Quick answer
Large power systems rarely fail because the power hardware wears out. They are taken out of service by small components the manufacturer stopped making: low voltage circuit breakers, the Ethernet switch that links the controller to the inverters, and control boards built on parts nobody sells. Each has a route back into service, usually repair or a like-for-like substitute, sometimes a redesign, and for control boards an upgrade to the current controller. The time to find that route is before the part fails.
Key takeaways
- Electronics go out of production in about a decade; the power hardware around them is built to run for thirty years, so the small parts always go first.
- A substitute part is not always a drop-in: a replacement Ethernet switch had to be proved on a test set because the inverters need their commands at exactly the same moment.
- Three moves protect any ageing system: list the single points of failure, find out who else can service each one, and hold a qualified spare.
A power system worth millions can be put out of action by a component worth a few hundred pounds that the manufacturer no longer makes. The inverters, transformers and cabinets in a UPS, a STATCOM or a battery system are built to run for decades. The electronics that control them are not, and when one of those small parts fails with no replacement on the shelf, the whole system stops.
This article is for the people who own or look after power systems that are ten or more years old: facility managers, plant engineers, wind and solar farm operators, and the contractors they call when something trips. It covers the three parts we see go obsolete most often, what each one does when it fails, and the repair, redesign and upgrade routes that keep the system in service without buying a new one.
Why the small parts go first
Industrial power equipment is designed for a thirty-year life. The semiconductors, network chips and microcontrollers inside it are commercial electronics with production runs measured in years, not decades. Somewhere around the ten-year mark, the components on a control board stop being made, the network switch is superseded, and the breaker family is replaced by a new range that does not fit the old cutout.
Obsolescence arrives in three ways. A manufacturer discontinues a product line and stops supporting it. A manufacturer leaves the market altogether, which is what happened to a battery range we wrote about in our Kokam battery replacement guide. Or the product is still listed, but the individual components it was built from are gone, so the manufacturer cannot build another one either.
The result is the same in every case. The system works perfectly until one small part fails, and then the fastest route to a running site is the one somebody worked out in advance.
The low voltage circuit breaker nobody sells
What it does. Low voltage circuit breakers protect the cables, transformers and modules inside a power system. A whole generation of systems was designed around one breaker family, so the same discontinued breaker now sits in a lot of sites.
What happens when it fails. A breaker that trips and will not reset, or fails to trip when it should, takes the section it protects out of service. If there is no replacement, the section stays out.
The usual route. A modern breaker of equivalent rating rarely drops straight into the old cutout. The engineering work is in the fit: an adapter plate or a re-drilled mounting, the wiring terminations, and then the protection settings, which have to be rechecked because a different breaker has a different trip curve. Done once for a system type, that route can be repeated for every site that shares the design, and a refurbished exchange breaker can be held on the shelf for the next one.
The Ethernet switch between the controller and the inverters
What it does. In a multi-inverter system the master controller does not wire to each inverter separately. It talks to one Ethernet switch, and the switch fans out to every inverter. That one box carries every command the system relies on.
What happens when it fails. The controller cannot reach the inverters. Nothing powerful has broken, but there is no working system, because the part that carried the instructions is gone.
Why it is not a simple swap. The commands have to arrive at every inverter at exactly the same moment, on several channels at once, not one after another. Most switches on the market cannot do that. Our engineers built a test set to try candidates against the real timing requirement and proved a replacement in the lab. The live trial waits for a planned outage, because a system that carries load whenever it runs cannot be experimented on during normal operation. That is the honest pace of this kind of work: a lab-proven substitute, a scheduled trial, and then a qualified spare for every site that uses the same switch.
What an owner can do now. Find out whether anyone has qualified a substitute for the switch in your system, and hold one. A spare that has been proved on a test set is worth more than a shelf of parts that might fit.
The control board with parts you cannot buy
What it does. The control board runs the system: it measures, decides and sends the switching commands. On older systems it is a generation of board whose components have been out of production for years.
What happens when it fails. The system will not run, and the failed board cannot be repaired if the parts it needs no longer exist.
Why printing new boards is not the answer. A new board can be manufactured from the original design, but the components that go on it are the problem in the first place. Substituting near-equivalent parts changes behaviour in ways that take a great deal of testing to prove, and it leaves you with a new board that is already obsolete.
The route that works. Upgrade to the current generation of controller. The power hardware stays, the cabling and enclosure stay, and the control platform is replaced with one that is in production and supported. It is the same decision we describe for the STATCOM fleet, and it is how a site with a twenty-year-old control system ends up with a fully supported one without replacing the equipment around it.
Repair, redesign or upgrade: which route fits which part
| Part | Usual route | Typical downtime | What you keep |
|---|---|---|---|
| Circuit breaker | Equivalent modern breaker, adapted mounting, protection settings rechecked | A planned outage | Everything else in the section |
| Communication switch | Substitute qualified on a test set, trialled at a planned outage, spare held | None, if the spare is ready before the failure | The controller and every inverter |
| Control board | Upgrade to the current controller | A planned outage, usually days | Power hardware, enclosure, cabling |
| Battery module or capacitor | Like-for-like or qualified replacement | Hours to days | The rest of the string or module |
Refurbishment and substitution are the first answers, not the only ones. Where condition testing shows the power hardware itself is worn out, or a site has outgrown the system, a new build is the right call and we will say so.
What to do before a part fails
- List the single points of failure. Walk the system and note every component that would stop it if it failed and could not be bought tomorrow: breakers, communication switches, control boards, cooling fans, contactors.
- Ask who else can service each one. The original manufacturer is one answer. A specialist who has already qualified a substitute is a better one. If the answer is nobody, that is the part to solve first.
- Hold a qualified spare. Not a part that should fit, but one that has been proved on the system type, so the failure becomes a swap rather than a project.
Is there a part in your system that nobody sells any more? Renewable Energy & Drives qualifies substitutes, repairs boards and upgrades control platforms on UPS, STATCOM and battery systems from other manufacturers as well as our own. Start a consultation, email sales@renewable-d.com, or call +44 116 279 6131 (UK) or +1 414 562 6559 (US).
Frequently asked questions
What does it mean when a part in a power system is obsolete?
The manufacturer has stopped making it and holds no stock. The part may still be in service in thousands of systems, but when one fails there is no replacement to order. Obsolescence usually hits the electronics first (control boards, network switches, breakers) while the power hardware around them still has decades of life.
Can an obsolete control board be repaired?
Sometimes, if the failed component is one that can still be sourced. Often not, because the board was built from a generation of parts that no longer exists. In that case the practical route is an upgrade to the current controller, keeping the power hardware, enclosure and cabling in place.
Should I replace the whole system if the manufacturer no longer supports it?
Usually not. Loss of manufacturer support is a supply problem, not a condition problem. A specialist can qualify substitute parts, repair boards, and upgrade the control platform while the power hardware carries on. Replacement makes sense when testing shows the power hardware is worn out or the site needs more capacity than the system can give.
Tags
Working on a project like this?
Talk to our engineering team about your site, your grid connection or your power quality problem.
Get in touch
