A Study to Evaluate Your Charcoal Yield
The Italian barbecue fuel market is quite abnormal. Out of 10 bags sold in specialized stores, 9 are briquettes. Quality coal, the so-called hardwood lump charcoal, to say it as Americans is still an unknown concept to the most. Even in northern Italy where there is more attention to new trends, finding good coal is still very, too difficult, lacking mass distribution. On the other hand, even if it was easy to find, who would be willing to spend more than € 20 for a bag of 10 kg of coal? The average customer has in mind the charcoal of the supermarket, the impalpable and full of sparks one, which in a few minutes already has become ashes and compared to it, a box of briquettes that says “up to 3 hours of combustion” it inevitably seems more convenient.
So hard is the life of charcoal lovers like me, who had to move on pilgrimages between the less worst of the great distribution, Cash and Carry and the local charcoal producer, in the few regions in which they still exist. In the midst of a thousand tests, it was very easy to realize one thing: there is a clear correlation between the weight of the fuel and its yield. Myself in my book Subito Barbecue, Among the supply-related recommendations, I suggest using the same volume weight as a parameter to discriminate the quality of the coal. From here I have come up with a series of questions: how close is this relationship? Is it linear? Is there a coefficient that applied to weight can allow us to know the expressed power and duration? Obviously in the construction of such a function, we are neglecting an important factor, the mass of individual pieces of coal. However, this is a parameter that can be uniformed during the testing phase and that on quality charcoal is not so different: except for transport problems, I rarely see coal from high specific weight with a large difference in the size of the coal, inside the bag.
So I decided to go through a comparative test: the idea is to use some chimney starters placed in parallel and to saturate them with coal, at the bounds of embossing. The method is the same we used to build the fuel base in the method of extreme minion we saw some months ago, or to alternate smaller pieces in larger pieces to maximize occupation of available space and maximize the mass. After weighing it, we will put it under each chimney a long lasting Starter Cube XXL by McBrikett so that even the toughest coal may have enough solicitation to be able to turn on properly. The chimney would thus act as quickened burners, making coal work in the maximum possible stress condition, minimizing the influence of external factors in this way. With a laser detection thermometer we will measure the temperature recorded in each chimney at 5 minute intervals, then construct a graph of the trend and relate it to the starting weight.
We do not even have to say it: this is a very empirical and very scarce scientific test, and it deserves to be measured with more appropriate tools and in more standardized conditions than those in which we are going to do this which does not want to be anything more than a game, from which we have as the sole goal to get a estimated indication, a trend. And in this game we decided to ask for the help of Eco Trade, an important national trader of the branch which, in my opinion, produces one of the best Italian coal I have experienced. First of all, the owner, Tommaso Francalanci, confirms our basic assumption:
The expression of power and the duration of coal are function of the amount of carbon residue resulting from the carbonization process. This factor is a consequence of both the genetic characteristics of the starting wood and the production process, where the traditional carbonaian method returns a better result than the ovens. Certainly weight is the expression of the carbon residue and consequently the performance but I can not say to what extent. I am also curious to see the results of this test
The next step is to look for 4 candidates for our test. Tommaso provides three of them: his toe product, a Holly Oak coal that I use often and of really high quality, a mix of 90% Alder and 10% Chestnut and Argentine Quebracho Blanco. To these I add a hardwood Lump Charcoal mixed oak / hickory where however the percentages are not declared.
At the level of homogeneity, the producers of the samples of Alder, Quebracho and Oak have chosen a dimension that is perfect for the purpose, from the first to the last piece with the mean size. 
With Holly Oak I had to work a bit behind, in the sense that the smaller pieces were similar to those of the other samples but inside there were also several significantly more important sizes, which is certainly a manna from the sky for those who buy it but for this specific test it did not help us in saturating the chimneys properly. So I broke the larger pieces (in some cases even very large) to bring them to the size of the other samples. I so arranged the pieces so as to let the chimneys keep as much coal as possible and then proceeded to weigh. The result is a bit flooring: the far lower weight (net of the chimney) is that of Oak / Hickory with 1,132 Kg, then comes the Alder / Chestnut with 1,566 Kg and finally almost identical Quebracho and Leccio, respectively 1,991 Kg and 1,995 Kg.
At start-up, the coal produces little smoke. The first to produce it is as foreseeable, the Oak / Hckory, then after about 5 minutes in the unison Alder and Quebracho and after another 5 minutes more the Holly Oak. Meanwhile, clearly, smoke or not, the timely detection of temperature has begun through the laser thermometer. I immediately notice some difficulty in reading: depending on where you are pointing you get quite different datas. In the first few surveys, I shoot a dozen readings per chimney and recorded the most recurring data. Then I noticed that if I was pointing the mouth of the chimneys, a few inches above the coals I had clearly more uniform datas.
We recorded the temperature values of chimneys until they did not fall below 50° C, after which we considered it as convention as off. As expected, the four charcoals produced quite a different trend: Alder did not produce extraordinary temperatures reaching the peak of “only” 406 degrees, but it had a good 75 minutes duration. Holly Oak and Quebracho had similar trends with very high temperatures (up to 550 degrees) and very long duration of 85 and 80 minutes, where in truth this last difference is only due to the fact that Holly Oak was in the last 5 minutes a few degrees above the 50° C threshold we set as a limit. The Oak / Hickory eventually produced good temperature tips (529° C) but a very short duration of no more than 60 minutes of combustion. This is the situation at the end of the Oak / Hickory Coals:
The first conclusion we can come up with is that apparently there is no true correlation between the weight and the expressed temperature: a very light wood such as the Oak / Hickory used in the test produced temperatures similar to Holly Oak and Quebracho that weighed almost double. 
On the contrary Alder, whose weight lay in the middle, produced poorly temperature peaks. Instead, there would seem to be a direct relationship between the weight of the coal and its duration in combustion. If we relate weight growth to duration one, two curves tend to converge at the raising of weight, especially if we consider that the last value on Holly Oak’s duration has spotted to the 85-minute step only for very few degrees. The ideal would be to deepen the test with a accurate detection of both temperature and time but as mentioned, ours was little more than a game. If this assumption were true, it would mean that there is a relationship between weight and duration that tends to decrease in raising of the value growth or in other words, a greater weight on poorer wood would lead to a benefit in terms of duration higher than an equal weight difference would lead to heavier woods. Of course this is just a supposition. In any case, we try to continue the game and take a fifth charcoal, 
of acceptable quality among those commonly distributed on the market but in which the essence is not declared and we repeat the test. The net weight of the chimney is 1,233 Kg, slightly higher than that of Oak / hickory. If the trend of the curve was true we would find it on the dotted line, which would correspond to a expected duration time of 65 minutes. Let’s try.
One thing that emerges from now is the amount of expressed smoke: definitely higher, at least triple than that of the previous 4 candidates. Despite this, the data found fully confirms our expectations: once again the weight did not express datas apparently related with temperatures, with a peak of 497° C, 8 degrees below Oak / Hickory despite weighing more. By contrast, the duration was exactly the 65 minutes that the curve suggested.
Conclusions:
One thing we can give virtually for certain or almost, is that the pure power expression of coal is influenced by other factors, which are not the weight. The other thing we understood is that there is some relationship between fuel weight and fuel duration and we suppose it can be a variable that tends to decrease as the values increase. The next step might be to try to introduce into the test some devices, of which we should consider the mass and volume beyond of course the working temperature.
The question is: given the mass of the smoker, its volume, the average working temperature and the weight of the fuel input, will it be possible to know what its duration will be?
