self build waterproof basement formwork membrane concrete engineer

Structural Engineers.

This page is for Structural Engineers about to design a basement for a client who wants to self build my way. A client who expects internal drainage to be completely unnecessary.

You might think every basement leaks. But mine don't. Nothing through concrete, joints or kickers and no ingress over the top of retaining walls either. None since 2013.

BS8102 states that waterproofing can be designed and chosen either before building commences or after the structure is complete.

Obviously waterproofing a Victorian cellar today requires the choice of waterproofing to be made after the structure was built.

It is perfectly acceptable to treat a new basement the same way. Build it then visit to see what water ingress there is and only then decide what else is needed.

My customers build with waterproof concrete with my experience of how to build without leaks and all the important workmanship supervised. It is the different methodologies and effective supervision that make all the difference.

They move in to their new house with it complete above ground and the basement still bare concrete. They can use the whole of their basement ventilated but not heated, clean but not decorated. If there is ever a leak I will fix it (none fixed in anything built since 2013). They can ask you and anyone else with an interest to inspect the basement after a period of heavy rain.

If they have external drainage as well, and the basement is totally dry, then they have two defences and they work.

If they don't have external drainage BS8102 states they need a second defence. If the basement is always totally dry this can be a simple vapour barrier - either bitumen paint or polythene.

Then they can continue with their wiring, plasterboarding and decoration.

Building a basement to be dry from the concrete alone means that some of the practices you believe are normal, but leak, won't be used. For instance tie bar holes, kickers and retaining walls that stop beneath outside ground level.

It means some of the usual ways of working won't be used, such as casting sections of walls to full height causing honeycombing at the bottom and requiring masses of joints.

self build basement

Quick links to sections of this page.
  1. How a floor over a basement is different to the ground floor in a normal house.

  2. Extra steel to restrict crack widths in concrete that gets hot.

  3. The truth about waterproof concrete.

  4. Corner bars.

  5. Cantilever not propped cantilever.

  6. Room to properly place concrete and capping bars.

  7. Why wall kickers could never be waterproof.

  8. Joints (New page in a new tab).

  9. No tie bar holes (New page in a new tab).

  10. Waterproofing is dealt with on the page for Architects. (New page in a new tab).

My name is Phillip Sacre. I have been building and helping to build basements for years. I sell my expertise and some products I sourced because they are better than the famous brands. My clients are self-builders, seemingly the only people who care that a basement is built dry.

When I used to have leaks, none since 2013, I investigated their cause and changed my method to avoid them. I know how to avoid every cause of leaks through a new basement structure. I have been sharing my "secrets" on this web site for years but very few tradesmen have copied them, only self builders, and I am very busy helping many of them.

During the year or so the client lives with a ventilated, bare concrete wall basement, if he sees any ingress of water I will fix it. (None fixed since 2013).

Even if I fail and there is some ingress of water, it won't be a torrent.

A damp patch or a few drips of water don't want a powerful pump or an outlet hose that takes 2 litres to fill before any water is actually pumped away.

Excavating extra deep for a sump is very expensive, especially if the extra depth finds the water table. And extra depth for a deep sump can be dangerous if there is not enough room to safely batter the sides.

Many basements have plumbing and a plumbing spill will be far worse than a leak through my work. If the plumbing spill is catered for, any drips leaking through are catered for as well. One client chose a Baked Bean tin cast in his floor slab and others something the size of a dinner tray 50mm deep. These could be emptied with a WetVac or aquarium pump. They make far more sense and of course the cost is tiny in comparison.

Building Control expect to see "internal drainage, a sump and a pump" on the drawings before work starts; and there is no need to enter into a battle, you can specify exactly what they expect to see.

Your client hopes to see "internal drainage, a sump and a pump" and "shallow depression for plumbing spills in the structural slab within the plant room".

In this way, provision has been made for a pump to be added if there is enough water to pump out and the outlet hose might be 3mm diameter and all but an eggcup of water can be pumped right out.

What do I do differently?

I tweaked my methodology by
  • Using a bit more steel to reduce crack widths further;
  • Using concrete proven to be impermeable to BS12390 part 8. You might specify C35A. I will buy C35A but with 350kgs of cement blend and even less water and I will add a powerful plasticiser on site. Links open in new tabs.
  • Scabbling all joints, getting them clean and keeping them clean (to BS8007);
  • Not using kickers;
  • Not leaving tie bar holes through otherwise waterproof concrete;
  • Only pouring walls 2m high in one pour yet all the way round in one go;
  • Making sure the choice of floor over the basement is one that prevents water ingress over the retaining wall.
I want you to understand how a basement completely waterproof from the concrete alone is actually very easy, but it needs a few tweaks to your design. Particularly the extra steel and a good choice of floor over.

A good choice of floor over is often the most contentious issue. Steel Beams, Precast Concrete and Composite Floor Steel all work fine well above outside ground level but cause incurable problems below outside ground level, as well as cold bridging issues.

Please see special note #2, which is the bottom section on this page, where I explain why these choices are nothing but unnecessary expense and trouble.

What does everyone else specify and build?

Seasoned professionals will all use steel threaded rods in a plastic sleeve. They leave the plastic sleeve as a hole through otherwise waterproof concrete; or else they fill the holes but some leak.

I was sent these images by a prospective client looking for a different basement builder, so I don't know where he found them. He is planning a second basement and he says his first basement leaks - so perhaps this is it.

typical basement leaks

And here we see the hole destined to get the Delta-type sump for a pump, backup pump and backup power supply. All very expensive and not guaranteed to work long-term either.

the sump in a basement

PLUS. HORROR !!!!    The pipes and ducts connecting the outside to the inside are underground not set in any waterproof concrete.

When the shingle you can see in the photo above was tipped in it probably broke the joints between the ducts you see and the next sections added on. When the shingle fills up with rain during a severe storm, water could flood in through these duct joints far faster than the pump could pump it out. Pushing the drainage membrane off the wall, flooding the basement temporarily and leaving plasterboard, decoration and furnishings ruined.

No fancy warranty from a respected waterproofing company will cover any of this and all the 10 year latent defects insurers all exclude waterproofing below ground.

These built-in errors are at the risk of the householder. No one else.

Sometimes you can't trust anyone. Whose fault is this? The architect, the engineer, the internal drainage supplier or the contractor? None of them seem to have realised the terrible mistake being made.

The client won't be happy. He will be furious.


The floor over a basement is completely different to the ground floor of a normal house

because water getting in under the ground floor of a normal house settles harmlessly on the earth beneath.

But in a basement that same water is a leak.

  1. The ground floor over a basement must prevent a horizontal ingress of water beneath it.

  2. The retaining wall must prevent a horizontal ingress of water over it.
Please do not picture in your mind sticky-back membrane as the solution. Sticky-back membranes are hopelessly unreliable and won't last the lifetime of the building.

Sticky-back membranes might stick fabulously well on the south side with the sun shining.

But they will not stick to a primed surface or to a neighbouring sheet on the north side where the atmosphere will be moist and these products do not work. Neither will the membrane stick to the neighbouring sheet up the sides of the slab where the tails have been sitting weeks in water.

After they are fitted the bricklayers or the scaffolders will damage a sticky-back membrane and just push it back into place without a repair.

Basements with a flat top to the retaining wall stop just beneath outside ground level, water gets in horizontally beneath the ground floor and over the top of the flat-top retaining wall running down the inside of the basement wall. It cannot be stopped except by internal drainage and only if that is lapped up and over the ceiling. It would be a shame if your client spent thousands of pounds on internal drainage and the only water it removed got in only because of your uninformed choice of floor.

Therefore, if you specify steel beams, a beam and block ground floor, precast planks or composite steel deck with concrete on top, either the client will have to pay for a complex way to support his floor, losing real estate or it will leak. More reasons not to choose these floors at the bottom in Please see special note #2.

I was the first specialist to promote an upstand which I now see has become common practice.

Here you see 4 options.
  1. A continuous, air-tight structure of reinforced concrete. Continuous insulation envelope outside the structure. Maximum thermal mass. No cold bridging. Exceeds Passivhaus Standard. (See my page about bare concrete walls inside here).

    The basement has to be a little smaller and an extra footing for face brickwork.

  2. Beam and block, precast planks or corrugated sheets against an upstand.

    Makes the basement a little smaller and the corbel is expensive.

  3. Engineered timber joists.

    Really cheap and simple.

  4. Cast insitu reinforced concrete podium deck.

    This might be chosen for an outside patio, a garage over a basement, beneath underfloor heating or if you are really trying to maximise thermal mass.
    Please see special note #2 lower down where I prove cast insitu reinforced concrete can be a lot cheaper than steel beams and precast planks; and I explain the issue with composite steel decks.
seasonal heating cooling envelope

beam and block sketch        insitu podium deck

Another way, that also reduces the size of a basement, is to build blockwork inside the basement to support beam and block independently of the retaining wall.

In this photo, the right is outside and the left will be inside. I helped with all the waterproof concrete, the formwork, the pouring and the external land drainage. The overall wall width is about 600mm. The insulation had to be put in before the floor over could be installed, it already looks sodden.

waterproof basement blockwork beam and block

In my experience a waterproof concrete upstand is a most essential waterproofing measure.

It is very much easier to accommodate wall construction above a basement with a waterproof concrete upstand if engineered timber joists are used instead of beam and block or precast slabs.

outside brick ledge
waterproof basement construction upstand
  inside wall plate
waterproof basement construction floor ledge
  top chord supported
waterproof basement construction floor ledge
  floor joists in hangers
floor joists on hangers

An engineered timber floor joist can be any width and any height and stronger than concrete (for the same overall floor construction depth including the service void beneath concrete). It can be a trus joist (OSB web), an easi joist (lattice steel sided) or plywood sided. Services can go through it.

a top chord supported floor joist      Nu Heat underfloor heating for timber floors. underfloor heating over basement

Clearly, engineered timber joists offer more choice than concrete and they are flat on top. With services through and not under, and without the need for a screed to overcome curvature timber floor construction depth is usually less than concrete.

Waterproof concrete gets hotter while curing than ordinary structural concrete so the engineer will be asked to have enough extra steel to control the extra issue of thermal cracking. Therefore there will be enough steel in the walls for their not to benefit from propping by a concrete floor over the top.

Please choose an engineered timber joist floor, supported off an upstand created out of waterproof concrete. The basement will cost far less this way. Indeed this way is often the only way that suits a tight budget.

Unless the ceiling over the basement has to be concrete, for instance it is a garage, outside or, sometimes, a kitchen. In which case concrete cast insitu is flat and absolutely waterproof all round.

Our rods and nuts make forming an insitu concrete podium deck very easy indeed. The top row of rods for the wall are left to use again for an edge shutter. The upstand will need to be formed on the podium deck afterwards.

frp threaded rods reused

Note added March 2018.

I have recently been investigating Near Zero Energy Buildings and am persuaded that a 'Fabric First' approach is best.

On this page I discuss how you might build with maximum thermal mass including a basement.


Extra steel to restrict crack widths in concrete that gets hot.

I have a lot of experience making sure that waterproof concrete stays waterproof as it cures.

I have found that slabs do not crack when cooling. I presume because the edges are not restrained and the slab as a whole can shrink.

I have never seen walls crack horizontally, presumably because they can shrink in height unrestrained.

Cracking is regular and vertical in walls. Not quite to the bottom where there is extra crack protection from starter bars. Nor right to the top of a wall, presumably because walls are restrained at corners but even corners give a little at the top of a wall.

This is typical of a crack where there is not enough distribution steel and the ends of the walls are restrained.

A 300mm thick wall with A393 and 12mm starter bars on 200mm centres in both faces will crack like this and may not fully heal, whereas the same steel in a 200mm thick wall seems to always heal completely.

We all want any crack to heal before we backfill, to heal before water could flow through and prevent the autogenous healing process.

In 250mm thick and 300mm thick retaining walls more distribution bars and thicker starter bars seem to provide the solutions.
  1. Two faces of 16mm starter bars on 200mm centres

  2. A393 to both wall faces

  3. Extra distribution bars horizontally in the wall between the A393 horizontal bars, meaning 10mm bars on 100mm centres to both faces.

waterproof concrete crack width

The old BS8007 has an appendix with a calculation for steel to reduce crack widths and it wouldn't need as much steel as my experience insists is necessary.

I think the reason for that is in the days it was written all concrete was a blend of OPC and PFA. But we shut our coal fired power stations and PFA is rare and an expensive premium, rather than free of charge because it was a handy way to get rid of waste.

The concrete in your client's basement might not be pure OPC and it might not get as hot as I expect it to. But you could never be sure. Concrete companies get plant breakdowns and send you your concrete from a different plant with different cement in stock. I think this additional steel is essential.


This section explains the truth about waterproof concrete.  There is a certified CPD I wrote, about 2014, somewhere on the Internet.

This is the quote from the Concrete Society report, 2012:
"Many water-resistant basements have been successfully constructed on the basis of the above alone. There are a number of admixtures in the market to modify the properties of the fresh and hardened concrete (e.g. porosity, permeability). Such measures should not be necessary if the basic principles noted above are observed."

The Concrete Society has published reports more than once expressing doubt that brands such as Caltite, Xypex, Pudlo and Sika do anything more than good concrete can achieve anyway.

BBA certification is completely inappropriate because their testing is completely inappropriate. Suppliers just added a bit more cement or reduced the water slightly, got a tiny improvement and got a BBA certificate in exchange for a lot of money. But none of the BBA certificates say that they tested the concrete that would be delivered to site or that site concrete will be waterproof.

Yet concrete can be made waterproof. There is not any admixture that waterproofs un-waterproof concrete used below ground for basement construction. But there is a test that measures the porosity of a cube of hardened concrete delivered to a project and it can prove that the concrete used on site is waterproof.

My evidence is also from another UK Concrete Society report, 2013* and the BBA certificates for Caltite, Pudlo, Sika etc..

The proper certification for waterproof concrete is concrete made to BS EN 206-1, and concrete tested to BS EN 12390: testing of hardened concrete.

Thousands of test cubes are tested to BS EN 12390 part 3 every week. This is the test for compressive strength.

BS EN 12390 part 8 is the test for depth of penetration of water under pressure.

The different parts of BS EN 12390 are listed here on the BSI web site.

Here is a photographed extract from my copy of "Advanced Concrete Technology", the volume called "Processes", the Chapter called "Concrete construction for liquid-retaining structures" by Tony Threlfall, 2003, Butterworth-Heinemann. Page 16.2.   Click on this image to be able to read the text from the whole page.   from Processes, water retaining BS EN 12390 8
It says that permeability less than 20mm is satisfactory. You will see on my certificate of permeability that my results in this case were 1mm, 1mm and 3mm. I have dozens of similar results going back years.

You can specify that cubes should be taken and tested for strength and permeability. The permeability test costs £600 plus VAT.

Every BBA certificate for a water-resisting admixture for concrete tells the user to buy an already water-resistant mix of concrete. This is a pump mix with 325 to 350kg of cement per cubic metre with reduced water.

C35A in BS 8007 is a concrete mix with 325kg of OPC and 55% water by weight of cement and this is a watertight mix.

In the text you will see that the appropriate test for waterproof concrete is to BS EN 12390 part 8. There is no mention that a BBA certificate will prove concrete won't leak.

Many, many large sites routinely test concrete for strength to BS EN 12390 part 3. I have concrete tested to BS EN 12390 part 8 as well. Depth of penetration of water under pressure on concrete. The pressure is equivalent to a depth under water of 30m and the pressure is maintained for 96 hours.

Click on either image to the left to see two original test certificates.
  I always buy concrete over-sanded, with 350kgs of cement and a target slump of 60mm. From experience, concrete suppliers will send concrete too stiff to pump or compact but wet enough not to induce a flash set.

The concrete chemistry books state that by adding a little more cement to C35A and reducing the water a little further that pores of water between remnants of cement grains will be closed off from neighbours by crystals of hydrated cement.

However, such a mix would be too stiff to pump or compact without a more powerful plasticiser than those usually available. We have to make sure that a particularly powerful plasticiser is used.

I add my powerful, PCE based water reducing plasticiser at site and I count the revolutions to make sure it is thoroughly mixed through. The concrete becomes runny, pump-able and easy to compact.

concrete compressive strength report to BS EN 12390 3 concrete permeability report to BS EN 12390 8

A working group of the UK's Concrete Society published a report in 2013* produced by consultant engineers, academics and other experts. They strongly suspect that it is always the concrete that will be waterproof, not any of the admixtures.

* The influence of integral water-resisting admixtures on the durability of concrete. P36. Concrete Society. 2013.

But to make matters worse, most of the BBA tests were on concrete samples that would have been too stiff to pump. On concrete that would have to be rejected on site or, more likely, have more water added which defeats the purpose of trying to waterproof concrete.

Most BBA certificates contain the evidence that the concrete that will be delivered to site was not tested.

Do all products have to have a BBA certificate or else they cannot be used?
(Not according to a statement here from the BBA web site 22 Oct 17 stating that a specifier needs to make a judgement, not just trust a product because it has a certificate).

On November 9th 2020 the Grenfell Tower enquiry was told "This reveals an industry in which Arconic, Celotex and Kingspan were content to push hazardous products into the marketplace and sought to market them dishonestly. These products should have been safe, they should have been tested and certified rigorously, and they should have been marketed in an honest and transparent fashion. None of that happened. The testing and certifying bodies, such as the BRE and the BBA, were quite happy to go along with this process."
(Source: Click on the linked text above, opens in a new tab).

ABI exposes the 'utter inadequacy' of the laboratory tests currently used to check and certify the fire safety of building materials. 25th April 2018.

On 26 February 2021, this article reported on a technician involved with fire testing the products used on the Grenfell Tower.
  1. He failed to notice in 2014 that a test rig for an insulation product from Celotex had been secretly altered to increase its chances of passing.

  2. He claimed that manufacturers could sneak extra components onto test rigs without inspectors knowing

  3. The reliance was very much on the honesty of the client.

  4. If you have got somebody who is going out of their way to deceive, then there was a possibility they could do that, if that was their intention.
On 23 March 2021, this article reported that the BBA's Chief Scientific Officer told the Grenfell Tower Inquiry that the BBA did not question whether a manufacturer had provided the unhelpful evidence as well as the evidence that helped their case.

He also accepted the BBA had made "a very basic failure of due diligence".

In my opinion, construction material producers are well aware how to cheat BBA and get away with it. BBA are well aware that they do not ask difficult questions. The whole process is one to make BBA wealthy in return for an oligopoly or monopoly commercial advantage. For these reasons, I do not think any BBA certificate should be trusted.

* In 2013, the UK's Concrete Society published a report by a working group of 18 members, experts in their fields, that states: " water/cement ratio .... primary measure of water penetration and hence the durability of the concrete."

The authors of this report had the bright idea of plotting all the BBA data on one chart. Here is the same data I plotted on to this chart. It includes my educated guesses where data is missing, see the legend to see which.

chart   Concrete needs to be the right consistency to be pumped into formwork and compacted properly without bleeding. 120mm to 140mm slump would be fairly usual.

So any BBA testing on much drier, stiffer concrete has not tested concrete that would be used on site. Site concrete would have to have more water, which would massively affect test results.

Seemingly, a very similar situation to Grenfell Tower cladding. The cladding resisted a flame in a lab but they (BBA) stated that they didn't know if it would work on a tower, on the certificate.
ABI exposes the 'utter inadequacy' of the laboratory tests currently used to check and certify the fire safety of building materials.

That is what the cladding BBA certificate states and every admixture certificate also warns that BBA didn't really find out if any of the admixtures would work on site.

Slumps of tested concrete
Pudlo 45mm
Sika 40mm
Xypex 35mm
Kryton KIM 45mm
Caltite no information
BASF 135mm. Hooray. Site concrete. But BASF performed very badly.
Only BASF has a certificate for concrete with the consistency used on site.

All these admixtures have to include a water-reducing plasticiser.

Apart from plasticiser what seems to be in them?

With a little more cement or cement replacement:
Krystol KIM

With liquid plastic absolutely useless below ground where it won't dry
Others with polymer.

There is no evidence anywhere that anything other than extra cement and less water can waterproof concrete used beneath ground where it won't dry. The concrete will need a particularly powerful plasticiser as well to make it workable.

If you specify any of the famous brands, you risk litigation after Grenfell enquiries rule that not reading and fully understanding a BBA certificate is negligent.

None of the BBA certificates provide any evidence that any of these products make enough difference to site concrete to be worth any money at all.

The proper procedure is to specify concrete made to BS EN 206-1, and concrete tested to BS EN 12390:8 permeability of hardened concrete.


Corner bars.

Half of you claim corners are a weakness and specify corner bars in 3s on 200mm centres. basement construction The other half claim corners are the strongest part and don't specify any corner bars. I never had one crack but we should have full crack protection. Corner bars and the extra distribution bars.

Bars in threes make concreting with a pipe a bit harder and threes should only be needed on reverse corners. Most basement corners should be fine with only 2 corner bars. basement construction Only the A393 bars need connecting with corner bars. I don't think that the extra distribution bars need corner bars as well. They might overlap by 20mm but their work is along the wall, not at the ends.


Cantilever, Not Propped Cantilever.

An apparent advantage of a concrete floor over rather than engineered timber would be the support a concrete floor gives to the top of the retaining walls by propping the walls one side against the walls the other side and the backfill beyond.

But this would mean that the floor had to be in place and complete before the basement was backfilled and that causes access and safety problems.

It might also mean that internal basement walls had to be complete before the floor over could be installed. Delaying backfilling still further.

If you look through the section above about floors over a basement, the upstand to the retaining wall is essential. An engineered timber floor works with the upstand far better than pre-cast concrete.

The extra steel I require in 250mm or 300mm thick waterproof concrete walls to restrict crack widths is often sufficient for the wall to just be cantilevered.

Trying to save the client money by specifying walls only 200mm thick costs the client money by the time you also specify a precast concrete floor and screed to overcome the curvature of precast beams.

Also the client would need a service void under a concrete floor so the whole excavation needs to be deeper and the walls taller, costing more as well.


Room to properly place and compact concrete.

The Basis Of Design in para 2.1.4 of BS8110-1:1997 states:

"Design, including design for durability, construction and use in service should be considered as a whole."

Your design has to allow, not prevent, the required standard of workmanship.

There must be room to get concrete to the bottom and compact it properly.

Capping Bars.
Some engineers like to see capping bars all along the top of a wall and, sometimes, even a central distribution bar.

Concrete cannot be poured through a tube to the bottom or properly compacted with a vibrating poker with this amount of obstruction.

In contrast, we like to have the same bar Shape 21 on 1.2m centres both horizontally and vertically throughout the wall (1m from the slab and 2.2m from the slab) to control the space between the steel and keep it all perfectly upright throughout, not just along the top which would leave the middle to curl and wave without control.

Neat steel gives everyone confidence.

Not having U bars along the top allows us to move the concreting pipe along the wall formwork without taking it right out every time - which greatly reduces the mess from concrete dribbling from the tremmie pipe.

Not having U bars along the top means that the poker does not have to be pulled right out of the wall every 200mm before it is lifted back in to do its next bit of compaction - reducing the amount of work the man on the poker has to do considerably, so the compaction is carried out better without a top line of U bars.

basement construction basement construction

basement construction


Why wall kickers could never be waterproof.

Kickers are notoriously un-waterproof.

The biggest enemy of waterproof concrete has to be a kicker, a small upstand on the slab to clamp traditional formwork to.

They are only there for the benefit of the formwork, not the client.
kicker that leaks kickers leak leaking basement wall concrete

In this photo you can understand why the concrete in the kicker sinks into the slab if the concrete is compacted properly with a vibrating poker.

Therefore, kickers end up getting made out of old concrete already setting or the wrong mix bought in for something else.

This poker is obviously being allowed to dwell in one place too long. This will knock the largest stones away first and eventually the sand as well so that when the poker is withdrawn there will be just a column of water through the concrete.

If the kicker was made after the slab was set, you have an extra joint to scabble and clean first but the poker would send the kicker concrete flying because a poker is too powerful for concrete only 100mm deep.

More discussion and solutions here.
  waterproof concrete basement kicker

This photo is from a BBA certificate for a water-resisting admixture.

As a practitioner, I think this supplier, and I presume BBA as well, are absolutely clueless.

This is a basement built my way, without kickers.

A building firm saw a self builder's basement. Studied these web pages then placed their order. I took this photo when I delivered the FRP threaded rods and nuts.

They have spaced the formwork with measured pieces of wood. These came out as they put in the threaded rods with thin nuts to control the width instead.

basement 3 weeks

3 weeks later I collected the steel waler plates they hired. This is the completed job.

basement 3 weeks later

Shame they have a flat top to the basement. Perhaps the house extension will be bigger than the basement and the soil level beneath the extension will be lower than the top of the retaining wall.

Special note #2. The solution I found when I had to is at the end, but it was ruinously expensive as you will see.

These 4 products are commonly used well above ground where they succeed. I searched for photos of their use over a basement, but all I found was this case study on a precast supplier's website that conveniently stops at the point where the precast went on. I couldn't find any blog to let me know whether the basement leaked later.

All these basement ceiling choices leak and some leak heat as well.

I try here not to repeat myself. Generally, the problems of the first product discussed are common to the second product discussed; and the problems of the second are common to the third, and so on. In other words I discuss least worse but still bad choice first and most worse bad choice last.
  1. Precast Planks.

    precast plank ground floor   It will always be true that the underside of a concrete ground floor, especially if it is to get insulation and screed on top, will be below ground level. if that concrete floor is not monolithic with the wall beneath the joints will leak.

    Precast planks might be used for the ground floor over soil. You can see they don't cover the cavity allowing for insulation and the cavity to drain.

    But if planks or beam and block are on a basement retaining wall the water ingress under the beams leaks into the basement; the cavity tray doesn't help because water will get under it and you might find insulation over the beam ends either impossible or requiring something unsightly to cover it.

  2. Beam and block.

    There is nothing wrong with beam and block if there isn't a basement underneath because water getting under the beam and block trickles into the vented space beneath; a cavity tray can stop water getting in any higher than the underside of the beams and insulation in the cavity covers the beam ends. Beam and block is marginally worse than planks because of their curvature that requires more screed on top to overcome.

    beam and block floor

  3. Composite Steel Deck.

    Every detail seems to show steel decking welded down to steel frame. Every detail below is of the edge.

    Because this decking needs to be concreted there is an edge shutter solution that prevents the concrete above the steel deck being joined to the concrete and reinforcement in the retaining wall. Where this joint is below ground it will leak and cause a cold bridge that will always be mouldy.

    If the steel deck only bears on the retaining wall in the same way as precast concrete and if the concrete on the deck is stopped at the steel edge then water will get in under the deck in just the same way as under precast.

    composite steel deck edge details

    But when an engineer wants to use steel decking over a concrete basement he wants reinforcing steel in the basement retaining wall to be continuous into the deck concrete.

    This creates a gap between the steel decking and the edge shutter.

    It also means that the underside of the steel profile is up in the air and all the floor concrete will pour through the profile into the basement.

    It also means that insulating around the edge of the deck is a problem.

    If you are a structural engineer considering composite decking over a basement avoid embarrassment in the future by drawing a section through every possibility the composite steel gives you. Drawing the edge adjacent to the profile is easy. Draw sections through the high and low spots of the profile and include your steel reinforcement and share them with the architect early on in case he warns that he cannot insulate it.

    I found this drawing amongst work I might do soon. More to follow ! !

    composite steel deck over basement

    I wrote in detail to complain about it but got over-ruled. But it turns out the practice went into administration or liquidation weeks later, so who knows whether it will be used or whether I will be involved.

    There is a lot of construction detail missing. If you look at the detail on grid B, how is the basement to be insulated? How do you prevent a cold bridge without smothering the inside of the deck in insulation - which would mean that the excavation was that insulation-depth deeper, and the retaining walls that depth taller, to maintain the headroom?

    Is the concrete over the steel deck really going to be stopped partially over the retaining wall?

    What is going to stop ingress of water from the topsoil getting in under the decking?

    • To stop water getting in the concrete needs to be bonded to the retaining wall.

    • But if the edge shutter isn't hard against the composite steel all the concrete will pour down into the basement.

  4. Steel Beams.

    I was approached about a basement here 7 years ago but didn't get or didn't want the work (I don't remember). After the first basement was built it was decided to carry on and build the buried swimming pool structure. The first basement got covered in beam and block a few years later. The choice of roof over the pool area was made 7 years ago and I was asked to install it recently. A good opportunity to experience what I thought was a bad idea.

    The design was for the steel beams and the precast planks to bear fully upon to the retaining wall all round. I was, at least, able to get that reduced to 200mm even if I couldn't sell cast insitu instead (because it was deemed quicker to stick with what they had rather than find an engineer to design a new alternative).

    flat top retaining walls   steel beams over a basement

    This meant I had the opportunity to

    • Scabble the top of the retaining wall,

    • Drill in some dowels,

    • Increase the mesh size in the structural screed that will be next over the top and use waterproof concrete.

    flat top retaining wall rescued   precast planks over basement

Special Note 2. Why cast insitu saves money as well as all the problems.

The client told me what the roof above was costing him. It was very time-consuming, frustrating, difficult and annoying.

Bear in mind, the precast plank people take full control. The steel beams above were down to me to get in as was casting the structural screed over the top of it.

9m wide deck   whereas

I built this 9m span roof over a basement in 2012 and I know what it would cost me today.

The cost of the steel beams and precast planks.

The steel beams were in but the precast planks weren't installed till the 4th date we were given. First time we were ready and the installers were ready the articulated trucks couldn't get near enough to the crane. The second time the same crane driver with the same crane refused to go down the drive, the third date was too windy and finally, with the planks already delivered and offloaded, a different crane did the job. Two weeks was lost and they tried to bill the client thousands of pounds extra. Their "trick" was, at every opportunity, to get the client to agree an email buried within which was a phrase making him completely responsible for design safety and cost. Completely unfair.

The steel beams and planks save the cost of falsework to support an insitu pour, but the steel beams and planks mean the concrete pour on top is small, possibly incurring small load concrete charges and definitely meaning you pay four hours for a pump you only need for one hour.

This roof cost the client:
£19,000 for 5 large steel beams
£1,000 for two small beams
£3,000 nearly for transport and galvanising the steel beams
£14,000 for 42 precast planks and installation, fortunately only the one crane hire despite their trying to make him pay for three
£2,000 my team of 3
£300 for a surveyor to mark the bolt holes
£350 for a concrete drilling specialist
£72 to hire the trolley we used to get the beams down the drive
£4,000 contract hire to lift the steel beams into place
£2,000 readymix concrete, and
£480 for the concrete pump.
£1,000? for the client's structural engineer.
Total £47,202.

The overall size is 9m x 17m to outside edges. With confidence I believe that just reinforced concrete instead, temporarily supported by Acrow props, timber and ply, my way, with my admixture would have cost me, even with profit if it were for a fixed price, £125 a square meter. £19,125.

If you add £2,000 for an engineer's design and VAT that amounts to £25,350.

In the past I have charged £175 per square meter plus VAT which was usually zero rated. That is £26,775 and I would have had a really good week.

Even £26,775 plus £2,000 for an engineer plus VAT is only £34,530. This roof could not have cost £47,200 if I had cast it insitu.

The client knowingly paid the extra to avoid any further delay, but all the manufacturing and installation delays added up to 4 weeks.

Because my time was paid for by this client (I worked for him, not on a price), even with VAT on design and materials, this roof should have cost him £17,600.

Some of the health and safety was indefencible. The big beams weren't to be bolted down but the plank people insisted at the last minute that they were. The bottom flanges are 400mm wide, 30mm thick steel. They claimed that with a plank one side these beams would twist.

They wouldn't twist if they were bombed. But another £1,000 was added to the client's costs at the last minute to get the holes set out precisely, drilled precisely, threaded rod and nuts sourced and supplied and resin-anchored-in all the day before the beams arrived.

mouldy steel beam   mouldy steel beam

Another memorable aspect was that when the project manager came to site to measure up and said different things to what the office had sent in writing, he said they haven't got a clue in the office sometimes.

He prepared a lifting plan that he wanted the client to sign off and take full responsibility for. When his crew eventually lifted the planks in - but in a completely different sequence to his plan, they said project managers don't know how to do it.

We lifted a 1.7 tonne steel beam to within 2.5m of the end wall with a 65 tonne crane. The project manager specified a 95 tonne crane because his planks each weighing 1.3 tonnes had to be lifted a bit further - but no more than 4m further.
We questioned this several times.
In the event, the crew lifted 2 planks, 2.6 tonnes, together almost to the farthest point. The project manager made the client pay for a much bigger crane so his crew could finish by 11.30am though, no doubt, the client paid for the crew and the crane for the whole day.

mouldy steel beam   mouldy steel beam   mouldy steel beam

At every stage and seemingly in every way possible the client was cheated, frustrated, robbed and delayed. It was as if because the plank people could take advantage of a timid private client (as distinct from a major house builder who probably calls more of the shots) they did their best to maximise the opportunity.

The reason steel beams are the worst choice is heat loss and cold bridging

made worse by the heat conductivity of steel pulling huge amounts of heat out of the room and creating cold spots for condensation and mould.

This work was in 2010 before I learned, 3 years later when I was asked back to see the mould, never to allow steel beams over a basement.

mouldy steel beam   mouldy steel beam

Mould is an awful problem. It must be avoided at the design stage. Never specify steel beams founded on the retaining wall.

You would not have any of these problems with a cast insitu RC ceiling with insulation over the ends. See the image at the top of the page reproduced in the menu image just below. The upstand in the retaining wall is the insitu deck over the basement. Room for insulation and face brickwork. Waterproof.

If insulation is a real problem, if the architect has specified internal insulation only for the basement, you might need to specify insulation inside the retaining wall then a second wall inside that to support the ceiling. But this will cost the client accommodation space and a lot more money. He might not like that.

Many of my clients changed architects while only a few changed engineers. I hope you argue for what is right for your client.

basement construction menu page   At the foot of every page you will find this image. Click it to go to our menu page where every page is listed.

Clicking the banner at the top will return you to the home page.

 Previous Page

basement architect

 Next Page

planning your basement floor slab and sump

 The Page After That

planning your resin rod formwork