WoodCentral Forums

Est. 1998 — 27 years of woodworking knowledge

Prototyping Rockers for a Windsor Chair

Posts

Prototyping Rockers for a Windsor Chair

#1

Prototyping Rockers for a Windsor Chair

Paul Leuba

Over the last few years I've built about a dozen Windsor Chairs, mostly for family and as gifts to grandchildren graduating from college. I've become interested in adapting the sack back Windsor format I've been building into a rocking chair.

The geometry of the undercarriage of a Windsor introduces some challenges into the rocker assembly because of the rake and splay angles of the legs. In a traditional rocker there is usually no splay. The legs do not angle out to the sides, they usually form a right angle to the seat. There is still usually a rake angle with the front and rear legs angling out frontwards and backwards respectively from the vertical. The joinery for this multiple angle is an issue to be dealt with at assembly time and not the subject of this current prototyping task.

This task is focused on determining the most comfortable rocker radius, and determining the relative lengths of the front and rear legs which is the primary factor in placing the inflection point of the rocker directly under the center of mass of the rocker and it's occupant. For this effort, the inflection point is the point on the rocker that is in contact with the floor when the rocker and occupant are at rest.

A rocker is an arc of a circle. Research of rocker design has revealed that the recommended rocker radius is about 3.1 times the distance from the seating surface to the floor. Using an 18" seat height as the guideline yields a radius of 56" for the rocker, rounding up to whole numbers. Being a doubter, I prepared multiple prototype rockers to test the recommendation at: 46", 48", 50", 56" and 70" for an extreme example. The prototype rockers are pictured below:


After assembling the prototype pictured below, sure enough, the 56" radius rocker proved to be the most comfortable rocker, at least to my primary tester, my wife.


The other important test objective was to determine the relative lengths of the front and rear legs in order to place the natural inflection point of the rocker at rest directly under the anticipated center of mass of the rocker and occupant. Using a center of mass height of 18" from the floor to the seat surface, after multiple adjustments it was determined that with the 56" radius rocker, the distance from the seating surface to the top of a 1" thick rocker is 17" for front legs, and for rear legs it is 15 1/2". This yields a front to rear ratio of 1.1. This ratio can be used to determine the leg lengths if it is desired to vary the height of the rocker from the 18" test height. The ratio would vary for chairs with rake and splay angles that differ from the test sample.

Now, the next steps are to prepare steam bending forms for the 56" radius rockers, bend and dry the rockers, and turn legs to the above specifications.

More to follow.

Paul in Hunt Valley, MD

Re: Prototyping Rockers for a Windsor Chair

#2

Re: Prototyping Rockers for a Windsor Chair

Barry Irby

Nice work.

Have you looked at Hal Taylor's site and materials? I am considering building a Maloof style rocker and Taylor has done a lot of work on them and offers plans. My understanding is he has even made them flexible so you can scale the chairs to fit the person.

Re: Prototyping Rockers for a Windsor Chair

#3

Re: Prototyping Rockers for a Windsor Chair

Paul Leuba

Yes, I have reviewed Hal Taylor’s web site, and I purchased his book which contains very clear and specific plans for building a rocker. It was a source of much helpful information, including the formula for computing rocker radius. The book has plans for building laminated rockers attached to the legs with 3” long wood screws.

I would prefer to steam bend the rockers and to not use screws to attach the rockers to the legs. All Hal’s rockers have the back posts as an extension of the rear legs, and all the legs are attached to the periphery of the seat. In a Windsor, the legs are mortised into the seat about 3 1/2” from the edges of the seat, and they are angled out to the front, back and each side making the undercarriage geometry more challenging when it comes to a fixing the rockers.

I do highly recommend purchasing Hal’s book if you plan on making a Maloof style rocker, or just to gain an appreciation of what’s involved in rocker construction. It’s a .pdf download.

Paul in Hunt Valley, MD

Re: Prototyping Rockers for a Windsor Chair

#4

Re: Prototyping Rockers for a Windsor Chair

Larry Barrett

Your experiment is very impressive. I have steam bent rockers for a few chairs. I started with a 2"x2"x4' riven oak billet, steamed it for 2 hours, then bent it on a form. Took 2 men and a boy to bend - the boy hammered in wedges as the bend progressed. When dry, I bandsawed the bent oak in half lengthwise, then planed each rocker to 1/2" or so thickness.

One unexpected but nice feature of bending the billet this way was that the ends of the billet that extended past the form formed a slightly reversed curve which I kept in the completed rocker - gave it the appearance of not rocking too far, even though no one ever rocks more than a few inches past the stable point.

Re: Prototyping Rockers for a Windsor Chair

#5

Re: Prototyping Rockers for a Windsor Chair

Paul Leuba

That is an impressive feat to steam bend a 2”x2” billet without a power press. The plans for this rocker are to steam bend a 1” x 1 1/4” x 36” piece of oak. That will be tough enough for me.

👍 This page answered my questions

Your vote helps other woodworkers quickly find the answers and techniques that actually work in the shop.