I'm a controls technician. I've been handling service orders and panel retrofits for 12 years. I've personally made—and documented—30+ significant mistakes, some in the hundreds and a couple in the thousands of dollars. I keep a team checklist now so others don't repeat them.
If you're staring at a GE air conditioner control panel replacement decision, you have two main roads: order a direct replacement board, or retrofit with an Allen-Bradley Micro820 PLC. And inside either panel, you still have to answer what is the difference between a relay and a contactor.
Before I knew what I was doing, I thought a relay was small and a contactor was big. That's not wrong, but it's incomplete. I know because I made that mistake on live equipment.
Two Roads to Fix a GE AC Control Panel
Option A: Buy a direct replacement for the original GE board. Option B: Build a PLC-based control panel around a Micro820. Both are valid. The comparison is about speed, support, and what happens after you install it.
Availability and lead time. A direct replacement board is fast when it's in stock. In 2023, I found a discontinued GE board with a six-week lead time from a wholesaler and zero stock at local distributors. A Micro820 was sitting on the shelf, and I had the program written in a few hours. That retrofit was done in a day. If you can get the original board tomorrow, that's often the most efficient path. If not, every week of delay makes the PLC path look better.
Support and future failures. A direct replacement gets a dying system back to where it was. A PLC retrofit gives you something you can actually debug. I'm not an HVAC engineer, so I can't speak to the compressor staging algorithms inside an OEM board. What I can tell you from a controls perspective is this: when a board fails and you can't read the program or the logic, every repair is a fishing expedition. With a Micro820, I can open the project on a laptop, force outputs, and watch the logic run. That's a huge time saver.
Cost comparison: Don't hold me to the exact numbers; vendor pricing moves around. In March 2025, I looked at distributor listings for a Micro820 and saw prices in the $300–500 range depending on I/O and communication options (based on distributor listings, March 2025; verify current pricing). A GE replacement board, when available, often runs $300–900. The PLC path is not automatically more expensive.
What Is the Difference Between a Relay and a Contactor?
A relay (meaning a control-circuit switch) is designed to switch relatively small electrical loads: a pilot light, a solenoid valve, the coil of a contactor. Contact ratings are often 5 to 10 amps. A relay doesn't need heavy arc suppression because it isn't routinely breaking motor current.
A contactor is, technically, a heavy-duty relay. The difference is that contactors are engineered for power circuits: motors, heaters, compressors, condenser fan motors. They have higher current ratings, better arc suppression, and are meant to make and break high-inrush loads repeatedly. If you use a general-purpose relay to switch a motor, the contacts can weld or bounce, and the motor circuit can fail in dangerous ways.
The coil matters too. A large contactor's coil can pull a lot of inrush current. Many PLC outputs should not be expected to switch that coil directly. In that case, you use a relay as an interposing device: the PLC output energizes the relay, and the relay energizes the contactor coil. This is normal, and it confuses a lot of new techs.
Here is the surprising part for me: in a small PLC panel, you often use more relays than contactors because the PLC outputs can't switch motor loads. But one relay/contactor mistake can wipe out years of time savings.
This gets into arc-flash and SCCR territory, which isn't my expertise. I'd recommend having a licensed engineer or qualified electrical designer approve the final contactor selection.
In 2017, I made the classic mistake: I ordered a replacement control board before checking the contactor. The board was fine; the contactor contacts were cooked. That cost $312 and a week of downtime. Then in September 2022, I saved $120 by choosing a relay instead of a contactor for a condenser fan motor. It worked at first. By the second week, the relay contacts welded and a weekend service call followed. The final bill was about $1,100. (Ugh.)
Allen Bradley PLC Training Program vs. Learning by Trial
If someone is thinking about an Allen-Bradley Micro820 PLC, another comparison comes up: the Allen Bradley PLC training program versus learning by trial on a real machine. I did both, and I have an opinion.
I started self-taught, watching videos and reading PDFs. It helped, but there is a gap between understanding an instruction and knowing why a panel burns up. After the 2022 contactor failure, I enrolled in a formal training course through a Rockwell Automation authorized training partner. The Allen Bradley PLC training program didn't teach me contactor sizing; it taught me how to structure logic and, more importantly, how to ask better questions before wiring something.
That is not to say self-teaching is worthless. I still learn more from a stubborn machine than from a classroom lab. But the classroom gives you a map; the machine gives you the terrain. The Allen Bradley PLC training program focuses on software workflows, tag conventions, and debugging moves that take years to discover on your own.
If you're a new tech, I'd argue the formal training program is worth it when you know you'll be working with Allen-Bradley PLCs more than once. It's not a replacement for field experience. In the classroom, you have predictable trainer boards and no heat. In the field, you're kneeling on a roof next to a compressor with a flashlight in your mouth. That's where the real lessons happen.
According to Rockwell Automation documentation (Source: Rockwell Automation, 2024), the Micro820 family is designed for small, standalone control applications and supports Ethernet/IP. The programming software is Connected Components Workbench, and the free version is a reasonable way to start before spending money on training. But don't skip learning the difference between relay and contactor before you do.
The Little Things That Wasted My Time
I keep a list of mistakes that have nothing to do with PLC brands or ladder logic. Two of them come up a lot.
First, don't power a PLC with a battery charger. A battery charger is not a regulated 24VDC power supply. I've seen a panel reset every time a contactor pulled in because someone used a battery charger to keep the processor alive during a swap. The PLC needs a clean, regulated supply. That mistake was on a 14-panel order—every single panel had the issue. A proper 24V supply cost $85.
Second, label every wire that goes to an old GE air conditioner control panel before unplugging anything. It took me years to understand that a good photo of the wiring is worth more than a perfect wiring diagram. I now take photos from five angles before I remove any board. It seems obvious, but in the middle of a hot rooftop, it's easy to skip.
The way I see it, efficiency isn't about moving faster. It's about not moving backward.
Which One Should You Choose?
Here's the practical advice, based on my mistakes:
- If the original GE control board is in stock and the system has been working fine, direct replacement is usually the most efficient choice.
- If the board is discontinued, on long backorder, or the control sequence keeps drifting, a Micro820 retrofit can be the more reliable long-term path.
- If the load is a motor, compressor, or heater, use a contactor. If you're only switching a pilot device or sending a signal, a relay is fine.
- If you don't have someone on staff who can support a PLC, budget for the Allen Bradley PLC training program before you commit to a retrofit.
I don't have hard data on every GE board's failure rate versus PLC retrofits, but anecdotally, the Micro820 panel we installed in 2023 has had zero logic failures. The only issue was a loose terminal, and that took twenty minutes to find because the PLC program made it obvious.
Personally, I've come to believe that the best control solution is the one you can support when things fail. A direct replacement board is simple until it's discontinued. A PLC is a teaching tool, a spare-parts strategy, and a high-tech headache all at once. Choose the path that matches your skills, not the one that sounds more modern.
If you ask me, the first step is always to answer the relay vs contactor question. After that, everything else gets easier.
