The Alpenrost is a fantastic idea that employs a simple, elegant design. It can roast coffee in ½ lb. batches in a reasonably good time, approximately 18-20 minutes. Experience showed that careful weighing was important, so having a decent scale (e.g., +/- 2.5 g) is essential. Green coffee is loaded into a heavy duty, stainless steel cylindrical chamber, which is rotated steadily over a heating element by a motor. A fan draws air through the unit throughout the process, and aids in cooling the roasted beans during the last phase of roasting. Angled fins within the chamber toss the beans to ensure even roasting; upon cooling, the motor reverses direction and the fins direct the roasted beans from the chamber into a collection bin. I was particularly impressed by this feature, which reflects thoughtful engineering. The chamber is perforated with large enough holes that chaff created during the roast can escape, collected neatly in a removable tray located under the chamber.
The only control is roasting time, determined numerically using settings (1-15). The idea is that, after loading the machine and entering the setting, one can leave the coffee to roast on its own. A cooling cycle is initiated near the end of the roasting time; internal doors open automatically to increase the airflow through the unit during cooling. This roaster produces a great deal of smoke. Initially, I roasted in our garage with the door open. Later, I added a section of clothes-dryer vent hose (flexible type), along with another fan to increase the draw, which enabled me to use the unit inside and vent through a window. The cooling cycle is initiated automatically, depending on the setting. However, another switch enables the user to override the programme, and initiate a cooling cycle at any point during the roast.
Several problems quickly emerged. First, different coffees take different amounts of time to roast, which required the settings to be adjusted for different beans. I found that Costa Rican Tarrazu, for example, required a much longer roast than some of the African beans (Kenyan, Ethiopian) I was using. Fortunately, first and second crack are quite distinguishable on this machine, and audible over the sound of the tumbling beans. I kept a journal of different roast times for quite some time, and it was at this point that the second problem became apparent. The unit is quite susceptible to ambient temperature, so roasting in the garage became a problem (I live in Canada, sometimes roasting at -20 C). The colder the weather, the longer the roast. Later, in moving the unit inside, I came to understand that electrical current may also be a significant factor; the roasting times were noticeably reduced inside, even compared to room temperature outside (19-21 C), possibly the result of line loss when using the unit in the garage (indoors I plugged in next to the main supply box).
Despite roasting at least once a week for three years, I never came to rely on the settings. This unit requires supervision from the first crack onward (usually ~13 mins into the roast). For this reason, some knowledge of roasting is necessary, and it should not be considered an “automatic” unit that takes the skill out of roasting. Keeping the unit clean is important; roasting times increased with dirtiness. Interestingly, when I added the vent hose and fan, the unit stayed quite clean for good periods of time (6 or more roasts; the instructions recommend cleaning after each use, which would add at least 10 minutes of finicky work each time). Likely this was the result of better smoke venting during roasting. A setting of 13 or 14 was usually reliable for espresso roasting Ethiopian Yirg (I did this naively, before learning that single origin espresso is a contentious issue). In addition, I found that it was very difficult to deliver a mild roast—this roaster tends to produce a robust and somewhat smoky (but by no means unpleasant) flavor. Finally, setting the unit at anything less than 12 resulted in under-roasting (i.e., second crack not achieved).
The cooling cycle is a problem. The unit cannot draw enough air to cool the beans fast enough, giving you less control over the roast. Once I knew what I was doing, I would open the lid for cooling (against the advice of the user manual), which worked quite well.
Now for the worst part. I have had three units in three years. The first I returned to the manufacturer because it could not reach second crack for most roasts, even on a setting of 15 (I had to stop and then re-start the roaster, often roasting longer than 30 mins). Initially I was told the unit was to be fixed; later I received a replacement. This unit lasted another year, but again the roasting times increased, despite keeping the unit meticulously clean (this was the “suggested” reason for the initial unit failing; however, being highly OC, I kept it very clean). The problem is rectified somewhat by roasting smaller amounts (e.g., 6 oz). In addition, on the second unit, the lid hinges failed. They are much too flimsy for the weight of the lid, which is quite solid (lined with stainless steel). I did not bother returning this unit, and instead passed it on to a friend after I “inherited” another (identical) Alpenroster from someone who didn’t use theirs. (It was this unit that I moved indoors and attached the dryer vent to). Sadly, this unit recently failed; it shut down suddenly after first crack of a batch of espresso choco, and will not restart/power up. I had been using it approximately one year.
The second unit is officially still working, though the roasting times are longer as noted. I have learned that the thermostat controlling the heating element (a single, wave shaped loop located under the roasting chamber) can be adjusted, though doing this modification will void the warranty. Instructions for this modification can be found on another discussion site. I have since learned that others have found that the roast temperature is quite variable from unit to unit. A review posted on another site indicated that a setting of “1” would have burned the beans, as the second crack was achieved by 12 or 13 mins. Given the thoughtful design of this product, failing to employ a quality element/thermostat seems a substantial and surprising oversight.