torrefriteuseeng


	Translated by Gary Seeman

Roasting is something that has always attracted me, but I had never had the opportunity to take the plunge. Maybe it was lack of time, commitment, space, or apprehension. But still there was something missing in my approach. I made myself espresso, coffee in general. I had everything, but…. It was time to fill that gap. So, one day on an espresso forum I read of someone who had acquired what he thought was a roaster, but in fact it was a rotary fryer without oil. Admittedly it was very like a Behmor roaster. So far I had not intended to invest a substantial sum in a roaster and the solution of a popcorn machine attracted me. I don’t know much about that, but it was the catalyst, I don’t know why. It was time to consult free announcements, so I found myself with a fryer at home in exchange for a handful of small coins.


	It’s not really discrete, this fryer. I’ll have to find a place to stash it.



There are two heating elements on each side.	The central drum is driven by a small motor.	A timer and a switch offer the possibility of running the lower heating elements or the upper ones or all at once. This creates interesting possibilities to manage roasting.





	Test heating.


	It's getting warm!
This test was chaotic as it usually is with first attempts. This will certainly require a finer grid because the drum is not designed for frying small coffee beans. They mischievously escaped and fell out onto the bottom tray. This was a precious loss. Unacceptable.
I adapt some mesh from kitchen splatter screens.



	It has a vicious, rough edge that bites me!



My other concern was to insert blades in the drum to initiate the mixing of beans. This is yet a very artisanal achievement, but quite effective. It will require another solution because some beans still get stuck in the grid. Only a few grains get stuck, but enough to annoy me.


Now for the real adventure, starting to roast….


	This was “Roast No. 1."



(There was a roasted 0, but it was zero from every perspective.) Not knowing what I was doing, I stopped roasting, then applied heat again, and so on. This yielded nothing worth seeing. The result must also be viewed with caution. It’s difficult to accurately photograph roasted beans.

	The second roast followed quickly and confirmed my achievements. I begin the third ...



	… and then one rejoices too quickly sometimes.



	After a few long days of waiting, I can finally brew and cup results that are very promising.


I also discovered that there are strange things in packages of green coffee …

At this stage of my tests I think it is necessary to modify my fryer to adjust some details. Namely, I need to change the blades. And why not create a real drum, increase the speed of rotation for better mixing and the ability to adjust the heating elements via an auto tuning controller (PID). This will be more suitable for roasting. And to think that I started this to be able to roast at low cost!


It's funny, If you would have asked my impression of motor speed, I would have said "whew! awfully fast!" when in fact it was crawling at 2.5 revolutions per minute! Then I changed to a powerful motor running twice as fast, at 5 rotations per minute! I finally found a unit that could rotate 30 times per minute, but 4 watts is far too little to drive it loaded with beans. The rest is another challenge. At the same time I began to understand the right quantity to roast. Attempts at 100 gm, 250 gm, until I found the right dose at about 160 gm. This allows one to roast enough beans anyway, with sufficient uniformity.



There is the need to control temperature and time. So I mounted the thermometer and timer for easy viewing.


The PID track was abandoned, as a newbie explains that the electrical load would be too high, so the controller would not survive for long. So I turned to the solution of a voltage regulator.

	Here are both units, one for each row of heating elements. Disassembly can begin.



	I place them in front, hoping for proper positioning.



	Before / After




Impeccable! They control heating very well and I find their arrangement very pleasing.

	Now I began seriously to have meaningful results.


And I’ve created a very pleasing cup, if you could taste it.




	My only regret is not to have started earlier.

	But what a joy to taste one’s own roasted coffee!



Always looking for improvement, I modified the drum several times,


removing grids where beans got stuck and replacing them with metal rods, like a staircase. Then I ended up with a drum roaster that seemed well suited to my needs. This meant that the size of the mesh would be less problematic than the current grid. The seller only wanted to let go of his drum for a few coins. But the drum was still at the other end of France. Super nice!

	tatata!! And by the way, I gained nearly 400 gm weight with this drum!

	Sealing the ends.

And making small stainless steel blades to improve mixing. They are placed so they move the beans from left to right and right to left with each turn.


	And thank you to Armand for making blades. It’s funny that when I first saw the blades in the photo and compared them to what I had imagined in place, I realized they must be aimed in the same direction to roast alternately from left to right. And there is nothing political in this!

	Roasts are now more consistent and no beans get stuck.




I also obtained a motor that makes 15 turns per minute, which significantly changes the movement of beans. I tried 30 rotations in vain; it is not powerful enough to roast a larger load.



Here you can appreciate the difference in speed. And I assure you, it is much less noisy in reality.



Now I roasted with a stainless steel cover. I could control some cooking phases. I quickly and easily achieved 70°, which is a significant improvement in efficiency. I fitted a handful of wood that grabs the end of the drum and allows me to remove it while it is hot without much trouble. I do not know what may turn up tomorrow, but I'd be surprised if there aren’t yet changes in the future ...