Riser length... I take this to mean the glued-on handle piece including enough wood to work the dips into working limb. Mine have varied anywhere from 9" on short bows where I'm pushing limits and need every inch of working limb I can get, up to about 15". But on an average bow of mine, I like 12-13".
Regardless of how long or short they are, I shape and tiller them so that flexing isn't brought to an abrubt halt in an isolated area of the fadeouts. Some guys choose to shape their fadeouts steep and rigid... some are incredibly so. These bows are probably better off with powerlams or pedestals to set the glued-on handle piece on... but they're not necessary for most bows.
I've made bamboo backed bows over 70 lbs with glued-on handle/dip pieces, no pedestal or powerlams whatseover, without an issue, in fact, in all the bows I've made where I added handle pieces on like we're talking about here, I've only ever had one come loose, very early in my bowmaking. It reiterated the lessons I needed to fully injest and apply.
I don't build my handle up with stacked thin slats. It would work, I just don't want to go to the trouble, and prefer the look of a single piece. Not only do I try to use the same kind/strength of wood that's in the limbs, I try to use wood with the same ring orientation(i.e. if the limbs are quartersawn osage, I try to use quartersawn osage handle wood) if possible, but I don't think it's a deal breaker if some other factors are properly addressed.
Some of those things that help ensure they don't come loose are:
Good mating of glue surfaces. Mine are often curved due to the blank's deflex, so I take my time to be sure it sits on there square and that there are no gaps. I hold it up to a light and look hard from every angle for any tiny gaps and correct them, even so, this usually only takes 10-15 minutes.
Proper glue surface preparation. I only glue onto freshly exposed, clean wood. After the handle piece is shaped/ground and mated to the blank, and immediately prior to gluing, I run a toothing plane blade by hand longways on both pieces, kind of like a scraper, the full depth of the teeth and completely past the very end of the fadeout. The toothing plane blade works better than a dragged saw blade or file edge for this because it cuts clean, even, and precise, leaving no tearing or fuzzy remnants. I do this on any horn tips, overlays, underlays, etc as well.
Use the best glue for the application. In this case, because of the grooves from the toothing plane blade, a glue with gap filling properties such as Smooth On epoxy or Unibond works well. I use Smooth On.
Proper mixing ratios and techniques. Make sure glue as at or just slightly above room temperature and mix very, very well... every speck of it, for a long enough duration, then mix for another full minute, and in proper amounts.
Apply to both pieces with a sufficient amount of glue. Fully wet both pieces to be adhered, running the applicator parallel with the grooves and in both directions. On the last pass or two, there should be enough glue on the surfaces that the applicator will 'float or glide' on glue, not felt to 'scrape' the gluing surface. After clamping, there should always be some squeezeout around all sides.
Proper clamping pressure. Using the toothing plane iron, it would be difficult to clamp the pieces tight enough to starve the joint of glue, but I just snug them so the pieces are seated and don't move while they cure. You're not trying to crush anything, and more pressure doesn't make for a better joint. The prep makes the joint.
Heat during cure if warranted. Epoxy likes a little warmth, so I throw a shop light near it and loosely drape a towel over it to keep the heat in. Not so close or tightly that it would burn of course. Let some heat escape an opening in the top.
One of the most important factors is how the whole thing is shaped. If it's short, steep, and angled into working limb, it's going to try to pry off. If it's long, gradual, and flows into working limb, it won't.
I think a couple of things that might also help me is my handle shape(round-ish or bulbous) and that I radius all of my bow's bellies. This means the handle piece doesn't 'fadeout' in a straight lateral line across the limb... rather, as it fades out, it does so as much longitudinally as it does laterally... which distributes the 'prying stresses' on the glue joint over a much greater area than a steep, flat dip coming down into a flat limb belly.
Anyway, there's a few things to think about.
Sorry this post was so long :rolleyes: