How I Sourced the Right Phoenix Contact DC UPS for Our Plant (A 5-Step Checklist)

Who This Checklist Is For

If you’re an admin buyer—or an engineer who’s been handed the purchasing task—and you need a DC UPS for a control cabinet or a small automation network, this is for you. I’m an office administrator for a 150-person electronics manufacturer. I manage all the MRO (maintenance, repair, operations) ordering—roughly $180k annually across 8 vendors. In 2024, I was tasked with sourcing backup power for a new PLC line. I’d never bought a UPS before. This is the checklist I wish I'd had.

This guide covers 5 steps. It's not about the theory of UPS topologies (there are plenty of white papers for that). It's about the practical steps to get the right unit, specifically a Phoenix Contact DC UPS (like the popular 2907918), ordered and installed without a headache.

Step 1: Know Your Load (Don't Guess)

Most people start by looking at the UPS specs. Wrong move. Start with what you're protecting. I learned this the hard way when I ordered a unit based on the panel's main breaker rating. Surprise: the PLCs and HMI only drew a fraction of that.

Pull the datasheets for every device you're backing up. Add up their steady-state power consumption in Watts (W). Don't forget the Ethernet switch, the signal isolators, and any safety relays. I had a list of 12 devices. Total draw: 47W. It's a small system. Simple.

Side note: The Phoenix Contact DC UPS units (like the 2907918) are rated for 24V DC systems. That's your base voltage.

Step 2: Determine Your Buffer Time

This is where people get it wrong. They ask 'what's the biggest battery you have?' The right question is 'how long do I need to stay alive?' For us, we needed 15 minutes to safely shut down a chemical mixing process. No more.

From the outside, you'd think longer buffer is always better. The reality is bigger batteries cost way more and take up space. For a simple alarm and controlled shutdown, 10-30 minutes is usually enough. For ride-through during a generator start-up, you might need 2-5 minutes. Know your scenario.

I worked with the electrical lead to define this. We settled on 20 minutes. That gave us a target energy capacity: 47W * (20/60 hours) = ~15.7 Wh. The Phoenix Contact 2907918 (a 24V, 10A UPS) with a standard battery module handles this easily, but verifying this number is key.

Step 3: Select the Model (Not Just the Highest Number)

Phoenix Contact has a ton of DC UPS models. The 2907918 is a popular one, but it's not the only option. The model number tells you the current rating (e.g., 10A, 20A, 40A) and the functionality.

Here’s the practical decision tree:

• Load current: 47W / 24V = ~2A. A 10A unit (like 2907918) is overkill in terms of current, but that gives headroom for inrush. That's fine.
• Communication: Do you need a relay output to tell the PLC that mains power is lost? Or do you need an IO-Link interface? The basic models have a relay. The advanced ones have more interfaces. We needed a simple 'power fail' signal.
• Battery type: Lead-acid or lithium-ion? Lead-acid is cheaper and fine for occasional use. Lithium is more expensive but lasts longer and handles higher temperatures. For a cabinet that gets warm, I’d lean towards lithium. Phoenix Contact offers both.

In our case, the Phoenix Contact 2907918 (10A, 24V DC UPS) with a lead-acid battery module was the right call. It was available, priced within our budget (prices as of Q4 2024 were around $200-250 for the UPS module alone; verify current rates), and had the simple relay interface we needed.

Step 4: Configure the System (The Part Everyone Skips)

You don't just plug a DC UPS in. You have to configure it. This is the step that tripped me up.

The Phoenix Contact UPS comes with a configuration tool (software or via DIP switches). You need to set:

• Nominal output voltage: 24V (default).
• Battery type and capacity: Tell the unit what battery module you're using so it charges correctly.
• Buffer time alarm: When should the relay trip? When the battery reaches 50% charge? Or 10%? We set it to 70% to give the HMI a 'power failing' warning early.

Honestly, I'm not sure why the default settings don't always work for our application. My best guess is they assume a standard backup scenario, which isn't always the case. Spend 15 minutes with the manual or the software once the unit is powered up. It saves a call to support later.

Step 5: Install, Test, and Document (The 'Trust But Verify' Step)

Mount the UPS on a DIN rail. Connect the input (from the 24V power supply) and the output (to the load). Connect the battery. Then run the test.

The test is critical. Simulate a power failure: unplug the input to the power supply. Watch what happens. Does the PLC stay running? Does the HMI show the alarm within the expected time? How long does it actually run before the battery dies? We found our estimated 20 minutes was actually 23 minutes in the real world (good news: we had headroom).

Document the settings. Take a photo of the DIP switches or the software configuration screen. This is what the maintenance team will need in 2 years when they replace the battery. If you don't document it, you're setting them up for a call to you at 3 AM.

Common Pitfalls & Reminders

I've hit a few of these. Learn from them:

• Inrush current: A device might draw 5x its normal current for a few milliseconds when it starts. A 2A load might need a 10A UPS just for the inrush. Check the datasheet for 'making current'.
• Battery storage: Don't store the battery module for 6 months before installing it. Lead-acid batteries self-discharge. If you let it sit, it might be ruined.
• Wrong voltage: Double check. 24V DC UPS is for 24V systems. Do not use it on a 12V or 48V bus.
• Shipping damage: Open the box and check the terminals and the display unit for physical damage. I had a unit with a cracked terminal block once. It's easier to file a claim immediately.

That's it. 5 steps. From 'I don't know what I need' to 'it's installed and working'. The vendor who helped me wasn't perfect—they couldn't tell me the exact inrush specs—but they admitted it and pointed me to the manufacturer's website. That honesty earned my trust. I'd rather work with a specialist who knows their limits than a generalist who overpromises.