Field Fortifications: How thick was thick enough?

British soldiers digging a piquet dug-out during the second Boer War 1899 - 1902 © IWM Q 72508

The ability to construct quick but effective fortifications in the field has historically fallen to the soldiers of the Royal Engineers. This dirty and backbreaking work may not be precise, but it is based very soundly on experience and scientific experimentation. During the 19th and 20th centuries, as weaponry and warfare evolved at a staggering rate, the data provided to frontline engineers also evolved. I have drawn on an extensive collection of military manuals spanning from 1888 through to the modern day to collate this data and publish it in a single place.

The data is for field fortifications such as trenches, and does not include permanent fortifications such as stone built forts. The majority of data reflects protection against small-arms or rifle fire.

Field Works by Major General C.B. Brackenbury (1888)

Protection in the field against artillery

The following table of data was extracted from the Manual of Elementary Field Engineering and updated for inclusion into the 1888 publication.

GunConcrete ButtSand & Clay ButtRammed ClayExtreme PenetrationAdvised Parapet Thickness
7-inch B.L. rifled gun11863 Newhaven experiments7ft 9in12ft 11in18ft 3in21ft 11in25 to 30 feet
70-pr B.L. rifled gun21863 Newhaven experiments6ft 4in14ft 3in17ft 0in25 to 30 feet
40-pr B.L. rifled gun31863 Newhaven experiments6ft 1in11ft 8in16ft 4in18 to 25 feet
20-pr B.L. rifled gun41863 Newhaven experiments7ft 0in10ft 3in13ft 3in15 to 18 feet
12-pr B.L. rifled gun51863 Newhaven experiments3ft 2in4ft 0in6 to 9 feet
10-inch S.B. gun61863 Newhaven experiments3ft 10in11ft 0in12ft 0in15 to 18 feet
8-inch S.B. gun71863 Newhaven experiments3ft 6in11ft 5in12ft 9in15 to 18 feet
68-pr S.B. gun81863 Newhaven experiments4ft 0in7ft 6in14ft 10in21ft 6in25 to 30 feet
32-pr S.B. gun91863 Newhaven experiments2ft 8in5ft 8in9ft 5in14ft 0in18 to 25 feet
9-pr M.L. gun101882 Shoeburyness experiments4ft 0in9ft (Sand)
12ft (Medium Soil)
15ft (Clay)
16-pr M.L. gun111882 Shoeburyness experiments6ft 0in9ft (Sand)
12ft (Medium Soil)
15ft (Clay)
13-pr M.L. gun121882 Shoeburyness experiments7ft 0in9ft (Sand)
12ft (Medium Soil)
15ft (Clay)
12-pr B.L. gun131885 Lydd experiments17ft 0in9ft (Sand)
12ft (Medium Soil)
15ft (Clay)
22-pr B.L. gun141885 Lydd experiments9ft (Sand)
12ft (Medium Soil)
15ft (Clay)
Mean penetration data for projectiles fired at 1,060 Yards.

Protection against rifle bullets

The following table gives the thickness of material which may be considered proof against rifle bullets of existing service arms (in 1888!) at any range:

Protective MaterialThickness Required
Earth2 feet
Iron Plate⅜ inch
Steel Plate¼ inch
Fir Logs12 inches
Fir planks, 3-inch thick6 planks (18 inches)
Pack Logs6 inches
Oak planks, 2-inch thick3 planks (6 inches)
Gabions filled with earth1 gabion
Filled sandbags, crossways2 sandbags
Filled sandbags, lengthways1 sandbag
Rope mantlet6 inches
Loose cotton4 feet
Compressed cotton in a bale2 bales

The image below from the Imperial War Museum shows a destroyed field fortification from the mid-C19. Experiences such as this would have helped inform measures for ballistic protection. This fort has been constructed primarily using wicker gabions that would have been earth filled. They have also utilised hessian sand bags for additional protection in areas.

THE CRIMEAN WAR 1854-56 (Q 71081) View of the ruined interior of the Great Redan, Sevastopol. Copyright: © IWM. Original Source

Manual of Military Engineering (1901)

The parapet of an infantry trench should be thick enough to keep out a rifle bullet – 2 feet will suffice. Overhead cover was not considered necessary at this time as the enemy’s artillery fire would be frontal, however to protect against oblique fire a thickness of 6 to 9 inches of earth supported by brushwood was advised. The oblique cover was intended to keep out shrapnel, bullets or shell splinters.


Military Engineering, Field Defences (1908)

The following table gives the thickness, in various materials, proof against a bullet from the short Lee-Enfield Service Rifle at 30 yards range. The bullets of some continental armies have, however, greater penetration.

MaterialThickness ProofRemarks
Clay5 feetVaries greatly. This is maximum for greasy clay
Earth, free from stones (un-rammed)3 feetRamming earth reduces its resisting power
Sand2ft 6inRather more than enough. Very high velocity bullets have less penetration in sand at short than at medium ranges
Sand between boards18 inches
Brickwork9 inchesIf well built.
Soft wood (fir)48 inches24 inch proof at 500 yards
Hard wood (oak)27 inches15 inch proof at 500 yards
Wrought iron, or mild steel½ inch
Hardened steel plate¼ inch1/16 inch proof at 600 yards
Special hard steel1/5 inch
Shingle6 inches
Coal (steam)2ft 6in
Chalk1 footWhen freshly excavated

The 1.457-in Vickers-Maxim gun (pom-pom), which may be taken as an example of light quick-firing gun, fires a 1 lb common shell, with a bursting charge of 340 grains of black powder. Its penetration into wrought iron is 2.25 inches at the muzzle. The bursting charge of the shell is so small that its effect against earthworks is insignificant.

While no data is given for the action of artillery shells, a useful plate was included in this manual showing the effect of anti-infantry artillery shells.

The caption read: Plate I gives an idea of the action of various kinds of projectiles. It will be observed that practically the only one which has any backward effect after burst is the howitzer common shell, fired at high angle of elevation.

Manual of Field Engineering (1911)

Rifle fire

The following table gives the maximum penetration of the pointed bullet in various materials. In order to obtain proof cover, a percentage must be added to these numbers (this number is unspecified in the manual).

MaterialMaximum PenetrationRemarks
Steel plate, best hard7/16 inchAt 30 yards normal to plate; 3/16 inches is proof at not less than 600 yards, unless the plate is set at a slope of 3/2 when 3/16 inches is proof at 250 yards.
Steel plate, ordinary mild or wrought iron¾ inchAs above
Shingle6 inchesNot larger than 1in ring gauge
Coal, hard9 inches
Brickwork, cement mortar9 inches150 rounds concentrated on one spot will breach a 9 inch brick wall at 200 yards
Brickwork, lime mortar14 inchesAs above
Chalk15 inches
Sand, confined between boards or in sandbags18 inchesVery high velocity bullets have less penetration in sand at short than at medium ranges
Sand, loose30 inchesAs above
Hard wood (oak, with grain)38 inches
Earth, free from stones (unrammed)40 inchesRamming earth reduces its resisting power
Soft wood (fir)58 inchesPenetration of brickwork and timber is less at short than at medium ranges
Clay60 inchesVaries greatly. This is maximum for greasy clay
Dry turf or peat80 inches
Logs are used to protect a building in Mercatel. This photograph was taken in 1915 © IWM Q 51085

Tests such as the one below helped to educate and guide the policy on levels of protection required for troops in the field and for equipment being developed for the battlefield. For these, it was essential to have examples of the enemy ammunition and weapons to conduct realistic tests.

MINISTRY OF INFORMATION FIRST WORLD WAR OFFICIAL COLLECTION (Q 14636) Experiment with 3/4inch mild steel armour plate for armoured cars, outside view. One German ‘S’ bullet penetrated point first, another failed point first, and a third ‘reversed’. Wormwood Scrubbs, 1915. Copyright: © IWM. Original Source

Artillery fire

Field guns

Both shrapnel shell and high explosive shell are fired by the field artillery of most foreign nations. Shrapnel with time fuzes can be used up to a range of about 6,000 yards. With percussion fuzes shrapnel can be used effectively against troops behind 14 inch brick or 2 feet thick mud walls as they penetrate before bursting.


Manual of Field Works (All Arms) (1925)

MaterialPenetrationMinimum thickness to be providedRemarks
Steel plate¾ inches1 inch
Shingle or broken stone6 inches9 inchesIf pieces do not measure more than 1 inch placed between boards
Coal (hard)9 inches13 inchesBetween boards
Coal (kitchen)15 inches23 inchesBetween boards
Brickwork in lime mortar14 inches21 inches
Chalk15 inches22 inches
Sand, confined between boards or in sandbags18 inches27 inches
Sand, loose30 inches45 inches
Earth, free from stones, unrammed40 inches60 inchesRamming earth reduces its resisting power
Sawn timber, hard wood (oak)38 inches57 inches
Soft wood (fir)56 inches84 inches
Freshly cut timber logs, 12-in diameter and over24 inches36 inches
Poles 4.5 to 8 inches in diameter38 inches57 inches
Clay60 inches90 inchesVaries greatly. This is a maximum for greasy clay
Dry turf or peat80 inches120 inches
Snow, rammed60 inchesVaries greatly. Soft snow has little power of resistance

Tunnelled dug-outs

Provided that the ground is suitable and that the site is not too close to the enemy, tunnelled dug-outs are the most satisfactory type of shell-proof accommodation on any scale.

Complete protection against shells of large calibre (8-inch and over) can seldom be gained without descending to impracticable depths. The thickness of cover required under these circumstances will be:

MaterialThickness
Made earth35 feet
Clay30 feet
Gravel25 feet
Chalk25 to 20 feet
Hard rock15 feet

Royal Engineers Pocket Book (1936)

Protection against the rifle bullet

The data in this publication is the same as the data published in the Manual of Field Works (All Arms) 1925, given above.

Artillery shells

Shrapnel

The bullets come down at a steep angle and have very little power of penetration. A brick wall 9 inches thick, a bank of earth 18 inches thick or the roof or floor of a good building will be sufficient to stop them.

High Explosive (H.E.) shells with instantaneous fuze

The shells burst directly they touch the ground. The effect is mainly lateral, and is stopped by a 9 inch wall or a bank of earth 2 feet thick.

H.E. shells with non-instantaneous fuzes

These burst after penetrating for some little distance, and are of more value against material than against personnel. The force of the explosive tends to shatter surrounding material. Good protection against the effect of splinters from shells can be afforded by slit trenches.

Little can be done in hasty defences to protect against direct hits of these shells, as the amount of material needed for safety is too great; in deliberate defences, dug-outs can be made deep enough to afford protection.


Field Service Pocket Book, Field Engineering (1944)

The following table shows the thickness of various materials which should be provided to give protection against bursts of 20 rounds armour piercing bullets from LMGs up to 7,92mm or against splinters from 100-lb bombs bursting not less than 30 feet away.

MaterialSafe Thickness
Earth of loam as in parapets60 inches
Chalk as in parapets60 inches (variable)
Clay as in parapets72 inches (variable)
Sand, loose or between boards30 inches
Brick rubble confined between boards18 inches
Coal between boards24 inches
Road metal 1.5 – 2 inches between boards14 inches
Sandbags, filled with rubble30 inches
Sandbags, filled with earth30 inches
Sandbags, filled with road metal20 inches
Sandbags, filled with shingle20 inches
Sandbags, filled with sand30 inches
Brickwork in lime mortar18 inches
Concrete, unreinforced12 inches
Mild steel plate1 ¾ inches
Timber60 inches (variable)

Royal Engineers Pocket Book (1979)

MaterialBulletsSplinters, shell fragmentsGamma radiation (to at least 75% reduction)
Ordinary soil1.5m45cm45cm
Compacted sand (sandbags)75cm45cm45cm
Brick45cm45cm25cm
Timber1.5m20cm (variable)90cm (variable)
Concrete30cm12cm30cm
Mild steel4cm1cm8cm
Loose snow3.5-4cm
Packed snow1.8-2cm
Packed and frozen snow1.2cm

This information was also replicated in the Commanders Pocket Book, Tactical Aide Memoire 1988.

Comparison Table

A good way to represent the advancement of weapons and ammunition is to plot the increase in protective measures required.

Material188819081911192519441979
Earth243640406060
Gravel / shingle666
Chalk12151560
Clay60606072
Sand loose303030
Sandbags1818183030
Soft wood184858566060
Hard wood6273838
Iron0.40.50.750.75
Mild steel0.50.7522
Hardened steel0.25
Concrete1212
Brickwork, lime mortar9141418
Brickwork, cement918
Dry turf, peat8080
All measurements in inches using the data tables in this article. The 1979 measurements in metric have been converted and rounded-up to imperial.