Shh! The Baby’s Sleeping!

My daughter is 1.5 years old and goes to bed around 7pm.

I get home from work around 5:30pm and after dinner, play time, and bath time, 7pm comes too quickly.

I’m not complaining that I don’t get enough time to play with my daughter, I’m complaining that I can’t make any loud noises after 7pm.

When she’s asleep, I can’t use the  table saw, circular saw, air compressor, bench grinder, angle grinder, sander, router, impact driver, or shop vac.

This cramps my style.

Want to break down plywood sheets? Can’t. Want to rip a 2×4? Can’t. Want to clean up the garage? No. But even if I wanted to, I can’t.

I do have some power tools that are a bit quieter that I can use after curfew.

My bandsaw, scroll saw, jig saw, milling machine, lathe, drill press, and hand tools are all available (although hammers are iffy).

Over the course of my pergola project, I have made extensive use of my hand saw and chisels to trim posts and timbers at night… in the cold… while it was raining… barefoot… uphill both ways.

Last night I needed to cut a 1.25-inch wide channel 1-inch deep along the length of a 4×4 post 40-inches long.

The tool for the job was the table saw with a dado stack, or the router table with a spiral cutter. But since it was 8pm, these were off limits. What was available was the milling machine.

I haven’t found a lot of examples of people using their milling machines for wood, but in my experience they work very well.

Since the post was 40-inches long and my milling machine only moves 18-inches on the y-axis I had to clamp the post in place in three different positions in the vise to cut the full length, but it worked really well.

The process was simple:

    1. move the table all the way to the left
    2. clamp the post in the vice with the start of the post near the cutter

Milling Wood

    1. turn on the milling machine and turn it up to full speed
    2. crank the handle to move the table to the right as fast as you can
    3. widen the cut with a second pass

Milling Wood

  1. repeat as necessary, shifting the post to the right in the vice each time until the full length is cut.

Milling Wood

The milling machine spins very slowly compared to a router (2000RPM versus 25000RPM, but the vise holds the piece securely and I was able to make full-depth cuts with a 3/4-inch end mill without bogging the milling machine down at all.
Milling Wood

Granted, I was cutting douglas fir, but my experiments with oak have not turned out any differently. Apparently wood just isn’t much of a challenge for metal-working machines.

And it only took about 8-minutes to cut the channel in the board, including setup time, which is really not bad.

Plus I didn’t wake the baby!

Full disclosure: when I turned off the milling machine my daughter was crying, but I’m reasonably certain that was coincidental.

Del Mar


Don’t Tell Me What I Can’t Do.

I am in the process of building a pergola off the back of my house (it might technically be an awning, but it will be open on top like a pergola so that’s what I call it).
backyard concept

It’s partially a design element in the backyard, but the real reason I’m building it is so that I can add a railing and a gate to allow my 1.5 year-old daughter to be outside without having direct access to the pool.

I got a quote for some 3/8″ tempered glass panes early in the design phase of the project and while it was less expensive than I expected it was still a significant cost (about $500).

We started looking at alternatives and eventually decided that we could use welded steel railing panels from Home Depot, cut to length and framed by wood, and it would still look nice but be significantly cheaper.

I was okay with this plan, but not thrilled. So the night before I was supposed to buy the railing from Home Depot, I looked on Craigslist for “Tempered Glass” and found a guy selling pieces that were close to the right size for $40 total.

I called him, he still had the glass, I picked it up the next morning.

The plan now was to build the railing and slide the glass into 3/8″ channels after cutting it to the correct length. Test fit below for proof of concept:

After building a section of railing to the point that it was ready to receive the glass, I went into the garage to cut the glass.

I have cut glass before, and it is a relatively simple process: score the glass with a glass cutter, then break the glass along the line. It’s worked pretty well for me in the past.

I set the glass on a flat surface, then used a straight-edge to guide the glass cutter and scored a line at 50″.



According to prior knowledge, I then placed a metal dowel directly under the scored line and pressed down firmly on one end of the glass.

The whole piece of glass pivoted on the dowel like a teeter-totter.

I placed a heavy log on the end of the glass to hold it down and tried again.

The whole piece of glass pivoted on the dowel like a teeter-totter with a log on it.

Time to consult the internet…

Me: “Google, how do you cut tempered glass?”

Google: “you don’t.”

Me: “Google, there must be a way to cut tempered glass.”

Google: “two options: A laser cutter, or heat the glass in your oven to 900-degrees Fahrenheit and let it slowly cool before scoring and breaking it.”

Me: “has anyone ever successfully cut their own tempered glass at home?”

Google: “No.”

Me: “Challenge accepted.”

I placed clamps on the end of the glass that was prone to levitation and then placed another pair of clamps on the other side of the dowel. I placed a third pair of clamps directly over the dowel. All three pairs of clamps had 2×4 lumber under them to spread the pressure across the glass more evenly.

My first move was to tighten down the clamps directly over the dowel. The thought here was that the 2×4 was soft enough to sort of wrap around the dowel, applying pressure to either side of it and eventually breaking the glass.

I maxed out the clamping pressure… nothing happened.

Next I methodically beat the 2×4 furiously with a dead-blow hammer… still nothing.

So I began to tighten down the clamps on either side of the dowel, gradually increasing the pressure on one side and then the next… nothing happened.

Tighter… nothing.

TIGHTER… nothing.

TIGHTER…  nothing.

<put on earmuffs in addition to safety glasses>

TIGHTER… nothing.

…maybe I should re-think my strategBANG!



Contrary to what the internet told me, I did get a fairly clean break at the score line.

I have since gotten a revised quote for tempered glass with the final dimensions and it will cost about $420. Not bad.

Precision Preschmision.

When it comes to manufacturing process, there is a direct relationship between precision and cost: the more precise the product, the more expensive it is.

The reason for this is simple: making precise products requires the use of even more precise tools, and it requires additional steps to measure and verify that the accuracy is within tolerance.

This applies to the home shop as well.

If you want to work precisely, and repeatably, you need to set up your tools perfectly, you need to measure and verify your results, and you need to plan your moves ahead of time, to make sure your plans for assembly and finishing allow you to maintain the accuracy of the components in the finished product.

All accuracy takes time, but there are two different classifications of precision in my mind, and one matters far more than the other.

  • Absolute Precision: making components exactly 3-inches long, or exactly 45-degrees
  • Relative Precision: making components that fit together exactly, with little concern for absolute dimensions

For example: I can make a table with a top that is exactly 6-feet long and 2.5-feet wide, made from 5 planks that are exactly 0.5-feet wide and 6-feet long: this table looks nice, is well made, and is absolutely precise.

OR: I can make a table that is about 6 feet long and about 2.5-feet wide, made with planks that are all the same length and width: The end result looks the same, and has the same quality, but all of my “measurements” are really just making sure the parts fit together. This is relative precision.

Sometimes absolute precision matters: In mass production you need to be able to provide predictable dimensions so that the packaging and usage can be consistent; replacement parts must meet exacting specifications.

But at home it’s often less important that the part be a certain number of inches, and more important that is fits, works, or looks good, whatever the length ends up being.

This is a liberating principle for me. I don’t need to spend hundreds of dollars on Starrett measuring tools to make six cuts of the same length; I can just make the first cut however I like and then keep the setup in place for the next five.

Now, this does require some forethought. If you are planning to repeat the same cut, you need a setup that will allow it. Your bandsaw and tablesaw should have a sturdy fence (but Biesemeyer and Kreg are probably overkill), and you should use stop-blocks and simple jigs where needed.

You may be thinking: “if you’re making something exactly 6-inches wide, isn’t that just as difficultas making something exactly as wide as something close to 6-inches wide?”, and the answer is: …sort of.

You see, I like to start building based on a concept in my brain, and progress through the project adaptively, with the results of each step determining the details of the next.

I don’t want to sit at my computer planning the minutia of the project in Excel when I could be spending valuable time at the bandsaw. And since nothing turns out perfect anyways, my measurements would invariably need adjustment as the project progressed, which then begs the question: Why measure in the first place?

As noted previously, the tape-measure has it’s place, and the blade of any saw should be square to its work surface, but there is nothing magical about hitting the inch within a fraction of a millimeter: These are arbitrary lengths that were made up in antiquity.

The real test of valuable precision is simple: Does it fit? Does it work? Does it look good? If so, it doesn’t matter what the ruler says.

Inca Bandsaw Fence – Part 3 – The Fence

And so we come to the conclusion of the Inca Bandsaw Fence series.

At this stage I had a rail for the fence to slide on, a carriage that follows and locks onto the rail, and a mounting bracket extending vertically from the carriage for the fence itself.

The best material I found to make a fence was extruded aluminum rectangular tubing (about 1.5 inches wide and 3 inches tall). Aluminum extrusions tend to be very straight and since my saw’s table is also aluminum, using a harder metal could possibly damage the table over time.

I picked up a 36-inch piece of the aforementioned tubing as a remnant from the local metal supply store ($2.70 per pound) and cut 16 inches off to make my fence.

The fence needed to be connected to the vertical bracket attached to the carriage. My first thought was to attach it with bolts on the backside of the fence, but then the width of the fence would be limiting the cutting capacity of the saw by a full 1.5-inches. So I decided to attach the bracket to the inside of the tubing.

In order to do this I cut a slot in the bottom of the fence, right along the front edge where I wanted to attach it to the carriage.

Then I drilled and countersank two mounting holes. I transferred the location of these holes to the bracket and then drilled holes in the bracket accordingly.
Bandsaw Fence - Mount
Bandsaw Fence - Mount
Note: If you are a particularly observant reader, you will notice that I showed a picture of the bracket in “part 2” of this series, but here I am claiming that they were drilled as a step in “part 3”.  Well done.

The arrangement of the holes is not merely aesthetic. If the holes were aligned with each other horizontally, they would provide only limited vertical support. Likewise vertical holes would provide limited horizontal support. My theory is that diagonally arranged holes will be the best of both worlds.

Bandsaw Fence - Mount

I inserted brass (because it’s pretty) machine screws through the holes and secured them with washers and locknuts (to prevent loosening during operation).
Bandsaw Fence
Bandsaw Fence

And that’s it.

IMAG1284Bandsaw Fence

There is a bit of flex on the far end of the fence if to apply a lot of lateral pressure, but since my workpieces will be small and kickback is not a concern on the bandsaw, I won’t be pressing hard against the fence, so I think it is rigid enough.

That said, I’ll probably tinker with it at some point to make the far end lock in place as well.

Inca Bandsaw Fence – Part 2 – The Carriage

For lack of specific knowledge and due to a general laziness when it comes to looking things up on Google, I’m calling this part of the bandsaw fence a “carriage”:

fence scribble

(actually, I’m not certain my affliction can be considered “laziness” since I just went through the trouble of drawing a picture instead of just looking up the correct name… maybe I just feel like it is a carriage, whether or or not that’s what it’s officially called).

In any case, after completing the rail, I turned my attention to the part connecting the fence to the rail.

Essentially this works like a C-clamp, with the rail being pinched between a screw and a pressure plate. The construction may vary, but the most important aspect is that the carriage must be designed so that it is rigidly square with the rail when it is tightened in place.

I used a small piece of angle iron to act as the pressure plate. I shortened one side of the “L” (the “pressure plate” side) so that it wouldn’t hit the bolts on the underside of the rail.

Then I attached a vertical piece of steel that I milled flat and square. I kept it parallel to the angle iron by clamping a piece of metal in between it and the short end of the angle while welding it.

I drilled and tapped 1/4-20 threads into the vertical piece of steel and ran a Rockler star knob through it.
IMAG1232Bandsaw Fence - Slide/Lock

Bandsaw Fence - Slide/Lock

This tested okay, but I didn’t want the screw to mess up the rail over time, so I added a thinner piece of steel that was flexible enough to allow the fence to move while the knob was loose and still bear down hard on the rail when the knob was tightened.

Bandsaw Fence - Slide/Lock

So now I had an assembly that could be locked into any location with a mounting surface that was consistently parallel to the rail. The next step was to make a bracket that the fence itself could attach to.

I made the bracket out of a piece of trailer hitch tube I had leftover from another project.

I started by milling two sides flat and square with each other.

Bandsaw Fence - Mount

After that I cut off the other two sides and shortened one side so that the other would stick up above the bandsaw table perpendicular to the shorter side which would be mounted to the rest of the carriage assembly.

Like so:

Bandsaw Fence - Mount

I mounted this with a single 5/16″ bolt through the carriage assembly, which allowed it to be pivoted as necessary to keep the upright portion vertical (parallel to the blade).

Bandsaw Fence - Mount
Bandsaw Fence - Mount

And now I could turn my attention to the final component: Part 3 – the Fence!

Coming soon.