Initial Assessment and Crack Documentation

Purpose

This procedure produces the baseline record: a dated, reproducible description of what the structure looked like the day the shop first touched it. Whether a crack grew, whether a door stuck before the piers went in, whether the homeowner was told the garage lintel was already cracked - all of that is answered from this record or it is not answered at all.

The record is only as good as the smallest change it can distinguish. A gauge reading to the nearest millimeter cannot prove half a millimeter of growth, so every measurement here carries its instrument, that instrument's graduation, and its practical repeatability. Without that, the shop's position in a dispute is the assessor's memory against the homeowner's, and the shop loses that argument every time.

Scope

Covers the first site visit on a residential or light commercial foundation complaint: safety walk, condition survey, crack mapping and width measurement, monitor installation, the photo set, and the scope classification that decides whether the job stays in-house or goes to a licensed design professional.

Does not cover relative floor elevation measurement, which carries its own instrument discipline and closure check - see the elevation survey SOP. Does not cover repair method selection. A test pit opened during an assessment is excavation work and falls under the excavation and shoring SOP and the utility locate SOP.

Roles and responsibilities

Procedure

Step 1: Read the file before you get out of the truck

Pull prior visits at this address, structure age, foundation type on record, and any prior monitor readings. Acceptance: before entering you can state what you expect to find and what would surprise you. Wrong looks like arriving with no prior record at an address the shop has visited, which means a monitor may already be installed and you are about to open a second unrelated baseline that will never reconcile with the first. Stop and call the office for the prior record rather than duplicating a gauge. Read the file with the engine off and the brake set, not while rolling.

Step 2: Safety walk before any measurement

Walk the whole work area before opening a case. If you smell gas anywhere on the property or in the crawlspace, everyone leaves the building immediately, nobody touches a light switch, a garage door button, a thermostat or a phone inside, and you call the gas utility emergency line from outside and upwind. If a basement holds standing water with energized receptacles, a plugged-in sump or a wetted panel, do not step into the water: kill the circuit from a panel outside the water or have the utility pull the meter, then enter. Acceptance: every hazard on site is named with the control you put on it. Wrong looks like a sagging girder crushing at a post, an active sewage leak, or a jack post out of plumb - stop, tell the homeowner the area is off limits, put it in writing, and escalate to the supervisor before further work.

Step 3: Walk the exterior perimeter as one continuous circuit

Start at a fixed corner and travel the same direction every visit so station numbers stay consistent. Record grade slope away from the wall, downspout discharge points and their distance out, irrigation heads within a wall height of the foundation, tree species and trunk distance, and prior repairs. Acceptance: a grade note at each elevation stating whether ground falls or rises away from the wall in the first stretch out from it. Wrong looks like a downspout dumping at the wall or negative grade; these change the diagnosis, so record them as findings rather than fixing them on the spot. Hazard: open window wells and unfenced prior excavations. Do not stand on a well cover before checking it takes your weight.

Step 4: Number and map every crack you will later measure

Give each crack a station ID tied to elevation and position in a scheme any tech can re-derive months later, so the third crack from the northwest corner on the north elevation is N3. Record orientation, whether it is through-thickness, whether the faces are offset in plane, and whether it steps through mortar joints or splits units. Acceptance: the map walks a different tech to the same crack without you present. Wrong looks like a photo of "the crack" with no station ID, which compares to nothing later; re-do the mapping before measuring. Hazard: crawlspace entry. Treat any crawlspace with a fuel-burning appliance, an open sewer defect or standing water as a possible hazardous atmosphere, test the air before entry, and do not crawl under visibly deflected framing.

Step 5: Measure crack width by a stated method at a stated resolution

Measure at a marked point, not at the widest point, and mark that point on the structure so the next reading takes the same spot. A comparator card graduated in tenths of a millimeter is the working tool for most readings. A plate-type tell-tale monitor reads to about a millimeter with practical repeatability nearer half of that. A set pin pair read with a digital caliper resolves finer, but only if the pins go in against a fixed-length setting bar and are read the same way each visit. Acceptance: two independent readings at the marked point agreeing within the instrument's repeatability, with the instrument named on the record. Wrong looks like two readings disagreeing by more than that, usually a spalled edge or a paint ridge being measured. Do not average them and do not record a number: record "edge unsound, width not measurable at this station", set a pin pair spanning sound material either side, and note the substitution.

Step 6: Install monitors only where a threshold already exists

A monitor is worth installing only if the movement threshold was set before it went on the wall. Shop default, tune it to your soils: re-read at 30, 90 and 180 days, and call movement real only when the change exceeds twice the instrument's practical repeatability, which on a plate monitor means about 1 mm rather than the 0.5 mm its graduations imply. Acceptance: gauge ID, install date, initial reading, next-read date, and a photo of the gauge in place with the station ID in frame. Wrong looks like a monitor bonded to loose render or to a patch, which will move for reasons unrelated to the structure; remove and reset it. Hazard: drilling masonry to set pins releases respirable crystalline silica. Use a drill with a HEPA-filtered dust collection system or water feed as specified for handheld drills in 29 CFR 1926.1153 Table 1, and follow that table's respirator requirement for the duration and enclosure you are working in.

Step 7: Record the interior distortion cues the customer actually feels

Walk every door and window in the affected zone and record which bind, on which edge, and whether the reveal is tapered. Record trim separation, drywall cracking at opening corners, and grout fracture lines. Acceptance: a door-by-door list naming the binding edge, not a general note that doors stick. Wrong looks like recording only the doors the homeowner mentioned; the ones they stopped noticing establish how long this has been running. Do not force a bound door in a masonry opening with a cracked lintel, because the frame may be carrying load.

Step 8: Shoot the photo set to a repeatable standard

Every crack station gets a wide shot establishing location, a mid shot with a scale in frame, and a close shot with the comparator or gauge readable. Every photo carries the station ID on a card or chalk mark. Acceptance: a second tech can stand where you stood from the photo alone. Wrong looks like close-ups with no scale and no context, which prove nothing about which crack was shot; re-shoot before leaving. Do not walk backwards while framing near an open well or a stair opening.

Step 9: Classify scope and say the engineer trigger out loud

Classify as in-scope, monitor-only, or design-professional-required, and write the trigger you used. Stop scoping and route to a licensed design professional on any of: a wall bowed past the shop's stated deflection limit, bearing loss at a footing, a previously altered load path, a change of use or added story, a retaining function on the wall, or any condition where you cannot establish what carries the load. Acceptance: classification and trigger written on the record and said to the homeowner in the same visit. Wrong looks like pricing a repair on a wall whose deflection you never measured. Stop, tell the homeowner plainly that this one needs an engineer's design before anyone prices it, and hand the record over unsummarized.

The record this produces

One assessment record per address per visit, in the job file: safety walk findings and controls; the exterior circuit with grade and drainage notes; the crack map with station IDs; a width row per station carrying value, instrument, graduation, repeatability and the marked measurement point; the monitor register with gauge ID, install date, initial reading and next-read date; the interior cue list; the indexed photo set; and the scope classification with its trigger.

The estimator reads it to price. The install crew reads the crack map to know what was already there before they arrive. The office reads the monitor register to schedule the 30, 90 and 180 day reads. An adjuster reads it years later, and at that point the gauge ID and the repeatability line are the two fields doing the work.

One worked pass

Two-story brick veneer over a poured wall, first visit, complaint of a widening exterior crack. The assessor walked clockwise from the northwest corner and mapped four stations. Step 3 logged negative grade of roughly a hand's depth over the first stride out from the north wall, and a downspout discharging two feet from station N3.

Step 5 ran as follows. Station N1, comparator card at 0.2 mm graduations: 0.6 and 0.6 mm, agreement inside repeatability, recorded. Station N2: 1.4 and 1.4 mm, recorded. Station N3: 2.8 and 3.4 mm. That 0.6 mm spread is three graduations wide and well outside the card's repeatability, and the reason was visible once the assessor looked instead of measured, a spalled arris on the upper edge. Under step 5's stop rule no width was recorded. The entry reads "N3 edge unsound, width not measurable by card", a pin pair went into sound brick either side, and the initial caliper span was logged with the setting bar length beside it.

That failure changed step 6. The job's threshold had been set at twice a plate monitor's practical repeatability, about 1 mm, but N3 now sits on pins read by caliper, which resolves finer. Rather than run one job on two unstated thresholds, the record carries a threshold per gauge: about 1 mm at the plate monitors on N1 and N2, and 0.3 mm at the N3 pin pair, each printed beside the instrument that sets it. Step 9 classified the job in-scope with a 90 day monitor gate, since no bowing was measured and the load path is intact.

The station the customer cared most about is the one the primary instrument could not measure, and the procedure's answer was to change instruments and say so, not to write down the more alarming of two disagreeing numbers.

References

• 29 CFR 1926.1153 Table 1, respirable crystalline silica specified exposure control methods including handheld drills, for any pin or anchor set into masonry or concrete.
• ACI 224.1R, Causes, Evaluation, and Repair of Cracks in Concrete Structures, for crack cause classification and evaluation practice.
• ACI 562, Code Requirements for Assessment, Repair, and Rehabilitation of Existing Concrete Structures, in the edition your jurisdiction has adopted, which reaches you through the permit.
• See related: the elevation survey SOP for relative floor measurement; the excavation and shoring SOP and the utility locate SOP for any test pit opened during an assessment.