Soakaway Size Calculator
For a catchment of 25m² or less, Approved Document H lets you assume a design rainfall of 10mm in 5 minutes. That means the water arriving is simply your area × 0.01 in cubic metres. A 20m² garage roof therefore needs to store roughly 0.2m³, less whatever soaks away during the storm. Run a percolation test to find how fast your ground drains, and this calculator does the rest.
Everything here comes from Approved Document H of the Building Regulations: the 1 in 10 year return period and the design rainfall from paragraph 3.27, the percolation test from H2 paragraphs 1.34 to 1.38, the infiltration rate from paragraph 3.28, and the storage rule from paragraph 3.29.
This calculator stops at 25m², and here is why
Approved Document H is only self contained up to 25m², because that is the only case where it gives you the design rainfall itself. Paragraph 3.27 says that for larger areas you must go to the sources in paragraph 3.30, which are BS EN 752-4 and BRE Digest 365.
Both are paid publications, and the rainfall map in BRE Digest 365 is not openly licensed. We are not willing to guess at that data, so above 25m² this tool tells you to get the standard rather than inventing a number. Plenty of soakaway calculators online will happily give you an answer for a 200m² roof. Ask yourself where they got the rainfall figure.
Your site
Vp works out at seconds per mm.
Approved Document H gives no void ratio. These come from the product, not the standard. Use your manufacturer's figure if you have one.
Proposed pit (optional)
Give dimensions and we credit the water that soaks away during the storm. Leave them blank for the conservative answer.
Your result
Too large for this calculator.
At m² you are past the 25m² limit where Approved Document H supplies its own design rainfall. Paragraph 3.27 sends you to BS EN 752-4 or BRE Digest 365. We will not make up a rainfall figure to fill the gap.
Enter your area and percolation test result.
No pit dimensions given, so no outflow is credited. This is the conservative answer.
Where you cannot put a soakaway
Approved Document H paragraph 3.25. Infiltration devices should not be built:
- within 5m of a building or road, or in areas of unstable land;
- in ground where the water table reaches the bottom of the device at any time of year;
- close enough to other drainage fields, drainage mounds or soakaways that the overall soakage capacity of the ground is exceeded;
- where contamination in the runoff could pollute a groundwater source.
How to run a percolation test
This is the test in Approved Document H2, paragraphs 1.34 to 1.38. It is the same test whether you are sizing a soakaway or a septic tank drainage field.
- Dig a hole 300mm square, to 300mm below the invert of the proposed pipe.
- Fill it with water to at least 300mm deep and let it seep away overnight. This pre-soak matters: skip it and dry ground will flatter your result.
- Next day, refill to at least 300mm and time, in seconds, how long the water takes to fall from 75% full to 25% full. That is a 150mm fall.
- Divide that time by 150. The answer is Vp, the average seconds for the water to drop 1mm.
- Do it at least three times, in at least two trial holes, and average the results.
- Do not test in heavy rain, severe frost or drought.
A trap worth knowing about
You will find sources saying a soakaway is only valid when Vp is between 12 and 100. That is not a soakaway rule. It comes from Approved Document H2 paragraph 1.38, and it governs whether a septic tank drainage field may be used at all. It is a different application that happens to use the same percolation test.
Approved Document H places no Vp limit on soakaways in Section H3. Applying the 12 to 100 gate here would wrongly reject perfectly good sites. This calculator does not apply it.
Why the base of the pit does not count
Paragraph 3.29 defines a50 as "the area of the side of the storage volume when filled to 50% of its effective depth". Sides only. The base is not included, so a wide shallow pit and a narrow deep one of the same volume do not drain at the same rate. Depth buys you more soakage than width does.
Frequently asked questions
How big does a soakaway need to be for a garage roof?
For a 20m² garage roof, Approved Document H gives an inflow of 0.2m³, so you need roughly 0.2 cubic metres of storage before crediting soakage. In average ground a 1m × 1m × 1m open pit comfortably covers it. If you are filling with rubble at a void ratio around 0.3, you need to excavate more than three times that volume.
What Vp is good ground?
Vp is seconds per millimetre, so lower is faster. Free draining sand or gravel gives a low Vp, and heavy clay gives a very high one. There is no pass mark for a soakaway in Approved Document H, but very slow ground will need a large device, and paragraph 3.27 notes that "where the ground is marginal overflow drains can be acceptable".
Can I put a soakaway near my house?
No. Paragraph 3.25 says not within 5m of a building or road. Water soaking into ground next to a foundation is how you undermine it.
My roof is bigger than 25m². What now?
Approved Document H paragraph 3.30 says soakaways serving larger areas should be designed to BS EN 752-4 or BRE Digest 365. Both are paid documents. The method in BRE Digest 365 is the same shape as the one here, iterating over several storm durations to find the worst case, but it needs rainfall depth and duration data that is not published under an open licence. Buy the standard, or use a designer.
Sources. Approved Document H (Drainage and Waste Disposal): paragraph 3.1 (2 hectare limit), 3.25 (siting), 3.27 (return period and design rainfall), 3.28 (infiltration rate), 3.29 (storage and outflow), 3.30 (larger areas), and H2 paragraphs 1.34 to 1.38 (percolation test). Contains public sector information licensed under the Open Government Licence v3.0.
Paragraph 3.29 states the storm duration D "in minutes" while the infiltration rate f is in metres per second. Taken literally the units do not balance, so we use seconds, which is the conservative reading. Over a five minute storm the outflow term is around one percent of inflow, so it makes no practical difference at this scale.
This is a specification aid, not a substitute for design by a competent person, and it does not replace Building Control approval. Void ratios are properties of the product you buy, not figures from Approved Document H. Read the full methodology.