographic signature validation.
Why can I download a PDF that will not preview?
Page copying and preview use different guards. Oversized embedded images can block rendering while copying works. Result preview re-inspects under a 50000000-byte source limit, although output permits 100000000 bytes. Inspect the saved result in a suitable PDF reader.
What happens when I cancel a PDF merge?
Cancel terminates the active export worker, retains the original sources in the queue and provides no partial successful packet. After cancellation, reduce or adjust the job and process again. Each worker job also has a 60-second deadline.
Can the merged PDF contain different page sizes?
Yes. Source geometry, crop boundaries and an explicit UserUnit when present are copied per page. Added quarter turns combine with existing rotation. There is no A4-normalization or margin control in Merge PDF.
Does merging make a scanned PDF searchable?
No. Copying a scan does not perform OCR or add text that was absent. Merging preserves the page-copy workflow; use an appropriate OCR task separately when searchable text is required.
Will merging reduce the file size?
There is no size target or shrink guarantee. The new PDF can have different byte size after copying and rewriting. Use Compress PDF as a separate, deliberate step and check its own preservation and size limits.
Can I merge Word documents or images here?
Merge PDF accepts PDF sources. Convert supported Word documents or images to PDF first with Word to PDF or JPG to PDF, then inspect those intermediate PDFs before merging.
Are my PDFs uploaded, and what does the next-tool action retain?
Selected PDF bytes are processed locally. Page, code and renderer resources can still load over the network. The next-tool action carries the result once in same-tab memory; it disappears on reload and is not a saved file or URL payload.