There are several threads about breaking down the roasting process into three phases/ramps (drying, ramp to 1C, ramp to finish). Riffing on the "3 phases" tune, I have recently started experimenting with lengthening/shortening of the amount of time spent going from room temperature to ~250F.

Prior to this experiment, I've consistently used a 400F pre-heat as measured at BT probe. While consistent, my temp readouts have not been accurate until I stole a page out of seedling's book and relocated my probe. In my last 5 roasts, I've been able to much more strongly correlate yellowing to ~300F and 1C starting ~400F; this leads me to believe that I'm heading in the right direction. While the roasts taste about right to my palate, I've noticed some tipping, and have reduced my pre-heat to 300F to see if that makes a difference.

Now that I've got more accurate & consistent measurements, I'm wondering what effect, if any, there is in lengthening the period of time to go from room temp to 250F. I've noticed that, when I turn up the heat to full power in an effort to reduce the total amount of time spent going from 60F to 300F, I blow right past 250F and don't settle into a slower, more controllable rise until I'm closer to 280F.

I just finished a couple of Full City roasts, both Brazil Poco de Caldas, with the only modification on my end being the "full power into drying" (approx. 90 sec) vs. "slow into drying" (approx. 3 minutes), and trying to keep everything else constant (3 min from 250-300F, 3 min from 300-400F/1C, and 3 min from 400F-420F end of roast)to end. Tomorrow is cupping, and I'll see if I can tell the difference. If not, I'll try a couple more roasts, dragging out the "slow" to ~ 5 minutes.

Has anyone else already tried this, and if you have, am I on a fool's errand?

Curiously,
-s.

Hey, Sherman. Sounds like you're lowering your drop-in temperature, then increasing the environment temp after drop in. Remember that your thermometer can absorb heat faster than the beans, which is why your "bean temp" can rocket to 250F without much change in bean color. I'm sure there are calculations with 1500W of heat at 73% efficient transfer ratio and specific gravity of coffee of 495... (made all the numbers up), but you get what I mean.

You could end up with beans that look done from the outside, but are underroasted inside - and the grinds will be lighter than the bean color.

One idea is to take a few different coffees and:
1) measure exact amounts
2) preheat to X temperature
3) load beans and give full heat
4) note when beans yellow

From that you can back off the heat during drying until you reach the desired heat input + no scorching + even doneness.

At some point we can calculate all the fun right out of this hobby! Sort of like testing the oil density of a bowling lane and measuring the amount of material absent from chips and scrapes on a bowling pin, then calculating the ball rpm and vector for a strike.

CHAD

Chad,

Thanks for the feedback.

My thermometer setup is a sheathed J thermocouple, placed very similarly to yours. You're right that I'm lowering drop-in, but it's that as well as being more mindful of heat input and limiting maximum ET to <450F to see if it has any effect on tipping.

My point about 'rocketing' was an observation that, given a constant heat input during the 'pre-drying' or 'drying' phase, the BT measured a jump from 250F to 280F much faster than it would if I modulated heat input.

I think that I understand your point about heat absorption with regard to the bean vs. temp probe; that implies that my BT sensor is more reflective of the localized environment in the bean mass area, rather than the temperature of the bean itself... right?

That said, maybe it doesn't matter. I don't know, though, and that's why I'm asking these questions. If you're telling me that it doesn't matter, then I'll take it into consideration and be mindful of your comments when I'm reviewing my data to see if it corroborates.

I'm still wondering if it's worth breaking down the 'drying' phase to segments (i.e. room temp to 250F, and 250F to 300F). If it takes 2 minutes to get from room to 250F, then another 3 minutes to go from 250F to 300F, is the duration of the drying phase 3 minutes, or 5 minutes? At the same time, this may influence the other ramps. I don't know... yet.

Again, I'm hoping that someone who has either the knowledge or the similar experience to chime in and provide their input.

Thanks,
-s.

At the risk of too much information, this research by Swiss Federal Institute of Technology and Nestle, showed that the rate of increase in heat does indeed have an effect on final flavor profile. The experiment was carried out with a lab fluid bed roaster and Probat at different roast profiles. Specifically, on page 9, it was stated:

"The results obtained from this study showed that attaining the same coffee bean color (which is most frequently referred to as the degree of roast) using different time-temperature conditions during roasting
does not necessarily mean that coffees are equivalent in terms of aroma and physical properties. High speed roasting with high hot air temperature led to different formation and elimination kinetics and in many cases to different concentrations of aroma compounds at roast color L* ) 21 (Table 4)."

The link to the original article is here:

http://pubs.acs.org/doi/pdf/10.1021/jf800327j

Stainless steel sleeve is a better conductor of heat than a coffee bean so it will be more responsive than the insides of the bean.

Tipping with the breadmaker roaster? The hopper in my breadmaker is pretty thin and I don't think it will retain heat enough to scorch the beans. If you're experiencing a condition similar to tipping it would almost have to be from the heatgun temp set too high... It can be over 1000F on high... probably above 600F at 5" away where the beans are.

I don't use full high heat at all during roasting. Start at 60% drying, go to 80% until 1C, then back off to 50~60% to finish (all approx based on where the dial is set).

To the point, there is some minimum time that a coffee bean can go from 65F to 300F with a given heat source - and that time will vary with each roaster. If you sacrifice some roasts on High from the beginning, you'll get some idea of how fast your setup *can* heat the beans. I can't comment as to the difference between heating the beans to 250F in 3 minutes, then taking 2 minutes to get to 300F vs. heating the beans to 300F in 5 minutes via another ramp.

Is there much going on in the bean below 250F besides getting the water inside up to boiling?

CHAD

The results of the first cupping were inconclusive. Unfortunately, I didn't log any of the roast data, so I've chalked it up to a learning experience. On to the next set:

More Poco de Caldas. 440g loads, ~350F BT charge temp. Both roasts finished at ~440F at the beginning of 2C, after outliers but before a full rolling crack. ET readings are a bit off, as the probe is attached to the lid and I have to peek in occasionally to see what's going on .

Roast 1 is "full power" - that is, heat on the highest setting up to 1C. The remaining ramps are as fast as I dare while trying to keep equal time:


Roast 2 is slower. Full power to 250F, then slowing ramp to hit 300F in 3 minutes. 1C and final ramps stretched to maintain some semblance of equal time in each ramp:


Cupping Results (so far...)
Both roasts were cupped after 1 day's rest, and pulled as 14.5g normales on my Rituale after 2 and 3 days' rest. The cupping results surprised me. Tasting while warm, I had expected to prefer the faster drying, but found it one-dimensional. The slower drying roast had wonderful floral notes and more complexity. After cooling, these preferences held; the faster roast turned a little bitter, but the slower roast had begun to taste sweeter.

As espresso, the slower roast definitely won out - the fast roast was a sharp, acidic pop that faded quickly, where the slower roast was a mellower, creamy milk chocolate.

I don't think that this batch has helped much in terms of determining whether stretching pre-drying has an impact as much as it has given me a solid profile against which to test future batches. The next round will have to wait until after Easter.

Lots to think about, many taste buds to abuse.

-s.

Four to five minutes drying is what I use on my air roaster, and I'm finding I need a few minutes more on the drum (that is as soon as I figure out how to slow down the drying and still "loading up" heat for a fast ramp to the first). If the HG/BM setup is intermediate in terms of convection, it would probably be best at 5 to 6 minutes.

On pre-drying. It takes 24 hours to completely dry green beans in a 220F oven, but the green coffee smells which develops around 275 to 325F in a standard roast disappears in the drying oven after about 2 hours. I'm not sure spending 1 to 2 hours to pre-dry to this no green smell level is a useful idea. Of course, if you get some sort of god-roast by doing this, useful doesn't count.