Easy day today...

I had thought I'd do 3 or more roasts but got side-tracked reading some info I'd gotten on Aillio's Roast.World forum related to heat-soak. So I managed a single roast of some decaf (still struggling to roast decaf I'm excited about!).

The roast I did sounded interesting: ETHIOPIA Organic Limu Kossa SWP (decaf). Not exactly a 'chocolate sundae', but still...

The Cupping Notes told me I'd have to go a little dark (as usual). I tried to merge that target with another attempt at incorporating a 'heat-soak' described by Glenn Junkert (@quartzglen on Roast.World; formerly the roaster at Hunter Bay Coffee in Missoula, MT). The heat-soak requires changes in drum speed (as well as power and fan settings) . Since I can control drum speed now, why not give it a whirl?

"Fragrance and aroma are potent. Complex molasses and sorghum sweeteners, pine syrup aroma, dried berry, raisin and prune, big cocoa roast flavors when roasted dark. City to Full City. Good espresso."

So, this would need a darker roast to highlight any chocolate flavors. But it's hard to track color with a decaf coffee as they start out darker than non-decaffeinated beans. For this kind of roast I have to roast using temperature since decaf beans are much darker at the same roast level making it hard to use color to track the roast is.

Using the heat-soak Glenn talked about requires pre-heating the drum for a long time to stabilize Bean Temp as much as possible (no biggy since I do that anyway). What I don't do is a throw-away roast; the first roast is in my experience what does most to thermally prep the drum & front plate for roasting. But I don't have crap beans laying around for that purpose.

When the drum is ready (45-60 minutes), the plan is to add the beans ('charge' the roaster) and drop drum speed to D3. D3 is really slow which leaves the beans in contact with the drum longer for each revolution. At this same time I have to set power to P1. When Bean Temp stops dropping and just starts to increase, Rate of Rise will become positive and appear suddenly on the screen (takes about 1 minute). At that point Glenn feels the beans have absorbed the heat in the drum without heating the bean surface too fast (a very un-Rao approach). When RoR turns positive I have to increase power to heat the drum to roasting temps and start the roast.

These steps which warm the beans initially using the residual heat in the drum add about 2 minutes to the roast. I think I came pretty close to what Glenn had described:

All-in-all it went pretty well. The RoR curve that dominates the profile is flatter than I've seen before and the tail-off after 1C was better than I've managed in the past: it moved smoothly down to about 10F° per minute at the end of the roast which finished at 441°F when I opened the door to drop the beans into the cooling tray.

On the down-side I had another communication buss alarm as I started the roast and power went to P0. It was right at the start and I recovered quickly so it didn't affect the profile. But... ?!