Buying Guides
Repair, Upgrade, or Source a New Control Panel? How to Decide
A buyer's decision guide to control panel failures: when to repair in place, when to upgrade the components, and when sourcing a new panel is genuinely cheaper.
A control panel that keeps tripping, a PLC the supplier stopped making in 2011, and a cabinet with rust at the base—each of these looks like the same problem from the maintenance queue. They are not. One needs a $40 component swap. One needs a controller migration. One needs the panel replaced.
Choosing wrong is expensive in both directions. Repairing a panel that should have been replaced means you pay for the same downtime twice. Replacing a panel that needed one relay means you pay for a new build, a new commissioning window, and a document set you did not need.
This guide gives you the decision framework: the signals that separate the three paths, a scoring matrix you can run on your own panel, and the exact information that turns a vague inquiry into a quotable RFQ.
For the wider commercial picture—what drives cost, how long builds take, what certifications to ask for—start with our control panel buying guide. Our custom control panels team handles new builds and retrofit packages.
The three paths, defined precisely
Before scoring anything, get the categories right. Most bad decisions come from calling a retrofit a repair.
Repair. Restore the panel to its original specified function. Replace a failed component with an equivalent, fix a termination, and reseal an entry. The panel's design capacity, safety rating, and documentation all stay as they were. Nothing changes functionally.
Upgrade / retrofit. Keep part of the existing asset—usually the enclosure, the field wiring, and sometimes the power section—and change its capability. Typical retrofit work: replacing an obsolete PLC and its I/O, adding a communications gateway, upgrading a starter to a VFD, adding instrumentation. The panel's function changes, so its documentation and its safety assessment must change with it.
Source new. Specify and build a replacement panel, then swap it in during a planned outage. The enclosure, layout, protection, labelling, and documentation are all new. Existing field cables are usually reused and re-terminated.
Buyer Decision. "Can you fix it?" is the wrong first question. The right first question is: after this work, will the panel be doing something different from what it was doing before? If yes, it is not a repair, and treating it as one will leave you with a modified panel and no modified documentation — which is an audit finding waiting to happen.
The seven signals that decide it
Score your panel against each signal. Be honest; the point of the exercise is to surface the answer you would rather not have.
Signal | Points repair | Points upgrade | Points new build |
|---|---|---|---|
1. Component availability | Spare in stock or available from distribution | An obsolete controller, but a modern equivalent exists. | No equivalent; platform abandoned. |
2. Documentation state | The current single-line diagram and I/O list exist. | Drawings exist but are are not as-built. | No drawings, or the panel has been modified many times without record. |
3. Enclosure / structural condition | Sound, no corrosion at the base or door seal | Sound but dated, some surface rust | Corrosion at the base, the door no longer seals, and water has been inside. |
4. Safety compliance | SCCR adequate for the present supply, safety circuit intact | Adequate after modification and reassessment | Unknown, or inadequate and not economically correctable |
5. Capacity headroom | Failure is isolated; capacity is fine. | Need more I/O, more axes, or a new protocol | Load has fundamentally changed. |
6. Downtime cost | Outage is cheap or schedulable. | Short outage available in the next window | Major outages can be planned months ahead. |
7. Programming / software | Program and software available, runs on current hardware | Program recoverable and portable | The program is lost, or the software will not run on any supported machine. |
How to read the result. Look at where your points concentrate—but treat three signals as near-absolute, because they override the totals:
- Signal 4 fails (safety compliance unknown or inadequate) → repair is off the table. A panel whose short-circuit rating, or whose safety circuit integrity, cannot be demonstrated is not a maintenance item.
- Signal 2 fails badly (no drawings at all) → an upgrade is a rebuild with extra steps. You will pay for the survey work either way, so a new build is usually cleaner and not much more expensive.
- Signal 7 fails (program lost) → any path requires rewriting the control logic. At that point the marginal cost of a new panel over a retrofit is small, and the outcome is far better documented.
Engineering Note. The single most common expensive mistake is retrofitting a panel with no drawings. The retrofit itself goes fine. What you end up with is a panel nobody can fully describe — new controller, unknown legacy wiring, no as-built record. When it next fails, you start again from zero.
What each path actually costs
The comparison buyers get wrong is comparing the invoice rather than the outage.
Cost element | Repair | Upgrade / retrofit | New build |
|---|---|---|---|
Component / hardware | Lowest | Medium | Highest |
Engineering and drawing work | Minimal | High (survey + redesign) | High, but done once and delivered |
Labor on-site | Shortest, but may repeat | Medium | Longest single window |
Commissioning and debugging | Minimal | Highest risk—new logic on old wiring | Predictable if FAT is done properly |
Documentation outcome | Unchanged | Often incomplete | Complete and delivered |
Probability of recurrence | High if the root cause was age-related | Medium | Lowest |
Typical outcome at year 5 | Same panel, older | Hybrid panel, hard to support | Supported, documented asset |
The pattern is consistent: repair wins when the failure is isolated and the panel is fundamentally sound. A failed contactor on a five-year-old panel with drawings and stocked spares is a repair, and anything else is over-processing.
New build wins when the panel's problem is its age or its documentation, not one component. If you are replacing the controller, the I/O, and half the power section, and you do not have as-built drawings, you are building a new panel while paying retrofit labor rates.
Buyer Decision. Price the outage, not the panel. If an unplanned stop costs you more per hour than the difference between a retrofit and a new build, the new build is the cheaper option even when its invoice is larger. This is the calculation maintenance budgets rarely make and operations always feels.
The obsolescence trap
Obsolete PLCs deserve their own section, because this is where most retrofit projects actually start and where the reasoning is most often muddled.
The real problem is rarely the CPU. When a PLC platform is discontinued, the CPU itself can usually still be found on the secondary market for a while. What disappears first is everything around it:
- Programming software licenses and the hardware dongles they need
- A PC operating system that will still run that software
- I/O modules in the specific mix your panel uses
- Manufacturer support for firmware and communications
What "still works" really means. A panel running an obsolete controller is not broken. It is one failed module away from an unplanned migration, with no software to do the migration in, and no drawings to migrate from. That is a schedule risk, not an engineering problem, and it is best handled before it becomes an outage.
Three migration routes, in increasing order of disruption:
- Like-for-like replacement with a current platform—new CPU and I/O, program ported, field wiring reused where practical. Lowest disruption, but requires the original program and a careful I/O mapping.
- Gateway / protocol conversion—keep the legacy controller, add a modern communications layer. Useful when the panel works and you only need data out of it. It does not solve the hardware risk.
- Full panel replacement—new controller, new I/O, new wiring inside the cabinet, existing field cables re-terminated. Highest cost, cleanest long-term position.
Engineering Note. If the original program is available, take a backup before you do anything else — before you even decide. A readable program turns a migration into an engineering task. A lost one turns it into a reverse-engineering project, and that difference usually dwarfs every other cost in the job.
When you should not repair
These conditions mean stop and reconsider, regardless of how cheap the fix looks:
- Water has been inside the cabinet. Moisture ingress causes corrosion and tracking damage that does not show up immediately. The component you replace works; the one next to it fails in three months.
- The safety circuit has been bypassed or modified. Any evidence of jumpered safety functions means the panel's safety assessment is void until it is redone properly.
- The base or the door seal is structurally corroded. The cabinet can no longer deliver its rated IP protection. Replacing internals inside a leaking box just moves the failure date.
- You cannot demonstrate the short-circuit current rating against the present supply. Upstream transformer changes over the years frequently invalidate a panel that was fine at installation.
- Repeated failures of the same component. A third contactor in eighteen months is not three failures; it is one root cause you have not found.
When you should not replace
Equally, these are cases where a new build is overspending:
- A single failed component with a stocked spare and a sound panel around it.
- A panel under roughly ten years old with good documentation, where the enclosure, wiring, and protection are all in good condition.
- A capacity problem solved by one additional module, where the controller platform is current and supported.
- A cosmetic problem—faded labelling, a scuffed door, outdated legend text. Refurbish, do not rebuild.
What to send us to get a real quote
Vague inquiries get vague numbers. This is what turns either a retrofit or a new build into a firm quotation:
For any path: - Photographs of the panel—closed, door open, and the nameplate/rating plate - Photographs of the base and the cable entry area - What it does, and what is failing or insufficient - Site conditions: indoor/outdoor, washdown, ambient temperature, chlorides present - Any deadline driving the decision, and what an outage costs you
Additionally, for a retrofit: - Existing single-line diagram, if one exists - Controller make and model, and I/O count - Whether the program is available and in what format - Whether you can accept a planned outage and for how long
Additionally, for a new build: - I/O list with signal types, or a marked-up sketch - Motor list with ratings and starter or drive requirements - Required standards—UL 508A, CE, or a specific SCCR - Instrumentation to be accommodated, and any network requirements
You do not need all of it to start. Send what you have; we will tell you what is missing.
Frequently asked questions
How do I know if my panel is worth repairing?
Run the seven signals. If component availability is good, documentation exists, the enclosure is sound, and safety compliance is demonstrable, a repair is usually right. If two or more of those fail, the repair is likely to be the first payment on a larger job.
Can you retrofit a panel without the original drawings?
Yes, but budget for a survey, and expect the outcome to be less well documented than a new build. If the drawings are entirely absent, compare the retrofit quote against a new build — the gap is often smaller than expected, and the new build ships with documentation.
What if our PLC program is lost?
It can be recovered from the running CPU in many cases, or reconstructed from the I/O behaviour and the process description. It is a real engineering cost either way, and it argues for doing the migration and the documentation at the same time rather than in two stages.
Is a new panel always safer than an old one?
Not automatically "safer" — a correctly maintained old panel is safe. But a new build is demonstrably compliant: you receive the drawing set, the test records, and the rating evidence. With an old panel, that evidence is usually the thing you cannot produce.
How long does a new custom control panel take?
As a sourcing partner working with build shops in the Foshan network: a complete custom panel starts from 1 unit, with drawings and BOM reviewed before we confirm — typically 35–40 days. Empty enclosures start at 5 units, around 15 days. Stocked components ship in 3–5 days. Where a UL listing is required, the build is placed with a partner holding the relevant UL file, and the file number is provided in writing so you can verify it.
Will a new panel reuse our existing field cables?
Usually yes, and it is worth planning for. Re-terminating existing cables into a new cabinet is standard; what you should confirm early is cable length and condition at the entry point, and whether spare cores exist for any additional I/O.
Can we witness the factory acceptance test remotely?
Yes. FAT records are included with the build, and remote witnessing can be arranged during the test window. Our pre-shipment checklist shows what is covered.
What does a retrofit cost compared to a new panel?
Common cost is lower for a retrofit; engineering and commissioning cost is usually higher, because you are designing against an existing, imperfectly documented panel. Compare total cost including the outage, and compare what documentation you end up with.
Engineering Takeaway
Repair when the failure is isolated and the panel is sound, documented, and compliant. Upgrade when the capability needs to change but the enclosure and field wiring are worth keeping — and only when you have, or can recover, the documentation. Source new when the problem is the panel's age, its missing drawings, or its undemonstrable safety rating.
Two rules hold across almost every case: back up the PLC program before anything else happens, and never retrofit a panel whose safety rating you cannot demonstrate. Both are cheap now and extremely expensive later.
Related guides
Get a decision you can defend
Send us what you have and we will tell you which path we think it is, and why:
- Photographs of the panel — closed, door open, and the rating plate
- Existing single-line diagram or I/O list, if you have one
- Controller make and model, and what is failing
- Your outage window and what downtime costs you
Email info@uniregal.com, or use the contact form. To send drawings for review, use upload drawing → engineering review.
You will get a reply within one working day. Complete custom panels start from 1 unit after drawing and BOM review, 35–40 days; empty enclosures from 5 units, around 15 days; stocked components ship in 3–5 days. FAT records are included. UL-listed builds are placed with a partner holding the relevant UL file, with the file number given in writing.
