If I remember correctly, I've compared bids for data center cooling equipment maybe eight or nine times over the past six years. Around $180,000 in cumulative spending across those projects is my best guess—though I might be misremembering one of the smaller ones. The pattern is always the same: someone presents a lower upfront quote, and unless you dig into the TCO spreadsheet, you end up paying more in the long run.
This isn't about picking a winner for the sake of it. It's about understanding why two solutions that look similar on paper—Munters' evaporative and desiccant systems vs. traditional compressor-based cooling—have wildly different cost profiles over a 5-year period. The comparison framework here is simple: we'll look at capital expenditure, operational energy cost, maintenance burden, and climate resilience. Each of these dimensions reveals a hidden layer of cost that a standard quote won't show you.
From the outside, the upfront price difference seems straightforward. A traditional chiller-based system for a medium-sized data center hall—say, 500 kW of cooling capacity—might quote around $200,000 in equipment costs. A comparable Munters system combining indirect evaporative cooling with a desiccant dehumidifier? Probably closer to $250,000. The knee-jerk reaction is to go with the cheaper option.
People assume the higher quote means the vendor is less efficient or trying to pad margins. What they don't see is what's not included in the chiller quote. When I audited our 2023 spending on a similar project, I found that the 'lowest' bid excluded:
The Munters quote had line items for everything. I should add that their bid documentation was noticeably more detailed—each component's installation cost and commissioning time were broken out. That transparency, while making the total seem higher upfront, meant I could forecast the actual spend within about 5%. The chiller vendor's 'surprise' costs came out to about $8,500 in the first year alone. (Should mention: that doesn't include the two emergency service calls we had during a heatwave in Q3, but that's for another dimension.)
This is the dimension where the comparison flips completely, and in a way that's counterintuitive if you've only looked at sticker prices. The most frustrating part of managing cooling budgets: energy consumption is the largest line item, yet it's the hardest to verify from a quote. You'd think efficiency specs like kW/ton or COP would be standardized, but real-world performance varies with ambient conditions.
Traditional compressor-based systems advertise a COP of 3.0 to 4.0 under ideal conditions. In practice, my experience tracking quarterly orders across three data center sites shows that actual COP drops to around 2.5 during summer months when you need cooling the most. The compressors work harder as ambient temperature rises, electricity consumption spikes, and the chiller's efficiency curve is at its worst when demand is highest. Our facility in Phoenix saw a 35% increase in kWh per ton in July compared to March.
Munters' indirect evaporative cooling, by contrast, gains efficiency as outside air gets hotter, up to a point. The system pre-cools ambient air using evaporation without adding moisture to the supply air—that's the 'indirect' part. At 95°F outside, it's still delivering supply air at around 75-80°F using just fans and water spray, no compressor. When the air is too humid for evaporative cooling alone, the desiccant dehumidifier handles latent load without needing deep cooling first. The net effect: our energy consumption for a similar cooling load was roughly 40% lower in the Munters-equipped test pod during Q2 2024, compared to the chiller-based pod next door.
I'm not 100% sure the savings will hold consistently across all climates, but based on our 18-month data from two locations—one humid, one dry—the pattern is pretty strong. The vendor who lists all operational costs upfront, even if the total system price looks higher, usually costs less in the end. After tracking six orders over two years, I found that over half of our 'budget overruns' came from energy—not unexpected equipment failures.
When comparing quotes for a $4,200 annual maintenance contract on a traditional system, it's easy to assume that's the baseline. What I've learned, after getting burned on hidden fees twice, is to ask what's NOT covered. Standard chiller maintenance contracts typically exclude compressor overhaul, refrigerant replenishment (which is getting more expensive as regulations tighten—take this with a grain of salt, but refrigerant costs have roughly doubled since 2020), and emergency call-outs outside business hours.
For the Munters system, the maintenance profile is different. There are fewer high-pressure components to fail; the main moving parts are fans and desiccant rotors. The water side does require scale management and periodic cleaning of the evaporative media, but that's a predictable cost we calculated at about $600 annually based on our water quality. The desiccant wheel itself has a service life of around 8-10 years before replacement, versus a compressor that may need major service at the 7-year mark. A single compressor replacement on a 200-ton chiller runs around $12,000 to $15,000. That's not hidden—it's just not on the quote. In Q2 2024, when we switched vendors for one of our maintenance contracts, the new vendor pointed out three compressors that were drawing slightly higher amperage than spec. We weren't in immediate failure territory, but it confirmed that compressor wear is a slow, expensive problem you don't see until you're paying for it.
Roughly speaking, I'd estimate the 5-year maintenance TCO for a traditional system at 30-40% higher than the Muntres alternative, primarily because of compressor-related risks and refrigerant handling. The 'cheap' option resulted in a nearly $6,000 redo on one site when a poorly maintained compressor failed during a heatwave and took out a power supply module.
Here's the dimension that most comparison charts skip entirely, and why I'm somewhat skeptical of any cooling analysis that doesn't include local weather data. Traditional compressor-based systems are at their worst when conditions are most extreme: high ambient temperatures and high humidity. That's exactly when a data center needs full cooling capacity. We had an incident in July 2023 where the chiller staged down because the condenser couldn't reject heat into 108°F outside air. We were running at 70% capacity on a 100°F day. Not a failure, but not ideal.
Munters' hybrid approach handles this split more gracefully. The desiccant dehumidifier pre-treats air to reduce latent load, and the indirect evaporative cooler handles sensible cooling without taxing a compressor. In humid climates, the desiccant rotor spins slower and regenerates using less heat than a competing active desiccant system we evaluated—Per FTC guidelines, I should say that claim would need substantiation, but our internal tests on one unit showed 18% lower regeneration energy. That's not a guarantee for every installation, but it's consistent with what Munters publishes.
The bottom line on resilience: if your data center operates in a climate with high summer humidity—which covers the eastern two-thirds of the US, per ASHRAE climate data as of 2022—a system that separates latent and sensible cooling (like Munters' approach) will maintain a more stable supply air condition during peak loads. That's not a 'compressor is obsolete' claim. It's a statement about which approach is more likely to avoid capacity staging when you need it most.
Based on what I've seen across multiple projects and 8 vendor comparisons over 3 months using our TCO spreadsheet, here's my fairly practical guide:
Consider Munters (evaporative + desiccant) if:
Stick with traditional compressor cooling if:
Don't hold me to this, but I'd estimate that for about 60-70% of US data center locations (using ASHRAE climate zones as a rough guide), a Munters combined approach will show a lower 5-year TCO, even with a 15-20% higher upfront equipment cost. The gap I've seen in our own tracking: saving around $8,400 annually on a 300 kW cooling load after switching one pod, which was about 17% of our cooling budget for that facility.
Our procurement policy now requires quotes from 3 vendors minimum for any cooling project, and we built a simple cost calculator that factors in local weather data and energy rates. Because I've learned that the number on the first page of a quote isn't the number you'll pay. The vendor who lists all charges upfront, even if the total looks higher? That's usually the person who understands what the actual cost of cooling is.