As far as we know, this must have been the absolute ultimate DIY project: one whole year from "thought to fruition" and not without a set-back once in a while. But determination has paid off and we can only congratulate our persistent DIY-er Cyril and have the deepest respect for the task he had set himself.
Early May 2010 (06.05.10) we received Cyril's first email about the above pattern: would this be feasible in wood strips glued down to a concrete base? Once we discussed the options there were (making sure the strips would be as long as needed to avoid joints in places not wanted) Cyril visited our showroom (driving down from the south coast on his bike) where Maple (his preferred wood species) grades were further discussed.
Back on the south coast Cyril had a rethink and thought, why keep things simple:
When straight lines are not simple enough
Now, I'm not a mathematician but that does look very complex, but nice. A bit like a 2 dimensional "Wovin wall".
I guessed (correctly, see Cyril's reply) the amount of saw-waste for this pattern would be tremendous:
I have worked out the saw waste. In fact a did this very early on. Your phrasing 'an over average percentage of "saw-waste"' made me laugh. When I worked out the percentage of the wood that is wasted by this design I almost cried. There will be an most as much saw dust as pattern wood.
Even when it turned out our Design Parquet manufacturer can do many shapes and patterns, as long as the lines are straight, Cyril pressed on.
It not ready-made, then hand-made is the way forward. He "simply" made his own jig to cut the desired pattern out of specified on width solid Maple tapis parquet.
The making of this bespoke jig turned into a project of its own.
As Cyril explained to me it he turned it into a complete "factory line":
The Jig is design to take a plank. When making the first cut of a plank, it is inserted into the jig (as shown by the dotted line) and the three sides of the piece is cut. I then slide the plank into the jig until it hits a stop (as shown by the dashed line). At this point I can cut the last side of the first piece and the first three sides of the second piece. With the first piece cut I can remove it and slide the plank to the stop. With the plank once more against the stop I can then cut the last side of the second piece and the first three side of the third piece. I continue in this manner until I am left with a off cut that is between 40 - 138 mm long. This off cut will be used for the soldier.
2 "prototypes" further down the line the jig was ready:
The first test of this clamp revealed a few teething problems. At first I was not getting a consistent shape tile. I worked out the wood must be moving very slightly dew to the force of the router cutting bit. This problem was fixed by gluing sand paper to the bottom of the Jig plate and to the jaw of the clamps. The next problem was that the wood would not sit flat against the jig plate. I believed this was caused by the clamps not applying even pressure to the wood when locked. I managed to get a more even pressure by applying tension between the two clamps (see the red string put under tension by a piece of wood under the jig).
In the process of getting to this point I found that using the router created a lot of dust. So I looked at how I could attach a dust extractor to my router. I found the simplest way was to attach the hose of the dyson to the router. I was surprised at just how well this worked. You can see wood chips on the floor of each picture so it did not capture everything. But it did capture all the small Particles of dust. So when I finish the test the wood was on the floor and not in the air.
So with all of this done the jig is ready to go. It still has one problem but I think once I start using the real wood it will not be an issue."
He hoped. But as it turned out (being September 2010 by now) the router bearings he used were not up to the tasks at hand. After waiting for what seemed like ages new bearings finally arrived only for him to discover there was a problem with the collar - holding the bearings in place. Off to order a new collar only to find out that had a defect and off to order a replacement.
All the while the "factory" stood still.
"So the saga continues. I am takeing bets on whether I will get it done by Christmas...... I am not saying which year"
Subfloor down - November 2010
We always recommend to install an Industrial Grade Oak 7-finger mosaic subfloor first when you plan to install a Design Parquet floor onto a level and dry concrete floor. This subfloor will enhance the stability of the complete floor, plus provides the smoothest surface you can have - sanding the subfloor smooth will save you time and effort sanding the design parquet floor, illuminating many height differences before you start the installation of the (valuable) top floor.
And besides, it's a simple job:
I have laid the wooden sub floor and plan to sand it this weekend. Can you thank your suppliers for me? Having the fingers lay out on the string latis make laying it very simple.
Christmas came and went
The bearing kept breaking, no matter what. Even the manufacturer couldn't find faults and a second cutter - different brand this time - broke down too.
So once more I am looking at the jig to try to determine what is causing the router cutter bearing to break. I have an idea about what is causing the bearing to brake on the jig. I will be making some minor adjustments to the jig. When I get some new bearings I will give it another go.
I may take you up on your offer at a later date. But for now I will wait to see what effect the change to the jig makes.
By the way I was able to cut 32 pieces before the bearing broke. I have laid the pieces out and I am very pleased with what I see. The light is reflecting off the pieces just as I imagined. Using the jig gives a very accurate cut so the pieces fit together very well. Keeping them in place when gluing will be a problem but I will cross that bridge once I have cut out all the pieces.
So when I said "I just wanted to update you on my progress over the holiday" may be I should have said "I just wanted to update you on my lack of progress over the holiday".
Two weeks later: hooray!
Well..... The revision I had made to the jig seem to have worked. I used the jig for over 8 hours and the router cutter bearing did not brake. I used a jig saw to cut the piece to a rough size before using the router to cut the finished piece. I am glad to say that about 95% of the pieces are as I expect. Hopefully this will be good enough.
I was hoping to cut 60 pieces an hour but because I have to cut each piece to a rough size first, I am making 27 pieces and hour. Therefore, it will take me a while to cut the 2584 pieces I need. At least now I seem to have worked out all the problems and can start the production line.
It was a minor change to the jig that make all the difference. The outside corners of the jig were pointed. I notice sparks flying off the corner with a double circle, in particular. So I got these corner rounded off.
If I am not careful it will still spark. But it looks like this simple change has made a big difference. Here is a picture of the pieces I have cut so far.
Cutter not cut out for the works:
"So just when you think the end is in sight the plot takes an unexpected twist"
A hundred working hours later
Then, end of April:
Well I am pleased to say that I have cut all the pieces for the central section. It has taken over 100 hours of work ( my fingers are feeling a bit stiff from the work out of holding a router trigger for so long). In fact I have finished gluing them down too.
Hiring a decent sander to finish the works came with its own problems Cyril discovered:
When I spoke to you on Friday the floor had been laid and glued. All I had to do was sand and polish. With a extra long weekend it looked like I would get it finished. What can I say, nothing ever seems to go as you would expect with me.
On Saturday I went round a few hire shops to hire a re-finishing sander. Only one shop had one available so I took it with some sand paper. I could not start on Saturday because I had an appointment but I thought I would set up the sander and test it ready to start work Sunday morning. Just as well I did. Would you believe the sand paper would not stick to the bottom of the sander!? I contacted the hire shop. They had given me the sand paper from a new batch they had just opened. They checked the remaining sand paper in the batch and sure enough the batch was faulty. They found 3 sheets of 60 grit sand paper. But this was no way near enough paper for me. So the floor is still unfinished. I hope to finish it this weekend. In the mean time here are some pictures of the floor in its un-sanded state.
One year and three days after initial email:
It has been a long while coming but it is here. I have now laid to main section of the floor and here are the pictures.
I just have to do the transitions and I am finished.
A friend came round and had a look at it. He suggested I rebuild the house and the floor, as the floor was too good for the house!
Thanks for all your help and advice. Your book was very helpful and I would recommend it is anyone.
Cyril, it has been an absolute pleasure to help you any which way we could (unfortunately we're wood people, and know absolutely nothing about bearings, cutters and collars I'm afraid). Don't think many would have continued a huge project like this with all the problems thrown your way, but you kept on going regardless and you have every right to be absolutely proud of the results: it looks stunning!
You've earned your place in our "Hall of Fame" more than anyone else as far as we are concerned.