Relay Notes

Allen-Bradley PLC Buying Decisions: OEM vs. Aftermarket, Scenario by Scenario

2026-11-06 · Pavel Novak

Short version: there is no single buy-OEM rule for Allen-Bradley PLC work. Anyone who hands you one answer is usually selling the premium markup on that answer.

I have been handling Allen-Bradley PLC spare parts and workshop supply orders for about nine years. I have personally made — and written down, because apparently I have issues — somewhere around 14 significant purchasing mistakes. Total wasted budget sits at roughly $38,000. Don't quote me on the 14. It might be 12. I stopped counting when the tracking spreadsheet corrupted and I never rebuilt it.

What that money bought me is a working rule of thumb: the same decision — buy OEM or go aftermarket — has a different right answer depending on what kind of thing you are buying. It is not about loyalty to the brand or hunting the cheapest line item. It is about what happens when the part fails.

So here is how I actually split it now. Three buckets, and each one gets a different answer.

The three buckets

  • Core control hardware — CPUs, chassis, Controllogix modules, I/O cards. The stuff that stops a line.
  • Software and knowledge — programming environments, licenses, and the training that makes your people useful on them.
  • Maintenance-side tools and consumables — meters, chargers, backup batteries, the small stuff that lives in a drawer until three in the morning.

Notice I did not split it by price. Price is the worst way to categorize these purchases, because the cheap thing is sometimes the right thing and sometimes the $4,000 mistake you find out about six months later.

Scenario A: Core control hardware — stay with the OEM, no exceptions

This is where I got burned the hardest, and where I am now the least flexible.

In early 2021 we needed a spare Controllogix I/O module for an Allen-Bradley PLC rack, and we needed it in a hurry. Lead time from our usual channel was nine weeks (that was accurate as of that quarter — the supply situation has improved since, so verify current lead times before you plan around them). A secondary market seller offered the same module, refurbished, at about $600 under list. I approved it. Checked the part number twice, actually.

What I did not check was the firmware revision or the date on the internal battery. The module sat in a cabinet for four months, then went into service when a line controller faulted. It ran for about six weeks. Then it dropped out mid-shift, took a two-station cell down with it, and we spent six hours diagnosing something that should have been a 15-minute swap.

Cost of the saving: roughly $600 on the module, then about $4,800 in overtime, lost throughput, and one very unpleasant Monday morning call with production. Plus the credibility hit, which does not show up on the invoice but definitely shows up in the next budget meeting.

Here is something vendors won't tell you: the word 'refurbished' on an Allen Bradley PLC module covers everything from a factory-tested unit with a fresh firmware flash to a board pulled out of a decommissioned plant in 2014 and wiped down with a cloth. Both get called refurbished. Only one is worth buying.

My current rule: for Allen Bradley PLC modules, CPUs, chassis, and anything in the fault-critical path, I buy through authorized channels only — full stop. We pay more. We also stop getting surprises. I would rather work with a supplier who specializes in Allen-Bradley hardware and honestly tells me when a part has a long lead time than with a general industrial marketplace that lists 40,000 SKUs and knows almost nothing about any of them.

That is actually the trade I keep making: specialists over generalists. A vendor who says 'we don't stock the older SLC line, here is who does' has earned more of my business than one who promises me the world in three days.

Scenario B: Software and training — official for the license, flexible for the classroom

Different situation. The official programming environment is non-negotiable — if you are supporting Allen-Bradley PLC systems, buy the real license through the real channel. Pirated or gray-market copies are a legal risk and, more practically, an update-and-support dead end.

The training around it, though? That is where I would loosen up.

In 2019 we spent about $9,200 sending three people to a five-day official course. It was good. It was also the wrong shape for what we actually needed — we did not need three people trained to build a system from scratch, we needed two people who could safely modify existing logic and one person who could troubleshoot at 2 a.m.

The next round, we brought in a certified independent instructor for two days of focused troubleshooting, on our own equipment, at about $2,800 total. Different result. The team retained more because the examples were our actual mess, not a classroom demo rack.

I want to say we saved around $6,000 on that switch, but do not quote me on that — the number depends on how you count travel and time off the floor. The point is the shape of the training mattered more than the name on the certificate.

Scenario C: Workshop tools and consumables — this is where you can save, and also where the sneaky costs live

Here is the reverse case, and it is the one that surprises people: the cheap, boring stuff in the maintenance drawer is where I have saved the most and also lost the stupidest money.

Two real examples.

First, chargers. In 2023 our shop had accumulated an unholy collection of charging bricks — a proper Blue Point battery charger for the 12V jump packs, a couple of generic units, and, for reasons I still cannot fully explain, a Nikon D5000 battery charger that a tech had brought from home because the camera battery in a loaner inspection scope kept dying. That one worked for the scope. It did not work for anything else. Someone once tried to recharge a PLC backup battery with it, not realizing those are primary lithium cells that are not designed to be recharged at all. That ended exactly how you would expect — the cell got hot, then useless, and we found out later that the module had been running on a battery that read fine on a meter while being completely unable to hold a load.

Second, and this is the one I would put on a poster: how to test a battery with multimeter is not the same skill as knowing whether that battery is any good. A meter tells you the open-circuit voltage. A tired cell can sit at a perfectly healthy 3.6V with no load and sag to nothing the moment you ask it to do work. According to Rockwell Automation's published battery replacement guidance (rockwellautomation.com, accessed December 2024), the module indicator flags low voltage — it does not tell you about remaining capacity, and it certainly does not tell you about a cell that looks healthy at rest. What actually caught our bad cells was a load test, not the meter reading. If you are only checking voltage, you are probably catching half the failures at best. Verify the specific replacement and testing procedure in current documentation before you rely on it.

So in this bucket I do shop around. A decent meter is a decent meter. A charger is a charger, as long as the chemistry and charge profile match — and that 'as long as' is doing a lot of work in that sentence. Verify the specification against the actual cell you are charging. Do not assume '12V charger' means 'charges my 12V pack properly.' It might mean 'charges it too fast and cooks it.'

So which one are you?

Three questions, in order:

  1. If this fails during a production run, what happens? If the answer involves a stop, a scrap batch, or a safety event — you are in Scenario A. Buy OEM, buy through a channel that honors a warranty, and stop optimizing the line item.
  2. Is this about capability rather than a physical part? Licenses, training, programming knowledge — Scenario B. Split it: pay for the license, negotiate the training.
  3. Would this sit in a drawer for months and be needed at 3 a.m.? Scenario C. Shop around, but spec the thing carefully, and stop trusting a voltage reading as proof of health.

If you are genuinely not sure, default to Scenario A. The whole reason I keep this list is that the expensive mistakes almost always came from treating an A-item like a C-item.

One last thing, and it is the part that took me longest to accept: knowing what you are not good at is a purchasing skill. I am not a battery chemist. I am not an automation systems integrator. When a supplier told me years ago that half my drawer was the wrong tool for the job, I was annoyed. He was right. That guy still gets my calls.

Prices and lead times in this piece are general reference points from late 2024 and early 2025 — verify current figures with Rockwell Automation or your authorized distributor before you budget against them. The supply chain has moved a lot in the last two years, and the old benchmarks age badly.

Pavel Novak

Pavel Novak

Pavel Novak is a generator and backup-power analyst specializing in standby, portable, diesel, gas, inverter, and commercial generating sets with transfer equipment. For reciprocating-engine sets, he applies ISO 8528-5 dynamic-performance criteria and IEC 60034-1 generator ratings to load steps, starting current, voltage and frequency recovery, cooling, and output; for transfer equipment, he uses IEC 60947-6-1 requirements. He helps facility teams and buyers select systems around emergency loads, runtime, commissioning, fuel strategy, service access, operating cost, and resilience.