Rebuild a two-sided document from single-sided scans
A scanner without automatic duplex support can capture every front side first and every back side in a second pass. Joining those PDFs end to end leaves all the fronts together, followed by all the backs. Interleaving instead produces front 1, back 1, front 2, back 2 and so on.
Turning a stack over can reverse the order of its backs. Back reversal is therefore enabled initially, but it is a setting to verify rather than a rule about every scanner. The preview identifies both the original scan page and its planned output page, making it easier to check the pairing.
For one combined file, enter how many pages at the start belong to the front-side pass. The remaining pages form the back-side pass. A PDF that already alternates fronts and backs does not need this workflow; use Merge PDF or Reorder PDF Pages for other arrangements.
Handle blank backs without losing the sheet sequence
Automatic blank-page removal can make a back stack shorter than its front stack. Counts alone cannot reveal which internal sheets had no back scan. Enter missing-back sheet positions such as 2, 5 or 3-4. These numbers refer to the reordered front stack shown in the preview, including any front reversal.
If there is exactly one extra front and the missing-back field is empty, the tool treats the final sheet as having no back scan and labels that assumption. If a different blank back was skipped, enter its sheet number instead. For larger count differences, explicit positions are required before the output can be built.
Missing backs are omitted by default. Enable blank insertion when you want to preserve physical sheet pairing for subsequent double-sided printing. Each inserted blank matches its corresponding front’s media box, crop and other page boxes, user units and final rotation. Blank insertion can increase the output count, which must stay within 500 pages.
Check rotation, content and output limits
Additional rotation is applied on top of each source page’s saved rotation. It changes how the page is displayed without re-encoding the scan image or rasterising selectable PDF text. Source media and crop geometry are retained, so mixed-size originals are not silently resized to one paper format.
The output-sequence table and downloadable JSON map record each output page, sheet, side, source scan page and final rotation. Inserted blanks are identified separately. These checks do not recognise printed page numbers or determine whether a faint source page is genuinely blank; visual review remains necessary.
The tool accepts static PDFs with a combined file size of up to 100 MB and 500 source pages. Editable forms, dynamic XFA documents, PDF signature fields and certification settings are rejected. Flatten a separate copy in its original application when a static scan is needed. Password-protected files must be opened and saved appropriately before they can be processed here.
Page annotation objects may be copied along with their pages, but document bookmarks, attachments and navigation are not rebuilt. Review any links and comments in the finished file. For consistent paper sizes afterward, use PDF Page Size Fixer; for photographs of paper, prepare a scan with Document Scanner. All processing happens in your browser.