2021 test13 productsHand Tools
Which Allen Keys Brand Wins?
We compared 13 allen keys options head to head. Wiha and Wera Hex+ (tied as the narrator's named top performing picks) came out on top. See the measured results, the runner-up, the budget pick, and a link to the full test video.
Winner
Wiha and Wera Hex+ (tied as the narrator's named top performing picks)
Price shown in test: $42 for 9 wrenches, or $4.67 each
Check price on Amazon
Affiliate link. We may earn a commission at no extra cost to you.
Budget pick
Craftsman
Price shown in test: $15 for 20 pieces, or 75 cents each
Check price on Amazon
Affiliate link. We may earn a commission at no extra cost to you.
The measured results
Every number below is read straight from the test. Scroll sideways to see all measurements. Products are listed in the order they finished.
| Product | Claimed specs | Country of origin | 5/32 test (inch pounds before losing grip) | Ball end test (inch pounds) | 3/32 test (inch pounds) | 5/16 test (foot pounds) | Fit | Corrosion resistance | Corrosion |
|---|---|---|---|---|---|---|---|---|---|
| 1Wiha$38 for 13 wrenches, or $2.92 each | 20 percent longer keys for hard to reach areas, precision machined for exact fit, hardened chrome vanadium tool steel, hex ball end allows up to a 25 degree working angle | manufactured in Poland, designed to German standards | 174, first place overall, 18 inch pounds more than the second place finisher; held up exceptionally well with only minor damage to two sides | 104, second place, broke | 39, suddenly broke with quite a bit of damage, including splintering | 138.3, first place overall, 15 foot pounds more than any other brand; did not begin twisting until around 130 foot pounds despite quite a bit of eventual damage | described repeatedly across every test as offering the best fit of all 13 brands, with very little side to side slop or play | not tested | not tested |
| 2Wera Hex+ Hex Plus$42 for 9 wrenches, or $4.67 each | patented Hex Plus design providing surface contact claimed to transfer up to 20 percent more torque than standard hex profiles | Czech Republic | 136, loose fit compared to top performers hurt its result here; held up extremely well with very little visible damage | 106, first place, best of the 9 wrenches with ball ends | 48, first place overall, looks nearly as good as new afterward | 121, fourth position, performed better than average and held up by far the best of any brand, looking nearly as good as new | 5/32 test noted a looser fit than the top performing brands at that stage; by the 3/32 test the fit was described as the best yet | not tested | not tested |
| 3Craftsman$15 for 20 pieces, or 75 cents each | strong steel hex keys with a black oxide coating for durability, durable handle turns L keys into comfortable T handles, full lifetime warranty | China | 155, third place, held up better than Kobalt | 95, tied for third place with Bondhus, broke | 35, held up the best of any brand tested to that point with only a very small amount of rounding | 123.2 to 123 (stated slightly differently in two places), nearly tying EPAuto, some twisting but less rounding than most other brands | not tested | not tested | not tested |
| 4Gedore$30 for 8 pieces, or $3.75 each | zinc plated to help prevent corrosion, chamfered nose ends for easy use, chrome vanadium steel | Germany | 156, second place behind Wiha at the time it was tested | not tested | 31, zinc coating showed some damage | 110.3, slightly below average, zinc coating began peeling away during the test | not tested | the zinc coating did a great job resisting corrosion in the aggressive 12 hour test, but the coating is soft and appears to hurt the tool's grip performance | not tested |
| 5Kobalt$15 for 23 pieces, or 65 cents each | larger dual action handle for stronger grip, easy to see size markings, ball end allows for 20 degree pivot, meets ASME specifications, hassle free lifetime guarantee | Taiwan | 153, fourth place, one of the six sides of the wrench experienced some damage | 87, performed nearly the same as Tekton, held up just fine | 41, tied for first place with Facom at the time it was tested, badly twisted with some rounding | 115.9, second position behind Tekton at the time it was tested, damage about the same as Tekton | not tested | not tested | not tested |
| 6EPAuto$13 for 30 pieces, or 43 cents each | durable, heat treated chrome vanadium steel, includes metric and SAE, corrosion resistant black oxide finish | Taiwan | 146, fifth place, the wrench is soft and experienced quite a bit of damage | not tested | 40, moved into the lead at the time it was tested, badly twisted with quite a bit of damage | 108 before struggling, broke at 110.9, seemed quite a bit softer than most of the other brands | not tested | not tested | already showed signs of corrosion forming under the black oxide finish even before the dedicated corrosion test |
| 7Facom$36 for 8 wrenches, or $4.50 each | black oxide finish to help resist corrosion | France | 145, better than average, held up really well with very little damage | 91, about average, the ball end broke | 41, tied for first place with Kobalt at the time it was tested, held up well but not quite as well as Craftsman or Bondhus | 111.7, gave up a little early, quite a bit of damage and rounding to all sides | not tested | not tested | not tested |
| 8PB Swiss$140 for 12 wrenches, or $11.67 each, the most expensive brand tested | tolerances held to the width of a human hair for dimensional accuracy, longer handle and shorter key for comfortable reach, PB spring steel for toughness and elasticity, unlimited lifetime guarantee | Switzerland | 143, very close to average, held up very well with very little visible damage | 90, about average in both the 5/32 and ball end rounds | 34, gave up a little early, but looked as good as new afterward | 108.2, did not perform as well as some other brands on raw torque, but experienced very little damage and looked nearly as good as new | not tested | performed very well in the 12 hour aggressive corrosion test, called out by name alongside Gedore | not tested |
| 9Bondhus$21 for 22 pieces, or 95 cents each | proprietary Protanium steel claimed to be up to 20 percent stronger than the competition, ProGuard finish claimed up to 5 times more effective at stopping rust | USA | 136, slightly better than Husky, held up better than average with a very small amount of damage | 95, tied for third place with Craftsman, did not break (unlike Craftsman) | 36, performed about the same as Husky, held up really well with very little visible damage | 119.4, performed nearly as well as Tekton and Craftsman, quite a bit of rounding like most other brands | not tested | not tested | not tested |
| 10Husky$18 for 22 pieces, or 82 cents each | precision formed to exact tolerances, does not have a ball end (unlike most other brands, intended to resist rotational distortion), chamfered edges to guide the tool into the fastener | product of Taiwan, finished in mainland China | performed slightly better than Eklind, but the transcript does not state the actual number; a caption gap where the figure was dropped entirely, kept as omitted rather than guessed | not tested | 36, performed better than average, damage about the same amount as Eklind | 117.2, third position behind Craftsman at the time it was tested, quite a bit of rounding | not tested | not tested | not tested |
| 11Tekton$15 for 26 pieces, or 58 cents each | ball end allows up to 25 degree offset, chamfered inserts smoothly into fastener head to reduce wear, claims to be precisely sized | China | 139, performed nearly as well as EPAuto, quite a bit of damage but not as much as EPAuto | 91, broke | 37, a little more damage than Pittsburgh | 123.6, by far the best result at the time it was tested (before Wiha's later 138.3), second place overall behind Wiha, held better than EPAuto with less rounding and twisting | not tested | not tested | not tested |
| 12Eklind$18 for 22 pieces, or 82 cents each (stated in the transcript directly after this brand's own price introduction of 92 cents each for $24 across 26 pieces; kept as stated, see note) | ball end allows up to a 35 degree angle, burr free short arm allows users to apply maximum torque, manufactured using Eklind alloy steel | USA | 130, or 25 inch pounds less than Craftsman, quite a bit of rounding and damage to all sides | 87, broke, struggled in both the 5/32 and ball end rounds | 31, quite a bit more damage than Craftsman | 107.5, gave up quickly, seemed to be made of a pretty soft metal with quite a bit of rounding and damage | not tested | not tested | not tested |
| 13Pittsburgh$8 for 36 pieces, or 22 cents per inch, the least expensive brand tested | includes SAE and metric, standard length and long reach, chrome vanadium steel, black oxide finish to resist rust, lifetime warranty | China | 121, lowest result among brands with usable data in this test (Husky's number is missing), held up fairly well with very little damage | not tested | 32, very small amount of damage | 108.5, quite a bit of rounding and damage | described as offering a poor fit with lots of side to side play, sold at Harbor Freight | not tested | not tested |
How it was tested
- 5/32 inch allen key gripping strength test: torque adapter measures maximum torque before the key breaks or loses grip in a fixture holding a 304 stainless steel fastener
- ball end torque test: maximum torque before the ball end of the key breaks or loses grip, only for the 9 of 13 brands that include a ball end design
- 3/32 inch allen key gripping strength test, using a different, smaller fastener
- 5/16 inch allen key gripping strength test, using a heavier duty torque adapter, measured in foot pounds
- corrosion resistance test: an aggressive corrosion agent of hydrogen peroxide, vinegar, and salt applied twice over 12 hours
“if you're looking for the top performing brands i really like the weira and the wheel now the wheel seems to provide the best fit but the wira seems nearly indestructible”
Data notes and caveats
This transcript has severe phonetic overlap between two separate brands, Wiha and Wera Hex+, both captioned with very similar sounding strings throughout (we a brand, the wheel, wease, wii, wia, weira, weir, rear hex plus). Each occurrence was resolved independently by testing order and by cross checking against the recap leaderboard sentence for each test, which in three separate tests (5/32, ball end, 3/32, and 5/16) lists both brands' names and numbers together in the same sentence, confirming they are distinct entries rather than one brand. Wiha won the two largest raw torque tests (5/32 and 5/16) outright and consistently had the best physical fit; Wera Hex+ won the 3/32 and ball end tests and consistently showed the least wear and damage of any brand. The narrator explicitly names both as his top picks in the closing verdict, which is why winner captures both rather than picking one. Husky's 5/32 test result is missing its number entirely in the transcript (a caption gap, not a garbled figure) and was omitted rather than guessed. The corrosion test gives no brand specific numeric ratings beyond naming Gedore and PB Swiss as standouts and stating that black oxide coated brands performed roughly the same as each other.