home  /  insights  /  soil-classification-governs-everything-downstream
Geotechnical & Foundations

Soil classification governs everything downstream

Slope angle, shoring selection and shield rating all follow from one field judgment: what type of soil this is. How that class is set, and why it does not stay true for the life of the trench.

July 30, 2026 · 8 min read

The short answer

A trench’s soil class — stable rock, Type A, Type B or Type C under OSHA’s excavation standard, 29 CFR 1926 Subpart P — is a field judgment the competent person must base on at least one visual and one manual test, and nearly every downstream decision on the trench follows from it: the slope angle, the shoring type, whether a shield is rated for the depth, and whether an engineer needs to be involved at all. OSHA’s excavation standard builds its whole protective-system framework on that single classification, a judgment made early and usually quickly. The class describes a material in a condition, and the condition moves, so the classification does not stay true for the life of the trench. When a trench wall lets go, the question is rarely whether the soil behaved strangely; it is whether the class assigned to it matched the material actually standing there, in the condition it was in that day.

What this article establishes

  • Under OSHA’s excavation standard, 29 CFR 1926 Subpart P, the soil class sets the maximum allowable slope — vertical for stable rock, three-quarters to one for Type A, one to one for Type B, one and a half to one for Type C — and indexes the tabulated data used to select shoring, so moving a soil one class in the optimistic direction changes both the geometry a crew may cut and the hardware the crew may rely on.
  • Appendix A of 29 CFR 1926 Subpart P removes soils from Type A irrespective of strength when they are fissured, previously disturbed, subject to vibration from heavy traffic, pile driving or similar effects, or in a layered system dipping into the excavation at four horizontal to one vertical or steeper; the previously-disturbed exclusion carries particular weight when a trench cuts an existing utility corridor through backfill placed by someone else, at an unknown time, to an unknown standard.
  • Laboratory soil classification under ASTM D2487’s Unified Soil Classification System does not map one-to-one onto OSHA’s four soil classes, and treating a USCS group symbol as though it were an OSHA class is a category error; what the laboratory supplies is an independent measure of plasticity, gradation, moisture content and shear strength.
  • A trench soil classification describes a material in a condition, and it goes stale as rain, desiccation and nearby equipment change that condition, which is why OSHA’s 29 CFR 1926 Subpart P requires inspection before each shift, as needed during it, and after every rainstorm or other occurrence that increases hazard.
  • Under 29 CFR 1926 Subpart P, a layered profile is classified by its weakest layer, with a narrow exception where a more stable layer lies beneath a less stable one. Separately, a trench run of several hundred feet can cross fill, native cohesive material and a saturated granular lens, and the soil class that governs is the one at the station where a worker is standing, not the one recorded at the launch pit.
  • Opinions about a trench’s soil classification are predictably challenged on the grounds that exemplar samples taken after a collapse no longer represent in-situ condition, that moisture content shifted before sampling, that a laboratory group symbol was converted into an OSHA class without justification, and that one classification was applied to a trench whose conditions varied along its length. Classification work that states which soil each sample represents, where it came from, and how close the result sits to a class boundary survives that scrutiny; a soil class asserted without its basis does not.

What does a trench’s OSHA soil classification actually decide?

A trench’s OSHA soil classification decides the maximum allowable slope and the shoring data a crew may use. Under 29 CFR 1926 Subpart P, soil and rock are sorted into stable rock, Type A, Type B and Type C, and that class sets the maximum allowable slope — vertical for stable rock, three-quarters to one for Type A, one to one for Type B, one and a half to one for Type C — and indexes the tabulated data used to select shoring.

The OSHA soil classification is therefore not descriptive; it is an input to a design. Moving a soil one class in the optimistic direction changes both the geometry a crew may cut and the hardware the crew may rely on.

How is OSHA soil type supposed to be determined in a trench?

OSHA soil type is supposed to be determined by the competent person, using at least one visual and one manual test, against Appendix A of 29 CFR 1926 Subpart P. Appendix A defines the soil classes by unconfined compressive strength with a list of qualifying conditions: Type A at 1.5 tons per square foot or greater, Type B above 0.5 and below 1.5, and Type C at 0.5 or less.

The manual methods for OSHA soil classification are deliberately field-scale: plasticity by rolling a moist sample into a thread, dry strength by crushing a clod, thumb penetration, and sometimes a pocket penetrometer. They are quick and approximate, and 29 CFR 1926 Subpart P accepts that because the realistic alternative is no analysis at all.

Where does laboratory soil classification fit alongside OSHA’s soil types?

Laboratory soil classification supplies an independent measure of the soil — plasticity, gradation, moisture content, shear strength — but it does not map one-to-one onto OSHA’s four classes of stable rock, Type A, Type B and Type C.

ASTM D2488 formalizes the same visual-manual approach used in OSHA’s field tests, and ASTM D2487 applies the Unified Soil Classification System (USCS) using laboratory index testing. Neither ASTM D2488 nor ASTM D2487 maps one-to-one onto the four classes of 29 CFR 1926 Subpart P, and treating a USCS group symbol as though it were an OSHA class is a category error.

Where borings exist, ASTM D1586 penetration testing gives a depth-referenced record of the ground that predates the excavation entirely.

What conditions disqualify a soil from OSHA Type A?

Appendix A of OSHA’s excavation standard, 29 CFR 1926 Subpart P, removes soils from Type A irrespective of strength when they are fissured, previously disturbed, subject to vibration from heavy traffic, pile driving or similar effects, or part of a layered system dipping into the excavation at four horizontal to one vertical or steeper.

The previously-disturbed exclusion in Appendix A carries more weight than its length suggests, because trenching an existing utility corridor means cutting through backfill placed by someone else, at an unknown time, to an unknown standard.

How is layered ground classified under OSHA’s excavation standard?

Where a trench profile contains more than one material, OSHA’s excavation standard, 29 CFR 1926 Subpart P, classifies the system by its weakest layer, with a narrow exception where a more stable layer lies beneath a less stable one. The weakest-layer rule follows the mechanics: competent clay above a running sand does not stop the sand from undermining it.

Does a trench’s soil classification stay valid for as long as the trench is open?

No. A trench soil classification describes a material in a condition, and the condition moves, so a classification made once goes stale. Rain raises moisture content and washes fines from the wall face. Exposure desiccates cohesive soil and opens tension cracks parallel to the trench, well back from the edge. Equipment alongside the trench introduces vibration the original assessment may have assumed away.

OSHA’s excavation standard, 29 CFR 1926 Subpart P, answers this with a cadence rather than a one-time determination: inspection before each shift, as needed during it, and after every rainstorm or other occurrence that increases hazard. A trench open three weeks through changing weather is not the trench classified on day one.

Can one trench contain more than one soil type?

Yes. Trench soil classification is often discussed as though a job has a soil type, but a trench run of several hundred feet can cross fill, native cohesive material and a saturated granular lens. The OSHA soil class that governs is the one at the station where a worker is standing, not the one recorded at the launch pit. Depth does the same thing vertically.

What does a trench’s soil class mean for sloping, shoring and trench shields?

Once a trench’s OSHA soil class is set, the protective system has to follow from it: sloping and benching checked against the Appendix B tables of 29 CFR 1926 Subpart P, shoring checked against the tabulated data in the Subpart P appendices or the manufacturer’s own, and manufactured trench shields checked against their rating for depth and soil type.

Two structural limits in 29 CFR 1926 Subpart P sit alongside the soil class. An excavation twenty feet deep or more requires a protective system designed by a registered professional engineer. And a trench shield may not be loaded beyond its design, including where the excavation is carried more than two feet below the bottom of the shield.

How are opinions about a trench’s soil classification challenged?

Opinions about a trench’s soil classification are challenged predictably: that exemplar samples taken after a collapse no longer represent in-situ condition; that moisture content shifted before sampling; that a laboratory group symbol was converted into an OSHA class without justification; and that one classification was applied to a trench whose conditions varied along its length.

Trench soil classification work that states which soil each sample represents, where it came from, and how close the result sits to a class boundary survives that scrutiny. A soil class asserted without its basis does not.

This guidance on trench soil classification is general technical orientation, not a failure analysis, an engineering opinion, or advice on any specific matter. Determining the cause of a particular incident requires hands-on examination by a credentialed expert.

For informational purposes only. Not engineering or legal advice, and not an opinion on the cause of any specific failure or on the conduct of any party.

Related

The practice area

failure-analysis assistanttriage · not a substitute for an expert
Happy to. Tell me what failed, how it failed, and whether the failed part and the scene are still preserved. That last one often decides what can still be established.