Rob, now that you mention it being possibly a known problem, I think I remember Mike at BAR mentioning the possibility of separation problems having to do with the coupler binding inside the other tube due to it getting a little sideways or angled. I don't really remember what he was saying about it.
I was a little concerned about how tight my coupler fit into the lower tube, so I used CA to harden the surfaces and lips of the tubes and then sanded them smooth. The first step was to really soak a lot of thin CA into the lip of the tube. I did that by getting a surface I wasn't concerned about ruining --- it was the lid to a cardboard box of some kind, like maybe a shoebox. I put a layer of wax paper inside the lid and folded up the edges so that liquid could not easily escape the wax paper. Then I squirted a little puddle of thin CA into the center of the wax paper. It wasn't very much, just a fraction of an ounce, and maybe a puddle a couple of inches across. Then I got the lower section of the rocket, flipped it over upside down, and placed the lip of the tube flat on the wax paper over the puddle. By moving the tube around. I was able to get all parts of the rim in good contact with liquid CA, and it just wicked up into the cardboard. I actually did this with the coupler too and also with the top lip of the payload section.
Next, I used a foam brush to "paint" a lot of thin CA onto the inner surface of the lower tube. After it was dry, I sanded the inside of the tube so that it was very smooth. With the lip of the tube reinforced with CA, I was also able to sand a little bit of a bevel or a rounded lip on the inside of the rim of the lower tube. I also painted CA on the inside of the coupler and sanded that too, mostly because I liked how smooth it could be, and also because I thought it might protect against moisture and mud in case of a sloppy landing site. Then I also sanded and rounded the reinforced lip of the coupler. Having both the coupler and the lower tube lips hardened and rounded made it much easier to get the coupler started into the lower tube when mating the parts together --- the rims did not tend to bind on each other as much.
Those improvements made the coupler slide into the tube a lot easier, while still providing a good seal. It was still a bit tight, so the last step was something that I am not sure I can really recommend, but I will tell you about it anyway. I have some silicone spray lubricant called Jig-a-Loo. It's a non-staining lubricant that is safe for use with things like fabrics, and it is moisture resistant and non-greasy. People use it for things like lubricating curtain rods, or zippers, or hardware inside auto interiors where you want lubricant, but don't want to damage the fabrics. I had just a trace amount in the bottom of a can and I sprayed some on the inner surface inside the lower tube. Wow! That made it really TOO slippery! If I tried to pick up the rocket holding only the top section, the bottom section would just fall right off. You really don't want it so loose it will drag separate. To tighten it up again, I had to add a little bit of tape for a tighter fit. I don't know if the lubricant has dried up a bit or soaked in or evaporated, but it is not as slippery now as when I first added it. The reason I would not necessarily recommend it is that it might be too slick, and also I've heard since then that some silicone lubricants are bad for certain paints.
Regarding the motors, at the last launch, I used a G77-4R, and a G40-4W. At an earlier launch, I used a G75-4DM and a G78 or G79. So far I have only flown it on AT single-use motors, so the ejection charges are set at the factory. I've only used 4 second delays so far.
I hope that all helps! Good luck with your testing, and please let us know what you find. I'd hate to crash this rocket. It's one of my favorites now!