Mitsubishi PLC Models: A Cost-Focused Guide (and How to Learn Mitsubishi PLC Without Wasting Money)

Searching for 'mitsubishi plc' or 'mitsubishi plc models' gives you a lot of spec sheets. What it doesn't give you is the number that matters most in industrial purchasing: the total cost of ownership. A CPU price is only the start.

I'm not a controls engineer and I don't work for Mitsubishi Electric. I'm the procurement lead at a 28-person machine builder. Over the last six years, I've tracked roughly $180,000 in PLC-related spending, compared quotes from a dozen vendors, and built a cost spreadsheet that has saved me at least as much as it has annoyed me.

So here's the honest answer to 'Which Mitsubishi PLC should I choose?': it depends. But 'it depends' is only useful if you know what it depends on. In my experience, it comes down to three or four situations.

Before the model, answer three questions

Before you look at model numbers, answer these:

1. Are you replacing an existing controller, or building something new?
2. How many I/O points does the application really need?
3. What programming software does your team already know?

The third one matters more than most buyers think. I've seen projects stall for weeks because the hardware was a 'better' model but the software skill wasn't there.

Scenario 1: You're keeping existing machines running

If you have Mitsubishi PLCs already running in your plant, the safest decision is usually to match the series you own. If the machine has an FX3U, buy an FX3U replacement module. If it has a Q, buy Q. This sounds obvious, but it's not what many teams do.

I've had suppliers suggest an 'upgrade' during an emergency replacement. They weren't wrong from a technology perspective. From a procurement perspective, they were proposing a new CPU, new software, a different program conversion, and commissioning time on top of an urgent breakdown. The upgrade might make sense eventually. During an unplanned stop is the worst time to do it.

There is one exception: if the series is officially discontinued or your distributor can't source spares. The MELSEC A series falls into this category for most applications. I'm not going to pretend I know exactly when every A-series module will disappear. But if you're still running A-series equipment, treat it as a red flag. Build a replacement schedule before the module fails, not after.

Looking back, I should have built a five-year spare-parts list for our legacy machines earlier. At the time, we were too focused on new projects to worry about old panels. Then a $400 A-series module cost us $1,800 in expedited shipping and an overnight technician call.

Add obsolete modules to your budget now. Holding one spare in a drawer is almost always cheaper than paying for urgency.

Scenario 2: You're building a new compact machine

For a new machine with fewer than roughly 100 I/O points, I normally start my vendor conversations with the MELSEC iQ-F series, specifically the FX5U. Ten years ago, the default answer was often FX3U. That answer isn't wrong today, but it's not automatically right either.

When I compared a small labeling machine quote side by side, the FX3U with a separate Ethernet module and the FX5U with built-in Ethernet landed in the same cost ballpark. The FX5U was more expensive as a bare CPU. But once I added the Ethernet module, mounting hardware, extra wiring, and the additional part number to track, the FX5U became the cheaper overall build. It also had fewer components that could fail.

This is one of those 'industry evolution' moments. What was best practice in 2018 is not automatically bad in 2025; the fundamentals haven't changed. You still count inputs, outputs, and spare parts. But the execution has changed. Built-in Ethernet is no longer a premium feature; it's the default expectation on a new machine.

That doesn't mean FX3U is dead. If your maintenance team is fluent in GX Works2 and you already stock FX3U spares, the older platform can still be the lowest-TCO choice for a repeat machine. I'm not suggesting a migration just for the sake of freshness. But if you're starting from zero, I'd learn GX Works3 and build with FX5U.

Scenario 3: You're building a large or networked line

Above 100 I/O, distributed remote I/O, servo axes, or data collection, the conversation shifts to MELSEC Q and MELSEC iQ-R.

The Q series is the mature platform. It has a huge installed base and a wide range of modules. The R series is the newer flagship, designed for modern networks like CC-Link IE TSN and tighter integration with GX Works3.

I'm not a controls engineer, so I can't tell you exactly which CPU has enough scan-rate headroom for your application. What I can tell you from a purchasing perspective is that the software ecosystem is the deal-breaker. If your integrator has years of Q experience, a Q quote may deliver faster and with less risk. If they're new to the project and using GX Works3, R might actually be a no-brainer.

In Q2 2024, I asked one integrator why their quote showed Q for a new line. Their answer was honest: they had spare Q modules in stock. That's a legitimate reason. It just isn't a TCO reason for me. I asked to see the same design kitted with R. The price was close—well, close for PLCs. The deciding factor wasn't the CPU; it was the engineering hours and the future spare strategy.

Honestly, I'm not sure why some new projects still get quoted with Q. My best guess is a combination of legacy parts inventory and the installer's muscle memory. Neither is a bad reason, but you should put them on the table.

Scenario 4: You're asking, 'How to learn Mitsubishi PLC?'

If you've ever typed 'how to learn mitsubishi plc?' into a search bar, this scenario is for you. If you're new to Mitsubishi PLCs, your first purchase should not be a PLC. A new learner can waste months with a used controller, a missing cable, and a software license issue.

Start with the official materials from Mitsubishi Electric and the IEC 61131-3 standard. The GX Works3 environment includes a simulator, so you can write ladder logic and structured text without a physical CPU. You still need to learn wiring and I/O hardware eventually, but you can learn addressing and program structure at your desk first.

Check with your local Mitsubishi Electric distributor about software trial options, because GX Works3 licensing varies by region. That's a classic 'I'm not the software licensing specialist' moment. What I do know is that a week-long online course is usually cheaper than buying the wrong hardware and giving up.

If you already know another brand of PLC, the biggest learning curve is the editor, not the logic. The first time you open GX Works3, you'll probably feel slow. That's normal. It doesn't mean Mitsubishi is harder; it means your muscle memory is in another tool.

Which scenario are you in?

Here's the shortcut I use when I'm confused by a spec request:

1. Existing installed base in your building? Match the platform unless there's a documented end-of-life issue.
2. New build, under roughly 100 I/O? Start with FX5U. Get a quote, then compare it with whatever your maintenance team knows best.
3. New build, over roughly 100 I/O, with motion or network complexity? Get at least one Q quote and one R quote in front of you before you commit to software.
4. No in-house programming skill yet? Spend on training and simulator access before buying hardware.

Bottom line: the best Mitsubishi PLC model for you is the one that fits your installed base, your team's software skills, and your future spare-parts reality. That's not a fence-sitting answer—it's a way to avoid buying a PLC that impresses the supplier and haunts your budget.

Sources and further reading

• Mitsubishi Electric FA Global — MELSEC PLC product page
• Mitsubishi Electric FA Global — GX Works3
• IEC 61131-3: PLC programming languages standard, International Electrotechnical Commission