Danfoss Pressure Switch Checklist: 9 Steps After $2,400 in Avoidable Mistakes

This one's for maintenance techs and plant operators swapping or setting a Danfoss pressure switch — KP series, RT, MBC 5100, whatever's on the shelf. Nine steps, roughly 45 to 90 minutes depending on whether you're doing a like-for-like swap or a first install. I built it after two callbacks cost my team about $2,400 in parts, labor, and a shutdown that wasn't on the schedule.

Seven years handling industrial controls and field installs. Four significant mistakes I've personally made and documented on pressure switch jobs, totaling roughly $2,400 in wasted budget. The checklist exists so you don't repeat them (note to self: update it after the next one, because there's always a next one).

When to use this checklist

Use it when you're:

• Replacing a failed or drifting pressure switch on a compressor, chiller, or air system.
• Commissioning a new install and setting cut-out/cut-in points for the first time.
• Tracking down a pressure fault that keeps coming back after a switch swap.

Skip it for bench testing — different rules apply there.

The 9-step checklist

Step 1 — Match the switch to the duty, not to shelf stock

Pull the datasheet for the exact part number before you open the box. The Danfoss publications (danfoss.com) list setpoint range, differential, reset type, and media compatibility per model — and the Danfoss pressure switch KP36 and the Danfoss MBC 5100 pressure switch are different animals even when the pressure ranges look close on paper. A low-pressure switch and a differential switch are not interchangeable, and neither is an auto-reset where the application calls for manual reset.

I keep a one-page matrix taped inside the panel door: model, range, differential, reset, media. It's saved me from grabbing the wrong spare at least a dozen times.

Step 2 — Check the system side before you blame the switch

This is the step most techs skip, and it's the one that turns a 30-minute swap into a four-hour callback. Three things to verify first:

Air filter direction and loading. If the switch is a differential type across a filter bank — common on furnace and AHU applications — check the filter's airflow arrow. A filter installed backwards restricts flow sooner, and the differential reading drifts off the setpoint. That's the classic which way to put the air filter in the furnace, up or down question, and it's not cosmetic. The arrow points with the airflow, always. While you're there, note the filter size (a standard 10x20x1, for example) and log the last change date. A clogged 10x20x1 filter mimics a failing switch almost perfectly.

Oil filter condition, on compressor systems. Oil differential pressure switches trip on the pressure drop across the oil filter, not on the switch itself. Run the oil filter lookup through the manufacturer's tool — Mobil 1 publishes one, and so do most filter brands — and confirm the filter on the machine is the right part. A restricted filter will trip a perfectly healthy switch, and I know techs who've replaced the same switch three times before checking the filter. (I was one of them. 2021. Not proud.)

Actual system pressure at the port. A gauge reading at the service port tells you whether the switch is misreading or the system is genuinely out of range. Those are different problems with different fixes.

I wish I'd tracked callback rates more carefully in the early years, because I suspect Step 2 alone would have cut them by half. No hard data — just a strong hunch from the jobs that went sideways.

Step 3 — Lock out and bleed down

Under OSHA 29 CFR 1910.147, servicing a pressurized control requires lockout/tagout before you break the connection (osha.gov). The federal rule is the floor, not the ceiling — follow your site's specific LOTO procedure, and bleed the line down to zero before you crack the fitting. Pressurized refrigerant or air through a loosened flare nut is a bad way to end a shift.

Step 4 — Inspect the port and the seal

Flare seat, O-ring, thread condition. A nicked flare seat will leak no matter how carefully you torque the new switch, and you'll spend an hour chasing a faulty switch that was actually a bad seal. Replace the O-ring as a matter of course — they're cheap and the callback isn't.

Step 5 — Wire it and re-check terminal torque

Match the wiring diagram on the switch body to the panel drawing. Normally-open vs. normally-closed contacts are the classic mix-up, and it only shows up when the system tries to start. Torque terminals to the datasheet spec — loose lugs arc, arc welds, and then the switch gets blamed for a wiring problem. I still kick myself for the 2020 job where a loose common terminal looked exactly like a failed switch on the meter.

Step 6 — Set the cut-out first, then the differential

Set the upper (cut-out) setpoint before touching the differential screw. Most Danfoss switches adjust cut-out with the main screw and differential with a smaller one — turning them in the wrong order means you'll chase the setting twice. Make small adjustments, let the system stabilize, re-read the gauge. Guessing at quarter-turns is how setpoints end up 15% off.

Step 7 — Test the reset and the manual trip

Force the switch past its setpoint and confirm it trips and resets as specified. If it's a manual-reset model, trip the reset button by hand and verify it latches. An auto-reset switch on a manual-reset application is a safety problem, not a convenience problem — the machine will restart itself after a fault.

Step 8 — Leak check under pressure

Bring the system back up slowly and check the port with soap solution or an electronic detector. Do this before you put the panels back on. Finding a leak with the panel open takes five minutes; finding it after reassembly takes an hour and a lot of swearing.

Step 9 — Tag, log, and hand off

Write the model, setpoints, differential, and date on a tag inside the panel. Log the install in your maintenance system with the part number and the filter change dates from Step 2. Future-you (or the next tech) will need that information at 2 a.m. on a Sunday, and it won't be in the drawing package.

Where the money actually goes

Here's the part nobody puts in the quote. The switch itself is maybe 15% of the real cost. The rest is labor, downtime, callbacks, and the parts you replace on a guess because Step 2 got skipped. A $180 switch that fails in six months costs more than a $260 switch that lasts five years — and that's before you count the lost production hours.

To be fair, budgets are real, and the cheap switch has its place on non-critical duty. But on anything that trips a line or a compressor, run the total cost math: parts + labor + expected callback rate + downtime per hour. The numbers usually point one direction.

I don't have hard data on industry-wide failure rates by price tier, and I won't pretend otherwise. My anecdotal read after seven years: failures cluster around installation errors more than part quality, which is exactly why this checklist exists.

Common mistakes to avoid

• Replacing the switch before checking the filter. Both air and oil filters mimic switch failures. Check the filter first.
• Ignoring air filter direction. Arrow points with airflow. Backwards filter, wrong differential reading.
• Setting differential before cut-out. You'll chase the setpoint twice.
• Auto-reset where manual-reset is required. Safety issue, not a preference.
• Skipping the leak check until after reassembly. Five minutes vs. an hour.
• No tag, no log. The next tech starts from zero.

Part numbers and prices shift; verify current datasheets and local regulations before you order or install.

Omar Rahman

Omar Rahman is a power-conversion and quality analyst specializing in UPS systems, inverters, rectifiers, battery chargers, power supplies, and bypass arrangements. He applies IEC 62040-3 tests to UPS efficiency, output tolerance, harmonic distortion, overload behavior, transfer time, and autonomy, while using IEC 62477-1 to examine safety boundaries for other power-electronic converters. He helps facility and renewable-energy teams match conversion equipment to critical load profiles, runtime targets, redundancy, thermal conditions, maintenance strategy, fault behavior, and total ownership cost.