-
Why "Best Pressure Switch" Is the Wrong Question
-
Scenario A: Single-Setpoint Alarm on Water, Air, or a Chilled System
-
Scenario B: Differential Pressure—Filters, Pumps, Oil Circuits
-
Scenario C: Refrigeration and Compressor Protection with Manual Reset
-
Scenario D: Panel Builders and MRO Stock
-
How to Tell Which Scenario You're In
-
What I'd Do Differently
Every month or so, someone forwards me a requisition and asks me to sign off on "the right Danfoss pressure switch." I've been reviewing these on the quality side for six years now—spec sheets, sample batches, first-article inspections—and I still can't answer that question without two or three rounds of clarification.
That's not me being difficult. Kp1 and KP35 switches look almost identical on a DIN rail, land within 20% of each other on price, and are absolutely not interchangeable in the applications they're designed for. Specify the wrong one and it won't show up at commissioning—it'll show up six months later as nuisance trips, or worse, a compressor that ran dry.
The framework I actually use splits the decision into four scenarios. Figure out which one you're in before you open a catalogue.
Why "Best Pressure Switch" Is the Wrong Question
People assume the pricier switch is the more reliable one. In my experience the causation runs the other way—models with a long field record can carry a premium because nobody wants to be the engineer who specified the cheap one. Price follows reliability, not the reverse.
What actually drives selection is three things: what are you switching, what should happen when it trips, and who's going to adjust it in service. Those three answers put you in one of the following buckets. I don't have hard numbers on how often misapplication causes field failures—nobody publishes that, and the vendors certainly don't—but anecdotally, in our own rejected batches, application errors outnumbered manufacturing defects by a wide margin.
Scenario A: Single-Setpoint Alarm on Water, Air, or a Chilled System
This is the simplest possible case. You need a switch that says "pressure dropped below X, sound the alarm or cut the pump." One setpoint, fixed or lightly adjustable differential, SPDT contact out.
KP1 and its low-pressure siblings are built for exactly this. Below-atmospheric protection, suction-side monitoring on refrigeration, pump cutout on a water line—the range runs from roughly −0.2 bar up to about 7.5 bar depending on model, with a fixed differential that you should not assume matches last year's datasheet. Check the revision letter on the PDF before you commit. (I've been burned by a revision change that narrowed a tolerance by 0.3 bar, which was enough to matter on a low-pressure cutoff.)
The KP35 covers similar ground for air and water where you need an adjustable range rather than a factory-set window. If your alarm point might move during commissioning, take the adjustable one. If it won't, the fixed version removes a knob that someone will eventually turn.
When this scenario doesn't apply: if the switch has to distinguish between two pressures rather than detect an absolute level, you're in Scenario B.
Scenario B: Differential Pressure—Filters, Pumps, Oil Circuits
Differential switches are where I see the most misapplication, partly because the difference between a low-pressure switch and a differential switch isn't obvious from the part number alone.
A differential model reads P1 against P2 and acts on the gap. That's what you want for filter-blockage alarms, pump flow confirmation, or oil pressure on a compressor. The KP15 is the one that comes up most often on the oil side—it monitors the pressure difference between oil pump discharge and crankcase, and it's almost always paired with a time delay. Skip the delay and you'll get a trip on every startup while oil pressure is still building. I learned that one from a rejection report, not a datasheet.
Two things I insist on in the purchase spec for this scenario:
- State the differential range, not just the setpoint. Half the disputes I've mediated came down to the vendor quoting a setpoint range when the application needed a differential range.
- Specify reset behavior in writing. Auto-reset versus manual reset is a safety decision, not a convenience decision, and it must be in the contract language.
Scenario C: Refrigeration and Compressor Protection with Manual Reset
Here the requirement isn't just "trip," it's "trip and stay tripped until a human looks at it." Refrigeration high-pressure and oil-safety circuits, compressor protection, industrial cooling loops—these want the RT-family behavior: robust enclosure, positive manual reset, contacts rated for the inductive load you're actually switching.
The counterintuitive part, and the one I'd push back on if a junior engineer proposed it: manual reset is sometimes the wrong call. If the switch protects a non-critical process in an unmanned plant, a manual reset means a three-hour drive to press a button. Auto-reset with a lockout relay upstream gets you the same safety outcome with a fraction of the downtime. I've specified both ways, and I'd rather have that conversation at design review than at 2 a.m.
Scenario D: Panel Builders and MRO Stock
If you're stocking for a maintenance department or building panels that go out to multiple sites, the decision is less about any single application and more about coverage. You want the fewest part numbers that handle the most range, with documentation you can hand to a technician who's never seen the part.
This is where the catalogue PDF earns its keep, and where I'd tell you to read it differently than most people do. Don't start at the product photos. Start at the range tables in the back—the columns for adjustable range, differential, and maximum working pressure. Those three columns tell you whether a series can replace four SKUs on your shelf. The MBC 5100 compact range is a good example of a series that covers a lot of panel-level work with a short list of variants, but I can't tell you how many of your existing part numbers it replaces without seeing your shelf.
A note on documentation: the official Danfoss catalogue and technical brochures are the numbers you should quote in a purchase specification, not a distributor's summary sheet. Distributor summaries are usually accurate and occasionally a revision behind. When we write a spec that references a pressure range, we cite the brochure revision letter.
How to Tell Which Scenario You're In
Answer these in order. Stop at the first "yes."
- Does the switch need to compare two pressures? If yes → Scenario B. Everything else follows from there.
- Is a human guaranteed to be on site after a trip? If no, reconsider manual reset before you go to Scenario C.
- Does one specific value trigger the action, on one specific medium? If yes → Scenario A.
- Are you buying for more than one application, or for someone else to install? If yes → Scenario D.
If two of those feel true, you're in Scenario D by default. Multi-application coverage usually beats a perfect single-application match when you're stocking.
What I'd Do Differently
Looking back at our 2022 procurement cycle, I should have written differential range and reset behavior into the master purchase agreement from day one. At the time it seemed like over-specification for a component that costs less than a hundred euros—and the vendors pushed back on it. I didn't push harder. We spent the next eighteen months resolving ambiguity case by case, which cost far more in engineering time than the spec work would have.
If you take one thing from this: the switch itself is rarely the problem. The specification is.
One quick clarification, since search traffic lands in odd places: if you arrived here looking for how to get an oil filter off without a tool, an Ezgo oil filter, or a Luber-Finer LAF1899 heavy-duty air filter, this page isn't about those—filtration parts and pressure switches share maintenance vocabulary but nothing else. Oil pressure monitoring is Scenario B above; the filter itself is a different purchase entirely.


