I just came back from doing a shot of the "waste" (remainings) of the experiment and it is really good for espresso.

This could be my best espresso ever. It has some kind of sweet and strong alcoholic flavor.

Cheers

Some observations about the accidental good light roast.

Loading heavy (270 grams) the roaster has helped maintaining a flat line in the final phase, which makes it easy to do light roasts. Had I used half the weight, BT would had climbed faster, and it would have been harder to control inside the light roast temperatures. With bigger more powerful roasters, I imagine they would have to cut power earlier on in order to coast in the F410s (thermal inertia) or use fancy/complex methods. With the Q3, I've used a heat gun as "elevator", and cut power at the desired spot. This "step" curve makes light roasts very easy to control/duplicate. I will venture into even lighter roasts now that I have tasted blood to explore territory.

With this light roast, I've lost the blue berry flavor that I was getting before with these same beans. Light roasts with "premature" endings had tons of blueberry flavor. Med/dark fast roasts had also a lot of blue berry flavor. What I have learned is not to chase the blue berry! Getting the most blue berry flavor is not near as good as getting the gold flavor. There is a lot of gold and untapped possibilities with light roasts. I believe this to be only applicable to espresso. The same flavor would be impossible in drip and I guess blueberry prevails there.

It is hard to describe the flavor. I tried to look for words but they didn't come out. It was like a liquor desert but not very sweet. The cream was gold light (not orange). I made three shots and the best was the first one. I believe the machine heated all the way and temp went up in the other two (wasn't really paying attention).

That's it.

Keep it up Rafael! I love doing stuff like this. All your friends think you are totally crazy but you know it makes sense. Can't think of a better way to learn than just testing, cupping, testing and adjusting.

I read the entire "Guide to Commercial Roasting . . ." and understood that post better (and missed a whole lot) than this post.

I do a lot of editing and advising in a field about which I'm very familiar, but sometimes reports, briefs, and research just make my head spin. I have to ask what on EARTH is the point here? And then I offer some (borrowed) advice:
1. Explain this as if you were talking to your grandmother.
2. Pretend you just got on the elevator with a legislative aide. You have until the 12th floor to describe your the value of your proposal.
3. Think of the judge as a very bright 10th grader. He or she knows nothing, but can learn. Now explain.

Often these questions help the writer clarify purposes and methods.

Thanks.

The main goal is to identify the best possible flavor (drip coffee) by taking samples from the trier at different times. The samples then are later cupped at another time (or day). The time of each sample is recorded as well as the graph (profile) of the roast. So, when I find the best flavor, I could later reproduce.

Another secondary goal is to learn how the flavor changes. I want to know how the time affects the flavor so that I can get experience both roasting and cupping.

This time I did three tests, one light, one light-medium, and medium. Tomorrow I'll cup them. They all are Ethiopia Amaro Gayo. Just in case, I tried not to make very long finish times so that I could drink the coffee instead of throwing it away (if it were to come out baked).

1 LIGHT ROAST 272 grams
Informal observation during spresso shot: Drinkable as espresso without need for aging. The fresher the better. No blueberry. Liquorish taste.





2 LIGHT-MED ROAST 263 grams
Informal observation during espresso shot: Acidic = good for drip, much more acidity than previous.





3 MED ROAST 284 grams
Informal observation during espresso shot: acidic = good for drip. Blue berry, winey.





Cheers

Informal observation; looking at the pictures of those beans, they appear to be yet another so-so mixed bag lot of Dry Process Ethiopia. These varied beans make for a bit of surprise in every cup. (even with a consistent roast practice) Why not use a zero defect wet process bean for your experiments?

Arpi wrote:....................... The cupping trays are made of aluminum (safer for brewing?).......................

I would not brew in any metal container (accepting a stainless steel portafilter basket)
Only use glass or ceramic.

Thanks Frost.

Well, my "experiments" are based on my usage. I am looking for ways to improve my current roasts. If I used a different bean then it would not apply to these beans. But I'll look into it.

I've heard about aluminum but then I see cooking pots and stuff. Some thermos are metal too. Is your recommendation based on food poison?

Cheers

Here are the results!

The test was successful this time in terms of differentiating the cups. It was a lot easier to taste once the cups got cold (from initial 212F). The flavors popped out and there were big differences. There were no defects. I modified the grind setting to 7, which in a previous test, gave me the most acidic flavor without encroaching in bitter territory. The blue berry muffin flavor was prominent during grinding. I also use a slightly higher dose to brew ratio this time.

There were noticeable differences between the different roasts. However, the differences inside each roast were more subtle (but still different). It is difficult/impossible to tell which one is best unless I were looking for something in particular a priory. There was a huge range of flavors.

Inside each roast, I could taste a difference in acidity, especially from first sample to last. There were differences in the "flavor spectrum". The last cups (sample at the end of each roast) had the lowest frequency, meanwhile the first cups were the brightest. Depending on your mood, you could like the higher spectrum better, or it could go the other way.

Ethiopia Amaro Gayo Roast Sample flavor test

First batch
147+125 = 272 grams

Event N Time BT

1 14:10 402.4
2 14:55 398.3
3 15:35 398.6
4 16:15 400.2
5 16:55 403.0


Second batch
141+122 = 263 grams

Event N Time BT

1 15:19 408.2
2 16:02 403.2
3 16:39 402.7
4 17:09 404.2
5 final roast sample

Third batch
140+144 = 284 grams

Event N Time BT

1 11:25 432.8
2 12:24 444.1
3 12:59 431.7
4 13:34 423.4
5 14:09 420.5

Conclusion:

I wasn't able to tell which one was best. But this test helped me get familiar with flavor changes during the final phase of the roast. Of the three roast styles, personally, I would pick the first, followed by the third. The second roast had the most "premature" flavor. While it had the most blueberry, the other premature accompanying flavors brought down the overall appealingness. The second roast was my past idea of how light roasts always taste. The second and third roast were more boring. Inside the first roast (remember this is for drip), I would pick the first (14:10) for its most peculiar/distinctive/attractive taste in drip. But I know this is a subjective maybe temporal opinion.



Cheers

That is interesting and I suspected you will post something like that.

I noticed that majority of my roasts when cupped taste good. Some are really good, but most are good and I would be happy drinking them all. Best I can say is that they are all different but all tasty.

However that changes dramatically when they are made as espresso. Some are just not good at all, even some of the darker ones...