On Friday, April 7, 2017 at 8:49:33 AM UTC-4, ND wrote:
On Thursday, April 6, 2017 at 5:49:39 PM UTC-4, Bob Kuykendall wrote:
...start by mixing up a small batch of filler, and squeegee it onto the wing (or whatever) with a plastic bondo spreader...

The problem I and others have had with that approach is that the filler does not wet out the surface of the pinhole very well. Even though it fills the pinhole, it does not establish a bond with the surface. Often, the pneumatic pressure of the air in the pinhole forces the thick filler back out, so you get only a very shallow fill over the top of the pinhole.

As it ages and the wing (or whatever) flexes, the filler is likely to break free from the interior surface of each pinhole. Then the pinholes start mirroring through the finish paint, and eventually might start cracking in a checkerboard pattern.

Squeegeeing on a couple quick skims of runny epoxy helps wet the interior surface of each pinhole, and forms a rounded meniscus over its sharp irregular internal contour. Filler sprayed or squeegeed over the epoxy skim has a better surface to bond to, and there is less internal volume in each pinhole for air that repels the filler and impedes the bond.

--Bob K.

that has not been my experience. it's worked pretty well for me for the last several years, during which time i've done a few dozen refinishes. also, filler seems to be less viscous than runny epoxy. and you can spray right over, immediately afterwards without any sanding. i also believe that it does create a good bond, especially since there is a mechanical bond in there. i have not had any problems with filler pushing back out of the pinholes.. in fact, there is a soft of capillary action, and it seems to soak in like a sponge. i remember doing a wingtip that was heavily crazed, by squeegeeing some resin. it was messy, and not so nice to sand afterwards. no question, your method works well, but i think mine does too. there advantages to both.

I've seen some times when trapped air pushes the filler back out of the pinhole to some degree. The reason I went to the roller method is that each individual area gets multiple passes from at least 2 different directions which I have found minimizes this. I also have found that I can get a more uniform surface with the roller than I can with a squeegee. I've done it both ways and still occasionally find that the squeegee helps.
Sanding this filler back flat does no new damage to the surface where sanding epoxy is harder and, I have found, more prone to adding more damage.
I have not seen any pinhole repairs come back to the surface with this technique.
It's worth noting that there are several ways to skin this cat.
UH`