Dang, that last post came off way too sharp and harsh. I apologize NeighborJ. I gotta stop posting before having coffee in the morning.
I didn't notice the glitter you said you were using. It contains a heck of a lot more charcoal than a normal glitter. I still think it's unlikely, but could in theory be possible for some of the basic materials to started to affect things. Another possibility you may want to consider is the purity or pH of the sodium nitrate. If you have something more on the order of fertilizer grade that may be part of the issue as well. Besides basic impurities, chloride and some other anions can be quite corrosive.
Most charcoals have some native amount of ash. This is just the ultimate fate of minerals from the soil and life cycle of the tree. Very clean woods will have ash contents in the realm of 3-5% by mass. Very "dirty" woods will have ash contents in the 20-30% realm. Commercial airfloat charcoal is probably right around 15-20%. There is a general relationship between ash content and speed. Lower the ash content, the faster the burn.
Two random thoughts related to this. While low ash content is correlated with a faster burn, I've wondered if a high ash content is correlated with a long hangtime. The components of ash really aren't that different that the carbonates, bicarbonates, and oxalates we already use to slow burn rates and add delays to glitters. There are some instances, the firefly effect specifically, where addition of sulfate greatly extends the delay effect as well.
If you really want to lower the ash content, wash your charcoal with a dilute acid like nitric or perhaps acetic. 50% or more of the ash is very poorly soluble in water. That said, crappy charcoal is not magically made significantly better by this treatment. There's more to speed than ash content.