[Homeroast] Roasting based on crack sounds

Jon Rosen jon at jonro.com
Mon Jun 16 13:03:38 CDT 2014

This has been frustrating to me as an amateur home roaster (I've been home roasting for six or seven years). I also started with an iRoast and have been using a Behmor for the last 2-3 years. It's difficult for me to tell when the various roast stages begin and end. 

I usually buy five pounds of a coffee and hope to "get it just right" by the second or third try. I keep a log of the coffees I roast and when I get a new coffee from a region in my list (i.e., a new Guatemala or an Ethiopian), I start with the last roasting profile I used for that country. This works pretty well, but it isn't always exactly right.

Home roasting will remain a niche activity until there is an easy, reliable way to dial in a roast. Based on previous engineering work, I'm pretty sure something could be designed with a combination of visual and auditory analysis of the beans. Perhaps temperature and spectroscopy would need to be used, too. Anyway, I look forward to that day. I think I could do it, but it would probably take the better part of six months to design.


On Jun 16, 2014, at 1:49 PM, Mike Davis <mldavis2 at sbcglobal.net> wrote:

> I start with a disclaimer that I am a relative newcomer to the home roasting scene, having started a few years back with an iRoast2 and recently upgrading to a Behmor.  My experience with both types of roasters has revealed some variables that make roasting more of an art than a science.  Here are some observations from past years of roasting small batches (1/3 to 1/2 lb.).
> The sound of 1C varies with the bean origin.  I suspect the moisture content and bean size is also a factor, with (for example) peaberry reaching 1C a bit sooner than normal beans.  Interesting that the paper was written by the Acoustical Society of America rather than experienced professional roasters.
> Following a back to back roasting of the Ethiopian Shakiso with the Rwanda Karenge, using the same 1# P1 setting with a 1:30 preheat and a 30 second door crack to extend 1C, the 1C of the Ethiopian was more gradual in starting, lasted a bit longer and seemed to transition smoothly into the beginning of 2C.  The Rwanda had a sharp 1C onset, lasted about 1:30 and ended abruptly.  2C was almost inaudible or at least undetectible to my ear aside from a few whimpering little pops.  Additionally, these origins are often roasted a bit differently with the Ethiopians a slight bit lighter. Most of the time, 1C is obvious.  What is less obvious with some beans is the transition from 1C to 2C, and /or actually hearing 2C begin.  Most of us don't roast through to the end of 2C, especially with Africans, so the onset is important, not only in being able to hear it, but in deciding where it started.  15-30 seconds one way or the other makes a huge difference.
> So there are some variables to consider which, for those of us who roast different batches constantly must learn to recognize and control.  I don't think you can write a set of guidelines that will reliably hit the sweet spot based on time, on temperature or on sound.  If there are guidelines, they are just that - guidelines - a starting point of departure in the quest for the perfect roast.
> The quest continues, and every little bit helps.  Interesting article.
> Mike Davis
> _______________________________________________
> Homeroast mailing list
> Homeroast at lists.sweetmariascoffee.com
> http://lists.sweetmariascoffee.com/mailman/listinfo/homeroast_lists.sweetmariascoffee.com
> <a href="http://www.sweetmariascoffee.com/forum/">Sweet Maria's Forum</a>
> <a href="http://www.sweetmarias.com/library">Our new Coffee Library</a>

More information about the Homeroast mailing list