Why Your Next Rink Renovation Should Start with As-Built Drawings, Not Trends

Every few months, I get a call from a rink manager who just spent $12,000 on a design for a new chiller plant. The plans look great. The renderings pop. Then the contractor breaks ground and hits a 10-inch concrete-encased refrigerant loop that nobody knew was there. The drawings were flawed.

In practice, the process breaks when speed wins over documentation: however small the revision looks, the pitfall is that the next person inherits an invisible assumption, and the fix takes longer than the original task would have.

That call is avoidable. Ice rinks accumulate undocumented changes like barnacles on a hull. A pipe moved three feet in 1998. A sump pit added in 2005. A dehumidifier relocated in 2012. Nobody updated the master set. Meanwhile, every renovation starts with a fresh CAD file that assumes the building is exactly what was built in 1972. It never is.

Start with the baseline checklist, not the shiny shortcut.

The Real overhead of Skipping As-Builts

According to industry interview notes, the gap is rarely tools — it is inconsistent handoffs between steps.

Nothing is more expensive than cheap information

I walked a municipal rink job three years ago — ice plant swap, new dasher boards, upgraded lighting. The owner had a binder of drawings from the original build, 1987. Looked official. Felt complete. The general contractor broke ground on a Tuesday. By Thursday they had hit a 6-inch chilled-water return pipe that wasn't on any sheet — rerouted sometime in the '90s to dodge a roof drain. That pipe wasn't dead. It fed the resurfacer room. The fix: tear out a fresh concrete slab, re-route the line, pour again, lose five days. That was a $140,000 mistake baked into a $1.8 million renovation. The irony? The old facility manager knew about the reroute — he just retired before bothering to mark it up.

In practice, the process breaks when speed wins over documentation: however small the revision looks, the pitfall is that the next person inherits an invisible assumption, and the fix takes longer than the original task would have.

How undocumented changes compound over decades

A single undocumented site shift — a feed line shifted six inches, an electrical panel relocated to clear a new chiller — costs almost nothing in year one. The problem is multiplication. A rink that opens in 1995 and gets three leasehold improvements, two equipment swaps, and a partial roof replacement by 2025 might have five separate layers of unrecorded modifications. None of them live on the same drawing set. Some exist only as pencil marks on a wall, or as memory in a foreman who left the trade. That hurts when you try to retrofit an energy recovery unit and discover the structural steel beams supporting the roof were notched during a 2003 dehumidifier install. The notch isn't on paper. The new equipment won't fit. You spend two weeks and $18,000 on condition surveys just to figure out what you're working with. The odd part is — most owners treat as-builts as a close-out document, something the contractor hands over after construction. faulty order. You need them before you price the job, not after.

The difference between record drawings and as-builts

One distinction kills budgets quietly: record drawings are the architect's final design — clean, idealised, often missing the three site adjustments the electrician made because the conduit run conflicted with a structural column. As-builts capture what was actually installed, including that offset. In rinks, where refrigerant lines, brine loops, and condensate drains share tight ceiling chases, the difference between record and reality is usually a conflict someone had to solve on the fly. That solution rarely gets drawn. I have seen a $12,000 budget line for "miscellaneous structural coordination" balloon to $89,000 because the as-builts showed a beam that didn't exist — it had been removed during a 2014 cooling tower replacement and nobody logged it. The catch is that most owners pay for record drawings at project close-out and assume they own as-builts. They don't. What they own is a set of intentions.

'We thought the drawings were accurate. We were faulty. That single pipe overhead us more than the entire as-built survey would have.'

— Facilities director, 4-pad municipal complex (speaking after a $140k bench conflict)

What Most Owners Think They Know

Misconception: 'Our CAD files are accurate'

The digital file looks pristine on screen. Layers are labeled. Dimensions snap neatly into place. That comfort is expensive. I have walked into rinks where the operations director swore by the BIM model—only to find the actual chiller trench runs two feet west of where the file shows it. The CAD files were based on the original design intent, not what subcontractors actually built during a late-night change order. That disconnect turns a simple duct reroute into an on-site demolition surprise.

The odd part is—most facility owners never verify their own digital records. They trust the file creation date. They trust the architect's stamp. But a rink that underwent three emergency repairs, two compressor swaps, and a hasty dehumidifier relocation over five years? That CAD model is a museum piece. It memorializes what someone planned, not what exists under your concrete slab.

Misconception: 'The original builder drew it correctly'

Builders do not draw for posterity. They draw to pour concrete, hang steel, and pass inspection. As-built redlines during construction are notoriously sloppy—scribbled on soggy prints, missing offsets, rounded to the nearest inch. "Close enough" looks heroic during a Friday pour. It unravels during a Tuesday renovation.

We found a header pipe that the original crew swore was twelve inches clear of the footings. It was touching rebar. That find alone saved us a week of excavation.

— Refrigeration consultant, Midwest retrofit project

The catch is that builders rarely face consequences for inaccurate site markup. They get paid. The inspector leaves. The slab goes down. Twenty years later, you are the one paying the exploratory core-drill bill. The original crew's pencil sketch? Long gone. What remains is the confident assertion that "it was drawn right."

Why rinks change more than offices

Office buildings accumulate new furniture and fresh paint. Rinks accumulate heat loads, brine leaks, dehumidifier retrofits, and zone reconfigurations. That mechanical room you think handles Zone A? It was re-piped two winters ago when the original chiller died mid-season. No one documented it because the team was fighting an ice-quality crisis. The electrical panel schedule? Obsolete the day the second-hand compressor arrived with different amp draw.

Most teams skip this: rinks undergo functional change every three to five years—refrigeration overhauls, dasher board swaps, humidity control upgrades—but documentation updates happen exactly zero times unless a renovation forces them. That gap is where budgets bleed. You plan new piping to mate with existing flanges, but those flanges were swapped for an oddball metric fitting nobody logged. Wrong order. That hurts.

The pattern is predictable: confidence in old drawings, shock when the tape measure disagrees, then a frantic scramble to source adapters or cut new openings. We fixed this by insisting on a laser scan before we touched a single demolition permit. Not because we doubted the builder—because we doubted the decade of undocumented survival decisions that followed him.

Patterns That Save Money and Time

According to internal training notes, beginners fail when they optimize for shortcuts before they fix the baseline.

site-verify everything before design starts

I watched a crew unroll brand-new dasher-board sections last spring. The wall they were bolting into was eighteen inches off from the original permit set. That mismatch overhead two weeks and a custom steel order. The fix is brutally simple: put a tape measure on every existing dimension before a single CAD line gets drawn. Most teams skip this because it feels like busywork — they already have drawings, right? Wrong. Old rinks settle. Concrete slabs creep. Pipe chases get bench-modified during the last emergency repair and nobody updates the record set. The pattern that saves money is a single, uninterrupted day of measurement before design starts.

Do it yourself or hire a junior architect. Either way, verify column locations, ceiling heights, and the exact position of every refrigeration stub-up. I have seen a renovation budget blow by forty thousand dollars because a beam that was supposed to be at fourteen feet turned out to be twelve-nine. That is not a design problem — it is an as-built problem. And it is entirely avoidable.

Start with refrigeration and structural loads

The second pattern is about priority order. Most owners want to talk about finishes first — locker-room tile, lobby flooring, spectator seating. Those are the fun parts. The catch is that rink refrigeration systems are unforgiving. If you change the slab profile or shift a header location, the structural steel above it may no longer work. The smart sequence: measure the refrigeration loop, model the structural loads, then design everything else around those constraints. Not the other way around.

One arena I consulted for tried to relocate its chiller plant after the new bleachers were already specified. The steel columns supporting those bleachers sat directly over the proposed pipe trench. Relocating the plant added three months and a structural re-engineering fee. That team now field-verifies mechanical and structural systems before they discuss paint colors. A simple shift in process, but it saved their next project roughly six figures in change orders.

Use 3D scanning for complex pipe chases

Tape measures work fine for open spaces. They fail in mechanical rooms. Rink pipe chases are notorious — refrigeration lines, brine loops, hot-water returns, condensate drains, electrical conduits, all jammed into a corridor barely wide enough for a maintenance worker. The old way is to send a junior staffer with a laser disto and a clipboard. They miss twenty percent of the penetrations. Then the sheet metal contractor shows up and discovers a six-inch brine line running exactly where the new duct was supposed to go. That is a stop-work situation.

The pattern that fixes this is targeted 3D scanning. Not a full-building scan — that is overkill and expensive. Just the chases, the ceiling plenums, and the slab edges. A half-day scan, maybe two thousand dollars, produces a point cloud that eliminates the guesswork. The ROI is immediate: zero field-fabricated ducts, zero emergency pipe relocations, zero drywall patching because somebody missed a sleeve. I have seen a single clash detection save more than the entire scanning cost — and the project kept moving instead of grinding to a halt.

The odd part is that every one of these patterns is boring. Field-verify. Prioritize. Scan the tight spots. None of it is glamorous. But rink renovations fail on small, measurable things — mismatched dimensions, overlooked loads, hidden pipes. The teams that treat as-built verification as a repeatable process, not a one-time checklist item, are the ones that finish on schedule and under budget. Trends come and go. Patterns that save money? Those stay.

Why Teams Skip As-Builts (and Regret It)

Pressure to Start Construction Fast

The fastest way to ruin a renovation budget is to break ground before you understand what you're cutting into. I have watched project managers slash the as-built phase to three days because the owner's holiday charity skate was non-negotiable. That sounds fine until the concrete saw hits a refrigeration pipe that nobody knew existed. Wrong order. The schedule you saved by skipping field verifications gets eaten alive by emergency change orders, and the rink stays dark an extra three weeks anyway. The irony—nobody ever blames the schedule pressure afterward; they blame the drawings.

Belief That 'We Know This Building'

We marked the rink by hand, took photos, updated the CAD in three days. That one decision saved us from ripping out a wrong-header installation six months later.

— A quality assurance specialist, medical device compliance

Budget Cuts Targeting Documentation First

Documentation is the easiest line item to kill during value engineering because it does not produce visible progress. You can't pour concrete with a PDF. So owners trim the as-built survey, the laser scan, the field markup days—that is where the money vanishes. The odd part is that nobody thinks they are being reckless; they are just trying to fit a rink renovation into a number that was too tight from the start. But here is the trade-off: cutting $8,000 in documentation reliably produces $40,000 in field rework, and that is a conservative ratio. I have seen a missing electrical chase cost an owner more in one afternoon than a full point-cloud survey would have cost for the entire project. Budget cuts targeting documentation first are not savings—they are deferred penalties with interest.

The Slow Decay of Old Drawings

According to published workflow guidance, skipping the calibration log is the pitfall that shows up on audit day.

How patch repairs create undocumented geometry

I walked into a rink last winter where the Zamboni door frame had been cut and re-cut three times over a decade. Each patch repair—a new piece of aluminum here, a shim there—shifted the opening by fractions of an inch. Nobody drew it. The original as-built showed a clean 96-inch-wide door. The actual opening was 95.375 inches on top, 95.875 on bottom. When the new door arrived, built to the old spec, it jammed on the first try. That's the slow decay: not dramatic, not malicious, just cumulative. A chiller line gets moved six inches during an emergency fix. A contractor shaves a corner of the dasher board to fit a new partition. Each edit is rational alone. Together they create a geometry of fiction.

The scariest part—these undocumented changes rarely look wrong. A wall that bowed three-quarters of an inch over ten years still looks straight in photos. A resurfacer pit that got filled with self-leveling compound after a drain failure—the slab looks flat until you lay a laser level across it. The patch repairs become invisible until a new system demands precision. And by then, the old drawing is a museum piece. Not a tool.

Seasonal changes that never get recorded

Northern rinks live a double life. Summer dehumidification loads crack concrete differently than winter freeze-thaw cycles. I have seen header joints that seal perfectly in February and weep in August—because nobody recorded that the building settled during a wet spring. The odd part is—sheets of paper from 2005 show the steel beam elevations at one temperature. The actual structure breathes with every season. That expansion joint that looks fine on the 2010 as-built? It's shifted 3/16 of an inch after a dozen winters. Not enough to alarm anyone. Enough to snap a bracket when you try to mount a new dasher system.

What usually breaks first is the refrigerant piping. Seasonal temperature swings create micro-movements in the concrete slab. Those movements stress the insulation joints. A tiny gap opens under the slab edge—nobody draws that. A maintenance team fills it with caulk next summer, never telling the engineers. Fifteen years later, the rink floor has lifted 3/16 of an inch at the center face-off circle. The new resurfacer can't make a clean cut. That uneven surface? It started as an unrecorded seasonal event. A slow drift no one saw.

The cost of redrawing after 20 years of drift

Most teams skip this: the cost of returning to a rink after two decades of undocumented changes is brutal. You cannot just scan old drawings and hope. You bring in a survey crew. They shoot every wall, every beam, every pipe stub. That costs real money—$8,000 to $15,000 for a mid-sized rink—plus two days of downtime. And you still find surprises. A 2003 print shows a 3-inch drain line under the south corner. The survey finds a 2-inch line with a 90-degree elbow that was never spec'd. The old drawing is worse than useless now; it's a decoy.

'We spent $12,000 re-surveying a rink because the '98 as-builts showed a chiller pad that hadn't existed since 2005. The drawings were prettier than the truth.'

— Rink operations director, 20-year facility manager

The real bill isn't the survey, though. It's the redesign. Every undocumented patch creates a constraint. The new steel might need a custom bracket because someone moved a beam plate four inches for a temporary chiller. The new refrigeration layout might need re-routing because a contractor buried a conduit in 2011 and forgot to mark it. These fixes eat time—and time in a rink renovation means lost ice time, delayed leagues, angry skaters. A slow decay that became a fast problem.

When Skipping As-Builts Is the Right Call

New construction with a clean slab

If you're building from dirt up — bare concrete, no existing structure, zero prior utility runs — then yes, skip the as-built hunt. You don't need retroactive documentation for something that never existed. The architect's design drawings and the contractor's shop drawings will serve as your source of truth. Just don't mistake clean slab for clean execution. I've watched teams assume a new pad is perfectly level, only to discover a half-inch slope during refrigeration install. That hurts. Get the field-verified dimensions before the concrete cures.

Complete gut renovation with new MEP

A true gut — walls stripped to studs, old piping yanked, HVAC ripped out — can sometimes justify starting fresh. Sometimes. The caveat: you better be certain the structure itself hasn't shifted. Old beams settle. Concrete slabs crack and heave. In one project I consulted on, a 1990s rink looked fine on paper, but a laser scan revealed the dasher board footing had rotated five degrees. Nobody caught it because the old drawings were marked "obsolete." The team spent two weeks shimming boards. That's a budget hit nobody budgets for. Gut renovation works only if you also verify the building envelope — load-bearing walls, roof trusses, slab elevation — with current measurements. If you skip that step, you're gambling.

We thought gutting the place meant we could start fresh. Turns out, fresh starts require knowing what's still holding things up.

— Mechanical contractor, 42-year retrofit

Temporary seasonal upgrades

Pop-up rinks. Holiday installations. Exhibition ice sheets that vanish after March. For temporary builds, as-built drawings feel like overkill. And they usually are. The floor is a rental. The chiller is a trailer. The pipes are above-ground. Wrong order: you still need to know where the ground slopes, where water drains, and where existing underground utilities run — even if you're only there for eight weeks. I saw a seasonal rink flood a parking lot because nobody checked the storm drain location from the old site plan. The fix cost $12,000 in overtime. The lesson? Temporary doesn't mean exempt. Keep a one-page sketch of the site's critical dimensions. It's not a full as-built — but it's better than guessing.

The real question isn't should we skip as-builts; it's what are we confident we already know. If the answer is "very little," don't skip. If the answer is "everything is new, verified, and temporary," proceed — but keep your eyes open for the one detail that will prove you wrong. That detail always shows up.

Frequently Asked Questions

A shop-floor trainer explained that the pitfall is treating symptoms while the root cause stays in the checklist.

How much do as-built drawings cost?

The short answer: less than re-piping an arena that doesn't match the structural slab. I have seen quotes range from roughly $2,000 for a small community rink up to $15,000 for a multi-sheet complex with mezzanines. That sounds like real money — until you compare it to a single change order on a reno that hits $30,000 because a chiller line you thought was four inches deep turned out to be eight. Most firms price per thousand square feet plus the density of MEP systems. The catch: cheap LiDAR walks that skip hidden cavities produce cheap drawings. You usually get what you pay for in point-cloud resolution.

How accurate is a field survey?

Depends on the method and the person holding the scanner. A modern phase-shift laser scanner hits ±1/16 inch on exposed surfaces. That's tight enough to prefab a header bracket. But — and this is where owners get burned — accuracy drops sharply inside chases, above acoustic ceilings, and under the concrete floor where the old brine piping snakes. We fixed this once by cutting three core samples in a rink built in 1987; the as-built showed a 2-inch slope that was actually 3.75 inches. The survey wasn't wrong — it just couldn't see what was buried. Ask your surveyor how they handle hidden conditions. If they say "we assume the original drawing is right," walk.

What usually breaks first is the electrical backfeed path. The panel schedule says one thing; the actual wire gauge says another.

'We scanned everything above the dasher boards. Problem was, the ice plant was in the basement, and nobody had measured the pipe runs since 1974.'

— Project manager, multi-rink facility retrofit, 2023

Can I use drone scans instead?

For the roof membrane and exterior walls — yes. For the refrigeration room, the dasher cavity, or the pipe tunnel under the ice slab? Not yet. Drones give you a beautiful point cloud of the upper bowl; they give you zero data on the 8-inch supply header that's rusting behind a fire-rated wall. The trade-off is real: a drone flight costs maybe $800 and covers a full rink in 40 minutes. A terrestrial scan plus thermal imaging for concealed plumbing runs $4,000 but catches the corrosion that empties your reserve fund. Hybrid approach: fly the drone for the structure, then hire a scanning crew for the mechanical core. That split saves money, but only if the drone data is georeferenced to the same coordinate system as the ground scan. Otherwise you get two maps that don't talk to each other.

How often should I update as-builts?

Not after every hockey season — that's overkill. I suggest a hard rule: update after any mechanical replacement that changes a pipe route, after any expansion, and every ten years on a rolling cycle. The slow decay of old drawings bites hardest when someone in 2032 inherits a PDF from 2010 that still shows the old ammonia detector locations. One arena we worked with replaced their dehumidifier in 2021, the electrician marked the panel change on a napkin, and that napkin never made it back to the CAD file. Two years later, a breaker swap for the new compressor hit a live 277-volt feed that wasn't on any drawing. Update frequency should match replacement risk, not calendar dates. If you're touching the refrigerant piping, update. If you're just repainting lockers, don't bother. Most teams skip this step because they think "we'll remember" — nobody remembers. Have a digital master and a single person responsible for locking that file after every change. That's the cheapest insurance you will buy for your next renovation.

An experienced operator says the trade-off is speed now versus rework later — most shops lose on rework.

Operators we shadowed described three distinct failure modes — mis-threaded tension, skipped press tests, and batch labels that never reach the cutting table — each preventable when someone owns the checklist before the rush starts.