Who this checklist is for

If you're the person who has to sign off on a control valve purchase before it ships, this is for you. That means procurement engineers, QA reviewers, and maintenance leads who source flow control valves, pressure control valves, hydraulic control valves, pneumatic control valves, or directional valves—and then get blamed when something leaks or sticks three weeks into service.

I've been the quality gate on industrial valve purchases for about six years now. I review roughly 200 unique valve specs and submittals a year, and in 2024 I pushed back on around 15% of first-article deliveries. Not because vendors are dishonest—most aren't—but because specs get loose and assumptions get made on both sides.

Here's the thing: there are 7 steps I run through before I sign anything. Do them in order. Skipping around usually means you miss what step 4 was supposed to catch.

Step 1: Lock down the service conditions first

Before you look at a single valve, write down what the valve actually has to do:

  • Media (clean water, steam, slurry, gas with entrainment—be specific)
  • Normal and maximum flow rates
  • Inlet and outlet pressure, including the worst-case differential
  • Temperature range, not just the setpoint
  • Line size and any straight-run pipe constraints

I can't count how many disagreements trace back to a vague flow spec. If the vendor doesn't have these numbers and neither do you, stop here. Anything downstream is guesswork.

Step 2: Match the valve type to the actual job

This is where "control valve sourcing" gets sloppy. People categorize by preference instead of application. Quick mapping:

  • Flow control valve — for regulating rate, not just opening and closing
  • Pressure control valve — for holding downstream (or upstream) pressure steady
  • Hydraulic control valve — high-pressure fluid power circuits
  • Pneumatic control valve — air or gas actuated, common in process control loops
  • Directional valve — routing flow path, not modulating it

If a rep is pushing one type for everything, that's a flag. A globe valve and a ball valve are not interchangeable just because both "control flow."

Step 3: Check Cv sizing and rangeability—not just the max

Everyone checks max Cv. Almost nobody checks the low end.

The question isn't "can it handle peak flow?" It's "does it still control at 15% of peak?" If the valve is oversized, you get hunting, poor resolution, and premature trim wear. If it's undersized, you never reach setpoint.

Ask the vendor for the Cv at your normal operating point, not just the catalogue maximum. And ask for the rangeability figure (typically 30:1 to 50:1 on a good globe valve). If they dodge the question, that's your answer.

When I first started reviewing valve submittals, I assumed "meets spec" meant the vendor had done the sizing math. Two years and one badly oversized pressure control valve later, I learned that sizing is the buyer's responsibility to verify—not the vendor's to volunteer.

Step 4: Verify the leakage class in writing

This is the step most people skip, and it's usually the one that bites.

Control valve seat leakage is classified under ANSI/FCI 70-2 (also referenced as IEC 60534-4). Class I through Class VI, with Class VI being the tightest for metal seats, and Class VI is used for soft-seated valves. "Low leakage" is not a class. "Tight shutoff" is not a class.

Get the required class in the purchase order. Get the vendor to state the class on the datasheet. Then verify against the test report (see Step 6). Don't accept "shutoff" as a description—it means whatever the vendor wants it to mean on a given day.

Step 5: Confirm body and trim materials against the media

Two things to check here, and they're often conflated:

  1. Body material — usually cast steel, stainless, or ductile iron depending on pressure and corrosion
  2. Trim material — plug, seat, and stem, which see the erosion and the temperature

A stainless body with a soft trim seat is fine for clean water. Put the same valve on steam or a slurry and the trim goes first. Match the trim to whatever is going to wear the valve out—not to what's cheapest.

Also confirm the gasket and packing materials. Those are the parts that leak after the first thermal cycle, and they're usually not on the datasheet front page.

Step 6: Demand the test documentation

Anyone can print "tested" on a datasheet. What you want is the actual test record, referencing a real standard:

  • API 598 for shell and seat pressure testing
  • ISO 5208 for industrial valve pressure testing
  • ANSI/FCI 70-2 for seat leakage class verification
  • Material test reports (MTRs) with heat numbers traceable to the mill cert

If they can't produce an MTR with a heat number, that's a documentation problem, not a valve problem. But it tells you a lot about how the shop is run.

Step 7: Check actuation, accessories, and fail-safe behavior

Last step, and easy to overlook because it's "just the actuator."

Ask what happens when the air supply fails. Fail-open, fail-closed, or fail-last? That's not a preference—it's a safety decision, and it should be on the spec sheet. The same goes for the positioner, limit switches, and any solenoid accessories.

Confirm the actuator sizing matches your actual supply pressure. A pneumatic control valve that's rated for 80 psi but installed on a 60 psi header won't hit full stroke. Nobody notices until commissioning day.

Notes, and the mistakes I see most

This approach has worked for us, but our situation is a mid-size process operation with relatively stable service conditions. If you're dealing with severe service (high cycling, abrasive slurries, extreme temperature swings), the calculus shifts—you probably want a specialist vendor, not a general supplier, and some of these checks get more involved.

The three mistakes I see most often:

  1. Sizing to the maximum and ignoring the normal operating point. Oversized valves do a poor job of controlling.
  2. Assuming leakage class is implied. It isn't. If it's not in the PO, it's not enforceable.
  3. Not matching the trim to the media. The body survives; the trim doesn't.

So glad I started requiring the test records before approval. I nearly signed off on a batch of pressure control valves based on the datasheet alone—was maybe one email away from installing trim that wouldn't have lasted six months. The MTRs came back with a heat number that didn't match the mill cert on file.

One more thing—ask the vendor for references in a similar application. Not general references ("we sell to a lot of plants"), but actual similar service. It's the fastest way to find out whether the valve you're about to buy survives in conditions like yours.

Pressure test standards and leakage classifications referenced above (API 598, ISO 5208, ANSI/FCI 70-2) are for general guidance. Verify current revisions with the relevant standards body before finalizing any purchase specification.