Making A Homemade Table Saw Homemade Machines & Jigs
For many woodworkers, the table saw is the one tool that gets the most use – I know that is true for me. Having a good table saw makes all of the cutting tasks a little easier, especially if it has advanced features. Unfortunately, good ones with advanced features usually have a high price tag attached. Being frugal (cheap) and always up for a challenge, I went to work to build my own homemade table saw.
Here’s how it turned out:
Some key features are a homemade Biesemeyer style fence that locks to the front rail, a max ripping capacity of 34″, a sliding table with max crosscut capacity of 24″, 3″ max depth of cut (with a 10″ blade) and an integrated outfeed table with dust collection.
The base is a simple plywood box with 4 sides. Made from 3/4″ maple plywood, it has 2″ x 2″ solid wood cleats to join the corners with screws. On top of this box is a plywood and oak frame that supports the motor assembly, sliding table and the top.
Shown here is the frame and the motor assembly being mounted:
The motor is from a portable Makita table saw that I was using before building this. It was a reasonably good saw, with the best part being the motor, but it had a few glaring shortcomings. I’ve never regretted my decision to scrap it for this build.
I did have to heavily modify the motor assembly to better work in this saw though. Shown here is how I extended the blade elevator mechanism to increase the arc of blade tilt control:
A problem with the Makita, when the blade was tipped to 45 degrees, the elevator crank was nearly touching the underside of the top, making it extremely inconvenient to raise or lower the blade. By extending the blade elevator crank axle down an extra 6″, I solved this problem.
Much steel was added to the original motor assembly to accommodate this longer arc. Holding the assembly rigidly during cuts is critical and I took no chances on the modifications being too weak.
Mounting the motor was a bit tricky. I had to fabricate a new trunnion part for the rear mount from solid oak:
I left the front mount adjustable from side to side, for alignment.
For the top of the saw, my original plan was to use two layers of 1/2″ particleboard glued together with a thin sheet of stainless steel on the surface, as seen here:
Only problem was that as soon as I took it out of the clamps, it started to bow upwards slightly. To avoid this, I should have used an equal thickness piece of sheetmetal on the bottom. Of course, by the time I realized this, it was too late.
I considered other options: solid wood (butcher block) and rejected that – I like the look in a kitchen, not on my table saw. I thought about a solid surface product like Corian but rejected that as well. Keeping costs low was one of the design criteria here, so the material would have to be economical. Having seen saws with granite tops, I thought that I could do something very similar: concrete. So, I made up a mold to cast the top with the surface down. With the top down, it would be as smooth as the casting surface of the mold – in this case, very smooth, very flat. The mold was melamine that was perfectly flat.
I mixed up a 30kg (66lb) bag of 3000psi concrete, added some black pigment and put it in the mold. It took the full bag with none to spare. I then smoothed it off, covered it with plastic and let it cure for nearly a week.
Here’s how it turned out:
Deadly flat and only a few small holes to fill around the edge. In the picture I have just hosed it down – concrete will cure harder if it is kept wet during the first week or two.
I mixed some more pigment with straight portland cement and floated that on to fill the tiny holes. After it had cured for nearly three weeks (on the saw by that time), I sprayed on two coats of water based polyurethane to seal the surface:
The ability to quickly and accurately crosscut larger panels has always been the task that a large radial arm saw would excel at, but who has one of those anymore? I thought that an accurate sliding table would be a worthy addition to my table saw project.
Shown here is the linear support system for the sliding table:
These are 28″ full extension drawer slides. A better way to orient these would be vertically, rather than horizontally, but as can be seen in the picture, there is limited space between the motor and the sliding table. Orienting the slides this way also makes the whole assembly easy to disassemble for cleaning and lubrication.
My main objective for the sliding table was to minimize lateral play to give an accurate cut, and the four drawer slides do this well. With the slides flat and with the other support mechanisms in place (aluminum stiffening rails on the bottom of the sliding table between the drawer slides) , there is nearly zero vertical play at the blade, where the maximum support is.
A view of the underside of the sliding table. The aluminum rails between the drawer slides stiffen the table considerably:
The fence for the sliding table pivots on a point and clamps to the edge, to set it for any angle. It is quick and easy to remove when not needed:
The table is marked for precise 22.5, 45 and 90 degrees. Maximum width of material for cutting at 45 degree is 11″. Very accurate cuts are easily made. Here’s an example of how it cuts 22.5 degree angles to form this octagon:
These pieces are 4″ wide and the joints line up perfectly. This is a pretty good test for the accuracy of the setup. Any significant error would accumulate and show as a gap when all of the pieces are put together.
I made a video of the sliding table in action:
My biggest complaint with the Makita saw was that the fence was nearly useless. Actually, the shortcomings of the fence was my prime motivation for this project.
At first, I considered just making a new fence for that saw but I immediately realized that it would probably be a wasted effort – I really wanted more.
I looked at various fence systems and settled on the Biesemeyer design as the best all round. Its simplicity belies its performance and I knew it would be the right choice for this project.
Much of the material I used to build it cost me nothing – recycled from various sources. The main front rail is a 2″ x 3″ steel tube. Another piece of the same material made up the fence itself:
There is a piece of angle that is welded to the fence and wraps over the front of the rail as the main ‘T’ (keeps the fence square to the front rail). The other parts hold the cam lock, which clamps the fence to the front rail. Once clamped in place, it takes a great deal of force on the fence to see any end play.
The fence rides slightly above the top and the only contact is at the end, where a sealed bearing acts as a roller.
Operation is very smooth and efficient:
Outfeed and Dust Collection
Two problems with smaller saws are lack of outfeed support and effective dust collection. I made a point of addressing both of these for this project.
The outfeed table is attached to the rear of the saw, in line with ‘main street’ on the saw (the area where the majority of material that is cut will be). It is adjustable up and down via two locking screws. It is attached to what is in essence a funnel that channels the dust that is expelled into a standard size recycle bin.
This, on its own, catches about 98% of the dust from the saw, with a very small amount spilling out the front into that bin (see earlier pics).
There is a hole for connection to my dust collector through a 2.5″ hose, but I only use this when cutting stuff that produces really fine dust (like MDF):
Here’s a look at the rear of the saw cabinet before the outfeed/dust collector was installed. The semi circular slot allows the dust port from the motor assembly to protrude. Nearly all of the dust produced in a cut is ejected from that port:
A few more details
There are many details that I’m leaving out: a lengthy text would be the result of a complete documentation on a complex project such as this. Indeed, there were several hours spent on what most would consider a minor detail – minor details that are comprised of many details themselves. The pointer for the degree scale below for example took no less than 4 hours to design, construct and configure. Seems out of proportion, doesn’t it?
The degree scale was laid out using scraps cut to each angle on my mitre saw. Given the larger radius of the scale, it was much easier to mark accurately than a smaller one would be.
The handwheel above controls the blade elevation and was made from plywood. The hand screw on the right locks the blade tilt.
Finding level ground can be a challenge and to deal with that reality, I made levelling legs for the saw cabinet:
These are on the four corners of the main cabinet, plus the two rear legs of the outfeed table. Large screws lock the legs in place after the saw is levelled.
The end of the saw cabinet is were I keep the sliding table fence when not in use. Next to it is a taper jig I made a few years ago. Handy to have both of these close.
When it’s all done, it’s good to see that it all lines up. The notch cut in the front fence rail is there to clear the sliding table. It is an excellent place to store the blade change wrenches.
February, 2012 Update
The saw has been a steady, reliable tool in my shop and has been maintenance free since I finished it. There was one thing that I regretted not doing though: installing a miter slot. I assumed the sliding table would take the place of a slot but that turns out to be incorrect. Since so many handy jigs and fixtures use a miter slot, I thought I would add one.
The saw, as it is:
Showing signs of wear and use, but still in great shape. I decided the best place for a miter slot is between the oak throat plate and the concrete top.
To get started, I removed the top. This was held in place with dabs of silicone under each corner, so it wasn’t very difficult to get off.
I cut a piece of hard maple that the slot will be cut into:
And carefully planed it down flush with the finished surface of the saw’s top.
Using the router table and a 3/4″ bit, I cut the slot:
A nice tight fit – the UHMW plastic strip is exactly 3/4″.
I used the fence to line it up and screwed it down on the front and back:
I added a spacer and another screw to help support the saw motor sub-assembly (red arrow). This was a problem area, due to the relatively long span of the maple brace. This extra support should help cut down on vibration:
To set the top again, I’m using more silicone:
One problem is that at the time I did this, it was -12 degrees Celsius in my shop. A 500 watt work light is used for a temporary source of heat, until the silicone cures.
With the slot in place, I marked where I would have to cut out the out-feed table:
This is to allow the bar that rides in the slot to extend beyond the table and is a bit wider than the miter slot.
I removed the out-feed table and cut the majority of the slot with the table saw.
I then used the router free-hand and finished the cut:
To replace the support that was taken away by cutting the slot, I glued on a piece of plywood:
Yet another piece from my cabinet door project (waste not). I cut it short of the end so that dust would not accumulate in the end of the slot.
Finished, I have coated the slot with petroleum jelly and let this sink in. I prefer this to wax as a lubricant for situations like this. A polyurethane finish would work also, but would eventually wear off and cause problems:
March 10, 2012 Update
Building the fence for this saw would qualify as a project onto itself – it took some time to make and near the end, I did cut some corners to finish it. My plan was to revisit the fence shortly after I finished the saw and redo some of the parts I wasn’t happy with. Same old story: days turn into weeks, weeks turn into years…Better late than never, right?
The first change was on the wear surface on the fence. I had been using a piece of Russian birch plywood (a lower quality variety of multiply plywood, similar to Baltic birch) that was less than ideal. To replace it, I used a piece of 3/4″ thick UHMW plastic:
This is 3″ high and was cut to length so that it lines up with back end of the blade. This is an anti-kickback measure. Having the fence extend beyond the end of the blade can cause the wood being cut to bind between the blade and the fence, and this stops that from happening.
Another place where the fence was lacking was the pointer. I made this one quickly, just to get the saw ready to use, thinking that I would improve upon it when I had the time. The main issue with this pointer is how high it is above the scale, giving a different reading depending on my viewpoint:
The new pointer bends down to nearly touch the tape. I notched the fence T to accommodate it. Getting the pointer close to the tape eliminates the parallax that was a problem with the other one.
A close up:
With this new pointer, I can reliably set the fence to within 10 thousandths of an inch, which is certainly accurate enough for me.
I have had some email enquiries about the bearing in the end of the fence and how it is attached. It’s bolted to an angle which is screwed to the inside of the fence rail:
With the addition of the miter slot, I was finding that the bearing would dip into the slot, and to remedy this, I replaced it with a hunk of UHMW plastic (you guessed it, I love this stuff!)
This slides just as easily across the top as the bearing, and bridges the miter slot without a problem.
The underside of the fence T has strips of UHMW plastic tape that need to be replaced occasionally:
You can barely see it, but it makes a world of difference in how easily the fence moves.
Someone asked me about the locking cam. I made the locking cam from a piece of 1″ round bar and drilled the pivot hole off centre by about 1/16″:
There is a flat spot ground on the cam for the fully locked position. It took a bit of trial and error to get the cam working optimally.
To properly locate the UHMW plastic on the fence rail, I shimmed it with paper that was folded twice and clamped it on:
I then drilled countersinks and pilot holes through the plastic and into the steel. These were then tapped with a #10-24 tap to receive a machine screws and washers.
The fence has a more precision feel to it now. The UHMW plastic wear surface is slick and feels more substantial than the thinner plywood one.