Right now I am testing some new green i have received US. Its the El Salvadore La Florida Gourmet and a Colombian microlot (Julio Gomez) Im working on right now.

Some coffees have a wide sweet spot, and some coffees have a very narrow sweet spot - and it can take quite a few roasts before getting promising results. I am wondering to what roast level you roast your new coffees, and what weight loss to expect? When testing new coffees I try to aim for city/city+ roast, and work my way from there. But in what way will weight loss and roast level go hand in hand? I found this when doing a google search:

Roast level Percent Weight Loss:
C = 16%
C+ = 17%
FC =18%
FC+ = 19%
Vienna = 20%

Would you agree on this numbers? My experience is that my roast levels tend to have lower weighloss than the numbers in this article. And how consistent will the relation between roast level and weightloss be? Will some coffees have very different weightloss taken to the same roast level?

I roast more or less exclusively for espresso. If I roasted for brewed coffee preparation I would probably roast some coffees somewhat lighter.

My "average" roast level is to just before the onset of 2nd crack, and very seldom more than a few seconds into 2nd. This is a fairly small range of roast levels that under standard classification would presumably be FC to FC+.

16% is a good average weight loss for this degree of roast. I weigh all of my batches before and after roasting. Given this set of criteria and the degree of roast I shoot for, I have never had a batch that was more than 1% off of 16% roast weight loss, that I liked. So for me, anything from around 15.5% to 16.5% loss is expected (although I generally am more in tighter range, say 15.7 to 16.3%). Straying further from this range usually means that I screwed up the roast in some way, using my equipment and thermometry and typical time - temperature profiles.

ken

I just received a shipment of a Sumatra. As I opened the bag I could tell that this coffee had a very high level of moisture. I have not measured the percentage yet but I still roasted a batch to tried it out. The weight reduction was 20% for a Full city+ roast!

Just posting this to highlight that weight reduction is a function of the original moisture level.

I found the questions and answers with great information. Coffee from the high lands in south america tend to retain most of their weight after slow roasting.

From Artisan-0.5.3 -- Artisan beta roasting software available

    # from RoastMagazine (corrected by substracting 1% based on experience)
    # http://www.roastmagazine.com/resources/Roasting101_Articles/Roast_SeptOct05_LightRoasting.pdf
    def roast_degree(self,percent):
        if percent < 13.5:
            return ""
        elif percent < 14.5:
            return QApplication.translate("Label", "City",None, QApplication.UnicodeUTF8)
        elif percent < 15.5:
            return QApplication.translate("Label", "City+",None, QApplication.UnicodeUTF8)
        elif percent < 16.5:
            return QApplication.translate("Label", "Full City",None, QApplication.UnicodeUTF8)
        elif percent < 17.5:
            return QApplication.translate("Label", "Full City+",None, QApplication.UnicodeUTF8)
        elif percent < 18.5:
            return QApplication.translate("Label", "Light French",None, QApplication.UnicodeUTF8)
        else:
            return QApplication.translate("Label", "French",None, QApplication.UnicodeUTF8)


I just roasted two batches of a Huehuetenango. One was City+ and one was Full City+. The C+ lost 14% and the FC+ lost 16%. My roasting skills are far from professional but this is just my experience. A rule of thumb I concluded was that roasted coffee is about 85% of the weight of green coffee. This has held true for me. Measuring according to this and assuming a typical 14 gm double, I can roast per number of doubles I drink in a week without running out and little amount of leftover beans that will mix into the next batch since I don't usually roast the same origin week after week. I don't worry too much about the actual percent weight loss and I end up enjoying the roasting and drinking just fine. ymmv

I'm on board with the previous comments on percent weight losses (Δ) and their corresponding roast levels. I would reiterate their dependence on the origin and on the water content of the beans. And I would add that Δ for a given roast level of a given bean is also a function of the speed of the roast, with a higher Δ for a slower roast.

Personally, I record and depend on the Δ as an objective measure that is much more sensitive than my eyes, and sometimes even my palate. Given two roasts of the same bean, whose Δ differs by 0.5% (corresponding to roughly 30 seconds of roast time), I often cannot see the difference between them, but I can usually taste it and can always weigh it. The Δ immediately tells me whether I've hit (or missed) my target, without waiting to rest and brew the beans. If the result is important (e.g. an SO Brazil that loses all acidity and turns to cloying root beer and licorice when overshot), and the Δ says I flubbed it, I can redo it right away.

Jeff wrote:From Artisan-0.5.3 -- Artisan beta roasting software available

    # from RoastMagazine (corrected by substracting 1% based on experience)..


I added that code to Artisan when I was still young. Since then I learned that weight loss is very much dominated by moisture level of the greens (usually they are in a certain standard range, but sometimes not). So I am now with Whale on this. There is no direct correlation between weight loss and roast level independent from the roast profile or bean properties. However, if you roast the SAME beans again, weight loss can be used to compare the roasts (to some extend). Having said this, please ignore the roast levels "computed" by Artisan for now. They are usually off.

M.

Its the El Salvadore La Florida Gourmet and a Colombian microlot (Julio Gomez) Im working on right now.

I do recommend testing the moisture of the beans. The two types of green beans I mentioned in my first thread, had very different moisture content. This will affect the weight loss alot. The coffee from El Salvador had a moisture content of 10.4. The Colombian coffee was at 12.4. Its quite obvious that these two coffees roasted to a city roast will end up with different weight loss. I was puzzled when trying to develop a roast profile for the colombian coffee, but when I tested the moisture content it I had to add some time to the drying phase and "voila" - the coffee really showed its true potential.

I have got me this moisture meter:

http://www.coffeelabequipment.com/QCCOFFEEPRO.html

Hello Gismar,
Have you had time to come to any conclusion re. the coffee moisture analyzer and how will you use it to determine profiles for given coffees?
Thank you
JL