Micro PDF417 appears on small parts, circuit boards, vials and GS1 composite labels, and it is the PDF417-family symbology that readers drop most often. A device that reads PDF417 and Macro PDF417 is not guaranteed to read Micro PDF417: it is a separate symbology, standardized as ISO/IEC 24728, with its own decoder. This explainer gives you the capacity numbers, the difference between Micro and Macro, the pixel arithmetic for camera-based reading, and a verification pass to run before a label format is committed to production. You will leave knowing exactly what to look for in a reader's symbology list and how to prove it on your own parts.
What Micro PDF417 is
Micro PDF417 is the truncated member of the PDF417 family: a stacked symbology built for spaces where a full PDF417 symbol will not fit. It stands on its own as ISO/IEC 24728, which defines the symbology characteristics, data encodation and decode requirements (ISO/IEC 24728 in the standards body's public browse page), the same role ISO/IEC 15438 plays for PDF417. The capacity ceiling drives most decisions: the biggest 4-column version holds 150 bytes or 250 alphanumeric characters, and every smaller version holds less (MicroPDF417 reference). If a payload can grow, design against that ceiling from day one, because the step past it is a full PDF417 layout and a physically larger mark.
Micro, Macro and full PDF417: three symbologies, not one
The family is easy to conflate because the names differ by one word. The table is the whole story:
| Symbology | Standard | Decode path | Maximum data | Typical job |
|---|---|---|---|---|
| PDF417 | ISO/IEC 15438 | PDF417 decoder | 1,850 characters or 1,108 bytes | documents, shipping labels, ID |
| Macro PDF417 | ISO/IEC 15438, macro mode | the same PDF417 decoder | one file split across symbols | long data over several labels |
| Micro PDF417 | ISO/IEC 24728 | its own decoder | 150 bytes or 250 characters | small parts, tight layouts |
Macro PDF417 splits one data file across several ordinary PDF417 symbols, each carrying a linking header. Because every segment is a normal PDF417, any reader with PDF417 support recovers the segments. Micro PDF417 took the other road: its own start and stop patterns and its own row structure, which is exactly why a PDF417 decoder cannot fall back to reading it. When a toolkit or reader omits one family member, it is always Micro.
The decision rule: the symbology list must name it
A reader's supported-symbology list is the contract. Search it for the exact term MicroPDF417 or the standard number ISO/IEC 24728. A line that says PDF417 covers PDF417 and, since its segments are ordinary symbols, Macro PDF417. It says nothing about Micro. This is the same trap as Industrial 2 of 5 barcode support in fixed readers: a symbology outside the list is invisible to the device, whatever the sensor quality.
The generator side has the same gap, and it has bitten real projects: a widely used open-source barcode toolkit has shipped PDF417 and Macro PDF417 encoding for years with no Micro PDF417 support and no announced plan to add it. So both ends of the chain get the same gate: the printer's generator and the reader must each state ISO/IEC 24728.
Which reader class carries the job
Three classes can host the decode, and each model states its own list, so qualify the exact model rather than the series.
At an inline station, fixed code readers are the default class: the MV-ID3013PM-12M-WBN is a compact model with a 12 mm focal length, a 1.3 MP sensor and an M12 mount.
- Interface
- Fast Ethernet
- Resolution
- Up to 3 MP
- Shutter
- Global
- Lens mount
- M12
When the station inspects features as well as codes, smart cameras run the decode on the device, and for manual or low-volume stations handheld code readers do the job. The MV-SC5060XM-08M-WBN (6 MP mono, 8 mm focal length) and the LS-HCR-0001 (1.0 MP, USB 2.0) stand for the two classes:
- Interface
- GigE
- Resolution
- 6-10 MP
- Resolution
- Up to 3 MP
- Shutter
- Global
Whatever the class, the deciding filter is the same: the model you order must list MicroPDF417 or ISO/IEC 24728 among its symbologies. Filter the shortlist by interface, resolution and mount first, then let the symbology list make the final cut.
Where Micro PDF417 earns its place
The symbology was built for marks with no room: laser-etched serial data on PCBs and metal parts, vial and syringe labels in healthcare, tickets, and small logistics items where a full PDF417 would dominate the face. GS1 composite codes also stack a MicroPDF417-based component above a linear symbol, so a retail or healthcare line can meet the symbology without ever choosing it. In every one of these jobs the symbol is small by definition, which is what makes the pixel arithmetic below part of symbology support rather than an afterthought.
Worked example: 96 bytes at 100 mm with a 1.2 MP reader
Inputs, each with its source:
- Payload: 96 bytes of serialized traceability data, a typical lot-and-serial string.
- Capacity: 150 bytes in the largest 4-column version (ISO/IEC 24728).
- Working distance: 100 mm, fixed by the mounting bracket.
- Array: 1,280 x 960 pixels, the 1.2 MP class.
- Field of view: 100 mm wide, to absorb label placement tolerance.
- Narrow module: 0.25 mm (10 mil), the printer's smallest reliable element.
Capacity first. The rule: payload bytes stay at or below 150 bytes. 96 bytes is under it with 54 bytes of headroom. Decision: the format fits Micro PDF417, and the field list gets frozen, because growth past the ceiling means a full PDF417 layout.
Then the optics. The rule: the narrowest module gets 3 pixels at the worst-case field of view; at 2 pixels the bar-to-space transition blurs and contrast reversals begin.
pixel scale = 100 mm / 1,280 px = 0.078 mm per pixel pixels per module = 0.25 mm / 0.078 mm per pixel = 3.2 px
That clears the floor by 0.2 px, which is thin. If the tolerance study widens the field of view to 120 mm: 120 mm / 1,280 px = 0.094 mm per pixel, and 0.25 mm / 0.094 mm per pixel = 2.7 px, a fail. Tighten the bracket to 90 mm instead: 90 mm / 1,280 px = 0.070 mm per pixel, and 0.25 mm / 0.070 mm per pixel = 3.6 px, a pass with margin. Decision: fix the field of view at 90 mm and hold the print house to 0.25 mm modules or larger. Narrowing the search window pays twice, and the gains are the subject of setting an ROI for inline code reading.
A compact reader with a 60-120 mm reading distance, such as the LS-CRD-0047, matches this station geometry.
- Resolution
- Up to 3 MP
- Shutter
- Global
- Interface
- USB
Prove support before the format is locked
- Generate sample symbols from a generator that states ISO/IEC 24728 conformance, in every version your format can produce, including the smallest column count.
- Print them on the production substrate with the production process, label stock or direct marking.
- Decode with the candidate reader at the real distance, angle and lighting. Every version must read. Judge borderline reads the way barcode read confidence describes, not by feel.
- Grade print quality to ISO/IEC 15415 and accept C (1.5) or better under the aperture and light your line runs; the workflow is in barcode reader verification for label inspection.
- Archive the reader configuration that passed, so a unit swap or firmware update can be re-qualified against it. Vision reader configuration backup covers the practice.
Vision platforms expose the PDF417 family as symbology toggles and retry settings inside the reader configuration, the same settings behind configuring PDF417 reading reliably.
When the usual answer is wrong
- "It reads PDF417, so the family is covered." Wrong for Micro: its own decoder, its own list entry.
- "The handheld gun read it, so the inline reader will." Firmware differs between classes and between models in a series; qualify each model.
- "It must be the camera." With an industrial camera plus your own decode software, support follows the software. If the decode library omits MicroPDF417, as several open-source stacks do, no sensor or lens change fixes it; change the decode path.
- "It read on a fresh sample." A laser-marked or low-contrast mark needs symbology support and marking-grade decode together; once the symbology is supported, the methods in improving decode rates on damaged, low-contrast codes close the rest of the gap.
Checklist: before a Micro PDF417 format goes live
- The exact model's symbology list names MicroPDF417 or ISO/IEC 24728.
- The payload sits at or below the capacity ceiling with headroom; past it, the format moves to full PDF417.
- The pixel arithmetic passes at the worst-case field of view, and is re-run whenever tolerance or print resolution changes.
- The generator states ISO/IEC 24728 conformance, and the sample symbols cross-read on a second conforming reader.
- A graded sample reaches C or better under the line's own aperture and light.
- The passing reader configuration is archived, and any replacement unit is qualified against it.
Shortlist by interface, resolution and mount in the code reader finder, gate the shortlist on the symbology list, or configure a station and bring your sample symbols along. Symbology questions for a specific model go through the contact page, where the applications team can check the list with you before you order.
Questions
Does PDF417 support in a reader automatically include Micro PDF417?
No. Micro PDF417 is a separate symbology with its own start and stop patterns, standardized as ISO/IEC 24728, so it needs its own decoder. PDF417 and Macro PDF417 share one decode path; Micro PDF417 does not. If a list says PDF417 family without naming Micro, treat it as PDF417 only and get the ISO/IEC 24728 line confirmed in writing before you order.
How much data fits in a Micro PDF417 symbol?
The largest 4-column version holds 150 bytes in byte mode or 250 alphanumeric characters; every smaller version holds less, because capacity falls with the column count. If your payload can grow past 150 bytes, design for full PDF417 from the start, which reaches roughly 1,108 bytes before you re-layout the symbol.
Why do some barcode tools offer Macro PDF417 but not Micro PDF417?
Macro PDF417 reuses the normal PDF417 encoder and decoder, adding only a header that links several symbols, so it takes little extra work to implement. Micro PDF417 is a distinct symbology with its own row structure, so a developer who finished PDF417 still has real work left, and several open-source toolkits have simply never done it. Build the label supply chain on a generator that states ISO/IEC 24728 conformance.
How many pixels per module does a camera-based reader need for Micro PDF417?
Plan on 3 pixels across the narrowest module at the worst-case field of view; below about 2, the bar-to-space edges blur. Example: a 100 mm field across 1,280 pixels gives 0.078 mm per pixel, so a 0.25 mm module subtends 3.2 pixels, a pass but only just. Re-run the division after any change to lens, standoff or print resolution, because all three enter it.
What print-quality grade should a Micro PDF417 label reach?
Grade samples to ISO/IEC 15415, the 2D print-quality standard, using the aperture and light your line actually runs, and accept C (1.5) or better overall. B (3.0) leaves margin for print-head wear and substrate changes. A symbol that decodes on the bench but grades under 1.5 will fail on your worst day, dirty or angled or backlit, not your best one.
Is Micro PDF417 the component used in GS1 composite codes?
Yes. GS1 composite codes pair a linear symbol with a stacked 2D component so both scan in one pass, and those components are built on MicroPDF417: CC-A uses a defined subset of the versions, CC-B the full symbology. That is one more reason a retail or healthcare line needs a reader whose list names MicroPDF417, not just PDF417.



