You might want to try using a nearby star in lieu of an extended object. Perhaps you are already aware of the following, but if not, here's how these masks work...
Each opening acts as an individual off-axis telescope, with its own unique image. As an eyepiece (or camera, etc.) is moved closer or further from focus, the OA image will move across the field of view. This is most apparent when the OA image source is a point source of light (a star). Using extended objects is more difficult, although a two-hole mask can be used for well-defined line-features (such as Rupes Recta on the Moon), rotating the mask so that the two images form lines that diverge/converge laterally (=) and are NOT overlapped inline (--) to determine Critical Focus.
Focusing a single image can be ambiguous, as there is often some latitude or "sweet" spot where small changes seem to have little or no effect. When multiple OA images are used (ie; from a multi-holed focusing mask), the center of each focused "sweet-spot" (even if it is difficult to tell where each individual one might be) intersect at one unique point, (drum roll please...) "Critical Focus"!
So... here's the scenario: However many holes you have, you'll have separate images when out of focus. As you approach focus, those images converge. When you have a single image, Voilá, you have attained Critical Focus. More holes or larger size is not always better though, nor is using a star so bright as to become a giant fuzz-ball when approaching Critical Focus.
The farther apart the openings, the more dramatic their effect (and hence making it easier to determine Critical Focus). Openings nearer the center will have little movement even with large changes in focus, while those nearer the edge will show greater movement even with small changes in focus. Generally, you want the bulk of the opening area as close to the edges of the aperture as possible.
Bigger openings = brighter targets, but to be bigger means that the bulk of their areas is moved closer to each other, diminishing the off-axis effects.
Soooo.... it is a balance of enough total (and individual) area to be bright enough to work with, but not so large as to minimize their off-axis effect. Too small = not bright enough; too big = hard to detect any changes.
The use of "Zoom" to enlarge the image scale also helps. When using a Nikon CoolPix, I would zoom to the max of 8X to aid in scrutinizing the masked images. With the Canon D60 SLR I don't have the luxury of pre-screen zoom, but use a 2.5X optical finder instead.
Ron W. has an excellent chapter on focusing with masks that is available as a *free* sample of his "The New CCD Astronomy" book:
http://www.newastro.com/newastro/samples.asp
He also illustrates an interesting triangular cut-out, which adds diffraction spikes to the help find Critical Focus.
The below image illustrates the level of focus that may be attained when a mask is used in conjunction with zoom.
Paul
Attached Image:
