Don't Make These 5 Mistakes When Buying a Mean Well Adjustable Power Supply

What I Learned the Hard Way

I've been handling component procurement for a systems integrator for about six years now. In my second year (2018), I made a mistake that cost us about $1,200 and a two-week production delay, all because I assumed the adjustable voltage range on a power supply was just a 'nice to have' feature. I was wrong. Since then, I've made it my personal mission to document every stupid error so the new engineers don't have to repeat them.

This FAQ is for anyone staring at a Mean Well spec sheet, wondering if the adjustable power supply is worth the extra few dollars, or if the EDR-120-24 is the right pick for their enclosure. I'll walk through the questions I get asked most often — and the ones people should ask but don't.

1. What's the single most common mistake when buying a Mean Well adjustable power supply?

Assuming 'adjustable' means 'universal.' I see it all the time. An engineer specs a Mean Well LRS-350-24 because it's cheap, then realizes at the bench test that the device they're powering actually needs 26.4V to hit peak efficiency. The LRS series has a fixed output. The adjustable version (like the RSP series) would have saved them the headache — and the re-order cost. Look, if your load might drift or you need fine-tuning during commissioning, just pay the 15-20% premium for an adjustable model. It's cheaper than a re-spin.

2. Is the Mean Well EDR-120-24 good for an 'enclosed' device running 24/7?

Yes, but with a caveat that I learned the hard way in a September 2023 project. The EDR-120-24 is a fantastic DIN rail supply for industrial control panels. However, if your 'device' is in a sealed enclosure with poor airflow, you need to derate it. I once cooked three units in a test because we crammed it next to a PLC and a contactor without any ventilation. The supply itself is reliable — it can handle a wide input range and has decent hold-up time. But if ambient temps in the box regularly hit 60°C, you're not getting 120W out of it. Maybe 100W. Check the derating curve in the datasheet. That's not a Mean Well problem; that's a physics problem.

3. Can I use a Mean Well to power something sensitive, like a blood pressure cuff or test equipment?

This is the one that really bit me. In early 2022, we had a prototype for a portable diagnostic device — think blood pressure monitor with a tiny display and a pump. The engineer grabbed a cheap Mean Well RS-15-5. It worked, but the noise on the output was terrible. The ADC readings were jittery. We chased the issue for a week. The fix? Using an RPS series (with a metal case and better filtering) and adding a small LC filter at the load. For noise-sensitive gear, an adjustable power supply isn't just about voltage; it's about ripple. The Mean Well adjustable models in the RPT or RPS families are usually spec'd with lower ripple, but always check the datasheet. Consequences: That mistake cost about $450 in wasted prototype parts and a one-week schedule slip.

4. Why would anyone choose the Mean Well EDR-120-24 over a cheaper competitor?

Real talk: you can get a cheaper DIN rail supply for maybe 30% less. I've tested them. The failure rate on no-name brands in my experience is roughly double over 18 months. But the bigger issue for me is the certification. A lot of system integrators in the telecom space need UL 60950-1 certification for their final assembly. The EDR-120-24 carries it. The no-name one? 'Certification pending' — which means you're taking a risk with your own approval. For a production run of 500 units, a failed compliance audit is catastrophic. I'd rather pay the premium for the known entity. Time certainty: In April 2024, I needed a quick replacement for a malfunctioning supply in a cell site. The Mean Well distributor had it in stock, delivered in two days. The cheaper brand had a '4-6 week lead time.' That's the difference between a $400 rush bill and a $15,000 SLA penalty for downtime. The certainty is the value.

5. How do I know if I actually need an 'adjustable' feature, or if a fixed model is fine?

To be fair, I get why people stick with fixed. It's simpler and cheaper. But I have a rule of thumb now: if your load draws more than 80% of the supply's rated current, or if you're running LEDs that need a specific current limit, get the adjustable. The adjustability isn't just for tweaking voltage; it's for matching the supply exactly to the load's sweet spot. For things like powering a fan or a relay, fixed is fine. For powering a motor driver or an analog sensor? I've seen too many 'almost perfect' fixed supplies cause weird noise or instability. Having a small trim pot on the supply has bailed me out three times in the last year alone.

6. What about the 'how to turn on' a Mean Well supply? Is it hard?

Honestly, if you're struggling with this, the issue is usually the AC input connection. I see engineers trying to power an EDR-120-24 with DC input because they misread the terminals. The EDR takes universal AC input (85-264 VAC). There's no 'on' switch on most models — it powers up when mains is applied. So 'how to turn it on' is literally 'plug it in.' But if you're building a prototype and you need to test it safely, use a variac or an isolation transformer the first time. I've personally blown the fuse on a brand new unit by plugging it into a mis-wired socket. That was a $25 mistake that cost an hour of troubleshooting. Look, I'm not saying it's common, but it's a simple thing that catches people off guard.

Here's the list of things I check now before ordering:

• Is the voltage range adjustable? (Check if it's 'V. Adj.' on the label)
• Does the form factor fit? (The EDR-120-24 is DIN rail; the RSP is chassis mount)
• Thermal derating: Figure out if my enclosure is hot. If yes, oversize it.
• Certifications: Match them to the final product's target market.

That's my two cents. If you've got a specific use case — powering a weird device or fitting into a tight space — double check the datasheet. And if you're ever in doubt, hit up a distributor's tech support. It's better to ask a 'dumb' question on the phone than to waste $1,200 in rework like I did.