Learning from mistakes

We all know the idiom about book-smarts versus street-smarts, and it holds just as true (if not more-so) for woodworking as any other pursuit.

You can read and read about a topic, watch all the videos, follow the forums and talk with experts, but I can still guarantee the first time you pick up a chisel and try to make a square object round on a lathe, you’ll discover in practice what a kickback (or chisel dig-in) is ALL about!

You can learn as much as you can from all other avenues (and that is a good thing), but the real learning curve comes from biting the bullet and trying it out for yourself.  However, jumping in the deep end without RTFM is fraught as well.  Tried that yesterday, and the result was, well, a learning curve.  No real harm – bit of time was wasted, and some scraps of timber, so that could be considered well worth the price.  As another saying goes (stolen from its association with fishing): “a bad day woodworking still beats a good day at work” (Of course you’d want to add a small suffix to that “so long as you finish with as many body parts as you started with!”)

Tried out the MLIS (multiple layer inlay stencils) from Tarter Woodworking, and while I didn’t finish with a result, the templates proved how well they would work once I refined my processes.  Trying to do it the first time and on camera just makes it that much more difficult!  I also started with a pattern that was perhaps a little more complicated than I should have, so the second round will be with a simpler form.

So the majority of the video footage is destined for the editing floor.  You can see a bit of timelapse footage that is left over.

It was a good test run of the multi-camera setup, and particularly the new audio recording arrangements.  Running a couple of high-end mics (NTG-3 and NT5 Rode mics) through a Beachtek DXA-HDV gave some great results.


Rode NTG-3

Rode NT5 Matched Pair

Rode NT5 Matched Pair

Beachtek DXA-HDV

Beachtek DXA-HDV


Blasterboyz Plaque

One thing I had yet to try on the CNC router, was other materials, and specifically aluminium.

As much as ‘they’ say that some Australian timbers are harder, and that I have both cut and routed aluminium manually, it was still with trepidation that I mounted a plate and engaged the CNC router.

There is more than hardness to the effects of machining different materials. Different materials form chips in different ways, different amounts of heat generation, and in the case of aluminium, a tendency for waste material to try to weld itself to the cutter if chips are not cleared adequately.

It is for this reason that aluminium router bits tend to be single fluted, allowing a much larger flute for more aggressive chip clearance.

Had a project come up that made it a great excuse to give it a try. The Blasterboyz are a group of JetSki riders, with a common tool of the trade- the Yamaha WaveBlaster. They often ride socially, right through to competitive rides.

They have asked if a plaque can be made, which will be used as a bit of a trophy. I won’t tell you what the trophy will actually be called – too politically incorrect!, but it translates as “go hard, or go home”

I set up to try a pretty standard V groove bit (before risking one of my soli carbide bits), and played around with feed and plunge rates to get one that cut sufficiently, without chatter or causing the CNC to move faster than the cutter could cope. The CNC Shark isn’t the most rigid, so when push comes to shove, there is some flex, which results in an imprecise job. The solution is to ensure the feed speed chosen suits the cutter and material.

So as a first attempt, this is the result. With some refinement, and a better, sharper, dedicated cutter this could be quite satisfactory. If the opportunity arose, it would also be very interesting to see a lasered result.


Routing Steel

Sound insane? It isn’t as uncommon as you’d think, but it does take a router bit that goes well above and beyond what is normally used in the woodworking arena. (Perhaps one suited to Australian hardwood!!)

You need a cutter that has a surface harder than the material you want to cut, and durable enough to survive the loading and impact. Start with a quality carbide, and not just carbide tips brazed to a steel core, but solid, micrograin carbide.


To dramatically increase wear resistance, apply an aluminium titanium nitride (AlTiN) coating, and with a few physical attributes (upcut, mill end, corner chamfer) you have a router bit that can rout and mill steel, stainless steel, aluminium, brass, copper, even titanium.

The router bits here are specifically designed for CNC machines, which is a good thing, as you can then accurately control feed rates to match the material being cut.

Lubricant/coolant is highly recommended, so things could get a bit messy.

Fascinating what materials engineering makes possible.



Seen while perusing the Toolstoday.com website. The concept caught my eye!

The Boris Goldberg Kerf Chair

This is pretty cool – a chair created on a CNC machine using kerfing as the technique to give it shape.  Opens one’s eyes to all sorts of possibilities.

SSYTC055 Routing the Japanese Dragon (CNC)

SSYTC055 Routing the Japanese Dragon (CNC)

Some Torque Functions

Getting ready for the jaunt across to the US in a week’s time, I put some time running through some of the basic Torque functions.

Firstly, I wanted to do some refinements to the overall setup – getting the centreline of the router as perpendicular to the workcentre top as possible.  I’ve touched on this recently, using a dial gauge mounted directly to the router collet.  That provides a significant level of accuracy, perhaps beyond that which is required.  I’ve also been using the Wixey Digital Angle Gauge directly onto a rod mounted in the collet, and that is a pretty simple system to use.

First step is getting the main arm parallel to the table.  Using the Wixey is the best way to achieve that – zero off the table (with the gauge parallel in alignment to the arm) then place it on the arm itself, and get it level.

Zeroing the Wixey

Next, repeat the process, this time zeroing the gauge then check the rod is at 90 degrees.  Repeat the process for the other axis of the router.

Aligning the Router

(You can see in this photo the router is out a significant 1.4 degrees around the Y axis).

I’ve also tried using the simplest technology – a square, and that is also quite successful.

Using a Square

The use of a straight rod in the router collet is so successful, I’d almost suggest it would be a useful inclusion in the Torque Workcentre package, and later it could become one of the set of copying attachments (although the copy attachments would need to be increased in size from 12mm to 12.4mm, but I’d see that as an advantage).

First job I did was finishing off the downsizing of the 4″ pre-separator bin.  It needed a new base, and although I’ve cut circles by other methods in the past, the TWC is by far the easiest method.  There are justifiable times that the bandsaw will still be better (thin kerf of the blade vs router bit width), but if that is not a consideration, then the TWC wins.

Circle Cutting

A circle is cut by placing the workpiece on the TWC pin (by drilling a pilot hole in the underside of the material), offsetting the router by the radius of the required circle and locking it’s X and Y axis.while rotating the workpiece, slowly plunge the router into the work, cut a circle, increase the depth, rotate again, rinse and repeat until the circle comes free.

New Base for Dust Bin (Pre-separator)

Next, I wanted to try some of the pin-routing technique.  This is a 2 stage process.  First you need to create the path that the pin is going to follow.  In this case, I have secured down the body of the racing kangaroo.  A couple of extra supports are added so there is no issue with the panel rocking on a too small-a pattern being duplicated.

Setting up the item to be duplicated

This is flipped over and run against the pin, with the router directly over the top of the pin, a channel is cut.

The path created for the pin

This is the path that the pin will fit into for the actual duplications.

All the parts for the racing kangaroo

Flip the pattern over, mounting it on the pin, and attach the blank board on top.  With the pattern captive on the pin, a perfect copy is made, and this can be repeated as many times as is desired.  The only thing I need to do is get a better match in size between the pin diameter and router bit diameter – they need to be a matched pair.

Pattern Copying

The other method is using the copying attachment.  This has a pin that follows the original path, and moves the router over the workpiece, rather than moving the workpiece relative to the router as happens with pin copying.

Resulting sign

In this case I used a router bit larger than the pin, so the thickness of the letters were increased.  Signwriting is only one use of this technique.  The ability to easily and accurately duplicate an object or pattern is a very powerful tool.  (As you can see, I was using the same workpiece to test this technique as I did for the pin routing)

Producing raised (bas-relief) lettering is as easy (if not easier!)

Movin’ In

It was definitely a weekend revolving around the new tool (understandably), from the pickup/delivery organising on Friday, the assembly (and getting it into the shed) Saturday, and commissioning it with some real tasks on Sunday.

If you were wondering how I was going to fit a 2.5m x 1m tool in my workshop, well, you were not the only one!  With a large pry bar!  Archimedes once remarked (paraphrasing) that with a firm place to stand he could move the Earth (with a long-enough lever that is).

Makin' a Hole

Clearing space was a combination of moving the jointer to where the router table was, removing the corner bench and in future the sanders will (potentially) be on roll-outs from under the Torque, and the router table met an untimely end, with the top being amalgamated with the TWC.  A hole was born.

The Torque arrives in its new home

I did try the TWC along the back wall, but lost too much access around the unit.  The MDF top has yet to be added – will do so some time this week.

That 1300mm arm is huge – I’m going to fit the 600mm (or 900mm) arms for the majority of the jobs, and the 1300 will get used only when I need that much extension.  For some workshops it probably would remain, but that is the one compromise on space I will have to make.  Sure looks nice here (by the end of the weekend it sure ended up rather dusty!)

Adding the Router Table

Here I have added the router table (with Incra Positioner/Fence) to the end where there is some dead space (because of the way the tools are carried on the left side of the arm), so this really utilises that area. I’ve moved the bandsaw further back so I have adequate access to the router table, and it isn’t hard to swing the entire unit out from the wall if I need more in-feed area.  Since this photo, I have also replaced the first module of the top with a plain cast-iron wing – it won’t be a two-router table now (well it is, but one is overhead), and I’ve shortened the cast iron top to 3 wings from the original 5.  The MDF of the rest of the bed will be at the same height, so there is no shortage of area for the positioner now!  At 2.5m long, this may be the largest router table in the world!

The clamp roof

The clamp wall was a problem, both getting access to the clamps, and also having them interfere with the X axis of the TWC, so they have been moved to the roof.  If the clamps are any good, then it won’t be a problem where they are!  Added benefit – partway through a clamp-up, if I need another I just reach up!

Temporary Triton Mount

One minor glitch I had (and it will be rectified by the end of the week), is the router mount had support posts for a Makita (I think), and not the Triton – the Triton needs both a larger diameter and a longer post.  Converting the Triton to fit the table took a few seconds, as did reverting it back to standard operation.  I was surprised just how easy it was – a single circlip.  So that I could make use of the router before the parts arrive, I added the old Triton quick-release plate from the original Triton router table.  The setup is rather flimsy like this, but it got the job done.

Oven Doors

First job was the oven doors for the kid’s kitchen – setting the stops for the size of the opening, then routing it out.  I also used the same method for cutting hinge mortises.

Oven Shelf

The second job was creating the shelf in the oven.  Rather than just a plain shelf, I didn’t want the inside of the oven to look like a cupboard, so I cut parallel slots through using a 1/2″ straight router bit, and moved the whole setup 30mm for each pass.  Again, something that could have been done other ways, but this was a very simple (and accurate) method.  It’s like using a machine with a built-in, adjustable jig.

So that was all I had time for in the end, but already it has been demonstrating for me just how functional the concept is.

Rockler Bench Cookies

Out of the mists at the recent AFWS show in Las Vegas they appeared – futuristic looking like they had been sent back in time to the woodworkers of today as a gift from the future.

Looking not unlike blue hockey pucks, the word was out – the Rockler Bench Cookies had arrived.  No more would woodworkers have to use antislip mats under their work when freehand routing, or sanding operations (and then having to try to get that little polka-dot pattern left by oil residues in the mat off their work).  The Bench Cookies are designed to lift, grip and protect both your workpiece, as well as the bench underneath.

With a surface on their top and bottom with a little give, and a lot of grip, one placed in each corner (of a smaller piece, or a couple more in the middle of larger ones), they allow the full edge of the workpiece to be exposed, so when running your router around the edge with a fence, or a bearing bit, it can protrude below the base, and still there is plenty of grip of the workpiece so you can safely route without needing to clamp the piece down.

When sanding, the piece is held firmly as you’d expect, and then finishing – the edge is fully exposed so you can get your finish right to the bottom of the edge, and not find a bead of it pooling in the lower corner. They work whether clean or dusty, as the YouTube video below will demonstrate.

Rockler Bench Cookie

Rockler Bench Cookie

What’s more, they are cheap – a set of 4 costing under $US12 (get the actual pricing from the Rockler Store) Check with them what shipping costs though if you are purchasing outside the States- I haven’t found that out as yet.

This YouTube video was produced by http://www.consultingwoodworker.com.  A quick tap and sweep with your fingers is what is recommended to clean it, but that is not even done here. I’ll make my own version of a video when I have a chance.

A Burl Clock for the Shed

To start the process, I’ve been preparing the burl slab itself, and the first part of that was the recent YouTube Chronicles video, running the burl through the drum sander.

Next, I took the random orbital sander to the surface, starting with the unusually coarse (for me) 80 grit paper (the burl is very hard), and continued through the grits to 400. For previous clocks I would normally oil the surface (with a burnishing oil), but in this case I didn’t think it would be needed to get the grain to show up, and I didn’t know how the Liquid Glass would respond to it.

I’ve then flipped the board over to mill out a cavity for the clock mechanism.

Creating the Template

Creating the Template

I needed a template to route out the opening, so started down the tradition path – marking out the opening, drilling holes, cutting with a jigsaw, filing off the jigsaw marks, and all the while I was thinking to myself – there has to be a better way. Then I remembered the Sonicrafter that I previewed for the manufacturers – one of the high vibrating speed cutting tools (takes different blades etc, the well known version is the Fein). This one is Worx brand (the bigger brother of the Rockwell that has recently hit the Aussie market) It will be in the marketplace soon fwiw. I gave it a try, and it worked like a dream – the perfect tool for the job. In future it will be the first tool I turn to for jig creation! I made the template out of MDF, and before you ask why I didn’t just cut the actual opening this way: burl is really hard, and I think any of these cutters would probably struggle, and secondly, and more importantly, I needed the opening in the burl to be a partial depth only.

A big reason for me using this tool, is I can cut a square opening, with straight sides a lot easier than my older methods!

The opening in the template is larger than the actual clock mechanism, as it needs to take into account the distance between the outside of the template and the router bit. I set the router bit depth, taking into account the thickness of the burl, the length of shaft of the clock, and the various components that are attached.

Router Bit Depth Set

Router Bit Depth Set

I used the Wixey Digital Height Gauge to set the height accurately. So once I had the template, this was clamped to the burl, and the opening created with the router.

Mechanism Opening

Mechanism Cavity

A perfect opening

A perfect opening (centre still to be removed)

The above-image has the outside routed to full depth, but as you can see the middle area needs another pass.

Back of Burl Clock

Back of Burl Clock

So this is the back complete.  I tend to leave it raw so I can see the difference in the finished front and the raw back when I want to.  I know this is not best practice, if for no other reason than it can encourage warping when the stock is thin.  Still, it’s a choice I make (in some circumstances).

Oh, and for the doubters, yes I do use my JawHorse, all the time, and for almost every project!

Next post will be about finishing the front.

%d bloggers like this: