Brent
Printing 2D Codes on PCBs and SMDs: Building Traceability with TIJ Inkjet
Table of Contents
Somewhere in the last few years, marking requirements in electronics quietly changed — and nobody sent out a memo. Labelling the box and the reel used to be enough. Now customers in automotive, medical, and industrial electronics want traceability at the part level: a DataMatrix or QR code on every PCB, every SMD carrier, every assembled module, tied back to a production batch, a test record, and sometimes the exact reel of material it came from. If a quote has landed on your desk asking for “unit-level traceability,” you already know where this is heading.
The difficulty is physical rather than administrative. A DataMatrix that is actually useful for unit-level tracking is typically 3 to 5 mm across. On a PCB edge, a connector body, or a moulded SMD carrier, that is a tight fit — and the code then has to survive handling, soldering or washing steps, and machine-vision reading at line speed.
The short answer: making 2D codes work on boards and components comes down to three things, in this order. First, an ink that survives your entire downstream process — solvent-based, matched to solder mask and engineering plastics, not a general-purpose aqueous ink. Second, a print engine with the resolution and drop control to render a 300-micron module cleanly, which is where TIJ at 600 dpi separates itself from CIJ. Third, a data loop that connects the coder, a verification reader, and your MES so every printed code becomes a record rather than a decoration. Get the sequence right and the marking station stops being a compliance headache and starts being the backbone of your quality system.
A 2D Code Only Counts If It Survives the Whole Process
The conclusion: a code on a PCB or component is not judged by how it looks coming off the coder. It is judged by whether it still scans after every downstream step — washing, soldering, handling, inspection, and years in the field.
The everyday version of this: the shipping label on a parcel. A label that looks pristine when you stick it on means nothing if it is shredded by the time the box reaches the customer. In electronics, that journey can include a wash cycle, a wave or reflow soldering step, and abrasive contact with trays, tubes, and feeders. The ink chemistry has to be selected for the journey, not for the first five seconds after printing.
Why the surface matters so much. For PCBs, the marking surface is usually solder mask — a cured epoxy layer — and occasionally bare board or copper. For SMD carriers and component bodies, it is engineering plastics: PA, PPS, LCP, and epoxy moulding compounds. These materials differ enormously in surface energy, and that is what decides whether ink stays put. A water-based TIJ ink, which performs beautifully on paper and on some plastics, will often rub straight off a solder mask or an LCP body after a light wipe. Solvent-based inks partially dissolve or swell the surface layer and lock the pigment into it, which is why they are the default choice for electronics marking rather than a premium option.
The same low-surface-energy problem shows up across electronics and packaging alike, and our guide to why ink will not stick to plastic film breaks down the surface-energy mechanics and the pre-treatment options when an ink simply refuses to anchor.
A discipline that keeps this manageable. Before any production commitment, print sample codes with the candidate ink and put them through your real downstream process — then test. Three tests catch most problems while they are still free to fix:
- IPA wipe test: a set number of rubs with isopropyl alcohol, agreed with your customer, then re-verify the code.
- Tape test: adhesion check after the surface has been handled the way your line handles it.
- Post-reflow inspection: confirm the code still decodes after the thermal excursion it will actually see.
One automotive PCB supplier discovered their codes yellowed after reflow at a 260 °C peak. The fix did not involve changing a single consumable — they moved the marking station to after the soldering step. Worth remembering before you spend money on a new ink: sometimes the process order is the problem.
Resolution and Drop Control Make Small 2D Codes Scannable
The conclusion: printing a scannable DataMatrix at 3 mm requires both high resolution and consistent drop placement, and this combination is where TIJ outperforms older coding technologies in this application.
The arithmetic is unforgiving. A DataMatrix is a grid of small dark and light modules. A 10 × 10 module code at 3 mm means each module is roughly 300 microns across. For that code to decode reliably, module edges have to be clean and module sizes have to be uniform. There is very little tolerance to spend.
Why CIJ cannot follow you here. Continuous inkjet builds characters from large deflected droplets in a dot-matrix pattern. That approach is superb at its job — printing “BEST BEFORE” on a beverage can at 120 m/min is exactly what CIJ was designed for — but it is structurally incapable of resolving a 300-micron module. TIJ raster-prints with picoliter drops at 600 dpi, producing modules with crisp, printer-like edges. The two technologies are not competing on the same task; only one of them is doing this one.
If you are managing the resolution question more broadly on your line, our breakdown of 300 versus 600 DPI explains how horizontal density is really bought with line speed — useful when a dense code and a fast belt have to coexist.
The camera analogy is the right one. Vision systems and scanners are cameras, and they read a 2D code the way a sensor reads an image. A blurry, low-contrast, or distorted code produces a failed read in exactly the way a shaky photograph produces a blurry portrait. High resolution plus uniform drops gives the camera something sharp to work with; contrast between ink and surface does the rest. It is also why dark, fast-drying solvent inks formulated for electronics surfaces beat general-purpose inks on decode rate even at identical resolution — the ink is doing work that the printhead cannot.
In practice this is not a lab exercise. A PCB assembly house marks a 4 mm DataMatrix on the board edge of every panel, carrying the serial number pushed from their MES. With 600 dpi TIJ, the AOI (automated optical inspection) system downstream reads the code and loads the correct test program for each board automatically. No operator scanning, no mix-ups between similar product variants. The code has become an active part of the process rather than a label attached to it.
The Payoff Comes From Closing the Loop With Your MES
The conclusion: printing unique codes is only half of a traceability system. The value appears when the coder, a verification scanner, and your MES or quality system form a closed loop.
How the loop should run. Your MES generates a unique serial or encrypted code for each part and sends it to the coder — over RS232, TCP/IP, or file transfer, depending on the equipment. The TIJ coder prints it as the part passes. A fixed scanner verifies the printed code immediately downstream, and that verification result flows back to the MES, which logs it against the serial number. If a code fails to verify, the part is diverted before it ever leaves the station.
Every unit now carries a birth certificate that follows it through test, assembly, and packing — and, if it returns from the field as a warranty claim, straight back to the exact batch and machine that produced it. That last part is usually what convinces management the investment is worth making. A verification step is only meaningful if its result is recorded somewhere durable, and our analysis of 2D code decode rates and what actually drives them covers how print quality translates into pass rates once a reader is watching every unit.
Questions to settle with the supplier before you sign. Be blunt and specific:
- What communication interfaces does the coder support?
- Can it receive variable data per part at line speed, or only fixed text per batch?
- Does it support print-and-verify with a downstream reader?
- Can print logs be exported for archival?
Suppliers who answer those four fluently have done this integration before. Suppliers who change the subject have not.
What it looks like when it works. A manufacturer supplying an automotive electronics Tier 1 was required to provide full unit-level traceability as a condition of IATF 16949 compliance audits. They implemented an in-line TIJ coder with a 600 dpi cartridge, an inline verification reader, and a MES interface that binds each serial number to test data automatically. The customer’s audit passed first time. In the customer’s eyes, the difference between a supplier with a closed data loop and one without is the difference between a traceability partner and a parts vendor.
Case Study: From Batch Labels to Unit-Level Codes in Ten Weeks
A mid-sized electronics manufacturer in Southeast Asia producing sensor modules for a European industrial client received a contract update: within one quarter, every module had to carry a 3 mm DataMatrix with an encrypted serial number, scannable after the conformal coating inspection step. Batch labels on the box would no longer be accepted.
The starting point was not encouraging. Two handheld coders and a label printer, three operators marking trays by hand, scan rates around 85 percent at best. The codes were essentially cosmetic — the data lived in spreadsheets that nobody could reconcile against physical parts.
The transformation ran in three phases.
- Phase one, physical: a positioning fixture presenting each module at a fixed height under the TIJ printhead on the conveyor, with a photoelectric sensor and encoder handling trigger and synchronisation.
- Phase two, marking: an in-line TIJ coder with a 600 dpi cartridge and a fast-drying solvent ink, validated through IPA wipe tests and a simulation of the customer’s coating-inspection handling.
- Phase three, data: a MES interface pushing unique serials to the coder and logging the inline verification reader’s results automatically.
The outcome after ten weeks:
- First-pass decode rate of 99.5 percent
- Two operators replaced by a single line attendant covering two lines
- Zero hours spent reconciling marking spreadsheets — the item the finance director noticed first
The customer who had applied the pressure subsequently moved additional volume to them, citing traceability maturity as a factor in the decision.
The lesson the team now repeats internally is the interesting part: unit-level traceability was not a compliance cost. Handled properly, it was the cheapest process upgrade they made that year — because it forced them to fix the physical marking discipline they had been papering over with manual labour.
At FirstColor we support electronics manufacturers in exactly this position. If you want to test the first two pillars — ink survival and print quality — on your own boards before committing to the data layer, we provide sample marking with your components and validation against your specific test protocol. It costs nothing but a few sample parts. Contact us here to arrange it.
Conclusion
Building real traceability on PCBs and SMDs comes down to three things, in order: an ink that survives your entire downstream process, a print engine with the resolution and drop control to make small 2D codes genuinely scannable, and a data loop that turns printed codes into living records in your MES. Get the sequence right and the marking station stops being a compliance headache and becomes the backbone of your quality system.
FAQ
Can TIJ really print a readable DataMatrix at 3 mm?
Yes, provided you are working with 600 dpi resolution and a solvent ink matched to the surface. At 3 mm across a 10 × 10 module grid, each module is about 300 microns, which TIJ can resolve because it raster-prints picoliter drops rather than deflecting large ones. The limiting factor is usually ink-to-surface contrast and drop consistency, not the print engine’s theoretical resolution.
Why do water-based inks fail on solder mask and SMD plastics?
Surface energy. Solder mask is a cured epoxy and SMD bodies are engineering plastics such as PA, PPS, and LCP — all relatively low-energy surfaces that aqueous ink struggles to wet and anchor to. Solvent-based inks partially dissolve or swell the surface layer, which locks the pigment in. This is why solvent chemistry is the default for electronics marking rather than an upgrade.
What should I ask a coder supplier before buying for traceability?
Four questions decide whether an integration is realistic. What communication interfaces does the coder support? Can it receive per-part variable data at line speed, not just fixed batch text? Does it support print-and-verify with a downstream reader? And can print logs be exported for archival? Fluent answers mean the supplier has done this before; vague answers mean you will be the one paying for their learning curve.
Do I need to move my marking station after soldering?
Sometimes, and it is worth testing before you assume otherwise. One automotive PCB supplier found their codes yellowed after reflow at a 260 °C peak; relocating the marking station to after the soldering step solved it without changing any consumable. Run your candidate ink through your actual thermal profile first — the answer is often cheaper than a new formulation.