busybee Food Coffee Talk

I ended up buying the starter kit, as for around $20 more than the base Popper* (“*” is short for “Is A Coffee Roaster”), you get a power meter and 4 pounds of beans, which individually would cost way more than $20.

The Popper* itself is basically a side-vented popcorn popper, with a few additions:

• Knobs for fan and temperature control
• A timer
• A chaff collection cup

The Popper* also comes with instructions (which I found a bit confusing until I actually used the thing, after which they became clear as day) and some color reference cards, which aren’t suitable for decaf (which always looks a bit darker than its caffeinated counterpart) but it also has a handy weight chart showing the approximate correspondence between roast levels and weight loss. Unfortunately, they formatted the it in terms of moisture loss rather than final weight percentage (yield), so it takes some extra steps to calculate. I generally just divide my final weight by my starting weight and then subtract 1, and ignore that it’s a negative number. But this is just a vague guideline anyway; flavor is the real way to tell.

Operating it is a little bit counter-intuitive the first time; basically, the timer can only go from 0:00 to 10:00. While the timer is between 0:00 and 3:00 it’s only running the fan, and from 3:00 to 10:00 it runs the fan and the heater. Typically, the way you work with it is to set your starting temperature, then turn the timer knob one click to the left to get it to the full 10:00. Then you slowly increase the temperature over the next 7 minutes in order to get the roast style that you want, and then you let it cool for the remaining 3. At any point you can also add more time (in case you’re going for a much slower roast) or put it right into cooling mode (in case you want it to stop roasting). However, adding time is a bit tricky, as each click on the timer knob is 5 seconds, but the timer keeps counting down as you add time, so you have to be good at math and counting.

The temperature knob doesn’t have any actual markings on it. The instructions always refer to things in terms of “clock position” (i.e. 12:00 is straight up, 3:00 is all the way to the right). But that’s where the power meter comes in: what you’re supposed to do is to find the point on the dial where the Popper* uses a certain amount of power, as a proxy for its actual roast temperature. I ended up using a Sharpie to mark three positions, at which point the Popper* is drawing 980W, 1140W, and 1200W, respectively, as those seem to be its stable power draw points for me (points in between would basically flicker back and forth between those draw levels on the meter which isn’t great for readability).

The UX of the Popper* could be a lot better, but it was clearly built to a price target, and it’s amazing that they were able to get such a good, specialized device together for only $80. Other similar roasters cost about 3x as much, and ones with better UX and automated profiling start at quite a lot more, so I’m willing to put up with a little jank and having to do a little mental arithmetic to use it.

That said, it would be really cool if it had some sort of serial interface such that one could control it from an Arduino or the like, in order to get some more automated profiling.

In my experience so far, staying at just 980W never quite gets to first crack, and results in a super-light roast which remains pretty acidic, and without a lot of flavor development. So, you really do want to increase the roast temperature over time. No unattended roasting here.

Lately I’ve been roasting some Indonesian Sumatra Mandheling Royal Select MWP which wants to be a bit on the darker side. I’ve gotten decent results by doing 3 minutes at 980W and then 2 minutes each at 1140 and 1200, although it’s a bit darker than I like. My most recent roast was 4/2/1, which strangely has reduced the mass a little more but it doesn’t seem to be developing the characteristic oils of a city+ roast, so who even knows what’s going on. In any case, for the next roast I’ll try 5 at 980W and 1 each at 1140W and 1200W, as I’ve consistently hit first crack 1:20 after increasing to 1140W and I feel like getting more development time before first crack is the important bit.

At this point I’ve stopped keeping detailed tasting notes, and am kind of not that interested in maintaining my spreadsheet either. I might take note of particularly good roasts so that I can revisit them later, but for my daily coffee drinking I just don’t really need to do the full scientific method. For now I’m just keeping the roast information on the lids of my containers:

• The date and variety
• Input and output weight
• Approximate roast level (as determined by the weight loss)
• Roast profile (time and power level)
• First crack time
• Espresso grind size

At its price point, the Popper* is a really good buy for someone who wants to save money on coffee. At essentially $8 per 14 ounces (more or less) it’s way more affordable than commercially-roasted specialty coffee, and with providers like Sweet Maria’s and Burman there are so many interesting and affordable varieties to try. Given that good specialty decaf starts at around $22/pound after shipping, my Popper* has already more or less paid for itself after having roasted around 6 pounds of coffee.

Of course this is only suitable for small batches for home-roasting purposes. Trying to do larger batches is a tedious process that is also likely to severely decrease the lifespan of the roaster (as you can only do around 90-100 grams at a time and need to wait at least half an hour between batches, but more is better), and the lack of built-in automatic profiling also makes it very much a tinkerer’s tool. It suits me just fine but for people who are extremely serious about roasting there are definitely better devices to go with.

Anyway, now that I have a better handle on using the Popper* I am really looking forward to trying out the Colombian Las Montanas EA Natural, and the Mexican Esmeralda EA Natural should make for a very nice light roast, which I’ll probably do at 980W the whole way through, or perhaps 5 minutes at 980 and then finishing it at 1140.

So, my popcorn popper is now back to just making popcorn. It’s a bit strange to run it without a concurrent stopwatch, but popcorn doesn’t need much in the way of calibration. Either it pops or it doesn’t.