For some of the background on this read my web pages on Reactive Targets–especially Project Nitro.  The recipe given there is obsolete, but it will still help you understand some of what follows.

The last year the targets just didn’t detonate as well as they should have.  We changed the containers we used for the targets, we ground the NH₄NO₃ much finer that ever before, we used a KitchenAid^R mixer instead of hand mixing, and we put in a small amount of MnO₂ as coloring to help us tell when it was thoroughly mixed.  Too many variables and we discovered the problem just a week before the event.  We suffered through the event and then last fall I bought some books on explosives and detonation mechanisms and have spent a lot of time reading and thinking about what is going on with our mixture.  Some of my hypotheses were shot down after doing a few thought experiments.  Some others died after some field experiments.  My latest hypothesis is that when we grind the NH₄NO₃ it absorbed the liquid fuel (proprietary) and the KClO₃ isn’t able to interact with the fuel as well during the detonation initiation phase.  This might also explain why the mixture becomes more and more difficult to detonate as the targets age–the fuel is absorbed more and more into the NH₄NO₃ and it’s also possible it is evaporating through the container walls.  Fortunately if this hypothesis is correct the fix is easy–add more fuel and perhaps seal the containers better.  For maximum sensitivity the mix has always been very oxygen rich and I have wanted to add more fuel just to get more bang for the unit volume of target.  If adding more fuel works out I’ll be pleased for more than one reason.

Most recently the ratio of NH₄NO₃ to fuel has been 24:1 by volume. This was based upon a density measurement from a long time ago with a different batch of NH₄NO₃ and I decided to measure the density again before doing these tests.  Previously I had measured a specific density of 0.96.  This time I came up with 0.875.  Ouch!  That could be making a difference too.  Time to redo all our sensitivity tests.  Tonight I finely ground four 500 mL batches of NH₄NO₃ and mixed in 20, 30, 45, and 60 mL of fuel in the respective batches.  That gives me ratios of 25:1, 16.7:1, 11.1:1, and 8.3:1.  I was amazed at how little of the fuel shows up in the 25:1 mix.  It’s almost as if it doesn’t exist.  I’m expecting that after letting it “soak” for a couple weeks even the 8.3:1 mix will look pretty “dry”.  After the next IPSC match (February 6^th) I’ll mix in 125 mL of KClO₃ in each mixture and do the sensitivity tests while at the range.  This year I’m going keep tweaking things until we get everything working right again.