From the Drawing Board
Occasional ramblings from
a Small Craft Designer

by John Welsford

Odd Jobs

I find myself, driven by a need to keep the wolf from the door, doing some odd things from time to time. I describe myself as an “Industrial consultant” which helps put the chargeout rate up a bit but I tell my friends that I am an “Odd job man”.

One of the more interesting “odd jobs that I have had of late is helping a friend rewrite some boating industry training guides. Guides that needed to catch up with the technical and social changes affecting the training of boatbuilders. An interesting job and one in which there is a lot of crossover between my life as a designer, and my Industrial Consultancy work, which is mostly in the field of wood remanufacturing.

One of those guides covers the making of laminated components from wood. Simple? Slice up some bits of wood thin enough to go around the curve, make a jig of some sort, smear on the glue and clamp it all into place until its not sticky any more.

Yeah! Simple! Fight with a dozen wobbly thin sticks all slippery with epoxy, all trying to escape , trying to clip you behind the ear as they spring loose and covering your best Levis ( if I get this done now before we go out Dear, it will be ready for me to fit in the morning) with permanent stripes.

Actually it's worse than that. I, as mentioned, do a lot of work in the industrial end of the wood business, and commercial lamination is a large part of that work..

When building a boat one hopes that the glue holds, as the result of failure may lead to wet feet, wet right up to your topknot! But a failure in a structural beam in a critical part of a several story high structure can endanger large numbers of people, and failure cannot be countenanced so there is a quite rigid set of testing requirements that go with the certification that a manufacturer must have.

A part of the process that few give consideration to, even among the big manufacturers, is the actual method by which glue bonds wood and the effect upon that bond of the surface preparation of the glueline.

I have on several occasions been engaged to trouble shoot situations where the glue tests within specification, where the process (such as mix ratios, clamping pressures, curing times and temperatures) is fine but there are occasional test failures.

Glue manufacturers are often blamed, and must go to every effort to try and find the source of the problem to avoid liability. That’s where I get involved.

Over the years I have learned a lot about this subject, and when I came across this very issue in the training manual that I am currently working on (or should be, it’s a couple of days past the deadline but I know that the guy dealing with it hasn’t finished with the previous one I sent him so I’m ok) I though that I would put it into column form for Duckworks readers to mull over.

It is in “training manual “ language so is a bit pedantic, but otherwise, I hope it is helpful.

Preparation of surfaces for glueing

* Note, a “Lamell” is a strip of wood to be included in a laminated component.

Surface preparation of the glued faces is critical to the success of laminating. Practically all of the structural glues used are mechanical bonding agents which means that the resins and their reinforcing materials penetrate the open cells of the wood to be glued, then set hard so keying the resin to the wood on each side of the joint. Surface preparation should be of a nature that preserves the openings into those cells cut during the preparation of the lamell.

Scientific tests have shown that the best surface for glueing is a freshly hand planed surface cut with a very sharp edge, however sawn surfaces cut with a very sharp circular saw blade, planed with a very sharp power planer ( buzzer, jointer or thickness planer) are almost as good.

Hand sanding or sanding with hand held power sanders is a less reliable but acceptable method of preparation but there are some common methods outlined below which are not satifactory and which may lead to failure of the glueline.

Wood purchased dressed or machine planed four sides has normally been dressed in a “four sider” or Planer Moulder. If run past its optimum knife life this type of machine, as with a “buzzer” ( jointer) or “thicknesser”( planer) heats and burnishes the surface of the wood leaving a very smooth glossy surface with a very good appearance. This appearance can deceive the user as it may be that the burnishing and heating has caused the lignin that binds the wood cells together to flow closing or filling the cut open cells preventing the glue from penetrating and making an effective key. Note that this is a known and serious problem in the production of industrial laminates such as structural beams.

Note also that in wood with a high resin content, (many common softwoods are in this category) resin may bleed into those open cells causing a reduction in the strength of the bond. It is good practice to prepare laminating feedstock, in fact to prepare any wood surface to be glued as soon prior to glueing as is practical. Certainly no longer than 24 hours prior to the application of the glue.

This means that it is good practice to resurface the outside edges of machine finished stock before inclusion in a laminate.

Industrial weight power sanding machines of the type used for finishing panels may also produce a surface not ideal for the penetration of glues. Much the same issue arises, that of blunt cutting edges which by virtue of massive power and pressure are able to produce a surface which looks acceptable but which may not glue reliably. This is particularly common with plywood and any such surfaces should be carefully checked and possibly prepared with cabinet scraper, random orbital sander or some other suitable means before inclusion in a glued structure.

John Welsford
Designer, and sometimes an odd job man.