We have produced and sold many gallery wrap canvases, both large and small. Small (e.g., 8x10) canvases seem to retain their tension but larger canvases are prone to buckling, either seasonally or with long-term changes in ambient temperature and humidity. While canvases can be tightened or re-stretched, that is a practical solution only for large collections of canvases, such as exhibitions, where several can be adjusted in a single trip. It is not a solution for individual canvases hung on customers' walls or shipped to distant locations. Even if it was possible, the need invariably to touch up spots where ink cracked or popped off during adjustment further complicates the matter. And this is all made worse by the fact that we are addressing photography, not high-value oil paintings for which the cost of a "touch-up" visit might be a minimal portion of the original sales price or value of the canvas.
All of our stretchers use mitered slip joints at the corners. For larger canvases we also install cross braces as well as either wooden wedges or turnbuckles to facilitate future adjustment. In addition, we varnish each wooden stretcher before assembling frames and stretching canvas, thus assuring that archival standards are maintained. But for individual, moderate size canvases, this is not much help -- the only solution seems to be replacement, and even that is no guarantee, depending on the ambient temperature and humidity.
We looked into various stretcher designs and have concluded that we are already using the best there is. Even Wunderbar (not available in the US) or other sprung-corner, patented designs do no more than to impose some initial tension on the corners. Although marginal adjustments can be made with turnbuckles and wedges, the canvas still needs in most cases to be unstapled and re-stretched in order to compensate for later buckling. Would a different method of attachment be less likely to accentuate problems? For example, a method by which the canvas was attached only to the center of each side, (or not attached at all), but held under tension between the center and the ends of each bar, with some kind of spring tension arrangement allowing the canvas to move along but not across the stretcher bars.
One other possibility is that manual stretching is itself at least the partial cause of later problems. Are machine-stretched canvases less prone to objectionable buckling?
Does anyone have constructive suggestions or a better approach?
By the way, a useful reference is Steven Saltzyk, Art Hardware, the definitive guide to artists' materials, Watson-Guptill, 1987