popper*, the coffee roaster was right on the verge of going into production.
… but one thing just wasn’t right. The factory had sent sample after sample, but the fan speed (and therefore the air flow) was not consistent. This impacted the potential batch size.
A late night video chat with the engineers in China made it clear to me though. They were testing under the principle that the roaster should not use the entire pre-programmed roast cycle of 7 minutes. They thought of 0 to 7 as a range so they were presupposing what I wanted was a 4 minute roast!
Well, a 4 minute roast is easy in a popper, but it’s rarely an ideal roast time. In fact, the whole idea of *popper is that the user can fluidly adjust roast time up or down on the fly. It’s a manual, low tech machine, unless you want to get in there with an arduino interface etc.
It’s a machine that rewards people who want to play around with the variables, so it’s simple to do a 7 minute roast just to first crack and add 2 or 3 minutes for development. You just have to be there and tend to your roast, make intuitive decisions and turn a knob. Simple.
(For me I actually feel the 7 minute pre-set roast time can work very well. For City roast I target 5:30 to hear the start of first crack, and the remaining 1:30 for further development. Remember, roast time is relative to the device and the type of thermal transfer. Higher air flow and more movement of the mass means more rapid levels of heat transfer via convection.)
Popper* will make coffee go pop, but probably pop corn too. And it is coming along nicely. No need to rush to get it made before Christmas, because it needs to do what it’s supposed to do. It’s going to be February I think.
The prototype samples were going well except in one regard: Each seemed to roast a little less than the previous. We ended up with a final sample straining to move 85 grams of fresh green coffee. No!
So we delayed production to balance out fan and heat and the roast chamber vents. I worked on the exit air vents thinking they were constricting flow … but it turned out it was the inlet air vents. And just minor tweaks made big changes in batch capacity.
But those changes impacted the heat curve too and, while I could roast 140 grams, too much heat was blowing by the coffee and out of the roaster. I couldn’t get darker roasts within a single cycle. (I could by adding time, but we need dark roasts to be in closer range).
So now, fingers crossed, last set of prototypes are on the way with new fan speed, and an additional set of roast chambers, 5 in all, to find the best combination.
I think this is it! What I want is a 120 gram target that makes use of the range of air and heat settings, allowing manual profiling with extended roast times, and full roast range.
And with that, I’m able to get some really striking profiles that don’t at all taste like fast popcorn popper air roasts. Low heat warm-up times, ramping up to first crack, and backing off to finish the roast with control. The 10 minute roast profile I did yesterday of Guji Uraga Hare Wato was really the best roast I have had of that coffee. Really fantastic, sweet and floral.