Advertising disclosure: the Amazon links on this page are paid links. As an Amazon Associate, top7tools.uk earns from qualifying purchases at no extra cost to you. Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on amazon.co.uk at the time of purchase will apply.
Rankings re-checked weekly · Last re-checked 19 September 2026
·Prices re-verified on refresh·AI-assisted, run by Jaco Olwagen
Body only. The wrench on this page that publishes both of its numbers - up to 406Nm fastening and 610Nm breakaway - out of a 125mm body at 1.1kg bare, with a hog ring anvil and four modes including one that stops hammering when the nut breaks free.
Why it's number one: The only wrench here that publishes both of its torque figures against a model code, 406Nm fastening and 610Nm breakaway, out of a 125mm body at 1.1kg bare, with a mode that stops impacting the moment a fastener breaks free and the one exact-model review on this page behind it at 9.7 out of 10.
8.4Top7 Score
Check price
✓Checked yesterday
★ 4.6(223 reviews)
✓ Ships from the UK✓ Free delivery with Prime
Ad· we may earn a commission if you buy through this link, at no extra cost to you.
This is the one to buy if you already run 18V XR, and it is top of the list because of what DeWalt is willing to put its name to. The instruction manual and dewalt.co.uk both publish up to 406Nm of maximum tightening torque and up to 610Nm of maximum breakaway torque against the model code, alongside 0 to 2,500rpm, 0 to 3,550 impacts per minute, a 415W power output and 1.1kg without a battery in a body DeWalt lists at 125mm front to back. Every one of those figures is checkable, which is more than half this shelf can say.
The four-mode switch is what makes it a wrench rather than a hammer with a square on the end. Speed 1 runs to 1,400rpm and Speed 2 to 2,500rpm. Precision Wrench drives forward at 2,500rpm until the moment impacting starts, pauses for half a second and then impacts at 3,550 per minute, which is the point at which most people overtighten; in reverse it impacts until it senses the fastener has broken free, then cuts the impacting and slows down so a loose nut does not spin off the stud and into the gravel. The fourth is Scaffolding mode, which winds a clamp nut down, impacts for a set period and then cuts off rather than keep going, and winds it back off the same way. Scaffolding work is the obvious use, but it is the behaviour you want on any fastener where the tool should stop by itself.
The anvil is a hog ring: push a socket on, pull it off, no pin to line up, which is the faster of the two retention systems when you are cycling through 17, 19 and 21mm on a driveway. There is also a through-hole so an O-ring and retaining pin can be fitted if you want belt-and-braces retention overhead. Three LED rings cycle low, high and off. It is a bare tool - no battery, no charger, no case, and a belt hook is what comes in the box - so it only makes sense if you own XR packs already. Pro Tool Reviews scored the Atomic 1/2in wrenches 9.7 out of 10 in October 2021, naming the DCF921 and DCF922 together, and that is the only published review of an exact model on this page. DeWalt gives one year as standard and three when the tool is registered.
Torque
7.5
Control
10
Build
9
Owners
8
Owners is the 4.6★ Amazon UK rating from 223 reviews, weighted toward the category average when there are few. Fewer than two published reviews of this exact model, so the owner rating carries that weight too. Tick marks show the category average.
How we scored it
Torque 7.5: DeWalt publishes 610Nm maximum breakaway, which is the card's 500 to 649Nm band at 7, plus half a point because it also publishes a rated 406Nm fastening figure against the model code rather than a loosening number alone.
Control 10: Four modes, and Precision Wrench meets both halves of the card's 10 anchor: it pauses half a second at first impact going on and stops impacting on break-free coming off. Scaffolding mode adds a fourth setting that cuts off by itself.
Build 9: 125mm body and 1.1kg bare clear the card's 10 anchor on size and weight, with a three-LED ring and a hog ring anvil, held one point down because DeWalt publishes no sealing rating and no LED run-on for this model.
+ Publishes both torque figures against the model code: up to 406Nm fastening and 610Nm breakaway, with 0 to 2,500rpm and 0 to 3,550ipm
+ Precision Wrench mode pauses at first impact going on and stops impacting when the fastener breaks free coming off, so a loose nut does not run off the stud
+ Scaffolding mode winds a nut down, impacts for a set period and cuts off instead of hammering until you let go
+ 125mm body and 1.1kg without a battery: it gets behind a wheel arch and inside an engine bay
+ Hog ring anvil swaps sockets one-handed, with a through-hole for an O-ring and retaining pin if you want more retention
+ Three-year DeWalt warranty once registered, and the widest professional battery platform on this list
Cons
− Body only: no battery, no charger and no case, so it costs more than the sticker unless you already run 18V XR
− 610Nm of breakaway is well under what the marketplace tools claim, and on a genuinely seized or over-torqued fastener that is the honest limit
− DeWalt publishes no dust or water sealing rating and no LED run-on for this model
− Vibration is declared at 13.0 m/s squared, the only vibration figure published for any tool on this page and a reminder that this is a hammering tool
18V XR Brushless406Nm Fastening610Nm Breakaway125mm BodyHog RingBody Only
Body only, and the same 406Nm fastening and 610Nm breakaway figures from the same 125mm body as the pick above - the difference is a detent pin anvil that locks a socket on and will not let go until you push the pin.
8.2Top7 Score
Check price
✓Checked yesterday
★ 4.6(617 reviews)
✓ Ships from the UK✓ Free delivery with Prime
Ad· we may earn a commission if you buy through this link, at no extra cost to you.
Mechanically this is the DCF921 with a different nose. DeWalt publishes identical figures for both: up to 406Nm maximum tightening torque, up to 610Nm maximum breakaway, 0 to 2,500rpm, 0 to 3,550 impacts per minute, 415W and 1.1kg without a battery, in the same 125mm body. It carries nearly three times the owner ratings of the hog ring version on Amazon UK, which is worth knowing when review depth is a third of the score.
The anvil is the decision. A detent pin goes through a hole in the socket and the socket does not come off until you depress the pin, which is what you want with a long extension, working overhead, or anywhere a socket falling off would mean fishing it out of a subframe. It is slower to change than a hog ring, and it only accepts sockets with a matching pin hole, which most 1/2in impact sockets have and some do not. If your day is wheels on a driveway, the hog ring above is quicker. If it is underneath a vehicle on axle stands, this is the safer end.
The mode switch also differs. Precision Wrench is here and behaves the same way - a half-second pause at first impact in forward, and impacting that cuts out on break-free in reverse so a loosened nut does not spin down the thread and off. Speed 1 and Speed 2 are the same. The fourth setting is Timber Screw mode rather than Scaffolding, tuned for lag bolts and structural wood screws, so it is a mode for a different trade rather than a second way of controlling a nut. Worth flagging: the Amazon UK listing title for this one reads as a machine translation and calls it a cordless impact screwdriver, which is exactly the confusion the buying guide below is about. It is a 1/2in square drive impact wrench, body only, three-year DeWalt warranty once registered.
Torque
7.5
Control
9.5
Build
8.5
Owners
8
Owners is the 4.6★ Amazon UK rating from 617 reviews, weighted toward the category average when there are few. Fewer than two published reviews of this exact model, so the owner rating carries that weight too. Tick marks show the category average.
How we scored it
Torque 7.5: Identical published figures to the DCF921: 610Nm maximum breakaway puts it in the card's 500 to 649Nm band at 7, plus half a point for also publishing a rated 406Nm fastening torque against the model code.
Control 9.5: Four modes with Precision Wrench meeting both halves of the card's 10 anchor, held half a point down because the fourth setting is Timber Screw, a wood-screw mode, rather than a second mode that controls a nut.
Build 8.5: Same 125mm body and 1.1kg bare as the DCF921 with the same three-LED ring, half a point lower because the card's 10 anchor asks for an anvil that swaps sockets easily and a detent pin is the slower of the two.
The only tool at this end of the shelf with a model code on the box, four modes and a stated hog ring anvil, supplied with two 4.0Ah packs - and a 682Nm headline that DCA publishes no fastening figure alongside.
6.8Top7 Score
Check price
✓Checked yesterday
★ 4.6(72 reviews)
✓ Ships from the UK✓ Free delivery with Prime
Ad· we may earn a commission if you buy through this link, at no extra cost to you.
This is the occasional-DIY pick, and it earns the slot by being the one tool at this end of the shelf that behaves like a product rather than a listing. It carries a model code, ADPB520, that appears on DCA's own catalogue and on distributor pages in several markets, so the tool exists as a line rather than as a photo. The Amazon UK listing publishes four working modes, a Smart Loosen function, a 13mm hog ring anvil, an LED and two 4.0Ah packs on DCA's own 20V platform. That is more published detail than anything else at this level here, and it is why it sits above the two tools on this page quoting bigger numbers than its own.
What it does not publish is a fastening torque. The headline is 682Nm of maximum torque, with no statement of whether that is measured going on or coming off, no test standard, and no data sheet on a UK site to check it against. The same 682Nm appears word for word on a different brand's listing further down this page, which tells you roughly how much of a measurement it is. Scored honestly, the number buys nothing, so the card holds it at the band for a tool that does car wheel nuts and general automotive fastening and takes half a point off for the missing fastening figure. For the job most people buy this for, that band is the truthful one: our own wheel nut chart puts almost every UK car between 110 and 140Nm, and a tool that reliably cracks those loose is doing its job.
Smart Loosen is the interesting claim - the listing says it stops once a nut is loosened, which is the run-off protection the DeWalts above do properly. Nobody publishes at what point it stops or how, so treat it as a listed feature rather than a measured one. Two batteries and a charger in the box means it works the day it arrives, which the DeWalt bodies above do not. It is a marketplace purchase from a brand with no UK service network, so returns run through Amazon rather than a repair agent.
Torque
5.5
Control
7.5
Build
6.5
Owners
7.9
Owners is the 4.6★ Amazon UK rating from 72 reviews, weighted toward the category average when there are few. Fewer than two published reviews of this exact model, so the owner rating carries that weight too. Tick marks show the category average.
How we scored it
Torque 5.5: 682Nm is published as maximum torque with no fastening figure, no test standard and no reachable data sheet, so the card's verification clause holds it at the 6 band for car wheel nuts and general automotive work, less half a point for the missing fastening figure.
Control 7.5: Four modes and a Smart Loosen function put it at the card's 8 anchor for three or four modes with one auto mode, less half a point because nothing published says at what point Smart Loosen stops or how it senses break-free.
Build 6.5: Held at the card's 6 band because no body length, bare weight or sealing rating is published, plus half a point for being the only tool at this level that states its anvil type, a 13mm hog ring, and publishes an LED.
Pros
+ Has a real model code, ADPB520, so the tool can be looked up rather than existing only as a listing
+ Four modes plus a stated Smart Loosen function, the most control published by any tool here outside the two DeWalts
+ Listing states the anvil type, a 13mm hog ring, which nothing else at this level does
+ Two 4.0Ah packs, a charger and a case: nothing else to buy
+ Compact design and an LED, on DCA's own 20V platform
Cons
− 682Nm is quoted as maximum torque with no fastening figure beside it and no data sheet on a UK site to check it against
− The identical 682Nm figure appears on another brand's listing on this page, which is not a sign of an independent measurement
− Smart Loosen is described but not specified: nothing published says at what point it stops
− No UK service network, so a fault is an Amazon return rather than a repair
− 20V DCA batteries fit nothing else you are likely to own
20V Brushless682Nm Claimed13mm Hog Ring4 Modes2 x 4.0AhMarketplace Brand
Makita DTW190Z 18V LXT 1/2in Impact Wrench with 3.0Ah Battery and Charger
18V LXT190Nm Fastening320Nm Nut-Busting
Premium
Rank #4
Makita DTW190Z 18V LXT 1/2in Impact Wrench with 3.0Ah Battery and Charger
Our Verdict
The gentle one, and the only pick here bundled with a battery and a charger. Makita publishes 190Nm fastening and 320Nm nut-busting, M8 to M16, and describes it as a tool for changing car wheels.
6.7Top7 Score
Check price
✓Checked yesterday
★ 4.9(10 reviews)
✓ Ships from the UK✓ Free delivery with Prime
Ad· we may earn a commission if you buy through this link, at no extra cost to you.
Makita's own product sheet calls the DTW190Z a handy wrench that is ideal for car wheel bolts, and the published numbers back that up rather than overshoot it: up to 190Nm maximum fastening torque, 320Nm nut-busting, 0 to 2,300rpm, 0 to 3,000 impacts per minute, a 1/2in square drive and a tightening capacity of M8 to M16 standard or M8 to M12 high tensile. Makita lists 1.4 to 1.8kg with a battery fitted, depending which pack you hang on it, with a rubber-coated grip, an LED and an electric brake.
That 320Nm of nut-busting is the lowest figure on this page and it is the reason to buy it or the reason to skip it. Our wheel nut chart puts almost every UK car between 110 and 140Nm and a Transit at 200Nm, and breaking a fastener loose takes meaningfully more than it took to tighten, so 320Nm covers a car that has been done up sensibly and runs out of road on one that has been seized by salt or leaned on with an air gun. What it buys you in exchange is a tool that physically cannot do the damage a 1,200Nm claim implies, on the LXT platform, with Makita's UK service behind it. For a household that changes to winter wheels and back and nothing else, that is a sensible trade.
Two caveats. First, the control is basic: one speed range on the trigger, forward and reverse, an electric brake and a light. There is no mode switch, no auto-stop and nothing that slows down when the nut breaks free, so the operator is the only thing between the gun and an over-tightened stud. Second, this ASIN is a seller-assembled bundle. The DTW190Z code is Makita's body-only tool and its own sheet says battery and charger are not included, so the 3.0Ah pack and charger here have been added to the box by the seller. Makita UK's extra two years of cover is registered within 30 days of purchase and explicitly excludes batteries and chargers, so the tool is covered and the bundled parts are not. The listing also carries only a handful of owner ratings, which the scoring shrinks towards the category mean rather than treating as a verdict.
Torque
6.5
Control
4.5
Build
7.5
Owners
8.1
Owners is the 4.9★ Amazon UK rating from 10 reviews, weighted toward the category average when there are few. Fewer than two published reviews of this exact model, so the owner rating carries that weight too. Tick marks show the category average.
How we scored it
Torque 6.5: Makita publishes 320Nm nut-busting, which is the card's under-500Nm band at 6, plus half a point for also publishing a rated 190Nm fastening figure and an M8 to M16 capacity against the model code.
Control 4.5: The card's 4 anchor exactly: one speed range of 0 to 2,300rpm on a variable trigger with no mode switch and no auto-stop, plus half a point for the published electric brake.
Build 7.5: Published 1.4 to 1.8kg with a battery, a rubber-coated grip and an LED put it in the card's 8 band, less half a point because Makita publishes no body length, no bare weight and no sealing rating for this model.
Pros
+ Makita publishes both figures for the exact model: 190Nm maximum fastening and 320Nm nut-busting, with M8 to M16 capacity
+ 0 to 2,300rpm and 0 to 3,000 impacts per minute, with an electric brake and an LED
+ Listed at 1.4 to 1.8kg with a battery, the lightest tool here in use
+ A battery and a charger in the box, which neither DeWalt above includes
+ 18V LXT platform with Makita UK service behind it, and three years on the tool once registered within 30 days
Cons
− 320Nm of nut-busting is the lowest on this page and will not shift a genuinely seized or over-torqued fastener
− No mode switch, no auto-stop and nothing that backs off when the nut breaks free: one speed range and a trigger
− A seller-assembled bundle, not a Makita kit code - the DTW190Z is body only and Makita's extra cover excludes the battery and charger
− Makita's own sheet describes the motor as brushless while UK trade retailers list the DTW190 as brushed, so this page does not state a motor type
− Makita publishes no body length and no bare weight for this model
18V LXT190Nm Fastening320Nm Nut-BustingM8-M16Battery and ChargerElectric Brake
A named manufacturer at the bottom of the shelf, with three-speed control and a single 4.0Ah pack - quoting the same 682Nm maximum loosening torque as the DCA above it, and no fastening figure either.
6.3Top7 Score
Check price
✓Checked yesterday
★ 4.5(120 reviews)
✓ Ships from the UK✓ Free delivery with Prime
Ad· we may earn a commission if you buy through this link, at no extra cost to you.
Dong Cheng is a real tool manufacturer rather than a brand invented for a listing, and that is most of the case for it here. The Amazon UK listing publishes a 20V brushless motor, a 1/2in drive, three-speed control, a 4.0Ah battery and 682Nm of maximum loosening torque, and that is the lot: no model code, no length, no weight, no impacts per minute and no fastening torque.
The 682Nm is worth pausing on, because the DCA listing further up this page quotes exactly the same figure. Two listings, two brands, one number. Whatever process produced it, it was not two independent measurements of two different tools, and it is a good illustration of why this page does not rank on the headline. What the listing does say honestly is that 682Nm is a loosening figure, which is more than most of this shelf bothers with, and loosening is the job an impact wrench is actually for.
Three-speed control is a genuine step above a bare trigger: it lets you run a nut down on the lowest setting rather than hammering it home at full noise. It is not an auto-stop and it will not sense break-free, so the last turn still belongs to a torque wrench. One battery rather than two means no swap when it runs flat mid-job, and the pack fits nothing else. Take it for what it is: the cheapest tool here from a maker with a name on it, for a car owner doing wheels two or three times a year.
Torque
5
Control
6.5
Build
6
Owners
7.6
Owners is the 4.5★ Amazon UK rating from 120 reviews, weighted toward the category average when there are few. Fewer than two published reviews of this exact model, so the owner rating carries that weight too. Tick marks show the category average.
How we scored it
Torque 5: 682Nm is a loosening claim with no fastening figure, no test standard and no model code to look it up against, so the card's verification clause holds it at the 6 band and takes half a point for the missing fastening figure and half for the missing code.
Control 6.5: Three-speed control sits above the card's 6 anchor for two-speed selection but below the 8 anchor, because none of the three settings is an auto or precision mode and nothing senses break-free.
Build 6: Held at the card's 6 band with nothing to move it either way: the listing publishes no weight, no length, no anvil type and no sealing rating, and compact design is not a dimension.
Pros
+ From a named tool manufacturer rather than an anonymous marketplace listing
+ Three-speed control, so a nut can be run down on a low setting instead of at full speed
+ Labels its 682Nm as maximum loosening torque rather than leaving it ambiguous
+ A 4.0Ah battery included, and a compact body for the money
+ Costs less than every pick above it
Cons
− The same 682Nm figure appears on the DCA listing on this page, so it is not an independent measurement
− No fastening torque, no model code, no weight, no length and no impacts per minute published anywhere
− One battery only, so there is no swap when it goes flat
− No auto-stop and nothing that backs off on break-free
− 20V platform that fits nothing else you are likely to own, and no UK service network
The complete-kit pick and the biggest claim on the page: two 4.0Ah packs, five sockets, a fast charger and a case, sold on 1200Nm that nothing published anywhere supports.
6.0Top7 Score
Check price
✓Checked yesterday
★ 4.6(413 reviews)
✓ Ships from the UK✓ Free delivery with Prime
Ad· we may earn a commission if you buy through this link, at no extra cost to you.
This is the one to look at if you own no battery platform and want everything in a box: two 4.0Ah packs, a fast charger, five sockets and a carry case, aimed by its own listing at cars, vans and machinery. It holds 4.6 stars across more than four hundred Amazon UK ratings, the strongest combination of rating and volume of any marketplace-brand wrench here, which is why it is ranked at all rather than left with the interchangeable listings named in the methodology.
The 1200Nm headline is the problem. There is no fastening figure beside it, no statement of how it was measured, no model code and no manufacturer data sheet anywhere to check it against. Written out, 1200Nm is roughly nine times what a Vauxhall Corsa wheel bolt is torqued to and six times a Transit's. Even taken as a breakaway figure in reverse against a rigid test joint, which is what these numbers usually are, it sits in the class the wheel nut chart reserves for commercial vehicle wheels and genuinely seized fasteners, from a tool that costs a fraction of what that class normally costs. It may well shift a nut that the Makita above will not. What it cannot do is prove it, so the score reflects what is published rather than what is claimed.
Treat what you actually get as the reason to buy: a 21V kit, sockets included, nothing else to purchase, and a brand with enough ratings behind it to suggest the tools arrive working. Check the sockets are impact rated before you put them on the gun, and read the buying guide below on why that matters. As with everything at this end, there is no UK service network, so a failure is an Amazon return.
Torque
5
Control
4
Build
6.5
Owners
8
Owners is the 4.6★ Amazon UK rating from 413 reviews, weighted toward the category average when there are few. Fewer than two published reviews of this exact model, so the owner rating carries that weight too. Tick marks show the category average.
How we scored it
Torque 5: 1200Nm is published with no fastening figure, no test standard, no model code and no manufacturer data sheet, so the card's verification clause holds it at the 6 band and takes half a point for the missing fastening figure and half for the missing code.
Control 4: The card's 4 anchor, because the listing publishes no mode count, no speed settings and no auto or precision mode - there is nothing here to score above a variable trigger.
Build 6.5: Held at the card's 6 band because no weight, length, anvil type or sealing rating is published, plus half a point for five sockets, a fast charger and a carry case being listed in the box.
Pros
+ The most complete box here: two 4.0Ah packs, a fast charger, five sockets and a carry case
+ 4.6 stars across more than four hundred Amazon UK ratings, the best rating-and-volume combination of any marketplace-brand wrench ranked here
+ 21V platform with two batteries, so there is a swap rather than a wait
+ Aimed at vans and machinery as well as cars, which is the heavier end of home vehicle work
Cons
− 1200Nm is a bare claim: no fastening figure, no test standard, no model code and no manufacturer data sheet to check it against
− No published weight, length, anvil type, no-load speed or impacts per minute
− No mode count or speed settings published, so there is no evidence of any auto-stop or precision control
− No UK service network: a fault is an Amazon return, not a repair
− Sockets supplied with a kit are not always impact rated, and a chrome socket can shatter on a gun
21V Brushless1200Nm Claimed2 x 4.0Ah5 SocketsCase IncludedMarketplace Brand
The cheapest way onto this page and the one with the most owner ratings of any budget listing here, quoting 700Nm and 515 ft-lb - and publishing almost nothing else.
5.1Top7 Score
Check price
✓Checked yesterday
★ 4.1(436 reviews)
✓ Ships from the UK✓ Free delivery with Prime
Ad· we may earn a commission if you buy through this link, at no extra cost to you.
The case for this one is volume and outlay. It has more owner ratings than anything else at the bottom of this shelf, and it costs less than every other pick by a wide margin, which makes it the honest answer for someone who changes a wheel once a year and does not want a tool sitting in a cupboard costing more than the job.
It also publishes less than anything else ranked. The listing gives a 1/2in drive, a brushless motor and a maximum torque of 700Nm, quoted again as 515 ft-lb, which at least converts correctly and suggests one figure rather than two invented ones. Beyond that there is no fastening torque, no model code, no weight, no length, no anvil type, no no-load speed, no impacts per minute and no mode count. There is nothing to verify, so the score says so rather than filling the gaps in.
Its owner rating is also the lowest on this page, and at this many ratings that is a signal rather than noise: the scoring shrinks ratings towards the category mean, and this one still comes out well below everything above it. Read that as a tool that works for a lot of people and disappoints a meaningful minority. For a once-a-year wheel change on a car that has not been seized up by a tyre fitter, it is enough. For anything that lives outdoors, gets salted, or has to come apart on a schedule, buy further up the page.
Torque
5
Control
4
Build
5.5
Owners
5.7
Owners is the 4.1★ Amazon UK rating from 436 reviews, weighted toward the category average when there are few. Fewer than two published reviews of this exact model, so the owner rating carries that weight too. Tick marks show the category average.
How we scored it
Torque 5: 700Nm is published with no fastening figure, no test standard and no model code, so the card's verification clause holds it at the 6 band and takes half a point for the missing fastening figure and half for the missing code.
Control 4: The card's 4 anchor: the listing publishes no mode count, no speed settings and no auto or precision mode, so there is nothing to score above a variable trigger.
Build 5.5: Held at the card's 6 band and taken half a point down because this listing publishes less than any other pick here: no weight, no length, no anvil type, no LED and no statement of what is in the box.
Pros
+ Costs less than every other pick here by a wide margin
+ The deepest owner review base of any budget listing on this shelf
+ Quotes its torque in two units that convert correctly, 700Nm and 515 ft-lb, rather than two unrelated numbers
+ Brushless motor on a 1/2in square drive, which is the right configuration for wheel nuts
Cons
− The lowest owner rating on this page, from a large enough sample that it is a signal rather than noise
− Publishes no fastening torque, no model code, no weight, no length, no anvil type and no speed or impact figures
− No mode count or speed settings, so nothing that backs off on break-free
− The listing does not state what comes in the box, so check before ordering
− No UK service network: a fault is an Amazon return
Nothing on this site is bought, tested or handled. Each wrench is scored against a published rubric - Torque and power 45 per cent, Control and modes 30 per cent, Build and ergonomics 25 per cent - from specifications published by the manufacturer or printed on the Amazon UK listing, plus owner ratings at scale and reviews of the exact model from named outlets where they exist. DeWalt's figures come from dewalt.co.uk and from the DCF921/DCF922/DCF923 instruction manual published on service.dewalt.co.uk; Makita's come from the DTW190Z and DTW300 product sheets on makitauk.com; the Milwaukee codes come from milwaukeetool.eu. Everything else comes from the listing itself and is labelled as such.
The torque rule carries a verification clause, added the way the circular saws card was extended for unbranded tools. A torque figure scores in the card's bands when a manufacturer publishes it against a model code. A marketplace listing's maximum loosening torque, with no maker standing behind it, no fastening figure beside it and no test standard named, is held at the card's band for car wheel nuts and general automotive fastening whatever the number claimed, with half a point off where no fastening figure is published at all and half a point more where there is no model code to look up. That is why a listing claiming 1200Nm scores below one that publishes 610Nm. Nothing here is scored on a figure borrowed from a sibling model, and where a maker publishes nothing - no body length or bare weight for the Makita, no anvil type or weight for most of the marketplace tools - the gap is written into the score note instead of being filled in.
Three candidates from the same shelf are named here rather than quietly dropped. An unbranded 800Nm kit (B0GDXF8KV1) and a 550Nm kit (B0CSYDFYYL) are interchangeable with the marketplace tools already ranked and publish less than either of them, and TaskStar's 1200Nm wrench (B0DRNSL86V) has the thinnest owner evidence of the ten candidates. Listings advertised as compatible with another brand's 18V batteries are excluded as a block: they are third-party tools, not that brand's. Owner ratings are shrunk towards the category mean, so the Makita listing with a handful of ratings cannot outrank one with hundreds on a rounding difference. One published review of an exact model on this list was found, Pro Tool Reviews on the DeWalt Atomic 1/2in wrenches, which names the DCF921 and DCF922 together, so it is recorded against both of those and against nothing else. Prices move daily and are not quoted in this text: tap through for the current figure on Amazon.
Scored on Torque and power 45%, Control and modes 30%, Build and ergonomics 25%, plus owner ratings and critic reviews
An impact wrench takes them off, a torque wrench puts them on
This is the whole rule and it is worth more than any specification on this page. Use the gun to crack the fasteners and to run them back down until they are snug, then put it away and do the final pass with a calibrated torque wrench, in a star pattern, with the wheel on the ground. What a gun actually delivers on the way in depends on battery charge, how long you hold the trigger, the socket and extension fitted and whether the threads are dry, none of which is repeatable and none of which is measured. Over-tighten and you stretch studs and warp discs; under-tighten and the wheel works loose in service, and neither error announces itself until something goes wrong. DeWalt says it in its own manual. The published figure for your vehicle, the sequence and the re-check interval are all on our wheel nut torque chart at /calculators/wheel-nut-torque.
Nut-busting torque and fastening torque are different numbers, and only one is on most boxes
Fastening torque is what the tool is rated to put into a joint. Nut-busting or breakaway torque is what it can generate in reverse to crack a seized one loose, measured on a test bench under ideal conditions, and it is always the bigger and more flattering figure. Brands do not all measure it the same way, so two tools quoting 900Nm are not necessarily comparable. DeWalt publishes both for the DCF921 and DCF922 at 406Nm and 610Nm; Makita publishes both for the DTW190Z at 190Nm and 320Nm. Most of the marketplace listings publish one number and do not say which kind it is. Buy on the nut-busting figure where you can trust it, because getting fasteners off is the job, and ignore the fastening figure entirely as a tightening guide.
An impact wrench is not an impact driver, whatever the search results say
These two tools are mixed together constantly on Amazon UK, and one listing on this very page is titled as an impact screwdriver when it is a wrench. An impact driver has a 1/4in hex collet, takes screwdriver and nutsetter bits, and is built for screws: DeWalt rates its DCF850 at up to 205Nm, which is the most compact one in the range and still nowhere near a wheel fastener. An impact wrench has a 1/2in square drive, takes impact sockets and is built for nuts and bolts. Put a wheel nut on an impact driver and the hex bit or the collet is what gives way, and 205Nm often will not crack a properly torqued nut loose anyway. If you want the driver, it is ranked on /impact-drivers. If you want wheels off, you want the square drive.
Impact sockets, not chrome ones
Chrome hand sockets are hardened to resist wear and they are brittle under impact loads, which is why one can crack or shatter on a gun with the fragments going somewhere. Impact sockets are black, made from chrome-molybdenum, and designed to deform rather than explode. They are also usually thicker-walled, so check clearance on alloy wheels with deep nut recesses, where a thin-wall impact socket or a protective sleeve is the answer rather than a chrome socket. Deep sockets matter on cars with long wheel bolts. The 1/2in drive impact sets worth pairing with these tools are on /socket-sets.
Hog ring, detent pin or friction ring
The anvil is the bit the socket goes on, and how it holds it is a real decision rather than a detail. A hog ring is a sprung ring that grips the socket: push on, pull off, fastest to change, which is what you want cycling through sizes on a driveway. A detent pin goes through a matching hole in the socket and will not release until you push the pin, which is what you want on a long extension or working overhead where a dropped socket disappears into a subframe. A friction ring, which is what Milwaukee uses on the code mentioned above, works like a hog ring. DeWalt sells the same wrench both ways here, as the DCF921 with a hog ring and the DCF922 with a detent pin, which makes the choice unusually clean. Only one marketplace pick on this page states its anvil type at all.
Body only, kit, and which battery you already own
Both DeWalts here are body only - no battery, no charger, no case - which makes them cheap if you already run 18V XR and expensive if you do not. The Makita listing is a seller-assembled bundle with a 3.0Ah pack and a charger added to a body-only tool code, which is worth knowing because Makita UK's extra two years of cover, registered within 30 days, excludes batteries and chargers. The marketplace kits come with one or two 4.0Ah packs, a charger and sometimes sockets and a case, on platforms that fit nothing else you own. If you already have XR or LXT batteries, buy the body. If you have nothing, price the whole box rather than the tool.
How much torque you actually need
Our wheel nut figures run from around 90Nm on an older Ford up to 200Nm on a Transit, with most UK cars between 110 and 140Nm. Breaking a fastener loose takes considerably more than it took to tighten it, because of rust, road salt, the seat bedding in and the fact that plenty of wheels were last done up by a gun on full noise. A 1/2in wrench with 400 to 600Nm of breakaway covers normal car wheel fasteners and 600 to 800Nm gives margin on corroded or over-tightened ones. The 1000Nm-and-up class is for commercial vehicle wheels and fasteners that have not moved in years. That is the frame to read the claims on this shelf against: the honest tools here sit in the first two bands, and the four-figure claims are for a class of tool that does not normally cost what these cost.
An impact wrench has one job on most British driveways: getting wheel nuts off a car that was last put back together by someone with an air gun set to maximum. It drives a 1/2in square, not a 1/4in hex bit, and it works by hammering rather than pushing, which is why it moves a fastener that a breaker bar and your whole body weight will not. It will also spin those nuts straight back on, and that is where people get hurt and wheels come off. An impact wrench is not a torque wrench. That is not this page's opinion: DeWalt prints "Impact wrenches are not torque wrenches" in the manual for the two tools at the top of this list, and follows it with an instruction not to use them for tightening to a specified figure. Our wheel nut torque chart at /calculators/wheel-nut-torque has the published setting for most UK cars, vans and caravans, the star sequence to work in, which cars use bolts rather than nuts, and the Vauxhall Astra generation trap that puts an Astra H owner 27 per cent over if they use the later 140Nm figure instead of 110Nm.
The second thing to understand is the number on the box. Half this shelf is marketplace brands advertising 800Nm, 1000Nm, 1200Nm and in places 1480Nm, and almost every one of those figures is quoted as maximum loosening torque: a breakaway number taken in reverse against a stiff test joint, with no standard named and no way to compare one brand's method with another's. It is not what the tool puts into a fastener. DeWalt publishes both numbers for the DCF921 and DCF922, up to 406Nm of fastening torque and up to 610Nm of breakaway, in the instruction manual and on dewalt.co.uk, tied to a model code anyone can look up. Makita does the same for the DTW190Z at 190Nm fastening and 320Nm nut-busting. Those figures are far below the marketplace claims, and they are the ones somebody can be held to. Torque is worth 45 per cent of the score here and it rewards a published, attributable figure over a bigger unverifiable one, because there is nothing to check an unverifiable one against. Where a listing quotes a number and no manufacturer stands behind it, the note under the score says so.
Three tools people come looking for are not ranked. The Makita DTW300Z, ASIN B08DHY6CVD, is the deepest-reviewed impact wrench on Amazon UK by a distance at roughly 4,500 ratings, and it has no buy box: no featured offer, nothing to add to a basket. Owners rate it 4.6 stars at that volume, Makita publishes 330Nm of fastening torque across four power settings and its XPT dust and water sealing for it, and none of that can put it on a page that links to Amazon when there is nothing on the page to buy. If you want one, it is a trade counter purchase. The listing badged "Milwaukee 2855-20" is a US catalogue code sitting on a UK page; Milwaukee's European code for that compact 1/2in wrench is M18 FIW2F12, and Milwaukee on Amazon UK is third-party only, with an extended warranty that cannot be registered against a marketplace purchase. A buyer gets neither the UK code nor the cover. And the DeWalt DCF850 turns up constantly in impact wrench search results because the listings blur the two tools: it is an impact driver, it takes 1/4in hex bits, and it is already number one on /impact-drivers.
Last thing before the list. What goes on the end matters as much as the tool. A chrome hand socket is not built for impact loads and can crack or shatter on a gun; impact sockets are the black CR-MO ones, and several kits here ship with sockets that are worth checking before you fit them. The socket sets page at /socket-sets covers the 1/2in drive impact sets that pair with these wrenches, including deep sockets for recessed wheel bolts and the larger axle and hub sizes.
Compare Top Picks
Can't decide between our top two? See how they stack up head‑to‑head.
How much torque do I need to get car wheel nuts off?+
Most UK cars are torqued between 110 and 140Nm and a Transit to 200Nm, but breaking one loose takes considerably more than that because of rust, salt and the seat bedding in. A 1/2in impact wrench with 400 to 600Nm of breakaway handles normal car wheel fasteners, and 600 to 800Nm gives you margin on corroded or over-tightened ones. Above 1000Nm is commercial vehicle territory. The DeWalts here publish 610Nm of breakaway and the Makita publishes 320Nm, which is why the Makita is described as the gentle one. The marketplace listings quote far higher numbers with nothing published behind them. Per-vehicle figures are on /calculators/wheel-nut-torque.
Can I tighten wheel nuts with an impact wrench?+
No, and this is the one thing on this page worth reading twice. Use the wrench to crack fasteners loose and to run them back down until they are snug, then finish with a calibrated torque wrench in a star pattern with the wheel on the ground. What a gun puts into a joint depends on battery charge, trigger time, the socket and extension fitted and whether the threads are dry, and none of that is measured or repeatable. Over-tightening stretches studs and warps discs, under-tightening lets a wheel work loose, and you have no way of knowing which one you just did. DeWalt prints the same warning in its own manual for the two wrenches at the top of this list. Re-check after 25 to 50 miles.
What is the difference between an impact wrench and an impact driver?+
The drive end. An impact driver has a 1/4in hex collet and takes screwdriver bits and nutsetters; an impact wrench has a 1/2in square drive and takes impact sockets. Drivers are built for screws and top out far lower - DeWalt rates its compact DCF850 at up to 205Nm - and the hex bit or collet is what fails if you ask one to shift a wheel nut. Amazon UK mixes the two constantly, to the point that one DeWalt wrench on this page has a listing title calling it an impact screwdriver. If the tool you are looking at has a square post on the nose, it is a wrench. If it has a hex socket you push a bit into, it is a driver, and those are ranked on /impact-drivers.
Why is the Makita DTW300Z not on this list?+
Because it cannot be bought on Amazon UK. ASIN B08DHY6CVD is the deepest-reviewed impact wrench on the site at roughly 4,500 ratings, and the page carries no buy box: no featured offer and nothing to add to a basket. A page that links to Amazon cannot rank a tool with no Amazon offer, however good it is. For the record, Makita publishes 330Nm of maximum fastening torque across four power settings for it, 0 to 3,200rpm, 0 to 4,000 impacts per minute, a 144mm body and its XPT dust and water sealing, with three auto-stop reaction settings in both directions. If you want one, buy it from a Makita dealer or a trade counter rather than waiting for the listing to come back.
Is a Milwaukee impact wrench from Amazon UK covered by warranty?+
Not by the extended cover, no. Milwaukee on Amazon UK is third-party only, and Milwaukee's extended warranty cannot be registered against a marketplace purchase, so what you are left with is the seller's obligations and your consumer rights. There is a second thing to check on those listings: some are US stock carrying US catalogue codes. The listing badged 2855-20 is one of them - the code Milwaukee uses in Europe for that compact 1/2in wrench is M18 FIW2F12. A US code on a UK page means an import, which affects the charger, the paperwork and any service claim. No Milwaukee wrench is ranked on this page for those reasons.
Can I use my normal sockets on an impact wrench?+
No. Chrome hand sockets are hardened for wear resistance, which makes them brittle under impact loads, and one can crack or shatter on a gun. Impact sockets are the black chrome-molybdenum ones, designed to deform rather than fracture, and they are the only thing that should go on a 1/2in wrench. Two practical points: impact sockets are usually thicker-walled, so check they clear the nut recesses on alloy wheels, and deep sockets are what you need for long wheel bolts. Sets are covered on /socket-sets. The same applies to extensions and universal joints, which should be impact rated too.
Should I buy the hog ring or the detent pin version?+
Hog ring if you are changing sockets often on a driveway: push the socket on, pull it off, nothing to line up. Detent pin if you work on extensions, overhead, or anywhere a socket falling off would mean fishing it out of a subframe, because the socket is locked on until you depress the pin. The trade-off is speed against retention, and the detent pin needs sockets with a matching hole, which most but not all impact sockets have. DeWalt sells the identical wrench both ways here, the DCF921 with a hog ring and the DCF922 with a detent pin, with the same published torque, speed, weight and body length, so the anvil really is the only thing you are choosing between.
When is the best time to buy an impact wrench in the UK?+
Black Friday falls on 27 November 2026 and is the deepest discounting of the year on power tools, followed by Boxing Day and the January sales. Late autumn is also when this tool gets bought for a practical reason rather than a promotional one: winter wheel and tyre swaps, and fasteners that have spent a year in road salt. The next window after that is the run-up to Easter on 28 March 2027 and the May bank holidays on 3 and 31 May, and Father's Day on 20 June 2027 moves a lot of mid-range tools. If you are buying through a business, the date that matters more is the tax year ending 5 April and what the Annual Investment Allowance lets you write off.
Found this useful? Share it with someone who needs one.
Share:
Weekly tool deals, in your inbox
We check prices on every tool we cover daily. Once a week we send the biggest genuine price drops we spot - the ones worth adding to your kit.