The purpose of these tests was to compare different flash/burst compositions for sensitivity, light output, sound, and power. Eight different compositions were tested. The aluminum powder used was the same in each, German dark, about 600 mesh. All compositions and ratios used except for #4 are from the Composition Database.
1) Standard ratio greenmix (blackpowder just dry mixed and shaken together)
2) H3 (75 KClO3/25 charcoal)
3) 50/50 KNO3/Al
4) 50/30/20 KNO3/S/Al
5) Whistle mix 70/30 KClO4/Sodium benzoate
6) 12/7/10 KMnO4/Al/S
7) 50/23/27 KClO4/Al/S
8) 70/30 KClO4/Al
Mixing and a mess in general:
The tests: Five different tests were done on each composition. They were friction, impact, spark, open air burning, and moderate confinement. Later I might test these same compositions in heavier confinement (cobs). These tests were rather crude, and should not be considered the definitive source for information, but they will give you some general information for comparison.
Friction/Impact:
A very small amount of each powder was poured onto a concrete block and scraped roughly with a very scientific instrument called a "broken chunk of concrete". Not a single powder ignited in this test. For impact, another small pile was poured onto the concrete, a piece of paper was put over it, and it was hit several times fairly hard with a hammer. Again, none of the compositions ignited. While this is comforting, you should still treat your mixed powders with care!
Friction and impact tests:
Sparks:
A small pile of powder was again poured onto the concrete and with a flint scraper, sparkers were directed onto the powder. The greenmix ignited without difficulty, but it took a pretty good spark to light it. Something seems to have gone quite wrong with my whistle mix, because it would not ignite. All other powders ignited easily, especially the KClO4 and KMnO4 based ones.
Burning:
About 0.5g of each powder was poured onto a piece of paper and lit. The greenmix burned fairly slowly, but completely. The H3 went very quickly compared to the greenmix. #3 was fairly quick, but not quite as fast as the H3. It was however, very bright and #4 was also about as quick as #3, but not as bright. The whistle mix (#5) would not ignite on its own and greenmix was added to ignite it. I'll have to try to improve the whistle mix because it should be much more sensitive to sparks and flame than it was. #6 burnt very quickly and brightly, the fastest burning of those tested so to that point. #7 also burnt quickly and brightly, but perhaps not quite as fast as #6. #8 was the fastest burning of all, and was very bright (but not as bright as #3).
Burning powders:
Confinement:
For the next test, 2g of each composition was used in a polumna (also called a triangle cracker). The same paper was used for each one, and I tried to use the same amount of tape for each so they would all be the same. You can't see it in the picture, but they were all labeled on the other side so I could tell them apart! I could have fit 3g in each, and that was my original plan, but I was a bit worried about the noise.
2g polumnas:
All the polumna tests were done in soft snow, and the craters were measured to get a very general idea of their power. I had pictures, but they seem to have been erasedÔø‡
1) The BP made a small pop, some sparks and a 10cm X 20cm crater.
2) The H3 made a similar pop and a 10cm X 15cm crater.
3) The KNO3/Al made a bright flash, some sparks, but was quieter than the H3. It made a 15cm X 25cm crater.
4) This one was a bit louder than #3, not as bright, and few sparks. 22cm X 15cm crater.
5) Surprisingly, the whistle mix that would not ignite from the
flame of a lighter, ignited just fine this time. It made a louder pop
than #4 but not much light. I forgot to measure the crater, but I think
it was pretty similar to #4.
6) The KMnO4 flash was fairly bright, made a pretty good bang, and a 22cm X 25cm crater that was also deeper than the others so far.
7) This flash was the louder than #6 but a little bit, made a deeper crater, and was 39cm X 25cm.
8) The standard flash used here was brighter than #7, but maybe not quite as loud. The crater was 28cm X 25cm.
Conclusions:
DonÔø‡t take these numbers to be the bible of powders because the tests were all very general. Just because, for example, #7 had a bigger crater than #8 doesnÔø‡t necessarily mean it is that much more powerful, there could be factors like variation in snow condition and depth of the polumna affecting crater size. 6, 7 and 8 were all pretty similar in sound, flash and crater size. Only use KMnO4 bases flash if you canÔø‡t get KClO4, because KMnO4 is more dangerous and unfavorable for several reasons. If you must use KMnO4, mix it shortly before use, donÔø‡t store it. #7 sounded a little louder than #8, so it might be better because it will also stretch out your Al a bit.
The purpose of this test was to compare different oxidizers for flash powder. KClO4, KClO3, and KMnO4 were used. Each was mixed in the standard 7/3 ratio with aluminum powder. 3g of each was then used in a cardboard tube, filling it only about half way.
The three salutes and comps:
The KClO3 and KClO4 were very similar in flash and sound, but the KClO3 seemed to make a little more smoke. For some reason the KMnO4 one didn't work twice, so it was tested another time in the day. It was surprisingly weak and quiet compared to the polumnas that I tested earlier, where the KMnO4 sounded almost the same as the KClO4. Perhaps it was because this time there was no sulfur, but I had done some tests earlier with KMnO4 with and without sulfur and they didn't seem too different.
KMnO4 salute going off:
Conclusions: KClO4 is the best flash oxidizer, KClO3 is very similar but you must be very careful with it to keep out any sulfur or sulfur compounds that will make it far more sensitive. KMnO4 will work if you have nothing else, but should be your last option.
Note: I read a while after this test about a test doing the exact same thing but using much more sophisticated testing. I don't know if they tested KMnO4, but from their tests they concluded that KClO4 and KClO3 have no noticeable difference in sound. Also, testing flash mixtures with and without sulfur showed no difference in sound apparently. The above conclusions then are then mostly accurate (except the sulfur one).