Archive for the ‘DIY Projects’ category


May 22nd, 2012

Most of what makes a good set of stairs or steps comes down to the initial planning  and there are a few important pointers that must be heeded to ensure this (most come directly from the SA National Building Regulations).

Here they are:

  • No rise of any step should exceed 200mm
  • The going or width of each tread should not be less than 250mm.
  • If the treads are open (no solid risers), each tread must overlap the lower tread by at least 25mm.
  • The variation in the risers or dimensions of the widths of treads should not be more than 6mm per flight.
  • Maximum slope 45 deg (preferably less)

It is amazing how often these simple rules are ignored and the result is almost always noticeable and often ruins an otherwise perfect deck, house etc.

So how do we go about this? Well it’s simple really:  first measure the overall horizontal and vertical distances you have. Divide the height by 200mm, this will give the number of risers you need. Round this number up to nearest whole number and then redivide the height by this number. This will give the actual riser height required. Now take the number of risers and subtract one, this will give you the number of treads needed (the first tread is the ground so you need to subtract one).  Remember each rise can only be a maximum of 200 mm, so if it’s not working out (number of treads multiplied by 250mm is too wide) then you will need to steepen the staircase. The only way to do this is to overlap the treads – don’t simply reduce the tread width to below 250mm as often is done!

Here is an example :

vertical distance = 1290 mm / 200 mm = 6.45 therefore we need 7 risers : 1290 mm / 7 = 184 mm per riser.

7 risers – 1 = 6 treads x 250 mm = 1500 mm overall width!  Check to make sure we have room. Either widen each tread (flatten angle) or steepen by introducing a tread overlap. Just remember guidelines above!

In terms of practical construction considerations there are a number of ways to go about constructing the stairs but I decided that cutting the rise and tread into each support was the way to go.  I used 2 pieces of timber and joined them to give sufficient depth of timber after each slot had been cut out (in South Africa you battle to get check timber wider than 225mm). I used a 225 x 38 + 115 x 38 to give it some strength and joined them by screwing 100 mm long screws though the 225 into the 115 at each tread at the riser interface (narrowest point).

Once you have one marked out and cut, you can use it as a template for the rest. (just remember to subtract for the thickness of the decking planks!)

To support each support I chose to cast a small concrete foundation to which I anchored some angle brackets. This made life easier as each support could be adjusted individually to make sure they line up.

Here are pics of what I did, enjoy:



April 11th, 2011

Here is a link to an article written by Dad (Dave Jones) about our pool for the Knysna-Plett Herald, which is one of the local newspapers where they live.

Obvious bias aside, its always fun to see something you have been involved in appear in the paper.

Here is the link:

Original Article


February 7th, 2011

We have had the idea of constructing a natural swimming pool for some time now, but finishing some of the other house necessaries have taken preference. However with the deck finally completed and some of the other renovations complete, we decided to tackle the pool next.

Finished pool

Finished pool

This is the third pool we have “self built” and we used much of what we had learnt with the first two in this one. One major difference this time is that we decided that we wanted to build a natural swimming pool. What is a natural swimming pool you might ask? Well my definition is a swimming pool that doesn’t require the addition of any chemicals, relying on natural processes to keep the water clean and clear. If correctly designed, the idea is that pool becomes an extension of the garden and becomes part of your garden’s ecology.

The basic principle of how a natural swimming pool works is, in order to keep the water clear, you need to deprive the algae which is always present in the water of nutrients. This is achieved by the large number of plants and large areas of gravel that the water flows through. The plants and the bacteria that colonise their roots and stones, compete with the algae keeping the nitrate and phosphate levels low. Low levels of these nutrients means the algae simply cannot bloom and the water stays clear.

The idea of natural pools is not particularly new and they do seem to be gaining popularity, however getting reliable and useful information on how to tackle a natural swimming pool turned out to be quite a challenge. Whilst there are quite a number of sites which showcase the projects completed by pool contractors, as well as companies selling bio-filters and related equipment. We found they were very skimpy on details. Some forums do exist which did offer some advice, but generally these were more orientated towards Koi ponds.

This lack of available info is something which I am sure will change as natural swimming pools become more commonplace, especially here in South Africa, although I suspect most experts will probably keep the info close to their chests. A book that did help a bit was “Natural Swimming Pools: A Guide to Building by Michael Littlewood”, however I found it did lack much of the nitty-gritty on how to actually build a pool. So after much reading both along and between the lines here are the basic principles for what I believe will make a successful natural swimming pool:

  • The planted area needs to be about the same surface area as the swimming area (ie. 50 / 50)
  • Swimming area needs to be as deep as possible, at least 1.5m (increases volume of water without increasing area exposed to sunlight)
  • The water needs to circulate slowly but constantly through the system.
  • Total size should be as big as possible (Littlewood refers to a minimum area of 50 sq meters – we broke this rule and opted for an area of 25 sq meters due to space constraints)
  • Water needs to be well oxygenated (plants and/or water action)
  • You need to provide maximum surface area for beneficial bacteria to colonise ie. roots and gravel beds.

Because our garden was not big enough to accommodate a large organically shaped pool and also there is a significant level difference between the floor/deck and garden level, we decided to opt for a more contemporary design.  Here is the final Sketchup drawing showing the planned shape prior to me drawing up the final details. The physical sizes are as follows:

  • Main swimming area : 2.4 m x 4.4 m – depth 2.4m with a continous step along the deck side.
  • Intermediate pond: 1 m x 2.2 m – depth 1.2 m
  • Lower pond : 0.8 m wide x +-10 m long – depth 0.6 m but varies from 0.1 m to about 0.4 m with the gravel inside.
  • Total volume 25 500 litres

Pool Concept

Because we wanted the swimming area to be close to deck level we decided to terrace the pool with the planted sections stepping down to the garden level. This way there was some continuity between the deck / pool level and the grass lower down. One thing we did pick up from other designs was that the water is almost always flows from under the planted area up and towards the deeper swimming area. In our design this wasn’t really possible, but since we couldn’t find any logical reason way it would make a difference reversing the flow, we decided to try it out in reverse.

Digging the hole

The Digging Starts

In terms of physical construction we decided on using standard 190mm wide masonry blocks, reinforced with reinforcing steel and concrete in-fill. The reason for this is we wanted to have narrow as possible walls, and this was the only way I could think of achieving this without going the shuttered reinforced concrete route. To waterproof the structure a glass-fibre reinforced resin system again proved to be a cost-effective solution. Prior to fibre glassing the block-work was simply plastered to the desired shape and detail and allowed to harden.

First course of blocks

First course of blocks

Block work complete

Block work complete

A large part of any natural swimming pool is obviously the water plants and we were  very fortunate that my lovely Mum was easily persuaded to start collecting and cultivating plants. I think she even started collecting before we had started digging the pool. She has very green fingers and very soon had quite a collection of plants happily growing in some temporary containers. Apologies to the neighbours who had to put up with all the mosquitoes who obviously loved the still, stagnant water. It’s interesting to note that there are no mosquitoes in the swimming pool, which I am sure is mainly due to the running water and our collection of eight hungry mosquito fish.

Gravel ready for planting

Gravel ready for planting

Green fingers

Green fingers

In the end she managed to collect quite a variety of indigenous water plants which I believe have contributed largely to the success of the project. We had to create shallow and deeper sections to accommodate the different plant types, but it worked out well in the end as it makes it all the more interesting. Special thanks to Robs & Vaughn who also let us raid their reservoir of additional water plants for the deeper intermediate pond.

In order to get the water to flow through the gravel of the planted area we placed three slotted 40mm pipes running the entire length of the lower pond. These join and then feed into a stainless steel filter box (actually an old dishwasher with a washing machine drum inside). This ensures a steady slow flow of water through the gravel and the plant roots keeping the beneficial bacteria happy and fed, and the water clear. To circulate the water we used a Hailea H9000 submersible pond pump which delivers about 5000 l/h. This means the water is turned over every 5 hours or so but the energy consumption is only 105W.

In terms of costs, because of the additional planted area and the plants, the initial capital outlay is obviously much higher. However in terms of running costs there is obviously no comparison to a conventional pool and there definitely is no comparing the finished product!

Some of the fun things we included was a pool window (at the request of my 8 year old – see ‘Mermaid’ pic below), rainwater top up system with level control and a pool light. The pool has been running for almost 2 months now and apart from one week in the first month when it was pea green, the water has been perfectly clear. Slight variations in colour do occur after rain but are hardly noticeable. Total water consumption (apart from rainwater overflow) has been an average of 150 litres per month, not bad for the height of summer .

For more details and a step-by-step building, watch out for my how-to guide coming soon.


January 19th, 2011

I was listing to a panel guest on SAFM  – think it was Jeremy Taylor, talking about rain water harvesting. I agreed with much that was said but one of his solutions struck me as a bit odd and it got me thinking that there has to be a better way.

He was talking about the problem of how to handle the tanks running dry when the house is setup to run off an automatic pump. Obviously if you have a Municiple supply the idea would be to connect back to this supply. His solution was to install a ball valve near the bottom of the tank running off the mains. Now whilst this will certainly solve the problem of the tanks running dry, it seems silly to waste the inherent  pressure of the municipal supply by filling up your tank, which you then have to pump out again to make usable. Not only does this waste electricity, but also will wear your pump out quicker. Also during a power cut you will be a bit stuck.

So this begs the question is there a better way? Well a less elegant system would be a set of two valves to enable a switch over between tank and municipal supplies, but that would be too easy and requires user intervention – so this is what I am going to do:

I am going to install 2 solenoid valves (1 ” irrigation type) – one on the municipal supply, the other on the tank. These I will connect via a transformer to a float switch. The float switch I have has 3 wires which effectively gives you 2 states, “tank empty” or “tank has water”. So by simply using this float switch as a kind of relay you have an intelligent system which will automatically switch between the 2 supplies without the need for any fancy control panel.

Another advantage of this system is that the pump can’t run dry and un-prime itself, disadvantage I suppose is the transformer is always pulling electricity. I doubt this is going to be a large cost factor as the solenoid valves are designed to very efficient, so I would be surprised if they use more than Jeremy’s system above. Also in a power-cut you would need to manually operate the municipal solenoid valve to get water – but this is easy enough as most solenoid valves have a manual overide.

Here some pics and a a diagram of what a mean (note non-return valve on municipal supply as precaution):

I am about halfway though my installation so I’ll keep you posted on how well it works, as always if you have a suggestion or comment post away.


October 7th, 2010

We are currently busy constructing a natural pool which I promise to post some articles about, however I though it might be interesting to show you what we did to solve two of our problems:

First was to find a cost effective way to hide away the very ugly precast concrete (Vibracrete) wall, the second being to try find a use for the copious amount of stones that came out of the pool excavation.

We hit on the idea of using a Gabion basket to create a dry packed stone look as the base to a wooden screen wall – because of space constraints we used a reno mattress (normally laid flat)  turned on its end. This reduced the thickness to 300mm but is only advisable if you can support it off something sound – in our case the existing wall.  The advantage of the Gabions is they are a wire basket that contain the stones, which means the stones don’t have to be as perfectly shaped as they would have been if it ws a conventional dry packed wall.

On top of the stones we installed a wooden screen constructed from Latte (basically treated alien saplings with diamter between 25 and 40mm).

Here is the wall about halfway constructed, the Vibracrete wall is visable to the left:

Gabion stone wall

Gabion stone wall

Some more pics now wall is complete:

Gabions and latte

Wall Complete Side View

Gabions and Latte Cladding

Wall Complete


August 12th, 2010

I am sure someone could make some money out of this one.

My lovely wife bought our two Jack Russel’s new dog baskets which are basically a canvas / wood construction. Although the dogs loved them they actaully took up a fair amount of space. The idea was to turn them into double bunk beds to halve the foot print.

As they have a wooden frame it was simply a case of screwing the legs together using a small block of wood. The result works really well – old man Oscar sleeps at the bottom and the young upstart Daisy takes the top. I can only see it being a problem if your dog cant jump  up – since our Daisy is regularly  caught on our much higher bed this isn’t a problem.

Here they are trying to look cheerful (actually bit suspicious of bosses intentions as it is nowhere near bed time) :

Dog double bunk

Dog double bunk


August 11th, 2010

I am cheating a bit on this post because truth is I am already quite far down in the construction. Mind you it has taken me a while to get to this point. So far the main deck is complete with one of the staircases, with the two bigger still ones to go.

The main deck is 4 m wide and about 15 m long with 3 sets of stairs (one wrap around). We opted for 90mm  Garapa decking which we chose to lay smooth side up and to leave untreated. Fortunately the wood seems to be properly kiln dried so has showed little signs of splitting.

With regards to securing the deck I found nifty fixing system which is almost hidden and doesn’t require any screws through the deck. The resultant look is a very smooth clean finish.

It terms of support I used 2 rows of 75mm pole droppers + the wall of the house where possible. The poles support two main beams (225x50mm), with 150x38mm beams running the 4 m width of the deck at about 450mm centers. The result is a very stiff and stable surface surface which is great – I hate it when there is give on a deck because the design has skimped on support beams.

The most difficult part was definitely getting the planks to run in straight lines as this would be very noticeable on a 15m run. The whole operation was made easier by using the plank lever which lets you squeeze the planks together, however the method that I eventually settled on was to lay 5 or 6 rows of planks with the clips installed but not screwed down. I would then run a gut line about 50mm ahead of the last row with same offset from the house. Using the gut line as a reference I was able to make a mark on each bearer representing a straight run. I then made a wood spacer equal to the smallest of the measurements made from the gut line mark to the last row of decking. Then working form one end with the plank lever and the spacer as a reference I then fixed all 5 rows to the bearer. This way any deviations are corrected over 5 rows as the clips tended to space the planks evenly. I have included some pics below showing progress so far:


June 27th, 2010

We hadn’t actually planned on doing a floating cabinet, but after the tilers had finished it seemed a pity to cover up the stunning floor – so we hit on the idea of doing a floating cabinet. One requirement though was that it needed to look like a smart box – but it needed to include nice big drawer to hide away all the usual bathroom clutter.

The only way I could think of pulling it of was to make a stiff steel frame which would be clad with the bamboo afterwards. Because it needed to be square, so the cladding would line up, I used 25 mm square tubing which made welding easier. The tubing was also pre-drilled at regular intervals so I could screw the cladding to the steel from the inside.

Here is the box getting its final coat prior to cladding.

Steel sub-frame

Once the frame was painted, the 16mm bamboo cladding needed to be screwed & glued together (at this point you will discover how accurate and square your frame is). Mine was far from perfect but nothing a bit of planing and wood filler couldn’t fix. Here are some pix of the gluing and clamping:

Bamboo Cladding Clamping

Bamboo Cladding Clamping

Once the cladding was on, the rest was relatively easy – I spray painted the bamboo with clear Acrylac paint and cut out the holes for the basins. To mount the box on the wall I used Hilti HIT 150 chem fix with M10 threaded rod – if you haven’t used chem fix before you should, as it is way superior to expanding bolts and if done correctly as strong as the substrate, without putting unnecessary stress on it.

Mounting Bolts

Mounting Bolts

Basins in

Basins in

As I was basically fixing a ridged steel frame to rather fragile porcelain tiles, I hit on the idea of using a glue gun to make a gasket on the steel to tighten against. It seemed to work as the box is totally solid and nothing cracked while tightening up the bolts!  With the box mounted the sinks could be installed plumbed and the large drawer installed. We omitted a handle to keep the illusion of it being a box. Here is the finished product:

Drawer Open

Finished Cabined

Finished Cabinet


May 30th, 2010

Part of the renovations for the house involved creating a entrance from the kitchen through to the scullery. The challenge was that the scullery was a previous extension to the house, so the door frame was through what was originally an external wall (280mm thick).

To compound things we din’t actually want a door through to the scullery as it was nicely hidden away round the corner. As a result we ended up with a rather large imposing opening which frankly looked a bit odd.

So my lovely wife came to rescue with the idea of turning it into a chalk board – come shopping list. I admit I was skeptical at first but once the frame was painted with chalk board paint (water based PVA) it looked stunning. It also is surprisingly practical – when I need to go shopping a simply take a picture with my cell phone and I am good to go.

I have included some pics below to give you an idea of how it looks – the lintel (top part) I use for recording my rainfall figures until I get time to capture them in a spreadsheet.

Chalk board Door Frame

Chalk board Door Frame


January 29th, 2010

One of the projects we had on the list to do was a fishpond. I have always liked the idea of having a couple of fish, but always thought the fish fund-i’s go a bit overboard with all the complicated filters, UV sanitisers etc. I wanted a pond that would basically look after itself, but support a couple of fish which will get fed occasionally.

Another hassle with ponds is that in our hot and windy climate they always needing refilling due to losses through evaporation – sometimes you think you have sprung a sudden leak after a hot and windy day! This means you either need to keep filling it up – or rig up some fancy ball valve arrangement from the mains.

So here enters the ‘Green’ pond idea – keep it topped up with rainwater. Now I know what you going to say this has been done before, but I think I have found a novel solution of keeping it neat and contemporary. In fact I challenge you to figure out how its done in the next picture:

Hidden rain filler

Hidden rain filler - pond on the left

So here’s what I did – I used the same design as for my rainwater tanks and set the level of the grate to the same level as the water level I wanted in the pond. Connected it under the paving with piping and presto you have a pond which is filled each time it rains. Once the level reach’s the top of the grate, it simply overflows like the downpipe would have anyway. I then cut the bottom out of a clay pot and put this over the grate to hide it. Here is a sketch to illustrate:



The design of the pond is simple rectangle  1.2mx1.5m  – I made it deep so the water level could fluctuate between rainy periods, about 1m. It is constructed of a 220mm 2 layer brick wall with brickforce wire reinforcing every 3rd coarse. It was then plastered and left to cure for a few weeks (mainly because I was putting off the nasty fiberglass job). Once it was nicely cured and dry, I then gave it a coat of normal resin with a second coat of black flow-coat to finish off. Those who have fiberglassed before will be saying but what about the glass matting – well I looked at the pond – there were no cracks and I didn’t have enough matting. So I decided to take a chance and only use the resin – touch wood it seems to be working and as long as it doesn’t crack it should stay that way!

To keep the water clean I put a good 4 inch layer of stone at the bottom to act as an anaerobic zone and then planted it out with some Irises and a water Lilly. Finally a submersible pump circulates the  water through 3 nozzles in the back wall. I was amazed that with 6 fish – 2 goldfish and 4 Koi that although it did initially go murky, after about 4 weeks the water stabilized was crystal clear. It has stayed that way for 2 months now and the only maintenance is I have cleaned the sponge filter on the pump twice. I know the system is healthy as 2 frogs have produced about a million tad poles which seem quite happy as well.

Now for the clincher – we are in the middle of a drought – it is in fact illegal to fill your fish pond with a hose and to date I haven’t filled it once. Enjoy the pics!