(Twice) Repaired Milking Stool

In my workshop for years there was a novel three legged stool that I’d picked up at the dump. I call it novel because the leg braces were quite clever. Where they adhered to the legs they had a vertical orientation and where they joined to each other they had a horizontal orientation. I’d tried to fashion similar braces as some sort of homage, with the intent of making a copy, but that was a little above my skill level back then, maybe still is.

I used it one day to stand on while trying to get a rope saw thrown over some high branch. This was beyond the service capacity. Result below.

The pieces were saved, probably fallow for four years.

Repair me

While normally I never use metal in woodwork, here I sank screws through the legs into the braces.

The dowels that had connected legs to seat were half inch. I replaced with 5/8 oak, hoping to gain a little strength.

All in all a most satisfactory exercise, a clearly useful outcome. So useful in fact, that I must add this little post-script.

We know that it is stated that virtue is its own reward. As a reward for this virtuous repair I got to use the stool. One thing I used it for was to hold up a 20 foot oak 6×6 that I was cutting at the twelve foot mark. When the eight foot piece came off the weight on the remaining 12 shifted in such a manner that the little milking stool was again sorely taxed. The leg assembly testified beautifully, as did its adherence to seat. It was the seat itself that broke in half.

I quickly harvested this opportunity to repair the seat. All is well in this corner of Mudsville,

Test Stool Two

The thread of building stools went dormant in May as the second test stool, having almost exactly the characteristics of the target stool, proved to be something of a failure.  The seat was a little bit like a surfboard, one could lean it a few degrees in any direction, and while that was cool it was not really a characteristic one wants in a stool.  It was asking to fail more profoundly and to this one must say ‘No’.

Initial version

S2 A S2 B

The cause of this wobbliness was that the mortises dug to accept the tops of the legs were too wide.  This, abetted by the fact that the four 3/8 inch dowels pinning the legs  through the seat really had no mechanical advantage, allowed the wobble.

Too wide mortises shown:

S2 C S2 D


The solution?  Epoxy-resin, perhaps overkill at 2500 psi.  Pouring that into the space where there was but an air joint really changed the performance of the joint.  It’s rock solid now, no wobble, no fail.  This is proof of concept that I’ll be able to go forward with the original idea – even though the original idea contains the same flaw, the over-wide mortises, because this remediation covers it.

S2 E

Stool 1 – Test Stool

This did not come about until mid-May, by which time I had almost ten boxes made and a feeling of some competence.  I’d started chiseling out the mortises for stool 0 at full width (as big as the leg diameter, albeit square) and stopped just because I did not want to get past the point of no return on stool 0 if failure was a possibility.

I had an old piece of wide fir lying around, and some scraps from the dump that I’d also used as Octobox pillars (I love getting the re-use) that I could easily rip into legs and cross braces.  I started chiseling out again full width mortises but halfway through that said heck, this is a test, gimme that router.  Then I niced up the edges and poof, I had four rectangles 1 inch+ deep and the shape of the legs.

IMAG0164 IMAG0165


I cut dados in the legs, pinned them through with 3/8 inch dowels, did the same between the braces with another cross brace, and came in from the corner with dowels through the side of the seat to keep the legs affixed.

It came out really strong, I was surprised.  I had avoided most of the real technical challenges.  A very short stool at 20 inches, legs perpendicular so no splaying, a triumph of simplicity and a confidence builder but not really proving much, methinks, so Test Stool 2 enters the picture – see subsequent post.

Stool 0 – an idea

Stool Zero is an idea based on a nice piece of wood.



It’s a piece of redwood burl 19 x 9 x 2 3/4.  I think it would polish up lovely so mentally I designed a stool for which this could be a seat.



I even designed the joining methods and built test joints to evaluate them.

IMAG0036 IMAG0040


I did not have a very clear vision though of the best way to attach the top of the legs into the seat or if the cross braces were sufficient or if I had the skill to execute the project – this was early, February of 2013.

Not wanting to screw up with such a nice piece I decided that some tests were in order