A 3D scan replaces an isometric that stopped matching the unit years ago with field-verified geometry of the tie-in points, pipe racks, and equipment clearances a brownfield project actually has to design against, rather than the drawing it was supposed to have.
W / 02
The problem: designing a tie-in against a drawing that's wrong
Brownfield process work runs on isometrics and P&IDs that were accurate once, before the third or fourth small modification stopped getting drawn. A flange gets replaced with a different orientation, a support gets added, a line gets rerouted around an obstruction — and the master drawing keeps saying what it always said, because nobody had the budget to redline it at the time.
That gap becomes a fabrication problem the moment a tie-in spool gets built off the old drawing instead of the actual flange. A bolt circle rotated a few degrees, a nozzle projection that's different than documented, a support that isn't where the isometric shows it — any of those is cheap to catch before fabrication and expensive to discover during a turnaround, when the fix competes with an outage schedule instead of a shop schedule.
Scanning replaces the assumption with a measurement: the tie-in points, the rack, and the equipment envelopes captured as they actually stand, on the schedule that works around your permits rather than the other way around.
What gets captured, and in what order
Process capture is driven by where new steel and new pipe are going to land, so the densest setups go where the fabrication decision actually happens.
Tie-in points first
Flanges, branches, and existing supports where new work connects to old get the densest setups, from multiple positions, so bolt circle orientation and flange face position are unambiguous.
Pipe racks and support steel
Rack levels, beam elevations, spare space in each tier, and existing shoe and guide positions — usually the actual constraint on what a retrofit can add, more than plot space is.
Equipment rows and clearances
Exchangers, pumps, columns, and vessels with their nozzle positions and platform steel, plus the maintenance envelopes around them — bundle-pull space, davit swing, and flange-breaking room.
Access, scaffold, and laydown routes
Ladders, platforms, and the ground space a scaffold or crane will occupy, captured well ahead of a turnaround so scaffold design and rigging plans can start early.
From cloud to something a fabricator can use
Process teams rarely need a whole unit modeled. Most projects want a small volume modeled to fabrication confidence and a large volume left as measurable cloud around it.
Dense cloud plus a modeled tie-in zone
Flange face position and orientation modeled at the connection point, with the surrounding unit left as measurable point cloud rather than fully modeled.
Point cloud with modeled rack steel
Existing rack members and lines modeled as solids so a new run can be clash-checked against them before routing is finalized.
Registered cloud with a coverage report
Documented control, registration residuals, and a statement of what was and was not visible, so planners know what the data can be trusted to answer.
Aerial site-scale data
Drone mapping and orthomosaics for plot, laydown, and access questions that are answered at site scale rather than flange scale.
What process and reliability engineers ask us
01Can you scan inside a classified area?
That's a site decision, not ours to assert. Area classification, your equipment rules, and the permit system determine what may be energized where, so classified zones get scoped with your safety group before mobilization. Where an instrument cannot go in, capturing from the boundary or during a de-classified window are the usual alternatives.
02How accurate does a tie-in measurement need to be?
Achievable accuracy depends on range, incidence angle, surface, and registration quality on your specific site, which is why tie-ins get their own setup plan and we agree the tolerance the fabrication actually needs up front rather than quoting a number in the abstract.
03Will the point cloud open in our plant design software?
Registered clouds are delivered in the exchange formats plant design and review tools read, with the coordinate system agreed before capture so the data lands where your existing model expects it. Which format is right depends on what your engineering contractor opens.
04Is it worth scanning before a turnaround, or only during it?
Before, almost always, where the unit is running and the geometry is accessible. That lets tie-ins be measured and spools fabricated ahead of the outage, so the shutdown is spent installing rather than discovering that a dimension moved. Anything genuinely only visible once the unit is down gets scheduled into the outage as its own short window.
Your next decision
Make the invisible measurable.
Tell us what is inside, outside, and on the line. We will help you scope the capture and the handoff — in New York, New Jersey, or Connecticut.