A screw can be original to a machine, a later repair, a standardized replacement, a piece of tooling, or a power-transmission component that was never a fastener at all. Those possibilities are easy to collapse when an old machine, building, press, tool or piece of infrastructure is encountered as one object.

The screw is not one clue. It is a stack of clues.

The useful question is not simply, “What kind of screw is this?” The better questions are:

• What geometry is physically present?
• What dimensions can be measured repeatably?
• Which thread families are compatible with those dimensions?
• What was the part doing in the assembly?
• Does the drive style or head form set a lower-bound date?
• Do neighboring fasteners agree, or does one part belong to a later repair generation?
• What documentary evidence connects this exact part—or this exact replacement campaign—to the object?

That is the purpose of this guide: to turn a small mechanical detail into a disciplined historical reading.

Contents

1. The Evidence Chain
2. What to Record Before Naming Anything
3. Thread Anatomy: Pitch, Lead, Handedness and Profile
4. Six Thread Families Worth Separating
5. Standardization as a Timeline, Not a Date Stamp
6. Drives, Heads and the Terminus Post Quem
7. Power Screws Are Not Fasteners
8. Mixed-Generation Assemblies and Repair Archaeology
9. Worked Plate A: A 60-Degree Thread on an Older Industrial Assembly
10. Worked Plate B: A Phillips Screw in a Circa-1890 Assembly
11. Worked Plate C: Magnolia Plantation’s Wood-Screw Cotton Press
12. Confidence Grades
13. Research Routing
14. Safety Boundary
15. Sources

1. The Evidence Chain

Use the same order every time:

OBJECT → OBSERVE → MEASURE → COMPARE → CONTEXT → CONFIDENCE

Do not begin with a label. Begin with the object.

OBJECT means the whole assembly, not only the screw. Record where the fastener sits, what it joins, whether it moves, whether it is accessible from one or both sides, and whether surrounding hardware appears uniform.

OBSERVE means photograph before disturbing anything. Capture the head, shank if visible, nut or tapped hole, washer stack, finish, corrosion, tool damage, nearby hardware, and enough of the surrounding object to preserve context.

MEASURE means record repeatable dimensions rather than relying on visual resemblance. Useful measurements can include major diameter, pitch or threads per inch, approximate thread angle where feasible, lead on multi-start or power screws, head dimensions, recess geometry, and whether the thread is right- or left-handed.

COMPARE means ask which families remain possible. A 60-degree profile does not automatically mean Unified National. A matching TPI does not prove the part was original. A damaged crest can make one form resemble another.

CONTEXT means move outward: manufacturer catalogs, maintenance records, patents, photographs, inventories, standards in force at the target date, surrounding fastener populations, and evidence of repair or re-threading.

CONFIDENCE means write a conclusion at the level the evidence can actually support.

A strong result may be: “60-degree inch-thread family supported; exact standard and installation date unresolved.” That is better history than forcing an exact label from incomplete evidence.

Field note

The biggest mistake in old-hardware identification is turning compatibility into provenance. Two parts can fit each other without being original to each other.

2. What to Record Before Naming Anything

Before you say Whitworth, Unified, Acme or Phillips, record a neutral description.

Minimum observation set

• Object or assembly name and location
• Fastener location within the assembly
• Head form
• Drive form
• Visible thread profile
• Major diameter if accessible
• Pitch or TPI
• Apparent handedness
• Nut, tapped hole or mating thread condition
• Washer or spacer stack
• Finish/plating/paint
• Corrosion pattern
• Tool marks or slot/recess damage
• Neighboring fastener types
• Evidence of inserts, bushings, welding, re-tapping or hole enlargement
• Documentary link, if any

Why neighboring hardware matters

One isolated screw can lie to you. A population is harder to misread.

If eleven fasteners share one head, finish and thread family but a twelfth has a different drive and plating, that outlier deserves attention. It may be a later replacement. If all fasteners differ, the object may have had a long maintenance life. If the tapped hole shows an insert or re-threading, even a period-looking fastener may be occupying a later repair.

3. Thread Anatomy: Pitch, Lead, Handedness and Profile

Several terms are often mixed together.

Pitch is the distance from one thread crest to the corresponding point on the next thread.

Threads per inch, or TPI, is the reciprocal way inch-series threads are often described.

Lead is the distance a screw advances in one full revolution. On a single-start thread, lead equals pitch. On a multi-start thread, lead is larger because several helical starts advance together.

Handedness describes the direction of the helix. Most ordinary fasteners are right-handed. Left-hand threads exist where operating direction, rotation or deliberate anti-loosening logic calls for them.

Included angle describes the angle between the thread flanks. It is one useful discriminator, but damage, corrosion, wear and imperfect measurement can make visual angle estimates unreliable.

Crest and root shape can help separate families. Rounded, flat, truncated and sharply formed features may matter, but they should be read together with dimensions.

Web table: terminology

4. Six Thread Families Worth Separating

Whitworth-family triangular threads

A 55-degree form with rounded crest/root geometry can support a Whitworth-family possibility when dimensions and context agree. It does not prove that the part dates to a particular decade, nor does it prove British manufacture.

Sellers / American National / Unified / other 60-degree triangular threads

William Sellers’s 1864 U.S. standardization proposal deliberately used a 60-degree form rather than Whitworth’s 55-degree geometry. Later American National and Unified systems also use 60-degree triangular geometry. That makes “60 degrees” a family clue, not an exact identification.

A measured 60-degree thread on an older machine can be:

• Sellers lineage
• American National lineage
• later Unified replacement
• another compatible 60-degree standard
• shop-cut nonstandard work

The dimensions, date context and documentary record decide how far the conclusion can go.

Metric triangular threads

Modern ISO metric thread families also use 60-degree triangular geometry. A visual 60-degree reading therefore cannot separate inch and metric systems. Diameter and pitch units matter.

Acme

Standard American Acme thread form uses a 29-degree included angle. Acme is common in power transmission and positioning where a strong, manufacturable thread is needed. Current definitions can identify geometry, but cannot date an old screw.

Square

Square threads use near-vertical flanks and historically offered efficient force transmission. Wear, damage and manufacture can round corners enough that the form should not be classified from a quick glance.

Buttress

Buttress threads are asymmetric. One flank is configured to carry high load primarily in one direction. Corrosion or wear can obscure the working flank, so document the full profile.

Wood power screws

Large carved or machined helical screws in timber mechanisms belong in a different mental category. They can transmit pressing or lifting force through a large wooden assembly. Their surrounding motive-power system may change while the screw survives.

5. Standardization as a Timeline, Not a Date Stamp

Historical fastener research becomes much stronger when standards are treated as institutional history rather than magic dating charts.

1864 — Sellers proposal

ASME identifies William Sellers’s 1864 proposal as the first U.S. system of standardized screw threads. Sellers changed from Whitworth’s 55-degree form to a 60-degree form and proposed systematic pitch, thread form, depth and nut proportions. The Franklin Institute endorsed the system on December 15, 1864. Adoption efforts extended through the U.S. Army, Navy and major railroads.

That does not mean every 60-degree thread after 1864 is a Sellers thread.

1918–1933 — National Screw Thread Commission

National Archives Record Group 167 preserves records of the National Screw Thread Commission. This matters because standardization became an organized federal and industrial process rather than a single inventor’s proposal.

1928 — “American National” usage expands

The National Bureau of Standards’ third commission report records the “American National” designation and expands treatment to additional thread forms, including Acme threads.

1939–1969 — Federal handbooks

NIST historical material traces federal screw-thread handbooks through H25 in 1939, H28 in 1942, and revisions in 1944, 1950, 1957, 1963 and 1969.

1948 — Unified accord

U.S., United Kingdom and Canadian conferences culminated in a Declaration of Accord on November 18, 1948, establishing Unified screw-thread practice.

Current continuity

DLA QuickSearch continues to list the FED-STD-H28 family as active, and ASME B1.1 remains a current thread standard. Current standards are useful for terminology and geometry. They are not proof of when an old part was installed.

The rule

A living standard can define a family. It cannot date a historical object by itself.

6. Drives, Heads and the Terminus Post Quem

A drive can sometimes tell you that an assembly cannot be entirely original to an earlier date.

This is a terminus post quem: a lower-bound clue. It is not a complete installation date.

Slotted

Slotted screws are long-lived and weak as precise date markers. Their presence rarely narrows the date enough by itself.

Robertson

Robertson company history records a Canadian patent milestone in 1909. A Robertson-type drive in a supposedly untouched nineteenth-century assembly deserves scrutiny.

Allen / internal hex

William G. Allen’s socket-screw patent was granted in 1910. Again, patent date is not widespread adoption date, but it creates a useful lower-bound clue.

Phillips

U.S. Patent 2,046,839 was filed in 1935 and granted July 7, 1936. A Phillips-type cross recess in an 1890 assembly is therefore strong evidence of later intervention.

The caution

Patent date is not adoption date. Adoption date is not installation date. Use the drive to identify an impossibility or a likely lower bound, then look for maintenance records, dated photographs, catalogs, repair invoices and neighboring hardware.

7. Power Screws Are Not Fasteners

A fastener’s job is generally to join parts. A power screw converts rotation into controlled linear motion or force.

This difference changes the evidence you should record.

For a power screw, document:

• thread family
• lead
• nut or moving block
• thrust direction
• guides
• bearings or friction surfaces
• handwheel, crank, gearing or other drive
• evidence of lubrication
• wear distribution
• travel limits
• surrounding frame

Examples can include screw jacks, presses, vises, feed screws and machine-tool lead screws.

A thread that looks “too coarse to be a bolt” may be doing exactly what it was designed to do: moving or pressing rather than fastening.

8. Mixed-Generation Assemblies and Repair Archaeology

Historic objects accumulate time.

An 1890 machine can carry a 1930s screw, a 1950s washer, a modern threaded insert and a still-earlier hand-fitted component without becoming “fake.” The task is to reconstruct the sequence.

High-value secondary evidence

Head markings can identify manufacturer, grade or system when a period-appropriate table exists. Do not apply modern marking charts indiscriminately to older hardware.

Plating and coating can reveal a replacement generation. Bright zinc on one fastener in an otherwise dark, oxidized population is a clue, not a verdict.

Tool marks can indicate removal and reinstallation. Damage is not automatically old.

Tapped-hole wear can show long occupation by a thread family. Inserts, bushings and re-tapping can reset that evidence.

Nut/washer populations are valuable because consistency across an assembly is more informative than one isolated component.

Catalog nomenclature shows what a manufacturer offered and how it named a part. Availability does not prove purchase or installation.

A useful web-native graphic

Create a four-layer exploded diagram labeled:

ORIGINAL STRUCTURE

EARLY REPAIR

LATER STANDARDIZED REPLACEMENT

MODERN MAINTENANCE

Beside it, place the rule: “Authenticity can include repair history. Date the intervention before judging the object.”

9. Worked Plate A — A 60-Degree Thread on an Older Industrial Assembly

Observation

A machined male thread is present on an older industrial assembly.

Measurement

The pitch is repeatable. The included angle appears close to 60 degrees. The crest appears truncated rather than rounded.

Source routing

Start with ASME’s Sellers history, then compare the target date with National Screw Thread Commission/NBS material and dimensional references appropriate to that period. If necessary, use current dimensional tables to define family geometry—but do not use current publication date as historical evidence.

Competing explanations

1. Sellers/American National lineage
2. Later Unified replacement
3. Other 60-degree standard
4. Shop-cut nonstandard thread

Conclusion

“60-degree family supported; exact standard and installation date unresolved.”

This conclusion is deliberately modest. Profile family is a geometry claim. “Sellers-era bolt” would be a provenance claim and requires contextual proof.

10. Worked Plate B — A Phillips Screw in a Circa-1890 Assembly

Observation

A cross-recess screw appears in an assembly whose principal construction is documented to about 1890.

Record

Head form, recess geometry, thread, finish, tool marks, washer/nut population, neighboring fasteners and the condition of the female thread or hole.

Chronology clue

The Phillips patent was filed in 1935 and granted in 1936. An untouched 1890 installation is therefore not a viable explanation.

Competing explanations

• later repair
• retrofit
• interpretive restoration
• reused component

Conclusion

“Later intervention is strongly indicated; the date and purpose of the intervention remain unresolved.”

This is the right way to read a mixed-generation object. “Not original” does not mean “not historic.” The replacement itself may document decades of use.

11. Worked Plate C — Magnolia Plantation Wood-Screw Cotton Press

National Park Service and HAER documentation identify the Magnolia Plantation wood-screw cotton press as an 1830s-era survivor. HAER LA-11 describes the press as structurally integrated into the gin barn. NPS interpretation explains that the lower press rotated around a stationary vertical screw.

National Register documentation adds useful mechanical detail: the carved screw has a near-perfect helix, the lower portion rotated clockwise, and the press likely produced bales in roughly the 400–500 lb range. The same documentation treats the original motive power cautiously. Draft-animal power is probable, but the exact original arrangement is not documented with certainty.

That uncertainty matters.

Observation

Massive timber frame and vertical carved screw integrated into a pressing mechanism.

Documentation targets

• screw orientation
• helix
• moving press box/platen
• structural connections
• evidence of rotation
• later machinery generations

Source routing

Use NPS site history, HAER LA-11 photographs/measured drawings/data, and National Register documentation as separate evidence layers.

Competing explanation

Do not turn “probably mule-powered” into a fact merely because the mechanism could accept animal power.

Conclusion

“1830s wood-screw press strongly supported; exact original motive-power arrangement unresolved.”

This is one of the most important habits in mechanical history: a mechanism can be confidently identified even when one part of its operating system remains uncertain.

12. Confidence Grades

TS-A — Directly documented object/part; dimensions and provenance agree.

TS-B — Strong converging physical and documentary evidence.

TS-C — Good family identification, incomplete provenance.

TS-D — Partial clues with live alternatives.

TS-E — Evidence insufficient or too damaged for responsible classification.

Example TS-C wording

“60-degree inch-thread family supported; exact standard and installation date unresolved.”

The grade is not a score for the object. It is a score for the quality of the conclusion.

13. Research Routing

When the object itself stops answering questions, move outward in a deliberate order.

First: standards history

Use standards organizations and federal handbooks to understand what systems existed and how terminology changed.

Second: maker and product evidence

Catalogs, manuals, parts books, patents and trade literature can show what was offered or specified.

Third: object-specific evidence

Maintenance logs, invoices, photographs, measured drawings, inventories, conservation records and institutional files can connect an actual part to an actual intervention.

Fourth: comparative population

Compare neighboring fasteners or surviving examples from the same machine, building, production run or institutional collection.

Fifth: state the limit

If geometry is strong but provenance is weak, say so.

14. Safety Boundary

The Turning Screw is an identification and history resource. It is not an engineering replacement specification.

It does not authorize reuse, load rating, torque selection, repair, fabrication, re-threading, re-tapping or replacement selection.

Screw jacks, lifting hardware, structural connections, pressure systems, vehicle components, rotating machinery, electrical assemblies and critical guards require current standards, manufacturer instructions and qualified professional judgment.

Do not disassemble energized, loaded, pressurized, moving or structurally critical equipment for the sake of identification.

15. Sources

ASME — The United States Standard Screw Threads
https://www.asme.org/about-asme/engineering-history/landmarks/234-the-united-states-standard-screw-threads

NBS Handbook H28 (1944)
https://nvlpubs.nist.gov/nistpubs/Legacy/hb/nbshandbook28-1944.pdf

NBS Miscellaneous Publication 89 — National Screw Thread Commission
https://nvlpubs.nist.gov/nistpubs/Legacy/MP/nbsmiscellaneouspub89.pdf

National Archives RG 167 — Records of the National Screw Thread Commission
https://www.archives.gov/research/guide-fed-records/groups/167.html

DLA — FED-STD-H28 umbrella
https://quicksearch.dla.mil/qsDocDetails.aspx?ident_number=53600

DLA — FED-STD-H28 Section 2 Unified Inch
https://quicksearch.dla.mil/qsDocDetails.aspx?ident_number=53604

DLA — FED-STD-H28 Section 12 Acme
https://quicksearch.dla.mil/qsDocDetails.aspx?ident_number=53613

U.S. Patent 2,046,839 — Phillips
https://patents.google.com/patent/US2046839A/en

U.S. Patent 960,244 — Allen
https://patents.google.com/patent/US960244A/en

National Park Service — Magnolia Plantation Wood Screw Cotton Press
https://www.nps.gov/places/magnolia-plantation-wood-screw-cotton-press.htm

Library of Congress / HAER LA-11 — Magnolia Plantation Cotton Gins and Presses
https://www.loc.gov/item/la0404/

← Return to The Library