Rank an RTU portfolio for repair or replacement surveys

How do you rank commercial RTUs for a survey?

Start with one record per unit and rank the evidence that would change the next spending decision. A failing unit serving a space without backup deserves attention before an older unit with sound components and a clean service history. Keep immediate response, planned diagnostics and capital analysis in separate work queues so a spreadsheet score never delays a hazard response.

The U.S. Department of Energy (DOE) Advanced RTU Campaign decision tree moves from inventory and preliminary screening to field evaluation, economic analysis, and project planning. It considers condition alongside age and size, then asks about sizing, improvements, and economics. Use that order for the portfolio: inventory the units, inspect them, then compare alternatives. Its historical screening bands are not replacement rules. [1]

The worksheet below is Anvilfield's proposed scheduling method. Its weights and cutoffs are examples for a service manager to adopt or change with the owner. They are not DOE scoring rules, equipment life predictions or permission to replace anything. The deliverable is a defensible survey order and a specific question for each visit.

Portfolio inventory worksheet

Give each unit a stable ID tied to a marked roof plan, a nameplate photograph and the space served. Keep the property's address and ownership record above the unit records. One roof can contain several purchasing responsibilities, and one owner can control several buildings. Neither relationship should be inferred from a tenant sign.

Separate installation age from building age, roof age and compressor replacement age. If the installation date is missing, record the manufacturing date as a different field. Resolve serial-number decoding with the manufacturer rather than assigning a familiar-looking year. Unknown is a useful entry when it carries a named follow-up action.

Keep an evidence date beside every finding. For this example workflow, review service events over the previous 24 months, but preserve older major repairs in the equipment history. That window is an estimating choice, not a maintenance standard. Agree with the owner on whether seasonal shutdowns and tenant changes require a different review period.

What evidence is enough to assign a score?

Use confirmed records for a final score. An operator's complaint is valuable reported evidence, but it does not establish the failed component. Ask for the ticket and the outcome. A photograph of corrosion establishes where corrosion was visible; it does not establish remaining capacity or a heat-exchanger failure.

Keep missing information out of the zero column. If a factor is unknown, show the score as a range: the sum of known points through that sum plus the maximum points for each missing factor. Also show completed factors, such as four of five. Do not sort a partly documented unit beside fully scored units as though the totals were equally reliable.

Assign unresolved units to an information queue. Request the nameplate, service history or operating check that could settle the missing factor. A potentially critical unit with uncertain backup gets prompt review even when its numerical range is broad. Saving a survey trip is useful only when the records actually answer the question.

Survey priority scoring worksheet

Add the five factor scores for a maximum of 100 points. State the evidence behind each selection. Condition receives the most weight; age receives the least. The age bands below are original planning inputs and say nothing about a particular model's remaining life. Use completed years in service when documented so every known age falls into one band.

For this example, put fully documented scores of 60 or more in the first planned survey group. Scores from 30 through 59 enter the next diagnostic group. Scores below 30 stay with the maintenance review unless a specific finding changes the plan. These thresholds allocate survey hours; none selects replacement automatically.

An active hazard, loss of required service or unresolved safety concern follows the site's escalation process outside these bands. Record that override in plain language. A unit should never have to collect enough points before somebody addresses a serious finding.

Worked portfolio ranking

The case uses four typical units at four separate properties. All findings except Unit D's installation age are confirmed, current enough for the owner's planning decision and free of an immediate hazard. Each unit has equal survey travel cost. The table uses the five factors in the preceding order and includes no additional units sharing the same failure consequence.

Unit A leads because several independent concerns line up. Its survey should compare a defined repair with a replacement scope. Unit C also enters the first group, but its repeated control-related failures still need root-cause diagnosis. Its score earns an earlier visit without proving that the cabinet belongs on a crane.

Unit B goes to the next diagnostic group for the damper repair and verification. Unit D needs the missing installation record; its known factors do not justify a replacement survey. Do not replace that missing age with the warehouse construction year. At building level, carry the highest unit priority, the number of affected units and any shared outage constraints rather than adding every unit's points together.

How do you verify condition during the survey?

Start with the components that could change the repair decision. Carrier's 48TC service manual separates fan and drive work, coils and refrigeration components, economizer checks, and heating inspection. It is a model-specific reference, not a checklist for every RTU. Build the evaluation from the installed model's instructions and record defects, tests, and work left incomplete. [2]

Agree on the visit's limits before dispatch. A visual inventory, a functional diagnostic visit and a replacement engineering survey have different labor and access needs. Mark cooling, heating or economizer operation as untested when weather, shutdown permission or missing instrumentation prevents a valid test. A checked box cannot stand in for an operating condition you never established.

For authorized testing, record the manufacturer's target, the measured value, operating mode, indoor and outdoor conditions and the instrument used. Compare airflow and external static pressure with the applicable equipment data before treating weak cooling as a refrigerant problem. Keep an unresolved symptom open until a qualified technician establishes the cause.

Use the existing RTU startup, preventive maintenance and economizer guides for the detailed work. This portfolio sheet needs the resulting findings, uncertainty and recommended next action. It should not become a second, conflicting set of universal charge, airflow or combustion targets.

Does R-410A equipment automatically need replacement?

No. EPA's HFC phasedown questions say equipment already in use can continue through its useful life. Do not turn restrictions on new equipment into a blanket replacement deadline for an operating RTU. Record the refrigerant and investigate the actual service options, parts, repair cost and applicable handling requirements. [3]

For a proposed replacement, check the current EPA restriction table against the equipment category and transaction. Self-contained products and field-assembled systems have separate tables and provisions. Confirm how the selected equipment is classified, its manufacture or import documentation, and the rules applicable when it is supplied and installed. A deadline copied from a split-system summary may answer the wrong question. [4]

The sector table reviewed for this article shows an August 28, 2026 update and reflects revised provisions. Use its current text and governing regulations when the purchase proceeds. Get written confirmation from the supplier for the selected model. Neither this survey score nor an old campaign decision tree establishes refrigerant compliance or authorizes substituting a different refrigerant.

What belongs in a replacement survey scope?

Confirm the load and required operating sequence before selecting capacity. Bring the owner a record of occupancy, ventilation requirements, schedules, complaints and changes to the space. Ask the design professional to resolve sizing and system changes. Repeating the old tonnage preserves an input that the survey was supposed to check.

Record curb dimensions, duct arrangement, unit and accessory weight, available electrical service, heating connections, condensate route, controls interface and maintenance clearances. Identify who verifies structural capacity, electrical compatibility and the roof detail. Put each unresolved item beside a responsible reviewer instead of hiding it in a general contingency.

Check the crane approach, pick radius, staging area, street or parking restrictions, tenant interruption and weather contingency with the people responsible for that work. Grouping replacements may share mobilization, but the actual lifting plan determines the saving. Two units on one roof do not automatically share a workable pick.

Include commissioning, airflow verification, controls integration, roof repairs and turnover records in the estimate boundary. State the permits and design reviews still required. Obtain a priced repair alternative where it remains technically acceptable, and identify whether temporary cooling or heating belongs in either option. The owner needs two complete scopes to compare.

Worked survey labor and cost example

Anvilfield research models a typical existing portfolio containing 12 buildings with exactly four RTUs each, for 48 units. An analyst spends ten minutes per unit reviewing the inventory at a loaded cost of $70 per hour. The review selects three buildings, containing 12 units, for an authorized survey.

One two-person crew spends three clock hours at each selected building and three total clock hours driving the grouped route. Both technicians cost $95 per person-hour. One technician then spends three additional hours preparing the combined report. The vehicle travels 150 total miles at $0.90 per mile; that allowance excludes paid driving labor.

The labor rates include wages, payroll burden and allocated company overhead, but exclude vehicle expense. The case uses owner-provided access and escorts, with no additional charges for parking, tolls, equipment rental, purchased lists, overtime, subcontractors or tax. The visit covers inventory and agreed condition checks only; repairs, teardown, engineering, inaccessible components and a second seasonal test require separate scope. All authorized observations fit these hours.

The total is $3,260, or $271.67 per surveyed unit rounded to cents. Use the amount as a campaign cost allocation; price the customer scope separately. If each site adds one clock hour of escort waiting, the two-person crew adds six person-hours and $570. Confirm access before promising the owner a fixed survey fee.

How do repair and replacement costs compare?

Compare alternatives over the same study period, with the same required service. DOE's Building Life Cycle Cost Programs describes comparing higher initial investment with lower operating costs over a project life. Use that discipline when moving from survey priority to a capital recommendation. The lowest initial invoice does not describe the whole ownership cost. [5]

One typical cooling-only RTU has two technically acceptable options: a $9,000 repair now or a $30,000 replacement now. Both prices include all necessary labor, materials, refrigerant work, access, permits, disposal and commissioning. The replacement also includes equipment, lifting, curb adaptation, roof work, controls connections and design fees. The existing structure and electrical supply need no upgrade.

Use a five-year study period in constant US dollars and a modeled zero real discount rate to keep the arithmetic visible. Electricity remains $0.15 per kilowatt-hour, with annual consumption of 40,000 kilowatt-hours after repair and 28,000 after replacement. The case uses identical demand charges, no heating energy, and no difference in other utility charges.

Annual maintenance costs $1,200 after repair and $1,000 after replacement. Include one additional $8,000 repair in year three for the retained unit and no other repairs for either option. The case uses a net residual value at the end of year five of zero for the retained unit and $8,000 for the new unit. Treat that value as an economic credit, not an assured resale payment.

The case uses no rebates, financing, tax effects, energy escalation or monetized downtime differences. The survey cost has already been incurred equally for both options and is excluded from this incremental comparison. Under those inputs, replacement has a $5,000 lower study-period cost. Remove its residual credit and replacement costs $3,000 more than repair. Remove the retained unit's modeled future repair instead and retaining it costs $3,000 less than replacement with the residual credit.

Those reversals belong beside the recommendation. Obtain support for energy use, repair forecasts and remaining value, then use the owner's study period and discount basis. For a longer horizon, include future equipment replacements consistently. A favorable example cannot establish that a real unit will survive five years or that a future breakdown will occur.

Access, isolation and survey limits

Confirm roof access, fall protection, electrical and mechanical isolation, weather limits and emergency arrangements before dispatch. The survey authorization should identify who can permit shutdowns and who must approve a return to service. Resolve an unsafe approach before assigning hours to the equipment itself.

Oregon OSHA's hazardous-energy fact sheet explains isolation and verification during service, including stored energy. Its designated-area fact sheet illustrates why roof arrangements depend on the work and the exposure. These are state guidance references; determine the federal or state rules governing the actual site and task. A roof photograph cannot establish a compliant access plan. [7, 8]

Keep visual observation separate from opening equipment or running a diagnostic test. Qualified personnel must plan operating tests under the applicable procedures, including any controlled change from an isolated condition. Do not improvise energized work to finish a portfolio form. Record the inaccessible item and the preparation needed to complete it.

What to document

Issue one portfolio summary backed by unit sheets. Show each unit's score or range, evidence completeness, operating consequence, proposed survey scope, access allowance and responsible reviewer. Put the required owner decision beside the record: authorize diagnosis, request missing documents, obtain design review or compare priced alternatives.

Write findings that the next person can reproduce. Weak: old unit, replace soon. Strong: Unit A, installation record dated 2009, two documented outages in the review window, damaged assembly confirmed in the attached evaluation, and supplier lead time beyond the owner's tolerated interruption. Next action: authorize a repair and replacement scope comparison.

Preserve changes after repair. Record what was corrected, the verification result and any revised factor score. Keep the original findings and dates visible. A repair that restores reliable operation should change the next capital review rather than leaving the unit permanently labeled for replacement.

Common mistakes

• Using the building or roof installation year as the RTU installation date. Request the equipment record.
• Counting repeat visits for one unresolved failure as independent failures while losing the total repair cost.
• Giving unknown condition zero points. Preserve the missing factor and its possible score range.
• Treating a high priority score as replacement approval. Complete diagnosis and compare acceptable scopes.
• Claiming energy savings from a rating ratio alone. Establish a baseline, operating conditions and a defensible forecast.
• Ranking by the sum of building scores. Show the highest unit priority and the actual shared outage consequence.
• Pricing equipment while omitting access, curb, controls, roof work and commissioning from the comparison.

Survey release checklist

Finding buildings worth an RTU survey

For an existing account, start with the owner's equipment records and service history. For a new territory, use property records, mechanical permits, management-company portfolios and dated roof views to locate candidate buildings. Keep observed rooftop equipment separate from confirmed RTUs and verified service need. A prospect record should survive a changed tenant or an updated roof view.

PorchRocket's rooftop HVAC destination offers Ground Control building lists based on visible rooftop equipment, with area, building type and size filters. Use that list to organize a compact group of inventory conversations. Its role is locating buildings to check; unit age, condition, purchasing responsibility and a request for work still need independent confirmation. [8]

Carry the building record into the unit worksheet after an authorized visit. Track buildings reviewed, buyers reached, surveys authorized and decisions supported as separate counts. A visible-unit band is useful for choosing where to ask the next question. It cannot supply the condition score or the replacement forecast.

Sources and verification limits

Sources reviewed November 4, 2026. DOE's campaign decision tree is a historical planning reference; its publication date was not established here. Use its inventory-to-analysis sequence, not its old calculators, incentives, or screening bands as current requirements. The Carrier manual is dated May 2014 and applies to the identified 48TC models.

EPA's linked sector table shows an August 28, 2026 update. The linked DOE ownership-cost page identifies the 2025 revision of the federal life-cycle costing handbook. The Oregon hazardous-energy sheet is marked December 2024; its designated-area sheet is marked September 2017. Use those safety sheets as explanatory references and check the governing rules for the actual work.

Confirm model instructions, local requirements, supplier terms and owner operating constraints before turning a survey recommendation into an approved project.

Methodology

Counts, prices, work rates, schedules and outcomes in typical cases are inputs to the Anvilfield research model. Calculations use the stated scope, formulas, cost inclusions and exclusions. Cited public records and technical documents support their attributed facts. The cost model uses US dollars. Named properties were not inspected; market totals, remaining life, conversion rates and customer savings were not independently verified.

Related tools

FAQ

People also ask

• How do you start up an RTU?
• What is HVAC preventive maintenance?
• What is an air-side economizer?
• What is air balancing?
• How do you estimate an HVAC job?