1. Schoolhouse Floor Plan

Table of Contents

Renovation Plan

Nothing structural is planned but the interior needs to be repaired and updated.

  • Replace ducted oil heating with ground-sourced HVAC ducted heating/cooling system.
  • Remove carpet and restore exposed floor boards.
  • Repair and restore all windows.
  • Replace exterior door.
  • Install skirting boards to replace the odd-sized quarter-round mouldings currently in place.
  • Update light fittings.
  • Update décor.
  • Interior repainting.
  • Remove library shelves.
  • Pull up non-original mismatched floor boards and replace with best boards from other rooms.
  • Sand and finish floor boards.
  • Strip all paint from beams and finish as natural timber.
  • Repair and repaint vaulted ceiling.
  • Replace wall render on all walls to level and smooth in traditional lime render.
  • Walls to be lime washed to allow lime render and stone walls to breath again.

Original flooring is 130x24mm T&G hardwood timber, possibly Tallowood. This is a very dense and hard wood. Floors are very worn in places but don't appear to have been sanded since originally laid. Various access points have been roughly cut into the flooring, at least one per room, for installation of the ducted oil heating. It was clearly intended that carpet would be laid to hide this work. It appears that no effort was made to lay ground sheets when repainting, for the same reason. Gaps in the flooring were roughly filled with a plastering filler.

Originally the most important room in the building, the classroom, includes a large open fireplace that is still operational. Access was originally only via an external doorway and later also via an internal doorway from the study, added after 1945 according to the recollection of an ex-pupil who saw the room recently. Four large double-hung sash windows provide light and views to the south, west and north. The 15 foot high vaulted ceiling is lined with painted timber paneling, with exposed tie beams extending across the width of the room. The uneven stone internal wall surfaces have been roughly rendered with lime render and then painted, with acrylic paint used in later coats. There are extensive library shelves covering all of the wall separating this room from the rest of the residence.


Old wall-to-wall carpet

The bulk of the wall-to-wall carpet and foam underlay was well past it's useful life and was removed as my very first inside task, so that I could see exactly what I was dealing with underneath.

Tie beams

The colour scheme for this room was "yellow". All walls, ceiling paneling, tie beams, window trim is painted yellow at the start. All but two of the seven low voltage spot light fittings wired into the beams were not working. The bulbs were not the problem, so I've bought replacements that I will install after the beams are finished.

My father taught me how to strip paint using a butane gas torch and scraper when I was 10 or 11 years old. This time I used an electric heat gun but it was still hard work. I'm aware there will be lead-based paint there as well.

To get the residue off I used an eco-friendly water-based chemical stripper (Dumond "Smart Strip"). The product is a foam that can be left on for hours and will not dry out completely. Before scraping the stripper and paint residue off I sprayed with a water bottle to soften it. Finally, I went back over the beam again, wetting it again in sections and scrubbing with steel wool. I alternated wads of steel wool that I kept soaking in a bucket of water. All up, with three passes from start to finish, it took between 6 and 8 hours per beam.

With the colour of the timber that has emerged (photo 6 below), I'm considering a sky blue/teal colour for the vaulted ceiling. I want the walls to be a white colour.

2. Starting with yellow painted tie beams. One of the new spot lights installed to test

3. One half of one tie beam heat-stripped. There is still a lot of residue but I can see the timber is a light colour with a slight pink tinge. I think it will look good natural

4. Three tie beams heat-stripped. It's a little precarious on the trestle and a bit more height would make the job easier. Decided to leave the old carpet down until the heat-stripping was completed, because the hot paint was soft and sticky and I did not want to tread it into the bare floor boards

5. Decided to invest in a scaffold. This one was just AU$200 new at Bunnings and does the job. No guard rails around the platform but the tie beams are now at shoulder height. I believe the design allows a second scaffold to be stacked on top, which would provide a guard rail if wanted

6. Completed. All sides heat-stripped then the residue was chemically stripped, applied according to directions then removed by scraping, followed by scrubbing down with water and steel wool

7. Carpet and bookshelves removed

8. Carpet and bookshelves removed (reverse angle)

Flooring - Part 1

All of the remaining carpet around the walls has been removed and the hook strips pulled up, which was hard work because almost every nail had rusted into the floor and had to be pulled out separately.

I especially want to retain the original timber in the flooring in this room even though it has obviously been pulled up and put back down at least once in it's lifetime.

Spent several hours punching floor nails down in preparation for sanding at some time. My regular nail punch and claw hammer weren't up to the job so I got myself a flooring punch and switched to the 4 lb (1.8 kg) block hammer pictured below. The heavier hammer and punch gave much better control and safety. I managed to not hit my knuckles at all. The boards pictured are 5 1/4 inches wide (and 24 mm thick).

9. Punching nails was easier with these

New skirting will need to provide a good seal against the cold air coming in through the wall vents, pictured.

10. Breezy

The floor boards in the lounge have pretty obviously been pulled up and re laid at some time. My guess is that this occurred when the room was still in use as a classroom because there are large spills of coloured paint on the floor, including the newer unmatched floor boards, that do not appear to match any colours used in the décor. present or past. They are kind of colours that you see in children's painting sessions. My guess is that the entire floor was pulled up and re laid to close gaps. The newer floor boards were laid in a continuous section, I assume to fill the shortfall resulting from closing the gaps and replacing boards that were damaged when pulled up.

Archeological Finds

11. Two pieces of wood used to pack a window frame. See right -->

This would have most likely been lamp oil. "Snow Flake" oil was sold in NSW in the 1910's according to one catalogue of historical Sydney signs. That would mean some repair work was done on the window after that time

12. An image of a box with matching print from a Google search.

If you had just glanced a look at the box you'd be forgiven if you'd thought you saw the words "Snake Oil"

13. Artifacts

Small artifacts found so far during the renovation. The fragment of a ceramic inkwell and the pencil were found under the floor jammed between a wall and the timber bearer supporting the floor joists. How they got there I can't say. The chalk came down when I was removing render from between one of the end tie beams and the end wall. Clearly it was thrown up there, so perhaps some rowdy classroom sessions were had. I don't recall where the marble came from but the game was a very popular passtime up until the 1950's. I remember playing marbles but wasn't very good so soon lost mine. Then "jacks" (knucklebones) took over in the early 1960's.

Fireplace conversion

Part of the interior makeover will replace the open fire in the lounge with a slow combustion wood heater. This is primarily for ambience and will provide a backup heat source if needed.

The insert will be a Lopi Flush Wood Large with Shadowbox fascia (bronze), chosen for it's efficiency and relatively large window in a fan-assisted unit. I decided not to go for a period styled insert in favour of the larger window size and simpler style. The original open fireplace quite possibly never had a mantle, just a simple brick and stone opening. I'm still looking for a mantle design that might suit the room.

The installation documented in the images below, as a one man operation, was a slow and very considered process. First, remove everything that could be detached to lighten the load. Complete this insert weighs 250kg and without door and liner bricks probably around 225kg. The main problem lifting it vertically was getting enough distance between the insert and suspension point for the ratchet block to work properly. After that it was a matter of ensuring the inserts centre of gravity stayed in line with the block, otherwise I would have a disaster on my hands.

The design for the insert surround allows it all to be taken apart for access.

14. Lopi Flush Wood Large with Shadowbox Fascia (Black)

15. Bronze Shadowbox fascia. The actual bronze shadowbox is not this bright (see below)

16. New Lopi insert hoisted off the trailer

17. Lopi insert hauled inside. The hoist mechanism was used again, this time horizontally across the floor inside, one click at a time

18. Old open fireplace cleaned out. Insert at far right under plastic waiting for it's final move

19. Repaired brickwork using high temperature mortar

20. The surround fascia (15mm compressed fibre cement sheet) will be hung off these blocks, made from the same fibre cement sheet material. The top edge is angled 15 degrees. Builders epoxy bog was used to fill behind to provide a level bed and anchor bolts used to fix to the wall

21. The floor for the Lopi insert is 15mm fibre cement sheet on Hebel blocks. The floor of the open fireplace was leveled with leveling cement mix and a wet lime mortar used between the blocks and cement sheet. The construction is not bonded so is free to expand and contract

22. 15 mm fibre cement sheet used to pack out around the opening behind the fascia. These all need to provide an even flat surface behind the fascia. Cement was used to fill behind the horizontal strip at the top, hence the timber supporting a temporary plywood form under the strip

23. The surround fascia rear view showing the corresponding hanging blocks. The fascia can be removed again if necessary once installed. More blocks to be added at the bottom

24. Surround fascia installed.Panel at the bottom will also be removable. The design of the insert floor (above) and packing strips allows wiring to be routed behind the surround fascia and bottom panel. The surrounds and hearth will be finished decoratively later

25. The Lopi insert installed. The power cable for the insert blower fans can be seen emerging at bottom-right. A mantle can be built onto the surround fascia and a shelf below the insert. They are planned but I'm holding off until I decide what I want for the final finish

For the final move of the insert into the new cavity I used the ratchet block hoist again. I assembled a platform in front of the cavity then suspended the insert via the hoist from the scaffold frame. To get the height needed I strapped two 12x2 inch hardwood planks edge-up across the top of the scaffold to act as a beam. The strap connected to the hoist passed down between the two planks to minimise the risk of the planks "tripping" (flipping 90 degrees). I could then very slowly roll the entire scaffold (on its casters) over the temporary platform drop zone and then lower the insert. Once on the platform with the scaffold out of the way the hoist suspension point was relocated to the beam above the cavity opening just to take some of the weight off the front of the insert as the insert was rolled from the platform into the cavity. The insert comes with tiny rollers at the rear just for this one-off operation (nice addition Lopi) but it is still necessary to lift the front half of the insert (about 115kg) to use them. Taking the load off also was to avoid tipping or damaging the front edge of the new floor of the cavity. That saved me having to build a more elaborate temporary platform.

26. Long shot to show insert relative to the room.

The surround height is 1525mm. Floor to beams is 3300 mm

Flue Installation

The 9 metre extension ladder in the photos was bought specifically for this job and an extra ladder will be useful for several projects I have planned. At a collapsed length of just over 5 metres it's tricky to get it to the vertical position even before extending it. Even though I'm above average height I am just short of reaching the ladder's mid point or centre of gravity so, without a second person to help me, I need to anker the bottom somehow to swing the ladder to vertical.

I spent the next weekend installing the solid tube stainless steel flue. First I needed to remove the old open fireplace chimney cowl, which I assume this was a latter addition.

The old cement flaunching was cracked and mostly loose and easily cleaned off. Then the old cowl had to be raised and chocked little by little until the 400mm section extending down into the chimney was clear and the bottom edge could sit on a couple of wood battens. Getting it down from there was easy. I attached a rope to the top and from a safe distance away gave it a solid sharp yank so that it would clear the ladder on the way down. Conveniently, it rotated a full 360 degrees on decent and hit the soft ground base first before coming to rest on it's side.

The cowl had a bullet hole in it clearly shot from one of the two hills across the main road (you know who you are if you are still living :-) A pretty deliberate and dangerous potshot.

Not incidently, for safety I have tied the ladder securely to the chimney at the top.

27. Old cowl

28. Old cowl

29. Old cowl gone, lying on the ground where it landed.

30. Old cowl, complete with [the obligatory] bullet hole

Someone clearly took a pot shot at the cowl at some point. The bullet presumably ended up in the fire place below because there is no exit hole. I'll retain it for its historical significance and turn it into either a garden feature or a lamp stand

The flue was pre assembled at ground level on the floor in the living room then, on the first dry windless day, partially disassembled again to drop it down the chimney from above. To support the drop down the chimney, a rope was threaded through the whole flue length and then attached to a short rod across the bottom of the flue. The first (lower) 2.5 metre section was raised to the top via a second rope, lifted and lowered into the chimney and hung via the support rope while an octopus strap with vinyl sleeve for grip held the flue as the next 1 metre section was reattached. Gently dropping the entire flue another metre the last (top-most) 2 metre section was reattached and the whole flue lowered and finally connected to the insert from below. 2.5 metres was the most I could manipulate safely.

Unfortunately I don't have any photos of that process because, having a fear of heights, I was focused on self-preservation and completely forgot about photos. (Before starting with this DIY project my climbing limit was just four rungs up a ladder.)

31. New cowl attached to a chimney plate collar on a plate that is now under new cement flaunching. Some white knuckles on the ladder while the other hand holds the phone/camera.

32. Finished.

Not much to look at but from a restoration perspective it's closer to the original appearance.

The first firing of the insert was a success. The flue drafted nicely right from the start. This initial firing is to bake the paint on the insert and an opportunity to test everything and get an idea of the heat it can generate.

33. First fire


Originally I intended to wainscot and possibly fully or partially line the walls but after removing some of the render around a bulging crack high in the wall and after experiencing how hard it is to drill holes in the bluestone (basalt) for the new fireplace surround I decided I'd commit to strip it all off and re-render.

I will be replacing the existing rough render and it's coating of non-breathing acryllic (a.k.a. plastic or latex) paint with traditional lime render to a smooth level finish and a lime wash. This will restore the two important abilities of the walls to:

  1. regulate the humidity in the room by allowing the walls to absorb/release moisture from/to the air, and
  2. allow the render and mortar to self-repair fine cracks that naturally occur with expansion and contraction.

Of course, there are other reasons to re-render in traditional lime.

  • the light that reflects off calcium carbonate is different. Softer but brighter. See box ->
  • the satisfaction from living with something closer to the original.

The beauty of lime



Limewash is uniquely beautiful, incredibly bright and gives off such a depth of colour that it appears to shine. It is impressively vibrant at night and can lighten up the darkest of interior rooms or heavily-shaded courtyards. On the Atlantic coast, light houses have been limewashed white in living memory, as have day marks for shipping. Whether used on stone and earth houses, cathedrals and dry stone walls, or palaces and field barns, limewash has stood the test of time and has protected and decorated the structures of the world.

So what gives limewash these wonderful qualities? The beautiful luminosity of a limewashed surface is due primarily to the reflection and refraction of light. The intense brightness of non-pigmented limewash is largely due to the reflection of light back into the viewer’s eyes. A more subtle, but startling quality, is the refraction or bending of light through crystals of calcite (the carbonate mineral that forms through the process of carbonation). Light refracted through calcite splits into two rays, one fast, one slow, and the visual effect is a doubling or twinning of the light emitted by the crystal.

The intensity of the light itself is not changed, but when compounded by millions of microscopically-small calcite crystals increasing over time as the surface continues to carbonate, the effect is a surface that appears bright and vibrant with subtle internal texture and variance. This mottled effect provides depth and interest to the viewer

34. The crack that started it all.

35. Three-quarters of the first wall stripped after approximately 12 hours of effort. The old render comes off easily when chipped at using a crowbar.

A horizontal 400 mm wide band of cement render applied at the level of the wood blocks was harder to remove but fortunately could be pried off in small chunks after a little loosening without any damage to the stonework. Similarly around the doorway, which is a later addition, and will be harder to remove

36. A close-up of wood blocks embedded along the wall at chair-rail height. According to a former student pupils faced this wall in class, so perhaps these are to support blackboards attached to the wall. I don't know yet if they go all round the room, in which case they may be to support wainscoting that has been removed in the past

37. Wall completely stripped of old render. Hard cement render around the door will be removed when the doorway is rebuilt. This is a stitched panoramic, things aren't as wonky as they appear here

38. Fireplace with render stripped from wall. Continuing to remove render around the room and still considering whether to flat render walls or leave some stone exposed.

39. Clues to previous colour schemes. The yellow plastic paint at bottom is the main reason for the complete strip and re-render. This patch of wall was around and under a timber moulding installed just before the yellow colour was applied. Under it is the off-white colour, which may be an old lead paint, it has that linseed oil look about it. Below that is the teal colour. I found other areas of this that were much thicker and much more the colour of copper oxide and penetrated right through the plaster so I'm imagining it may be a lime wash

I decided to buy matured lime putty rather than make my own. The saving in time to make and store and mature it is worth the additional cost and I get a known quality product.

40. Just an idea but increasingly unlikely

Any stone left exposed will need some kind of sealer applied to the lime pointing to minimise dust. Traditionalists suggest "lime water" as the recommended stabiliser. This is a saturated solution of calcium hydroxide in water, about 0.2% or 2g per litre. It is usually made by adding some lime putty to water, mixing really well and then leaving the excess to settle before decanting the clear liquid that is the lime water.

I guess it works the same way lime scale deposits work and anyone who has cleaned a shower stall knows how hard it is to remove lime deposits.


I started out assuming and not properly understanding what "suction" referred to in plastering but fortunately I haven't started rendering yet. It is the action of the substrate in drawing water out of the render and high suction causes the render to dry too quickly making it brittle etc. It is not what makes the render stick to the substrate. High suction requires treatment to reduce it, usually by spraying with water enough to dampen but not leave the wall dripping wet.

The sequence on the lounge room walls will be:

  1. dubbing out
  2. a haired scratch coat
  3. a coarse stuff coat
  4. normal finish coats

It's very difficult and/or expensive to get hair in Australia so I'm going to use a synthetic fibre instead. It's a polypropylene fibre used in cement work including stucco and it has some advantages, including not needing to be slowly teased into the mix.

I have bought a couple of 15kg bags of fly ash to use as the pozzolan if I need it.

I've trialed some mixes of render to get a feel for the preparation, application and the effect of the dampening treatment through the initial curing stage of the products in different situations. Also looking at the colour change of the product as it cures.

  • 3:1 course sharp sand and undrained lime putty. This is a well graded clean sharp sand pale yellow in appearance. I'll measure the actual void ratio for this material and adjust, but as it was it would work as a scratch or course coat.
  • 3:1 river sand and drained lime putty. Predominant grain sizes vary from 1mm up to 5mm, with a few pieces over that size, medium sharp to smooth grains. Yielded a very grainy mix and not very sticky on the trowel. As with the sand, I'll measure the actual void ratio of this material. I think this mix would be ok for filling between stones and dubbing out.

To measure the volume ratio of sand to space between the grains, or void volume, I used the saturation method. I put 150 ml of bulk material into a small bottle and then slowly poured water into it until the water started to saturate the top surface of the material. For this measurement a tall narrow vessel is more accurate than a wide flat vessel. The volume of water used is the volume of the voids. For both materials this was within one or two ml of 50 ml, so the 3:1 basic ratio is confirmed for both types of sand.

After that I tried a 2:1:1 mix of river sand, sharp sand and lime putty. As expected this mix is a little smoother than the straight river sand mix. I used this to do a minor repair to a corner of the rendered plinth around the outside of the bluestone walls where some render had broken off, so I will see how it looks after a few days.

I don't need to colour-match any of the interior product but will for the exterior re-pointing where it's clear that ground charcoal and ash has been used to blend with the bluestone. This accentuates the straight lines of the pure white lime ribboning that was applied originally.

Results after a couple of years - yes it's been that long

The sharp-sand-lime mix has worked very well, it is hard and takes some wear while the river sand mix does not take wear as well. The corner repair did not work. It came away so I need to work on getting the bond right.


With the old render removed, now is the time to repair any windows that need it.

I'll start by removing the architraves and jamb extensions, clean out all the debris I find in the frame-to-stone gap. I'll retrieve the sash counter-weights as well if they've come loose.

Visible below the inside timber stool is the white block of limestone that is part of the exterior sill and showing all the chiselling marks. I'm thinking this should be left exposed.

At least two windows have majorish problems:

In one window the timber interior stool appears to have dropped a little at one end but, in fact, the whole frame is slightly lopsided and appears to have been this way forever. So this will be fixed but it will require removing and disassembling the entire window including the frame.

On another, which is the only one with a working sash, and the one I'm starting on, there's a gap on the outside of about 15mm at the top left corner between the limestone block cement filler and the frame (photo 41 below shows the corner with gap closed). It was fixed by cleaning out the gap, pushing the top frame corner out as far as possible and re-wedging the frame tight (photo 46 below).

This has exposed another problem with that window. The timber lintel is skewed and protrudes out from the plane of the wall at one end by 15mm (right-hand side of lintel photo 42 below). It has always been this way and the architrave head moulding had been cut (chopped) to taper at the back to allow for it. Simply putting things back as they were will cause problems when I re-render the wall flat, so I've added 15mm width to the side and head jamb extension to bring the architrave out 15mm.

I used strips of wood cut from one of the skirting boards that I removed from another room (photo 43 below) and is the same cedar as the existing jamb extensions. I will then attach a tapered strip of timber to the lintel to give the architrave moulding a flat backstop against the lintel and an edge and seal to render to. The stool has enough depth to take the extra 15mm width on the jamb extensions and will not look any different to the other window stools when finished. Once the wall is rendered there will be no gaps and the lintel will be rendered over and not be seen.

41. There was a 15mm gap in this top corner between the masonry fillet and the green frame. Simply cleaning out the gap, some pushing and re-wedging inside has reduced this to 1mm

42. Architrave and jamb extensions removed. Three wedges can be seen at the top between the frame and lintel and one wedge is used midway up the right side between the frame and the timber nailing block embedded in the wall

43. The jamb extensions made 15mm wider using strips of an old skirting board. Glued, sanded and filled waiting for final sand and re-installation

While cleaning out behind the window jamb extensions another artifact emerged from the rubble in the form of an ink pen. There have always been gaps at the top of the window between the architrave and the wall and this pen has landed in the gap somehow, perhaps thrown there.

44. Another artifact: Ink pen, with fragment of ink well found earlier

Now that I have a solution for the wonky lintel-to-trim interface for the first window and have prepared its jamb extensions for reinstallation, the next job before is to remove both sash, clean them up and open up the access to the counter weights, which have been painted over. The top sash had been painted closed and nailed in position through the horns.

Images 45 and 47 show the multitude of different colours used on the window.

45. Lower sash removed. The upper sash was nailed through the horns and partially painted shut. Required a hacksaw blade between the sash and the frame to cut the nails

46. Both sash removed. A tapered timber strip has been fixed to the bottom edge of the crooked lintel to give the architrave a parallel edge to sit against. The new wall render will cover the lintel and come level with the back of the architrave all around

47. Counter weight access panels removed. The panels (shown on the sill) where so badly mangled by previous maintenance I've decided to make new ones

48. New access panels made from a section of the old skirting board removed from the other rooms because it's the same timber (cedar) and age, made entirely on the table saw

49. A simple innovation I've added this time. Removing the parting bead next time will reveal a hole where a hooked implement such as an Allen key can be inserted to pull the panel out without damaging it

50. The four pulleys with paint removed and cleaned up. The wheel and housing look like they are steel and cast iron respectively and only the face plate is cast brass. The screws are steel also

51. The pulleys are quite heavily built and look like they'll last forever. Because of the potential for oil to collect dust and clog I'm using graphite as the axle lubricant. It doesn't look like they've ever been oiled

52. Paint heat stripped from all surfaces. The slot for the parting bead is a bit mangled also. I want to try to fill and shape the slot edges before new beads go in

53. Sanded, filled, primed and first top coat with new counter-weight access panel temporarily fitted

54. New 8x19mm beads cut from a length of original skirting board out of the "den" and shaped, making enough for five or six windows. Standard beads today are different dimensions at 9mm thick and about 4mm less in height

55. First top coat. I decided to use oil-based full-gloss paint. I'm leaving at least a week between all coats, including the undercoat

The entire frame will be painted before the sash, beading etc. are replaced. Both sash will be painted before reinstalling. The sliding faces of the sash will be left bare.

Sash Repair

56. After initial attempt at chemical paint stripping

57. Original exterior colour

58. Original interior colour

The easiest and quickest method I found to remove the old glazing putty is to use a multi-tool oscillating saw. Tried a heat gun but cracked too much glass, not from the heat but from the pressure applied removing the softened putty due to uneven support behind the glass where putty had fallen out.

These sash are in very good condition except for one muntin section where the fillet between the glazing rabbets was so damaged there was mostly only putty and no wood.

I decided to cut the whole fillet away back to a flat face, cut a 5mm deep centre slot using a sharp knife and glue a new strip of matching wood to create a new fillet. This worked very well - photos 59 to 61 below.

59. Damaged muntin

60. Repaired muntin - the fillet is a new strip of wood glued into a slot cut into the muntin, shown primed ready for glazing

61. A reverse angle of the repaired muntin (left of centre) after glazing. This was my first attempt at glazing and I discovered how sensitive linseed oil putty is to temperature

62. Upper sash re-glazed and repainted

63. Lower sash re-glazed and repainted. Replaced a total of four panes

64. Upper sash - interior side

65. Lower sash - interior side

Flooring - Part 2

About one third of the flooring in this room was replaced at some time with cypress. The planks are a much narrower width and a lighter colour. It can be seen in several of the fireplace reno images above (e.g. photo 26). I suspect it was laid when the building was still operating as a school, i.e. prior to 1946. My only evidence for that is the colours of the paint stains, which I have not found used anywhere in the rest of the building and which look like childrens water based finger paint colours. These odd planks will be replaced by the best original boards taken from the other four rooms, which will all be getting new floors.

The odd boards were all pulled out then replacement boards were selected, positioned and cut to length. Once all the replacement boards were fitted loosely they were all stacked in order out of the way and nailing began. I used whatever worked to close gaps and straighten boards as I nailed them down: clamps, car jack, and even a manual hydraulic ram and jaws kit that I bought cheap. I spaced and aligned each strip working towards a final gap that would be consistent the full length of the room, which turned out to be 20mm. The last step was then to cut some 19mm strips of flooring to lay in the gap. The 19mm strips had no tongue and what was left of the tongue on the strip adjacent to the gap was removed. One edge of the 19mm strips was glued then they were fitted. See image 75 below for the end result. I could have rebated the edges to overlap but decided against it. The floor will be sanded and then filled all over using a mix of epoxy and final stage sanding dust before being finish coated. If shrinkage causes a gap to appear I can always create a rebate along it and drop in a strip of timber, or brass if I want to make a feature of it.

Other than replacing the odd boards the aim is to retain the character of the original floor, so there will be no major structural additions to the subfloor. The subfloor will not be levelled, for example, except where gross adjustments are found to be needed. There are none to this stage.

66. Out of the floor and into the fire to heat the workspace (the smaller pieces, the rest is in the woodshed). The cypress was so brittle not a single board came up without breaking and long lengths literally snapped under their own weight

67. Revealing the structure beneath the hearth

68. All done. Revealing a glimpse of the centreline foundations supporting the floor

69. All done. There were a lot of large basalt stones from where they created a gap in the centreline foundation wall to pass the original oil heating duct through. I've collected the stones for use in other projects.The green duct is for the new HVAC

70. Cutting and arranging the replacement boards

71. All finished. The white on the boards is from where gaps were filled between boards in the rooms they were taken from. It will disappear when the floor is sanded. These boards are 23mm thick so I can take a few mm to get a smooth level floor

72. Where tongues were too damaged the section was removed and a slot routed into the edge then a plastic tongue glued in. This is the same 15mm x 4mm plastic tongue as used in structural particle board flooring only the product I bought was sold as electicians cable snake. They sell it online as such for about AU$18 a 3.6m length. I bought mine at Bunnings for AU$4.10

73. Router and bit - 4mm slot

74. There are small gaps because the boards were not all exactly the same width and I did not want to reduce it by dressing them. I used various methods to compact and straighten, including my car jack and finally a cheap hydraulic chassis straightening kit with various attachments including some spreading jaws that I used for the last row with just a 20mm gap. The main thing was to finish with the last board parallel to the existing edge. I started on the right and you can see where I finished on the left.

75. Because I was not starting or ending at a wall there was always going to be a gap to be filled at the end. It turned out to be 20mm that I filled with a 19mm strip of flooring.

76. The problem with running vents across boards is the lack of support for the short lengths of boards between the vent opening and the wall. The vent in the photo is for the new HVAC, and is shorter than the original oil heating vent. You can also see the effect of time, heat and moisture on the particle board used here to fill a gap. It has disintegrated.

77. Vent relocated

Both of the original oil heater vents in this room had the same problem. Either they were much bigger then reduced in size or their locations were adjusted slightly. They are both below a window and I suspect were relocated further from the wall and away from drapes before the room was carpeted. I relaid boards around the other vent and have moved the vent in this photo to another wall rotated 90 degrees.

Repairing the hole left behind required cutting and removing boards so that the replacement boards could span at least three joists. With the replacement boards around the vent, to reduce the width slightly to match the existing boards I chose to cut the tongue side then cut slots with the table saw and glue new plastic tongues in place.

78. Repair underway.

Boards cut and removed so that replacement boards will span at least three joists. The replacement boards are 1 or 2 mm wider than those below so this time I will dress them a little to match. The joists have been planed and the first board is in place.

The way the joists have been joined to extend across the width of the room seems odd to me. The centreline foundation wall supporting the joists is just one or two board widths to the right so I don't know why they didn't lap both sides over the plate

79. Finished.

The replacement boards are also thicker than the surrounding boards.

The rooms that the boards came from had dark perimeters a few feet wide leaving a lighter coloured square in the centre of the room, as if a rug had occupied that space.

Sanding should fix both those problems

80. New tongues

Reassembling the [first] repaired window

All beads were painted separately prior to reassembly.

New sash cords were installed during reassembly following a method that feeds the cord through all four pulleys before cutting to length.

Starting by feeding the cord in through the outer left pulley down to the weight access opening then across to the next pulley in the order: left outer, right outer, left inner, right inner. Then start cutting the cord to length and attaching the weights for the outer sash: left then right. Pull the weights up as far they will go and clamp the cords at the pulleys then attach the cords to the sash as it sits vertically on the sill. There should be minimal slack in the cord. Once fixed, carefully remove the clamps and gently let the weights take any slack. Position the sash in place and check that it has full travel up and down.

At this point the parting beads normally go in but because I have a single access panel on each side that the parting bead runs through I needed to attach the inner weights and fit the access panel before I could replace the side parting beads. Now attach the cords to the inner sash and check for travel.

Finally reattach the interior staff beading.

This method of threading the cord requires attaching a weighted "mouse" only once to the cord at the start and therefore saves some time. In my case I had removed all four pulleys and was able to drop the cord down and out the weight access opening without the "mouse". I also have a single access opening for the two weights on each side and no internal parting slip panel to separate the weights, which simplifies things but just needed to be sure the cords weren't wrapping around each other inside the weights pocket.

81. Cord threaded through all pulleys before cutting to length saves time.

82. Outer sash with cords and weights attached. In this position the weights will be hanging just below the pulleys at the top.

83. Outer sash in its normal closed position.

84. Inner weights and sash reinstalled. Access panels and parting beads reinstalled.

85. Window fully functional again - the new parting beads fitted perfectly in their slots and did not need nailing.

86. With widened jamb extensions reinstalled. Originally these were nailed to the box frame but this time I've used hidden screws. These and the stool will be painted white leaving the staff beading "Wheat" as a pin-stripe border.

87. Externally the colour is all "Wheat" - to match the windows from the extension renovation.

88. Lever catch installed. Not all catches are designed to pull the meeting bars together to close the gap but that is its job.

89. Lever catch installed - oblique view.

Both sash were a very loose fit so draught seals were fitted to the running edges of both sash before final positioning (Raven RP61 - woven pile strip). The seals were fitted toward the outside (of the cords) on the upper sash and toward the inside (of the cords) on the lower sash to avoid having them run across the weights access panel on each side, only because I felt it might otherwise increase wear and reduce the effective life of the seals.

The colour scheme for the windows is "Wheat" on the exterior and white inside but I've decided to leave some parts "Wheat" even though they are visible from inside. They would normally be painted half and half depending on which parts are visible with the windows closed. When the time comes to repaint I disassemble the window again because it's easier overall and the end result looks better.

The jamb extensions originally had been nailed to the box frame so this time I've used screws that will be hidden once the archtrave goes on. This is consistent with the philosophy of allowing for easy disassembly and reassembly during maintenance. Starting with the head jamb extension I used an auger bit to drill two deep pocket holes (about 120mm) at the interior-facing edge through the width of the panel (about 150mm) to take a screw the rest of the way into the top of the box frame. The side panels were then fitted and screwed at about 45 degrees through the interior-facing edge at the top into the head jamb extension and at the bottom to the stool. The architrave will be brad nailed to these edges to hide the screws and the holes.

Similarly, I've used screws instead of nails to fix the frame wedges in place. With expansion and contraction cycles and vibration etc the wedges can come loose and the frame can shift, as it did in the past to open up a 15mm gap (image 41 above). However, rather than screw through the wedges, all that is needed is to drive a course-threaded screw between the wedge and the lintel as in the images below. If they do come loose it will be possible to tighten the wedges and the screws without removing the jamb extensions, just remove the architrave.

90. Screws driven between the wedges and lintel hopefully prevent the wedges from shifting.

91. A screw driven between the wedge and the embedded wood nailing block prevents the wedge from slipping.

92. New mortar fillet between the sills.

Compare to image 87.

93. On the back of one of the jamb extensions is the name of the building contractor (Samuel Cox) to whom the completed window was delivered.

"Cox Bros"

One of the window jamb extensions from bedroom two was also labeled with this name and I expect every window has the same.

So imagine a cart arriving at the site loaded with eight completed windows (frames, trim and sash). The openings in the stone walls was complete and the frames made to measure but they still needed to be shaped roughly in places to fit some stone protrusions, probably using a hatchet or chisel.