Document separator sheets are printed pages you slip between stacks of paper before scanning so the capture software knows where one document ends and the next begins. Each sheet carries a machine-readable mark, either a patch code or a barcode, that triggers the scanner to close the current file and start a new one. Law firms, medical practices, financial institutions, and government offices use them to turn thousands of loose pages into properly named digital files without clicking “save” on each one.
Patch Codes vs. Barcodes
Two kinds of marks appear on separator sheets, and they do different jobs.
Patch codes are patterns of parallel black bars printed at fixed widths and spacings. Patch 2 tells the scanner to start a new document file. Patch 3 triggers a higher-level break; in Kodak’s Capture Software, for example, it starts an entirely new job rather than splitting within the current batch. Patch T, the “transfer patch,” marks the split and then tells the software to discard the separator image itself so it doesn’t clutter the final files.1Wikipedia. Patch Code
Barcodes carry data as well as the split instruction. A one-dimensional symbology like Code 39 can hold a case number, client ID, or file name. A QR code can carry enough information to populate several index fields at once, such as patient name, date of birth, and account number inside a single square. When the scanner reads the barcode, it extracts the data and uses it to name the file and route it to the right folder.
Pick patch codes when you only need to split a stack. Pick barcodes when each resulting file needs to be named or filed automatically. Many workflows put both on the same sheet: the patch code triggers the split and the barcode supplies the metadata.
How a Batch Scan Runs
A technician builds the stack by hand, placing a separator sheet in front of every group of pages that should become its own file. The whole pile goes into the scanner’s automatic document feeder.
As pages move across the sensor, the capture software watches for recognition marks. When it sees one, it closes the current file and opens a new one. Everything scanned after that separator feeds into the new file until the next separator appears. If the mark is a Patch T, the software throws the separator image away. If the mark is a barcode, the software reads the encoded data and applies it, naming the file or writing to index fields.
Routing happens automatically from rules you set in advance. A 200-page stack with 10 separators can produce 10 PDFs, each named with the right case number and dropped into the right client folder without anyone sorting afterward.
Printing Separator Sheets That Scan Reliably
A separator the scanner can’t read is worse than no separator, because it silently merges documents that were supposed to stay apart. A few physical details decide whether that happens.
Standard 20-lb copy paper works in nearly every desktop and production scanner. Thinner paper risks jams; glossy or textured stock can cause feed errors and ink smearing.
Print on a laser printer at 600 DPI or higher. Barcodes depend on precise bar-to-space ratios, and printers below that resolution sometimes distort the pattern enough to cause read errors or complete misses. Low toner and inkjet bleed cause the same problem.
Leave a clear margin around every barcode. This blank space is called the quiet zone, and the scanner uses it to locate where the code starts and ends.2GS1 UK. What is a Quiet Zone in Barcodes? For Code 39, the quiet zone must be at least 10 times the narrowest bar width or 0.10 inches, whichever is greater, on both the leading and trailing edges. QR codes need clear space on all four sides. Generous margins are cheap insurance against missed reads.
Keep marks away from the page edges. Even a small amount of paper shift during feeding can push an edge-placed mark out of the scanner’s detection zone. Centering the code, or putting it in the upper third of the page, gives the most consistent results. Some offices print separators on bright yellow or pink paper so staff can see batch breaks when assembling stacks; colored stock is fine as long as the printed bars still contrast sharply with the background.
Configuring the Capture Software
The software side is where separator sheets become useful rather than just functional. In Tungsten Automation’s capture platform (formerly Kofax), batch separator sheets carry both a patch code for separation and a barcode identifying the batch class, and the next scanned page starts a new batch. Unattended scanning requires a separator at the start of every batch except the first.3Tungsten Automation. Adding Batch Separator Sheets
Setting up barcode-based indexing involves a few settings worth understanding, because weak configuration here is what sends documents to the wrong folder in production:4Tungsten Automation. Configuring Bar Codes
- Barcode type selection. You tell the software which symbology to look for. Only one barcode type can be active per document separation job, though multiple types can be read for indexing.
- Index field mapping. Barcodes are linked to index fields with placeholders like {BarCode1} and {BarCode2}, with a page indicator telling the software which page carries the code.
- Checksum verification. Available for Code 39, Codabar, and Interleaved 2 of 5. A barcode that fails its checksum is discarded rather than written as a bad index value.
- Minimum length filtering. Any barcode shorter than the threshold you set is ignored, which catches partial reads from smudges or misalignment.
- Format filtering. Predefined patterns reject barcodes that don’t match the expected shape, so stray marks aren’t misread as data.
A single misread barcode in a production run can attach a document to the wrong client or create an indexing error that takes hours to trace. Time spent on checksum and format rules upfront saves far more time downstream.
When Separator Sheets Stop Working
The symptoms are consistent: documents that should be separate end up merged, or the scanner creates extra breaks where none belong. The common causes:
- Low print quality. Low toner, inkjet bleed, or a printer below 600 DPI distorts bar widths. Reprint on a laser printer.
- Encroached quiet zones. Text, staple shadows, or stray marks in the blank space around the barcode confuse the decoder. Widen the margins or move nearby content away.
- Wrong orientation. Patch codes and many 1D barcodes have to face a specific direction relative to the feed. A sheet loaded upside down or sideways can be invisible to the software.
- Paper skew. Pages entering the scanner at an angle shift edge-placed marks out of the detection zone. Snug the paper guides against the stack.
- Mismatched barcode type. The software only reads the symbology it’s configured for. A Code 128 barcode fed to software set for Code 39 does nothing.
- Dirty scanner glass. Dust, toner residue, or adhesive buildup can obscure part of a mark. Wipe the glass with a lint-free cloth.
For intermittent failures, scan a few separators by themselves. If they read cleanly alone but miss inside a large batch, the problem is almost always mechanical: skew, feeding speed, or pages sticking together.
Handling Separator Sheets That Contain Sensitive Data
A separator sheet with a patient name, Social Security number, or medical record number encoded in its barcode is still protected health information, even though a human can’t read it at a glance. HIPAA’s Privacy Rule requires covered entities to apply reasonable safeguards to PHI through disposal.5eCFR. 45 CFR 164.530 Acceptable disposal methods for paper separators that touched PHI include shredding, burning, pulping, or pulverizing, so the information cannot be read or reconstructed.6U.S. Department of Health and Human Services. Frequently Asked Questions About the Disposal of Protected Health Information Dropping them in open recycling violates the rule if unauthorized people could access the bin. A disposal vendor working with that paper is a business associate and must follow secure destruction procedures under its agreement.
On the admissibility side: Federal Rule of Evidence 1003 says a duplicate is admissible to the same extent as the original unless a genuine question exists about the original’s authenticity or admitting the duplicate would be unfair.7Legal Information Institute. Rule 1003 – Admissibility of Duplicates Scanned files produced through a separator-sheet workflow carry the same evidentiary weight as the paper they came from, provided the process is defensible. If opposing counsel can show the workflow dropped pages or merged unrelated documents, the authenticity of the whole batch becomes vulnerable. Quality checks after each batch, logged at the time, matter more than any argument made later.
Broker-dealers and other firms covered by SEC Rule 17a-4 face a further requirement: electronic records must sit in a non-rewritable, non-erasable format, or the system must keep a complete time-stamped audit trail of every modification and deletion along with the identity of whoever made the change, and the system must verify the completeness and accuracy of its own storage.8eCFR. 17 CFR 240.17a-4 – Records to Be Preserved by Certain Exchange Members, Brokers, and Dealers If the separator-sheet process introduces missing pages or broken indexing, the resulting records can fail a regulatory examination. Retention runs three to six years depending on the record type, which means errors compound if nobody catches them early.