Checking an Old Masonry Wall Before You Load It

The Reaction Turns Up Before the Evidence Does

A steel schedule lands on a Tuesday. Beam B4, 203 UC, end reaction 96 kN one side, and against it a note in six words: bear on existing brick pier. The pier is on the survey as a rectangle 340 mm wide. Nobody has written down what it is made of, how thick the joints are, whether it is one wythe or three, or what it is carrying already. The fabricator wants the beam released for cutting on Friday, and somewhere between now and then a person has to put their name to a sentence saying the pier can take that reaction.

What governs whether that sentence is even required is the existing-building code in force, not the structural engineer's appetite. Alteration provisions in the International Existing Building Code set the point at which an existing element carrying more gravity load than before has to be demonstrated adequate rather than left alone, and the threshold is a small percentage rather than a large one — read the adopted edition, because jurisdictions amend it and rules of thumb about it circulate freely and wrongly. The load side comes from ASCE/SEI 7, Minimum Design Loads and Associated Criteria for Buildings and Other Structures, and the combination matters: a capacity quoted in allowable-stress terms cannot be checked against a factored reaction, and the mismatch is not always visible in the units.

Underneath that sentence sit three separate claims, each capable of being wrong on its own. First, that the pier's cross section is what the drawing says. Second, that the masonry's compressive strength is what the report says. Third, that whatever produced the second claim was a valid test rather than a plausible-looking number. Most disputes about old masonry that end in a wall being taken down are failures of the third claim, because it is the one nobody photographs.

Opening It Up Before Assuming a Section

The width on the survey is a gross dimension, and every capacity equation on this site wants a net one. A nominal 13-inch or 330 mm solid wall in nineteenth-century work is routinely a face wythe, a collar joint, and a backup wythe of softer common brick — three things pretending to be one. Whether they act together as a single section depends on whether the collar joint is filled and whether shear can actually cross it, which is the composite versus noncomposite distinction TMS 402/602, Building Code Requirements and Specification for Masonry Structures, treats as two different design cases with two different sets of rules. Assuming composite action because the wall looks solid is the single most common way a capacity gets overstated by a third.

Rubble-cored walls are worse and more common than most estimators expect, particularly in stone and in thick basement work. Two dressed faces with loose or lime-grouted fill between them can have a net structural area far below the gross, and the fill can have voids in it that no amount of tapping will find. Chases cut for pipework in the 1950s, blocked-up flues, ash-filled cavities, and old timber bond courses that have since rotted out all subtract section from a wall that measures full thickness on both faces.

You get the section by looking, and looking costs money and makes a mess, which is why the argument about whether it is necessary happens on every job. It is necessary. Take a unit out and inspect the bed and the collar joint directly; drill a small hole and put a borescope through it where removal is unacceptable; run ground-penetrating radar to the approach in ASTM D6432, Standard Guide for Using the Surface Ground Penetrating Radar Method for Subsurface Investigation, to locate header courses, voids and old openings across a length of wall you could never open up. Where units are being removed for later testing, take them out to ASTM C1532/C1532M, Standard Practice for Selection, Removal, and Shipment of Manufactured Masonry Units and Masonry Specimens from Existing Construction, because a specimen cracked during extraction is evidence of nothing.

Record the section as a drawn cross-section with dimensions, not as a note. It is the input to everything downstream, it will be argued about, and in eighteen months nobody will remember whether the 40 mm you allowed for the collar joint was measured or assumed.

Where f'm Comes From on a Building That Is Already Standing

For new work there are two routes to a specified compressive strength and both are written into TMS 602: the unit strength method, which reads f'm off tables against the net-area compressive strength of the units and the mortar type, and prism testing to ASTM C1314, Standard Test Method for Compressive Strength of Masonry Prisms. The first route is unavailable to you on an existing building. It requires a documented unit strength and a known mortar type, and a wall built in 1898 has neither — the units were fired in a local clamp kiln to no standard that survives, and the mortar is whatever the gang was gauging that season.

That leaves in-situ measurement and extracted-specimen testing, and they answer different questions. Flatjack work to ASTM C1196, Standard Test Method for In Situ Compressive Stress Within Solid Unit Masonry Estimated Using Flatjack Measurements, tells you the compressive stress the masonry is carrying right now, which is a fact about this building that no laboratory can supply. Its companion, ASTM C1197, Standard Test Method for In Situ Measurement of Masonry Deformability Properties Using the Flatjack Method, uses a pair of jacks to load the masonry between them and gives you a stress-strain relationship, and therefore a modulus. Neither hands you an ultimate strength unless the test is deliberately pushed to damage, which on a loaded wall is a decision with consequences.

How many tests are enough is not a matter of taste. ASCE/SEI 41, Seismic Evaluation and Retrofit of Existing Buildings, ties the number and distribution of tests to the level of knowledge you are claiming, and to whether the properties you carry forward are lower-bound or expected values — take fewer tests and the permitted strength drops. BS EN 1998-3, the assessment and retrofitting part of Eurocode 8, does the same thing through knowledge levels and confidence factors that divide the material strength down as the evidence thins. Both codes are saying the same sentence in different dialects: the strength you are allowed to use is a function of how hard you looked.

The Condition Every Number Quietly Assumes

Every capacity above assumes sound masonry, and the wall gets a vote on that. Crack mapping comes first, classified rather than described — BRE Digest 251, Assessment of damage in low-rise buildings, gives categories that let one surveyor's note mean the same thing to the next person, and the distinction that matters is whether movement is live or dormant. That takes monitoring over time, not a single visit: tell-tales read across a season, or a set of measured points revisited, because a crack photographed once is a crack of unknown age. Cracks that pass through units rather than tracking around joints indicate the units themselves have been overstressed, which is a different and more serious finding than a stepped crack following weak beds.

Mortar is the variable that decides most of the rest, and it can be identified properly rather than guessed at. ASTM C1324, Standard Test Method for Examination and Analysis of Hardened Masonry Mortar, will tell you the binder and the aggregate, which settles whether you are looking at a lime mortar that has been quietly accommodating movement for a century or a hard cement repointing from 1965 that has been driving spalling into the face ever since. Joints eroded back from the face reduce the effective bearing area of every course they occur in, and where a bearing pocket sits over a run of raked-out beds the local stress under a padstone is nothing like the average you calculated.

Then the deterioration mechanisms, all of which reduce a section you have already measured. Frost damage and salt crystallisation delaminate faces; sulfate attack expands and softens mortar in walls that have stayed wet; fire leaves clay units looking sound while their bond is gone; and embedded ironwork — cramps, old lintels, wall ties, gas pipes cast into a chase — expands as it corrodes and splits masonry from within, which is why rust staining is a structural observation and not a cleaning problem. For facades that are being inspected on a cycle rather than in response to a project, ASTM E2270, Standard Practice for Periodic Inspection of Building Facades for Unsafe Conditions, is the framework that keeps those observations comparable year to year.

Holding It Up While You Find Out

Investigation removes material from a structure that is already loaded, and the sequence in which you do it is a design problem in its own right. Three prisms taken out of one pier at one level is not a sampling strategy, it is a demolition; spread the extractions across the elevation and stagger them vertically so no single course loses a run of section. Where beam pockets are being cut or a wall is being needled to work under it, that temporary condition is often the most highly stressed the masonry will ever be, and it is designed rather than improvised — BS 5975, Code of practice for temporary works procedures and the permissible stress design of falsework, sets out the procedural side including the temporary works coordinator role, and OSHA 29 CFR 1926 Subpart Q, Concrete and Masonry Construction, governs the site conduct of the work in the United States.

The conservation principle worth carrying into all of this is proportionality. ICOMOS/ISCARSAH's Recommendations for the Analysis, Conservation and Structural Restoration of Architectural Heritage argue for the minimum intervention that achieves an adequate level of safety, and for basing decisions on evidence gathered from the structure rather than on assumptions imported from new-build practice. On an old wall that usually means testing more and rebuilding less, which is also the cheaper answer once the scaffold is already up.

The Number Belongs to a Place and a Date

Write the capacity down with everything it depends on attached: the measured section with its collar joint condition, the extraction locations marked on an elevation, the specimen proportions as tested and the correction applied, the specimen count and the scatter, the mortar analysis, the load combination the capacity is expressed against, and every assumption you were unable to test. A capacity quoted alone is a number that will be reused on the next project by someone who never saw the wall, which is how a value derived for one pier at ground floor ends up justifying a beam at third.

Then say plainly who signs it. The calculators on this site put first numbers on a page quickly and they are useful for exactly that — sizing an investigation, sanity-checking a consultant's figure, deciding whether a scheme is remotely feasible before anyone is paid to draw it. They do not carry slenderness, eccentricity, the axial term in shear, or the judgement about whether the masonry in front of you resembles the masonry the equations were calibrated on. A licensed structural engineer who has stood in front of the wall closes that gap, and on any load path that ends with a beam bearing on a hundred-year-old pier, that engagement happens before the steel is released for cutting rather than after.