Danfoss Pressure Switch Troubleshooting: Differential Settings, Electrical Faults, and the First 15 Minutes of Diagnosis

Here's the short version: if you're staring at a dead AC compressor, a dark RV outlet, or a refrigeration rack that won't run, don't start by ordering the most expensive part. In my experience, the first 15 minutes of diagnosis should be spent on the pressure switch setting and the electrical path that feeds it. A Danfoss pressure switch that's mis-set or has a corroded terminal produces exactly the same symptom as a burned-out compressor, and it's a fraction of the cost to fix. Check those two things before you condemn anything.

Why I get to be grumpy about this

I'm a quality and compliance manager at an industrial controls company. I review every technical document before it reaches customers—roughly 200 items a year. In 2024, I rejected 12% of first deliveries because the specs didn't match what the field actually needed. Over the past 7 years, I've watched the same $6 mistake turn into a $6,000 service call more times than I can count.

Here's the thing: most field failures are boring. Not 'easy' boring. It's not an exotic electrical fault. It's a pressure switch set to the wrong range, or a screw that was never torqued, or an outlet installed with push-in terminals that vibrated loose. The fix is usually simple. Diagnosing it correctly the first time is the part that's worth paying for.

Danfoss pressure switch basics: know your series before you set anything

I keep the current Danfoss pressure switch catalogue PDF on my phone. Not because I'm a fanboy—because the model numbers are not interchangeable. The catalogue covers KP, RT, MBC, KPI, and ACB series, and each one has different ranges, differentials, contact ratings, and enclosure ratings. It also lists the approval markings. If the switch lacks the marking your system spec requires, don't install it, no matter how good the price is.

For instance, a differential pressure switch is not just a switch with two ports. It reacts to the difference between two pressures, and the setting means something different than a single-pressure cut-out. You need to know which series you're dealing with before you start turning springs.

Setting a Danfoss differential pressure switch without guessing

The first step is to read the label on the side of the switch. It lists the range, the factory differential, and the electrical rating. The second step—or rather, the simultaneous requirement—is to get a calibrated pressure gauge on the system. Setting a pressure switch by the gauge on the wall display is a good way to chase ghosts.

For the Danfoss differential pressure switch setting, the key question is whether the differential is adjustable or fixed. On many KP models, you set one pressure point and the differential is fixed. On some RT models, there's an extra scale for the differential. If you turn the main spring and the differential changes, you need to set the start point and the stop point together, not as two independent numbers.

I usually do it in this order:

• Verify the actual operating pressures with a calibrated gauge.
• Set the cut-out point first, then check the cut-in point.
• Adjust the differential spring in small steps, maybe half a turn at a time.
• Cycle the system at least twice before you call it done.

That last one is the step that gets skipped when someone is in a hurry. And skipping it is how you end up with a switch that trips at the right pressure on paper but not in service.

The electrical lookalikes: RV outlets, transfer switches, and the case of the untripped breaker

Now for the part that seems unrelated until it's not. I get asked about a set of RV electrical outlets not working but breaker not tripped at least twice a year. The usual suspects, in order: a tripped GFCI upstream, a loose neutral, a backstabbed outlet that has worked loose, and a failed transfer switch.

Look: a breaker that hasn't tripped doesn't mean the circuit is sound. A GFCI can open without moving the branch breaker. A neutral wire can be loose enough to kill the circuit but still show 120 volts from hot to ground. And a push-in terminal in an outlet can fail after years of vibration without ever blowing anything up. It just stops making contact.

The same logic applies to a manual transfer switch for portable generator. If the generator is running but the RV outlets are dead, check the transfer switch's internal contacts before blaming the generator. I've seen a manual transfer switch for portable generator deliver voltage on the utility side but only one leg on the generator side after an install error. No breaker tripped. The outlets just went dead.

Why does this matter for a Danfoss pressure switch? Because the pressure switch is also a set of contacts. It starts the compressor only when the electrical path is complete from the supply, through the switch, through any safeties, and back. A corroded terminal on the switch's normally closed contact will produce the same 'compressor dead, breaker not tripped' symptom as an open motor winding. The switch is part of that electrical path, and it deserves the same electrical test as everything else.

Testing an AC compressor with a multimeter

When people finally get a multimeter out, they often go straight to the compressor. I'd rather start at the pressure switch contacts, but if you're already there, test it right. Testing an AC compressor with a multimeter means checking the winding resistance between the common, start, and run terminals, and checking each terminal to ground.

For a single-phase compressor, you're looking for three things: continuity between C-R, C-S, and R-S; a reasonable resistance that matches the datasheet; and no short to ground. If you see a low reading from any terminal to ground, the winding is compromised. But here's the counterintuitive part: a compressor can pass all those resistance tests and still be mechanically seized or have a failed start component. When that happens, the pressure switch is likely fine. The problem is downstream.

What I learned the hard way about time certainty

I went back and forth for two days on a replacement pressure switch for a refrigeration rack. The cheap option was $42 and 'should arrive by Friday.' The expedited option was $86 with guaranteed next-day delivery. The rack was down, and each hour of downtime was costing roughly $200. I chose the $86 option because the premium wasn't for speed. It was for certainty.

Everything I'd read about procurement said to compare prices. My experience with hundreds of service calls says the real cost is the probability of being wrong. When a deadline is involved—a production run, a store opening, a test cycle—the difference between 'probably on time' and 'confirmed on time' is worth every dollar of the rush fee.

Looking back, I should have specified the exact Danfoss pressure switch model and required a wiring diagram with the order. At the time, I assumed any 'equivalent' would work. It didn't. The substitute had a different differential, and I spent another day re-tuning the system before I replaced it with the actual KP model. That day cost more than the switch and the rush shipping combined.

Boundary conditions: when this advice doesn't apply

I don't want this to sound like everything is always electrical. Sometimes the compressor is just dead. If you've confirmed that the pressure switch contacts close cleanly, voltage is present at the contactor, and the multimeter tests say the windings are good, stop guessing and get a manifold gauge set on the system. The problem could be a stuck mechanical valve, lost refrigerant, or a failed start relay.

Also, don't assume a Danfoss switch for one refrigerant family works on another. The catalogue lists compatibility for a reason. And if you're working on a live electrical panel, don't. This is the part where I remind you that I'm a quality inspector, not an electrician. My job is catching problems before they reach customers, and I'd rather catch you doing something stupid.

One more thing: if the switch has been through a short cycle or a compressor lockout, replace it. Cleaning contacts with sandpaper is a temporary patch, not a repair. I've rejected more 'cleaned and reinstalled' switches than I can count. They fail at the worst possible time.