Independent comparisons. We may earn from shop links. More info

Spirit level buying guide

Updated: 22 October 2026

You have a shelf, a door or a steel stud in mind, and the listing is already shouting 0.5 mm/m before you know how long the tool should be. Length comes first. A true bubble on a short sole still misses a bow, and a printed tolerance is a claim you prove on that level, not a certificate. This spirit level buying guide does not name a winner. Prices and review counts sit on the [spirit level comparison](/hub/measuring-and-marking-tools/spirit-level). Peter F. used the amazon.co.uk files captured on 21 October 2026, plus public notes from BuildWiz, Stabila and legislation.gov.uk. There is no lab test. A qualifying purchase can earn a commission at no extra cost to you.

You have a shelf, a door or a steel stud in mind, and the listing is already shouting 0.5 mm/m before you know how long the tool should be. Length comes first.

Our top picks

Grettz Grettz GL120 1200 mm digital level

Best overall

Grettz Grettz GL120 1200 mm digital level
9.6/10about £155.99
View offer
STANLEY FATMAX Stanley FatMax 250 mm torpedo

Rank 2

STANLEY FATMAX Stanley FatMax 250 mm torpedo
9.4/10about £14.99
View offer
Firecore Firecore 230 mm magnetic torpedo

Best value

Firecore Firecore 230 mm magnetic torpedo
9.2/10about £6.53
View offer
On this page
  1. 1.What 0.5 mm/m does and does not promise
  2. 2.Match the length to the work
  3. 3.Body, magnet and the units on the box
  4. 4.What to do if the claim is wrong

What 0.5 mm/m does and does not promise

Half a millimetre over one metre is what ±0.5 mm/m means. On a 250 mm torpedo it says nothing about the timber the sole missed. Stabila publishes 0.5 mm/m in the normal position and often 0.75 mm/m to 1.0 mm/m in reverse. That is not a UK statute. The Stanley bullets state 0.5 mm/m and both Teronbeta listings state ±0.5 mm/m. The Firecore, Franklin, XunHe and Urban Trends listings state no millimetres-per-metre figure. The Grettz states degrees. The URASISTO set prints 0.0005-inch/ft, about 0.042 mm/m, which is not 0.5 mm/m. The Measuring Instruments (Material Measures of Length) Regulations 2006 class tapes and rules, not bubbles. The method is on how to use a spirit level.

Match the length to the work

About 200 mm to 300 mm covers a picture, a socket or a short bracket: the Firecore at 230 mm, the Stanley at 250 mm, and the Franklin at about 305 mm. The XunHe is only 95 mm. About 400 mm to 600 mm suits a shelf. The Teronbeta 400 mm has a clean title. The one titled 500 mm has a dimension line of 50 cm by 50 cm, so measure it. A door or a stud wants about 1200 mm. This set has no classic 1200 mm bubble box. The only tool of that length is the Grettz, with six ratings. Tight gaps are on the torpedo spirit level page.
Grettz Grettz GL120 1200 mm digital level

Testsieger

Grettz Grettz GL120 1200 mm digital level
9.6/10about £155.99
View offer

Body, magnet and the units on the box

This set has no classic box beam, so you are choosing aluminium against ABS, and a 210 g Teronbeta against a 90 g Firecore. One UK review found the 400 mm body too thick for blinds near a ceiling. A magnet grips ferrous steel only, and paint can weaken the hold. Do not infer a V-groove from the word magnetic. The Amazon bullets here do not state one. A 45° vial is the wrong tool for a 1:40 drain fall, which is 25 mm per metre in England's Approved Document H. Wales, Scotland and Northern Ireland use their own standards. The URASISTO set at £56.99 is titled in inches and prints 0.0005-inch/ft.

What to do if the claim is wrong

Note the bubble, then turn the level end for end on the same marks. If the bubble walks, the tool does not agree with itself. One verified UK Firecore review describes that failure and a return. Do not average the two positions. A digital pair should read equal and opposite. If the listing states 0.5 mm/m and the level cannot hold a reading, section 11 of the Consumer Rights Act 2015 says goods must match the description. This is not legal advice. Named picks, once the length is settled, are on best spirit level.

Common questions

What does 0.5 mm/m mean on a shelf and on a 2 m wall?
It means half a millimetre of rise over one metre. A 1 m shelf at that limit is out by about 0.5 mm. A 2 m run is out by about 1 mm only if the sole spans the run. A 250 mm torpedo does not span it. Overhead readings are often allowed a looser tolerance than a vial read the normal way up. The figure is a maker's claim, not a UK legal class for spirit levels.
Is a 230 mm level enough for a door?
No. It can be honest about the patch under it and still miss a bow in a lining or a stud. That work wants about 1200 mm. This set's only listing of that length is the Grettz digital beam at £155.99, with six ratings, not a bubble box. Use a torpedo for the socket and a long sole for the frame.
Do I need magnets on timber and plasterboard?
No. A magnet grips ferrous steel. It does not hold on timber, plasterboard, copper, plastic or most stainless. Hold the level on those surfaces. Buy a magnet for a steel stud, and expect paint to weaken it. Stanley reviews here say the magnet is not very strong.
Is a printed accuracy a certificate?
No. A bullet that says 0.5 mm/m describes that listing. It is not a British Standard approval, and DIN 877 on a manufacturer page is not a UK legal requirement. Spirit levels sit outside the 2006 rules for length measures. Prove the tool with a reverse test. If it misses the description, the Consumer Rights Act route matters more than the star line.
Are inch-marked sets a problem in the UK?
They are, when your marks are in millimetres and you read 0.0005-inch/ft as 0.5 mm/m. Those rates differ. The URASISTO set is titled in inches and has a tiny UK review sample, including one buyer who found untrue bubbles. It can work if you already think in inches and you test each piece. It is awkward next to a metric tape.
What if the level I receive fails a reverse test?
Stop. A bubble that moves when you swap ends should not be averaged into a shelf or a door. Return it inside the seller's window and quote any printed tolerance. One UK Firecore review in this set is that case. If you are mid-job, check against a second level you already trust.

Related pages

Digital spirit level vs bubble

Spirit level

Best spirit level: Stanley FatMax ahead of a 4.8 star

How to use a spirit level: read it, then reverse it

Torpedo spirit level: short gaps, not a door frame

Sources

Public sources from the digest, then the maker pages named there. Shelf prices are the amazon.co.uk figures stored for this set. Scores are the listing stars scaled for the table. This is not a laboratory test. Retrieved 22 October 2026.

  1. Approved Document H (England, 2015), Table 6 (retrieved 22 October 2026)
  2. Consumer Rights Act 2015, section 11 (retrieved 22 October 2026)
  3. Measuring Instruments (Material Measures of Length) Regulations 2006 (retrieved 22 October 2026)
  4. GOV.UK, laser radiation safety advice (retrieved 22 October 2026)
  5. Stabila, spirit level accuracy (retrieved 22 October 2026)
  6. Stanley UK, FatMax 0-43-609 (retrieved 22 October 2026)
  7. BuildWiz, spirit level (retrieved 22 October 2026)
  8. Johnson Level, reverse test (US method page) (retrieved 22 October 2026)