Charcoal Tests

 AKA "The Pyro Film Can Tests"  (and a couple of burn rate tests)

This experiment was not precise. When using this information to evaluate relative merits of different charcoals, compositions and procedures remember that the tests had a wide time variance. The same powders shot on different days might have significant variances. 

Using this data to say that white pine is generally faster than weeping willow would be incorrect since there were just minor differences between the two when they were tested. On any given day, the experiment could flop the other way. However, it would be safe to conclude that white pine is much faster than briquettes since briquettes tested several seconds slower on multiple shots.

The intent of these tests was not to advocate using the fastest BP possible. Super fast BP is not very useful for most pyro needs. 

Additional Reading Recommendations:

"Effect of Different Charcoal Types Upon Handmade Lift Powder", Charles Wilson,  Journal of Pyrotechnics, Issue 10, Winter 1999. 

"A Systematic Study of the Performance of Charcoals in Pyrotechnic Compositions", Roger O'Neill, Pyrotechnica XVII, November 1997

The results of the following tests seem to support, in cases where similar charcoals were used, the studies cited above. 

Overview of Tests

The initial tests describe the flight time of a film can shot from a 15" schedule 80 PVC tube and the speed of various BP mixes on a 2.5 foot burn trough. The burn trough tests have become the predominant means of testing since they seem to have the highest relevance and repeatability. 

The film can was weighted with sand to 35 grams prior to gluing on the lid with hot glue.  Each shot used a 2 gram charge that was bottom fused.

The burn trough was 2.5 feet between measuring posts. It was a piece of angle iron and the BP was loosely poured and leveled to about 1/2" wide and about 1/4" deep for the length of the trough.

For the first series of tests, 200 grams of BP green mix was made from each type of charcoal. Each mix was milled for 24 hours in a Harbor Freight mill using 2.5 pounds of 50 caliber lead shot and then 

Example of mill that was used

granulated/riced through a window screen. Half the mix was pressed to 1.7 grams per cc, granulated and 2fg (16 mesh) screenings were taken for lift. The charge for the powder was 2 grams. The flight time was measured with a digital video camera. The launch tubes were bottom fused - lift cups were not used.

Mills like the one above were chosen to highlight the differences in reactivity of the different charcoals - mills like the one above are NOT the most efficient and will not get the best results. BP made with a 6" PVC container, using 30 pounds of good milling media, and turning at about 60 rpm will be much better and faster. See http://www.wichitabuggywhip.com/fireworks/misc.html#ballmill for an example of a pyro mill that can swing such a container.

Here is a burn trough example of Skylighter charcoal in 68/22/10 configuration that was milled for 12 hours in a Sponenburgh mill - the burn trough time was .5 seconds. That is faster than commercial powder: rp.wmv In addition, the same mix put a film can up for 5.295 seconds - the best a 75/15/10 Skylighter pulverone could do (using rock tumblers to mill it) was 2.47 seconds. This mixture would be considered unsuitable for lift powder if made in a mill like the one above but when made with an efficient mill, it produces good lift powder. 

For the second, third and fourth series - generic charcoals (briquettes/lump), milling times and binders were used as variables

For the fifth series, some powders were tested on a burn rate tester. The tester was a 3 ft track of angle iron. The burn of the powder was timed between two markers the distance between which were 2.5 feet. This test is similar to the Dan William's test (Dan William's site went down so the URL is not listed right now)  only a digital camera was the timer. This has become the predominant means of testing.

It was noticed during testing that the results were more repeatable than the mortar tests - at least during the same time frame. Only a couple hundredths of a second difference between tests of the same powder were seen. Pine, a readily available charcoal, made good BP that easily beat the commercial brand. Balsa BP was very fast as was Tree-of-Heaven (Ailanthus altissima) and wild grape.  Black willow was superb (got it from Steve Baron at csworks@netzero.net ) and the overall winner was paulownia (also from csworks). However, subsequent tests with different mills produced even faster burn times. See 'white ash' in the tables (items 25 and 26).

Other black powder tests were run with varying results. A fun test was using an Eprouvette pistol - which looks like the picture below - click on the picture to see a test with BP made from willow charcoal.

Burn troughs and film can mortars give more detailed differences than a pistol Eprouvette. The range on the Eprouvette was limited from 1 to 11 with only half steps between. For the most part, powder tests resulted in half increment values of 1 to 4 (7 values).

Differences of up to 25% were noted between film can shots of the same powders on different days during the same week. This would indicate that the exact time measurement system should not make one think that the tests were precise. The powder burn rate tests, on the other hand, were more repeatable but still not very scientific - things like humidity, temperature and powder height in the trough were uncontrolled variables. A remarkable difference was noted after the commercial powder had been stored for a year in an outside shed. The powder was burned on a cool October night and it was *very* fast compared to the same powder test the year before. See note 28 on the burn trough tests.

 Three granulated versions were slower than their respective riced versions -  the granulated versions were pressed to 1.7 grams/cc. The grain size was about the same for both the granulated versions and the riced (unpressed) versions. Some authors/webs have implied that one must press and granulate the BP to get the most power from it. However, that isn't always the case. One theory might be that riced powders are fluffier and lumpy as opposed to smooth faced grains - thus some riced powders may burn faster due to more area being exposed. Another is that different charcoals may have different optimum compression factors and 1.7 grams/cc may not be appropriate for all grains. Consistent advantages to granulation are more robust and less dusty powders than riced-only mixes.

Of the generic charcoals (briquettes, lump wood, and garden), it seems that garden charcoal would make a good lift powder. It won't be as hot as commercial but certainly serviceable. 

Kroger briquettes made milled and riced BP that was about the same power as Skylighter airfloat. 

Increasing the milling time increases the power of the BP but the increase is not uniform. Different charcoals improve at different rates and percentages. A range of about 10 to about 27  percent improvement was had by increasing the mill time from 6 to 18 hours. In the one case tried, milling past 18 hours did not seem to affect the power of the BP.

There is at least some evidence that the small mills do not get the full potential from the BP mix. See the white ash times in the burn trough tests. White ash milled with a hobby rock tumbler was tested and it delivered medium speeds. The same charcoal, milled in a Sponenburgh mill (a heavier duty mill) for just five hours was much faster. See note 26 in the tests.

Dextrin decreases the power of the BP and adding it might have some trade-offs - what you get in stronger grains is traded for a weaker BP. However, small amounts of Dextrin might be useful. Here is a film can launch with 2.5 grams of pine charcoal with 5% dextrin. This is different than the test launches in that it has a live payload (daylight launch): mortarDextrin.wmv

 Sodium Carboxymethylcellulose (CMC) seems to be a better binder than dextrin. The BP made with CMC was much sturdier than similar dextrin grains. It also seems to be a better fuel. Adding 2% to the BP did not result in noticeable burn trough differences. See a movie of the burn rate test here: paulownia+cmc.wmv.

For this equipment setup, flight times of over 3 seconds seemed indicative of a usable lift powder. Three seconds wasn't high enough for a safe break but increasing the 2 gram charge usually resulted in enough altitude. Powders that could not consistently beat the 3 second flight time were burning too slowly and adding more only push unburned BP out the tube; thus, they could not be used for lift powder for this small mortar. Powders with flight times of 4 seconds or more showed the best consistency, economy and repeatability of flights. Occasionally, one of the slower powders would just fizzle - thus barely pushing the shell out of the tube - this did not happen with the faster powders. Experience has shown that larger bore mortars with longer tubes are able to use the slower powders more effectively and in some cases the slower powders are desired to prevent slamming of the shell on ignition.

Variables such as wind, humidity, tube wear, flash hole wear, differences between the two mortars that were used and probably a bazillion other things were not taken into account. The intent was merely to show the relative merits of different charcoals and processes when making BP. In particular, one should not extrapolate these results or powder sizes to different size tubes or loads without testing. Tests using the same powder on different days netted different results but relative strengths between powders seemed to remain about the same. For instance, note the Elephant and pine tests on August 27th and August 30th.

Click on the above image to see a comparison between the slowest and one of the medium speed BP mixes. While the tests can only be used in a relative way, the information does appear useful - one can easily see the merits of pine vs. briquettes in the above video. 

In addition, take a look at the burn rate tests (last series) for another way of looking at the relative strengths of the BPs. Click on the image above to see a video of the Our Family Briquette burn compared to the Balsa burn.

¹ Skylighter airfloat made from a combination of hardwoods
² Skylighter CIA then ball milled and pressed to 1.7 grams/cc
³ Benzolift made with Skylighter pulverone
⁴  Shultz Garden Charcoal (meant to richen/filter the soil)
⁵  Our Family generic briquettes
⁶  Milled for 6 hours
⁷  Enchanted Garden Bamboo garden stakes
⁸  Maple from lumber yard - specific kind of maple unknown
⁹  White pine (soft pine) was used. White pine is found in smaller construction lumber such as 2 x 4  or firring strips. Yellow pine (hard pine), found in larger size construction lumber, was avoided because of resin problems.
¹⁰  Kroger Briquettes made by Hickory Specialties of Tennessee. They say the only ingredient other than wood is corn starch. The BP has a faint brown tint.
¹¹  Blackwood Canyon Lump Charcoal
¹²  Sieved to about 2fg
¹³  Texas Smoke Apple Chips
¹⁴  Second batch of pine charcoal made to verify first results
¹⁵ Balsa charcoal is very reactive - four shells were lost in testing the BP made from it. One shell broke up at the muzzle and the other three just disappeared into the sky. Heights for two of the successful shells exceeded 400 feet on just 2 grams. The time posted was the most conservative - two shots flew more than 10 seconds but noise precluded a positive identification of the flight time.
¹⁶ Tree-of-Heaven (Ailanthus altissima) is considered invasive - a weed - and it grows nearly everywhere. Only 1"-2" trunks were used for charcoal. The older trunks were much tougher to cut so were discarded in favor of the softer wood which was believed to make a 'fluffier' charcoal. 
¹⁷ Wild grape was made into charcoal using bark and all. The trimmings available were small so a good deal of effort would have been expended in getting only wood and no bark.
¹⁸ High winds during test made timings somewhat unreliable but subsequent burn-rate tests seem to validate flight times
¹⁹ Black willow and paulownia purchased from csworks@netzero.net  already milled to airfloat
²⁰ White ash, golden locust - note golden locust makes lingering sparks - more so than pine from Ron C (Smiley)
²¹ Black cottonwood and scotch pine from Tad Kolwicz
²² Red alder from Kelly Jones
²³ Red elm supplied by Jeff Grey
²⁴ Unidentified willow (not weeping but large so probably black willow) 'Smiley Willow' made with 'Sponenburgh' mill - 4" jar milled 3.5 hours
^{25, 26} White ash. This result should NOT be compared to the other charcoals. It was milled for five-six hours in a Sponenburgh mill. This is same white ash charcoal BP used on 4/16/06 - which was made with a rock tumbler. This is easily the fastest BP tested and indicates the small mills are not getting the full potential out of the BP mix. See: whiteash3a.wmv for a video of the burn test. Also, see rp.mpg for a video of Skylighter charcoal that has been milled using a Sponenburgh mill. The heavier mills definitely do a better job.
²⁷ Manzanita, tipuana, rotten willow, mesquite and California sage courtesy of Bob Forward and the California Desert!
²⁸  Note the huge difference between the 2Fg burn trough test from the year before (the first entry in the burn trough table) - it was the same batch of powder. The last test was done on a cool October night and both the Goex and the Elephant had been stored in an outside shed for several months. This effect was also noted by Dan Williams in his burn tests.
²⁹   White poplar charcoal donated by Konstantinos G.
³⁰ Yellow birch submitted by Nicolas G.
³¹ Red Alder from CustomCharcoal.com - provided by Brian Welch
³² Kiawe wood from Hawaii - provided by BungeeMike. This charcoal is offered commercially in Hawaii. It is similar to pine in reactivity. The test is a little skewed because the charcoal was too hard to break down in the small mill. It was re-milled for 1.5 hours in a large mill and probably gained a bit from the experience.
³³ Grapevine from Nitroparis. (Spain) courtesy of John Reilly
³⁴ Courtesy of Steve La Duke - the cedar was locally grown in Wisconsin. Update on proper name needed.
³⁵ Courtesy of Dan Kaiser
³⁶  Courtesy of Bob Forward
³⁷  Courtesy of Stewart Irsik
³⁸  Courtesy of Dan Chubka
³⁹ Courtesy of Roger Hankins
⁴⁰  Courtesy of Dave Taylor
⁴¹  Courtesy of Summit - all charcoals from trees grown inside the Arctic Circle
⁴²  Courtesy of Doug Knight
⁴³  Courtesy of David Cross
⁴⁴  Courtesy of Gary Robinson
⁴⁵  Courtesy of Steve Johannessen
⁴⁶  Courtesy of Mike Beyer
⁴⁷  Courtesy of Lee Bussy
⁴⁸  Courtesy of Ron H.
⁴⁹  Courtesy of Alan Brennen - 80 mesh commercial charcoal from a small distributer. Compare with the 80 mesh charcoal from Service Chemical.  Note both charcoals were reground to airfloat and then milled into BP.
⁵⁰  Courtesy of Bill Kimbrough
⁵¹ Courtesy of Dale Miller and Ron Christensen
⁵² Courtesy of David Forster
⁵³ Courtesy of Greg Martin
⁵⁴ Courtesy of Derrick (Mumpyro)
⁵⁵ Courtesy of Allen Collins