A 5-year recert on a 13-5/8" 10K Cameron Type-U — and an annular that came off the bench faster than it went on.

A 13-5/8" 10,000 psi Cameron Type-U double-ram BOP arrived on a low-loader at our Jebel Ali yard on a Tuesday morning, with a 5-year API 16A major-inspection clock that had run out 11 days earlier. The operator was an ADNOC Drilling rig in Abu Dhabi and the message on the pre-intake form was three words long. "Recert. Cheapest. Rig."

API STD 53 doesn't care about your spread rate. Once the 5-year window on the BOP closes, the asset cannot be put back on a wellhead without a full strip-down, dimensional and NDT scope, soft-goods replacement, hydrostatic shell test and function test — witnessed, certified, packaged. Skip a step and the well-control insurance walks. The operator knew that. The operator also knew the rig was off-hire by the day. The economic pressure said "minimum legal." The standard said "everything in the procedure, in the order written." We work for the standard.

The "minimum legal" recert and the recert that this BOP actually needed were not the same recert. We rang ADNOC Drilling at 14:20 on the third day with photos, dimensional sheets, and a single-page PDF that listed every finding alongside the standard reference that flagged it. The PO came back inside 36 hours. Nobody argues with API 16A in writing.

We took the BOP down to component level — bonnets off, ram blocks out, operating cylinders separated, locking screws backed out, every fastener tagged by position. Every metallic part went through the full API 16A Annex C NDT scope: magnetic-particle inspection (MPI) on the ferrous bodies, ultrasonic thickness on the body wall and bonnet flanges, dimensional verification against the original Cameron datum, dye-penetrant on every critical sealing surface, and Rockwell-C hardness on each forged component to confirm ≤ 22 HRC per NACE MR0175 / ISO 15156. The body and ram blocks are AISI 4130 alloy steel forging with Inconel-625 cladding on the metal-to-metal sealing lands; bonnet bolting is ASTM A193 B7M studs with 2HM nuts, all sour-service rated. Every certificate is EN 10204 3.1 minimum, 3.2 on the body and ram blocks.

Every elastomer in the BOP got replaced — no exceptions, no "looks fine." Bonnet seals, ram packers, top seals, annular packing element, piston seals, wear rings — all renewed in HNBR sour-service grade from the Cameron OEM kit, with FKM back-ups where the design calls for them. The scored hub face on the No. 2 pipe-ram block was reconditioned to the OEM datum and the seal land re-machined. The blind-shear ram with edge wear was retired and replaced from OEM stock. Reassembly was torqued in sequence with calibrated wrenches and recorded line-by-line.

Hardness control is the safety story on a sour-service BOP. Twenty-two HRC isn't a target — it's a ceiling. Every forging, every stud, every nut, individually verified. If the file isn't there, the part doesn't go back in.— Aisha Al Suwaidi · QA / NDT Inspector · Jebel Ali

A 5-year API 16A recert on a 13-5/8" 10K stack is not a job you finish faster by working harder. It's a procedure. We run every Cameron Type-U recert through the same four phases.

Below is the abridged SOP that governs a Cameron Type-U 5-year major recert in our shop. The full document runs 78 lines and four annexes. This is the subset that was checked off, dated and signed on this work order.

Below is what we measured on the BOP against API 16A and OEM Cameron tolerance. Four findings would have failed an audit if the unit had been put back on the rig as-is — every one was closed out in Phase 03.

The BOP came off the test bench at 11:40 GST on day 41 with the ADNOC Drilling rig engineer counter-signing the hydrotest chart. The FAT package — 96 pages, two paper copies, one PDF, one ZIP of NDT records and hardness logs — was issued on the morning of day 42. The unit went on a low-loader at 14:00 the same afternoon and was on the rig floor in Abu Dhabi by the next morning. The new API 16A 5-year recert is valid until February 2031. The annular closure time came off the bench 0.6 seconds faster than it went on — a small finding, but in BOP work the small numbers are the ones that matter.