If you have a dead controller board from a machine that’s been running since 2003, and the OEM no longer sells the spare, an industrial control PCB copy gets you a manufacturable replacement built from the physical board itself. We recover the layout, netlist, schematic, and BOM, rebuild the Gerbers, then build and test samples against your original. Send clear photos of both sides and we’ll come back with feasibility, lead time, and a price.
What we actually do to an industrial control board
Industrial control boards are their own animal. They mix a low-voltage digital section with 24 V field I/O, opto-isolation barriers, relay drivers, sometimes 110/230 V AC on the same panel. The copy work has to respect all of that, not just trace the copper.
Here’s the order of operations on a typical PLC I/O card or drive control board:
- Intake and photography. High-resolution scans of both sides before anything is touched, plus close-ups of every marking. If the board came out of a failed machine, we note the burn marks and the failure mode — that matters later.
- Component identification. Every part gets read, cross-referenced, and measured where the marking is gone. Sanded-off house numbers on custom ASICs and PALs are common on older industrial gear, and this is where a project gets interesting or gets stuck.
- Desoldering and layer work. Parts come off, the board gets cleaned, and inner layers are imaged one at a time. On a four-layer I/O card with mostly through-hole and a couple of QFPs, that’s fast. On a six- or eight-layer motion controller with fine-pitch BGAs and blind vias, the inner-layer imaging takes meaningfully longer because each layer has to be separated and registered against the drill pattern.
- Netlist and schematic reconstruction. The copper image becomes a netlist, the netlist becomes a readable schematic with functional blocks — power supply, isolation, field I/O, MCU section, comms.
- Gerber and stack-up rebuild. Copper, mask, silkscreen, drill file, and a stack-up that reproduces the original dielectric thicknesses and copper weights. Heavy copper on the relay and motor side gets measured, not guessed.
- Build and verify. Bare boards go out, samples get assembled, and then they get compared to the original — continuity against the extracted netlist, power rail voltages, isolation resistance across the barrier, and functional behavior on the bench or in your machine.
That last step is where a lot of cheap copy work falls down. A board that matches the Gerbers but has never been powered up next to the original isn’t a verified copy. We treat verification against the original board as part of the job, not an extra.
The industrial-specific details we check
- Creepage and clearance. If the original was laid out with mains or 24 V field wiring on a specific spacing, the copy holds that spacing. We don’t shrink a board that was drawn to IPC-2221 clearance tables for a reason.
- Copper weight on power traces. Relay commons, motor outputs, and 24 V rails often use 2 oz or heavier copper. Getting this wrong produces a board that works on the bench and cooks in a hot cabinet. IPC-2152 is the reference for what a given trace can actually carry at your ambient temperature.
- Conformal coating and potting. A lot of industrial boards are coated. Removal has to happen without lifting pads or damaging silkscreen. Fully potted modules are a different, slower job.
- Isolation barriers. Optocouplers, digital isolators, and transformer barriers get identified by part, not by footprint, because the creepage rating lives in the part number.
- Connector and mounting fidelity. A replacement board that doesn’t drop into the same DIN rail carrier or mate with the same ribbon header is worthless, so mechanical dimensions get measured with calipers and checked against the enclosure.
What to send us, and what comes back
You don’t need documentation to start. Most of these projects begin with nothing but a board and a machine that’s down.
Send this to get a real quote:
- Clear photos of both sides, whole board in frame, laid on a dark non-reflective surface. Even, diffuse light — no flash glare across the silkscreen.
- Board dimensions, roughly, and layer count if you know it.
- A readable close-up of the main processor or controller marking, plus any custom or house-numbered parts.
- One line on what you need: bare boards only, assembled and tested samples, a schematic set for your maintenance team, or all three.
- What the board does in the machine, if you know. “Servo axis feedback card for a 1998 injection molder” changes how we approach it.
What comes back from us first is a feasibility read — whether every part is identifiable or sourceable, whether firmware is involved, an estimated lead time, and a price. Then, on approval, the deliverable set: Gerbers and drill files, netlist, schematic in a standard EDA format, BOM with manufacturer part numbers and suggested alternates, assembly drawing, and physical samples if you ordered them.
Where industrial control PCB copy gets hard, and where it stops
This is the part most vendors leave off their pages. Honest limits save you weeks.
| Situation | How it affects the project |
|---|---|
| Two-layer or four-layer through-hole I/O card, all parts marked | Straightforward. Fast layout recovery, easy sourcing. |
| Six to twelve layers with fine-pitch BGA and buried vias | Doable, but inner-layer separation and drill registration add real time. X-ray helps confirm buried via structure. |
| Heavy conformal coating | Adds a cleaning step; slight risk to fragile silkscreen and older pads. |
| Fully potted or epoxy-encapsulated module | Slow and sometimes destructive. We tell you before we start whether we think the board survives. |
| Sanded or house-numbered custom logic (PAL, GAL, ASIC) | Sometimes we can characterize behavior, sometimes we can’t. This is the most common reason a project is not feasible as a like-for-like copy. |
| Obsolete parts with no drop-in replacement | Copy is fine; you’ll need a small redesign of that circuit block, which we can quote separately. |
| Board arrives severely burned or delaminated | Missing copper can’t be read. A second reference board, even a non-working one, often rescues the job. |
| Programmed MCU with no source code available | The hardware copy is straightforward; the firmware side is a separate feasibility question entirely. |
On that last row: a copied board with a blank microcontroller does nothing. If the original firmware is gone and you hold the rights to it, firmware recovery becomes part of the scope, and feasibility depends heavily on the specific part — an old 8051 with lock bits set behaves very differently from an STM32 sitting at RDP Level 2, and MSP430 parts with a blown JTAG fuse or an unknown BSL password are their own conversation. We’ll give you a straight read on your chip rather than a blanket promise.
Why the copy usually beats a redesign on a legacy machine
When a 20-year-old drive controller fails and no spares exist, engineers often reach for “we’ll just design a new one.” Sometimes that’s right. Usually it isn’t, for one reason: the original board embeds a decade of undocumented tuning — filter values chosen to match a specific sensor, timing that the PLC upstream expects, a watchdog interval that keeps a safety relay happy.
A copy carries all of that forward without you having to rediscover it. A redesign asks you to re-validate the whole machine. For a single production line that has to run next month, that math almost never favors a clean sheet.
The related work we see most on the industrial side is motor control board duplication for VFDs and servo amplifiers, and instrumentation cards where test and measurement board copy practices apply because analog front-end layout is doing real work. If what you need is documentation rather than boards, reverse engineering an industrial control board into a schematic set is a narrower and cheaper scope.
What drives the price
Nobody can quote an industrial board from a description alone. The cost moves with layer count, physical size, total component count, whether there are BGAs or fine-pitch parts, whether coating or potting has to be removed, whether obsolete parts need substitution research, whether firmware is in scope, and which deliverables you want. A four-layer I/O card with 80 through-hole parts and a two-layer relay board are worlds apart from an eight-layer motion controller. Send photos and you get a number instead of a range.
Legitimate use and confidentiality
The industrial work we take on is repair and continuity work: keeping out-of-production equipment running, replacing spares an OEM has discontinued, recovering documentation for a design your own company owns but whose files were lost with a departed engineer or a dead server, dealing with component obsolescence, and failure analysis after a field return. You are responsible for holding the rights to the design you send us and for complying with applicable IP law in your jurisdiction. Every project runs under NDA — your board, files, and firmware stay yours, and nothing leaves the lab.
FAQ
Can you copy an industrial control board if I don’t have the schematic?
Yes — that’s the normal case. The physical board is the source document. We extract the netlist from the copper, rebuild the schematic from that, and produce Gerbers from the imaged layers. No original files, no vendor cooperation, and no design database are needed. A second sample board helps if the first one is badly damaged.
How long does an industrial
Working on a board like this?
Send the chip marking or two photos. You get feasibility, lead time and price within 24 hours, and the check costs nothing.
Get a free quote
Related reading
Working on a board like this?
Send the chip marking or two photos. You get feasibility, lead time and price within 24 hours, and the check costs nothing.
Get a free quote