Archive for the ‘Armor Testing’ Category

Over the past six months, there has been a great deal of both excitement, and lately concern, regarding the Armour Wear AR680 plate. Touted as a “level III+” plate, it is claimed to stop the extremely dangerous M193 high-velocity threat.

In the past few months, extremely un-scientific tests on Youtube seemed to “prove” that it was prone to failure when shot by M193 @ 3200 fps.

Unfortunately, Armour Wear did not originally release a very scientific test video themselves.

At this juncture, I have not seen proof either way, either validating or disproving the efficacy of the AR680 plates. Simply because both the proponents (the company in particular) and the detractors (youtube channel) did not take the small amount of extra time and effort to arrange a proper test.

A proper test is *NOT*:

Setting up a bunch of plates on a berm at a 45 degree angle and blazing away willy-nilly.
Setting up a huge sheet of the steel (again, at a range), and (again), blazing away.
Clamping the plate to a rigid fixture, with no backing, and shooting it.

To properly test body armor, hard or soft, requires the use of a backing. The NIJ specifies no.1 Roma Plastalina modeling clay. Any semi-flexible backing will do, as long as it is close in consistency to a human body. The reason for this is two-fold: first, to be able to determine how much energy (backface deformation) is being imparted to the wearer. Secondly (and for the purpose of this post, more importantly), to mimic the physics of the armor being worn.

A plate that is clamped to a rigid fixture will behave differently than one that is resting on a flexible surface. A rigid plate will have no give, and the round will transfer more energy to the plate. With a proper backing, the initial impact will be reduced ever so slightly.

For some armor (soft armor in particular) this will make the difference between complete penetration, and performing as designed (setting a soft armor vest against a plywood or other hard surface enables it to be penetrated with ease). This will also have relevance with hard armor, especially if it is near its failure threshold.

In the same way, propping a plate at an angle will allow it to stop far more than at 0 degrees of obliquity. MBT armor is sloped for this same reason.

As a result of the above, I will be performing a scientific (or at least, much more so than has been performed so far) comparitive shoot test on the Armour Wear AR680 and Maingun Patriot 2 Advance plates. I had contacted Spartan Armor in an attempt to source one of their level III+ plates to include in the test, but have not heard back from them.

It is my hope that this test will settle any arguments once and for all regarding M193 high velocity protection. Stay tuned!

In the previous post, an oft-repeated Internet legend regarding .22LR and light body armor was examined. .22LR has a reputation as a very high penetrating round, more so than .45 ACP. In this post, the results of an objective shoot test to determine the validity of that legend are posted.

The outcome was quite informative

As mentioned earlier, four panels (two 7-layer Level I equivalent, and two 12-layer Level IIA equivalent) were constructed. A block of #1 Roma Plastalina modeling clay was used as the backing, both to provide the requisite yielding surface for proper functioning of the armor, and to act as a witness panel for purposes of backface deformation/penetration evaluation.


Round used was the Remington Viper Hypervelocity 36gr. copper-washed truncated cone round, with a listed MV of 1410 FPS (out of a 20″ barrel).

Test platforms were a 4″ barrel and a 16″ barrel.



First up was the 4″ barrel and level I panel. Not surprisingly, the round was stopped by the first layer of material. Backface deformation was 11.65mm (for reference, the NIJ allows soft armor up to 44mm of backface deformation and still pass). Note the unburned powder near the impact.



Next up was the 4″ barrel and level IIA panel. Even less surprising, the round was stopped in the first layer. Backface deformation was 11.23mm. Note the crater was wider than the level I impact, showing that the force was spread over a larger area due to more fibers being involved in the arrest of the round.



Next up was the 16″ barrel and level IIA panel. Out of a 16″ barrel, this round is really moving (at least 1300 fps). The round penetrated four layers of material, and was stopped by the fifth. Backface deformation was 12.55mm.




Finally, the test everyone was waiting for: the level I panel and 16″ barrel. To dispense with the suspense, the round penetrated. It penetrated all 7 layers, with major fragments caught by the 7th. A surprisingly deep cavity was created (most likely due to fragments and expanding muzzle gases) 68mm deep into the clay.

So, thus ends (hopefully) internet rumors surrounding soft armor penetration by .22LR. What can be gleaned from this test: level I armor will stop what it is rated to stop, at least as far as .22LR.

Even though HV rounds were used, out of a 4″ barrel they cannot achieve a full powder burn, and so the velocity does not exceed the 1050 fps limit stipulated by level I. However, level I SHOULD NOT be relied upon to stop .22LR from a barrel longer than approximately 10″ (the point at which the velocity threshold is exceeded). As this test demonstrates, reading the specs for your armor is VERY IMPORTANT.

I do not recommend the use of level I armor, unless there are NO other alternatives. As can be seen, the extra layers of level IIA make a tremendous difference in terms of round-stopping ability. IIA should be considered the absolute MINIMUM for soft armor, and level I be retired as a ballistic rating.

For several years it has been debated whether or not .22 LR could or could not penetrate soft armor vests.

Originally, in the 1970’s, Kevlar soft armor was developed to protect Officers against common street threats, which typically were .22 LR LRN, .32 ACP LRN, and .38 SPL LRN. Threat level I vests were certified to stop these ubiquitous, low-velocity threats. As time went on, the threats escalated, and more powerful rounds became common, necessitating thicker vests. But, with the advent of the internet, rumors persisted and spread that called into question whether or not the .22 LR could pose a valid threat to lower level soft armor (I and IIA).

Level I soft armor is seldom seen in this day and age. Typically comprising between 6 and 10 layers of woven aramid, it lacks the thickness to provide sufficient protection against backface deformation. Surprisingly, Level I armor will often stop rounds such as .45 ACP hardball @ 850 FPS, or even .40 S&W. However, these rounds leave a very large backface signature, regardless of whether they are stopped by the armor.

.22 LR has the distinction of being a very good penetrator, primarily due to basic physics- it has a very small frontal area, and can achieve relatively high velocities (1400 FPS from a 16-20″ barrel is not unheard of in certain loadings, I.E. CCI Velocitors). However, it is not a jacketed round, and therefore deforms fairly easily.

The NIJ specifications for Level I call for it to be able to stop .22LR LRN at or below 1050 FPS. Now, it is very important to note the velocity threshold- most longer barrels (above 10″ or so) push .22 LR above this velocity, and therefore can be expected to defeat level I armor. The ongoing debate on the Interwebs rages, but without paying much attention to the distinction between .22 out of a short barrel (handgun) vs. a long barrel (rifle). In order to (hopefully) put this debate to rest, I am posting a test.

This test is aimed at settling the longstanding debate on whether .22 LR is a threat to lower rated soft armor (I and IIA) Furthermore, it seeks to establish whether it is a viable threat only in longer barrels, or both long and short barrels, when faster ammo/more pointed rounds are used.

For this test, the ammo used is Remington Viper 36gr. Hypervelocity round, which features a solid copper washed truncated cone lead bullet and a stated MV of 1410 fps (out of a 20″ barrel). This round was chosen as the shape is more conducive to penetration (smaller frontal cross section). This is fired from a 16″ and a 4″ barrel, from 12 inches.

The panels (two level I and two level IIA) are identical, and will each be shot only once to allow for their full ballistic potential to be evaluated vs. each barrel length. The level I panels comprise 7 layers of Kevlar 29, while the level IIA panels comprise 12 layers of the same material. The backing material is Roma Plastalina #1 modeling clay, to allow backface deformation/penetration to be evaluated and observed.

Stay tuned for the results.

In several of my posts, I mention that while UHMWPE UD armor is an excellent choice for certain applications, and has material advantages over woven or laminate Aramid ballistic fabrics (higher potential V50, positive buoyancy, UV resistant, waterproof), it suffers from several glaring weaknesses (degrades to complete ineffectiveness above 170F, no breathability, delaminates/curls, and is WEAK AGAINST CONTACT SHOTS).

It is important to reiterate that last weakness: a large number (if not the majority) of self-defense and duty scenarios take place at 0-5 feet, where contact shots are a high likelihood. Woven Kevlar soft armor has shown to provide EXTREMELY good protection against contact shots (defined as the muzzle of the weapon being in physical contact with the vest or armor panel). The point at which Kevlar chars is around 500F, and it will retain its strength below this temperature.

The failure mechanism for UHMWPE in contact shots is the high temperature propellant gases that exit the muzzle microseconds after the bullet. These gases heat the area surrounding the muzzle and bullet path, and cause the laminate to melt/denature. This allows the bullet to penetrate much further than would normally be possible. In the case of large caliber revolvers (with a large muzzle blast footprint), this can allow the round to completely defeat the vest.

Test Round

Test Round

For the sake of the test, the round chosen was the .357 Magnum, rather than a .44 Magnum, as I wanted to see if the (relatively!) more modest caliber would still defeat the level II ballistic panel. The panel consisted of 15 layers of Dyneema SB-38. Round chosen was Hornady Custom 158 gr. XTP @ 1250 fps muzzle velocity, from 6″ barrel. The level II panel is specced to stop an equivalent round.

Test Panel

Test Panel



The panel was placed against a backing material consisting of 8 layers of bubble wrap, covered in a dish towel. In retrospect, this was probably a bit too “springy,” giving the panel an advantage by permitting it to move away from the hot muzzle blast faster than if the armor was being worn.

Test Panel Ready To Shoot

Test Panel Ready To Shoot

The test panel and backing were placed upon the ground, and the muzzle pressed firmly (but not forcefully) against the surface. The round was discharged into the center of the panel.

First shot

First shot

First shot, through the backing

First shot, through the backing

First shot, rear of panel...

First shot, rear of panel…

The panel was defeated, showing that the muzzle blast had melted a moderately large area around the point of contact. The round penetrated the backing and buried itself into the dirt (as shown).

To verify, a second round was fired into the lower left area of the panel (away from the heat affected zone of the first round). The second round performed identically, burying itself into the dirt beneath the panel.

Second Contact Shot

Second Contact Shot

Second Round, showing penetration

Second Round, showing penetration

Second shot, layers peeled back

Second shot, layers peeled back

Second shot, showing exit into backing...

Second shot, showing exit into backing…

...And into the dirt

…And into the dirt


The results of the test shows that UD UHMWPE laminates are at risk vs. contact shots. Heat from the muzzle gases (especially medium to large caliber revolvers) “blazes a trail” so to speak, for the round to penetrate further than it normally would.

As many of my readers know by now, steel plates offer tremendous advantages (low cost, extreme durability/multiple hits, thin profile), but do suffer from issues with front-face spall (fragmentation caused by defeated rounds) and weight. Most solutions that attempt to address this issue exacerbate the weight, and negate the advantage of a thin profile.

D-Rmor Gear spall guards address the issue while remaining lightweight and keeping the thin steel plate profile, while stopping 95-98% of all front face spall. The newest version exhibits improved extreme angle spall capture, as shown by the test below.

The plate used was the excellent Maingun Surplus Patriot plate, in flat profile. The plate was placed at a 12 degree angle (to focus the splash pattern more strongly toward the upper edge where the neck/throat would be). This is closer to a worst-case scenario shot, since the residual velocity of the spall is higher.

The test rounds were M855, 20″ barrel, 3000 fps @ 10 FEET. The witness method was a cardboard box. The shoot consisted of two shots, the first with the D-Rmor Gear Version 4.2.3 spall guard installed, the second with the bare plate. The pictures below show the witness box after the first and second shots for direct comparison purposes.

Spall guard and plate mounted in witness material.

Spall guard and plate mounted in witness material.

Seconds after the strike.

Seconds after the first round impact.

Immediately after shot.

Immediately after first shot.

Right side of spall guard after shoot.

Right side of spall guard after first shot.

Top edge of spall guard after shoot.

Top edge of spall guard after first shot.

Upper left edge of spall guard, showing capture of M855 steel core.

Upper left edge of spall guard after first shot, showing capture of M855 steel core.

M855 steel core capture.

M855 steel core capture.

Interior of witness material, first shot, spall guard installed.  Note near complete absence of any spall whatsoever.

Interior of witness material, first shot, spall guard installed. Note near complete absence of *any* spall whatsoever.

Top side witness material, immediately after first shot, SPALL GUARD INSTALLED.

Top side witness material, immediately after first shot, SPALL GUARD INSTALLED.

Right side witness material, spall guard installed.

Right side witness material, first shot, SPALL GUARD INSTALLED.

Left side witness material, spall guard installed.

Left side witness material, first shot, SPALL GUARD INSTALLED.

As you can see, the shot with the guard resulted in nearly total capture of all spall, INCLUDING THE STEEL CORE. The core was captured by the spall arrest material, as can be seen in the upper left angle of the plate. M855 is particularly nasty as a threat, because the spall is lead, copper, AND the steel core, which tends to remain in one piece. In a plate carrier, none of the spall would have had sufficient energy to escape.

Some things to consider when viewing this test: front face spall will punch through wooden target stands, both sides of steel aerosol cans from several feet away, and deeply indent angle iron. The spall guard caught the overwhelming majority of this extremely energetic frag while being UNDER 5 oz. weight and containing NO metal itself.

Top edge, post shoot, NO GUARD INSTALLED- note large amount of spall cutting in chin and throat area. Compare to first pic with guard installed.

Top edge, second shot, BARE PLATE NO GUARD INSTALLED- note large amount of spall cutting in chin and throat area. Compare to first pic with guard installed.

Right edge of witness material, showing massive spall cutting.

Right edge of witness material, second shot, BARE PLATE NO GUARD INSTALLED, showing massive spall cutting.

Left side of witness material, second shot, BARE PLATE NO GUARD INSTALLED, showing massive spall cutting.

Left side of witness material, second shot, BARE PLATE NO GUARD INSTALLED, showing massive spall cutting.

Another shot of top edge, showing high energy of spall fragmentation leaving the box.

Another picture of top edge, second shot, BARE PLATE NO GUARD INSTALLED, showing high energy of spall fragmentation leaving the box and spall cutting.

Right side of witness material.

Right side of witness material, NO GUARD INSTALLED, again note sheer volume of fragmentation and spall cutting without the guard to catch it.

The second shot (bare plate/no guard), in contrast, shows massive spall/fragmentation. So dense is the pattern that it often looks like a saw or cutting tool was used on the cardboard. The majority exits along the edge of the plate, and as can be seen, would have impacted the neck/throat and arms.

Moral of the story? Spall guards are essential when running steel plates.

The Maingun plates are quite impressive. The two M855 rounds impacted almost in the same spot, with only a slight indentation and barely detectable bump on the backside. I will be performing a torture test on this same plate to see how many M855 rounds it can soak up before penetration or cracking, and I have a feeling it will be in the hundreds.

Overall, the new version 4.3.2 guard performed exactly as designed, capturing and mitigating the cloud of high velocity metal that is produced when a round hits steel. Check them out here:

The new style curved Patriot plates can be found here:

Stay tuned for further shoot tests!

It is no big mystery that I am at best a reluctant advocate of UHMWPE in soft armor. While the material itself *does* have incredible properties, these properties come at a steep price if the end user is not aware of the limitations and weaknesses inherent in the material.

These limitations and weaknesses are exacerbated by (in MY OPINION, ahhh, I promised you would see that word hurled around here!) a tendency to “softball” the armor test protocols. Even the current “best practices” protocols (The FBI and DEA tests for soft armor), have this same inherent kid glove treatment when it comes to UHMWPE containing vests.

“How can this be?” you may ask. Well, let’s review:

UHMWPE (“Ultra High Molecular Weight Poly Ethylene”) is an exceedingly strong material made up of long chains of ethylene molecules. The tensile strength is astounding, exceeding para-aramid (Kevlar/Twaron) and steel easily. It is positively buoyant, waterproof and does not degrade with exposure to UV light (three of the Achillies heels of aramids). However, as has been mentioned before, UHMWPE (regardless of brand- both Spectra and Dyneema are at their root the same basic molecule) will denature when exposed to temperatures exceeding ~168 F.

Think of it as exposing hardened steel to it’s normalization (annealing) temperature. The hardness disappears, and it becomes soft again. Unlike steel, it is impossible to change the UHMWPE back to its “super” state. In its denatured state, the material is identical to the stuff used to make milk jugs.

This denaturation temperature is well-known.

In the real world, temperatures often climb to well above this temperature, in both storage and incidental use (especially hotter regions of the world where .Mil users often find themselves). Why then are the test protocols seemingly designed to AVOID this issue?

Take the current NIJ 06 protocol. It incorporates many new rigors that an armor must pass in order to be certified (a GOOD thing, no doubt), including environmental conditioning. However, the temperature does not exceed the KNOWN denaturation/transition temp of the UHMWPE (highest temp in the conditioning phase is 149 F):

Even the FBI protocol, which excels the NIJ 06 standard in many ways, still only exposes the armor to a MAX temp of 140 F:

Quite frankly, this is a ludicrous state of affairs. Since temperatures can *regularly* reach 200 F in a car trunk or APC on hot days in CONUS or OCONUS, to not expose armor to these realistic circumstances could be perceived as softballing.

Even recent tests done by DSM which supposedly showed that their UHMWPE material did OK in elevated temperatures STILL tiptoed around the transition threshold temp:

In the paper, “temperatures up to 90 C (194 F)” were used. However, these temps were *only* applied to hard/rigid armor, which does not experience the same sensitivity as soft armor, most likely due to an insulative effect of the material in thicker section. The soft armor samples were still only exposed to 70 C (which is 158 F, 10 degrees BELOW the known transition temp). The test should have applied the same max temp to ALL samples, regardless of whether they were hard or soft.

In conclusion, test protocols should be designed to apply REAL WORLD rigors to life saving equipment. The current protocols SEEM TO incorporate temperature threshold requirements that allow the limitations and weaknesses of a specific material (UHMWPE) to pass. Designing tests with less rigorous standards so as to avoid excluding a material does not live up to the purpose of testing in the first place. Providing the absolute best lifesaving gear, regardless of any other considerations, should always be the goal.

Ever since the late 15th century, with the advent of powder-propelled projectile weapons (and indeed, pre-dating that time with crossbows), armor smiths have sought a way to ensure their product will resist (within reasonable limits) the injurious tendencies of fast moving bits of metal.

Smiths making plate armor would often shoot the finished product (with crossbow bolts prior to lead muzzle loader round balls arriving on the scene), and upon visual confirmation of a successful “stop,” would engrave their maker’s mark, showing the armor had passed. This methodology was (and is) referred to as “bench testing” or “proof testing.” An article of protective gear is subjected to one or more ballistic events, and when it either passes or fails, provides the maker, and the end user, with evidence that it is effective, or “proofed.” Many modern firearms bear similar proof marks after being subjected to equivalent testing.

Bench testing is still used today, and while it has certain advantages, it also has some notable drawbacks. For instance, if large numbers of finished articles need to be produced, it becomes ungainly to batch test each lot (a neccesary requirement to verify efficacy of the finished product). It is also subject to the whims of either the maker or the end user. Bench testing is much more suited to custom, or small-batch manufacturers of armor.

Bench testing remained the norm until the 70’s, when it became necessary to certify large numbers of concealable vests. The National Institute of Justice (NIJ) scrambled to come up with a way of testing and certifying large quantities of vests. The “NIJ Rating” methodology was created, which should be familiar to everyone with any exposure to armor. The rating levels go from I to IV, and are directly proportional to what threats are stopped. I-IIIA are for soft armor, while III and IV relate to hard (rifle) armor. The tests, in a nutshell, subject a batch of test armor articles to successive rounds of ballistic testing. This ranges from 2 to up to 240 rounds fired, and newer iterations of the tests have become more stringent. However, there are some issues with NIJ testing, which will be discussed in a later post.

In just the past few years, two new test protocols have appeared on the scene: the FBI and DEA test protocols. Publicly released in 2006, the FBI protocol was a vast improvement over the NIJ protocol, subjecting the armor to much more realistic and useful tests. In addition to simply shooting the armor, the FBI/DEA protocols subject the armor to extreme heat, cold, immersion as well as conditioning the armor and subjecting it to flash/flame. While there are still some issues (one in particular that is shared by the NIJ tests), they are far superior to the earlier protocols.

Moving forward, there is still room for improvement regarding armor test standards. This has been a brief overview of the most typical test protocols for modern body armor. In future posts, more detailed analysis of each protocol will be given. Thanks for reading.