It happens to all of us from time to time. I made the same mistake with a roast the other day. One of the hardest things to predict is just how much you can slow down development without stalling it out.

So, I hesitate to post this, as I really don't want to see this thread end up in another Rao related flame war. But hopefully we can stick to substance here, and keep the personalities out of it. [Fingers firmly crossed.] If people show up here and start slinging mud, I'm going to delete this post, and I won't post about my other experiments with Rao-style profiles.

As I mentioned in one of the locked threads, I thought Rao's advice was interesting, but I was surprised that he neglected to discuss the importance of paying attention to the ratio between "drying" (meaning the time to yellowing, or 300Fish for those of use who can't see into our roasting chambers) and "ramp," meaning the time from yellowing to FC-start. Ever since I was read the Chris Schooley posts about the effects of stretching out the drying phase, I have generally tried to make sure that, at a bare minimum, I spend more time in drying than in ramp. As a shorthand, the drying/ramp ratio should always be >= 1.

Rao instructs that one should maintain a "constantly declining ROR," which he labels as his "Second Commandment" for roasting. He offers, as primary evidence for this claim, a correlation he has observed between flat RORs and "flat" flavors such as "paper, cardboard, dry cereal, or straw." If you read this in isolation and don't pay attention to the drying/ramp ratio, it would be easy to end up starting quite fast, as this makes it easiest to execute the declining ROR in a reasonable overall roast time.

So I played around a bit with roasts in which the ROR declined constantly. In a subsequent post I will do a side beside comparison between a "Rao style" city roast and a "steady ROR w/ sharp bend" style roast, which has given me good results with some coffees in the past. But for now, I just wanted to share the results of two roasts, both of which follow the Rao Commandments, but which taste very different in the cup. FYI, the cupping comparison between these two roasts was performed blindly, but with me as the only taster. YMMV.

---

Roasting Info - Profile #1:

Bean: Kenya Kirinyaga Kiangoi AA
Roaster: HG/BM
Charge Mass: 200g
Charge Temp: 350F
Dry/Ramp/Development: 3:30/4/2:30
Finish Temp: 405F
Overall Roast Time: 10:02
Moisture Loss: 14.5%

Profile Plot:



Cupping Notes:

Rest: 4 days
Brewer: Cupping bowl
Grinder: Lido 2
Water: 150 ml, 200F
Coffee: 9g

Overall Impression: Sadly, my detailed cupping notes were deleted, so I have to go off of memory and the score that was saved in my cupping app. This roast had a pleasant, complex aroma, but tasted muted in the cup. It had plenty of sweet chocolates and a nice amount of body, but only muted, non-specific acidity when warm. As it cooled, the acidity resolved a bit more clearly into citrus. The finish was drying and short. I scored it at 86.5 overall.

---

Roasting Info - Profile #2:

Bean: Kenya Kirinyaga Kiangoi AA
Roaster: HG/BM
Charge Mass: 200g
Charge Temp: 325F
Dry/Ramp/Development: 4/3/2:15
Finish Temp: 408F
Overall Roast Time: 9:17
Moisture Loss: 14%

Profile Plot:



Cupping Notes:

Rest: 4 days
Brewer: Cupping bowl
Grinder: Lido 2
Water: 150 ml, 200F
Coffee: 9g

Overall Impression: Once again, detailed tasting notes were deleted. (I'm still figuring out the mechanics of the Catador app.) This cup also had a rich fragrance and wet aroma, but also retained the pleasing acidity I know this bean is capable of. I got the rich tangerine notes that I think of as the Kiangoi signature, along with some pleasant spices and florals. It had slightly less body, but a sweeter, more lingering finish. I gave it a 91.5 overall.

---

TL;DR:

These two roasts used identical charge weights, and had close (although not identical) overall roast times and finish temperatures. Both followed all three of Rao's "commandments," but the first used a higher charge temperature, and had a slightly inverted dry/ramp ratio of .875. The second roast was charged 25F cooler, and had a more typical dry/ramp ratio of 1.33. The difference was dramatic, as I discussed above, with the slower start and faster ramp bringing out more of the coffee's acidity and subtler aromatic notes.

So in short, you can get very pleasant light roasts according to Rao's Commandments, but you shouldn't throw out the whole HB playbook just yet. You can end up with some pretty bad results if you start your roast too fast in an attempt to achieve his declining ROR recommendation.

Soon to come: A side-by-side comparison of flat vs. decelerating ROR profiles in a city roast.

[creative nickname] wrote:Soon to come: A side-by-side comparison of flat vs. decelerating ROR profiles in a city roast.

Looking forward to it.

Does Roastmaster have a way to show RoR? It'd be nice to see it more clearly than can be discerned by looking at the BT only.

Roastmaster will show ROR of a measured curve as text at the top of the screen while roasting if you highlight the curve, but I haven't found a way to make it draw real-time derivative curves.

Thanks Mark. Very helpful. Although, the challenge for me now is to try to reproduce.
Dick

So this is just a quick update; I'll try to post more detailed notes when I can, but unfortunately the semester has started and I'm a bit swamped. I did get a chance to do some blind comparisons between the "slow start, fast finish" profile I described earlier (Roast and Learn Together - August 2014) and the two "Rao-style, declining ROR" roasts I described just above this post. I did a blind tasting last night, and another one tonight. Each time, the slow-start, fast-finish profile (which did not follow Rao's Second Commandment to steadily reduce ROR through a roast) beat the others on overall points. It was notable as having the most sweetness and the brightest, juiciest acidity, along with enough body to hold everything together.

I hope everyone will take this in the context in which I am sharing it. I do not mean to imply that you cannot get very good roasts following Rao's protocols. Nor do I mean to deny that for many coffees, in the kind of roasters he works with, his suggested method might be better than the one I describe above. And finally, some of this is dependent on individual preferences (personally, I prize acidity and sweetness over most other elements in brewed coffee). But I do wish to note that one can execute steady-rise, short-finish profiles (breaking two of Rao's commandments) without getting any of the taste flaws he warns of, at least with the right coffees. In particular, Rao warns that constant ROR profiles may "destroy sweetness," or create "'flat' flavors reminiscent of paper, cardboard, dry cereal, or straw." At least with this coffee, in my roaster, rather the opposite was the case.

All of which is to say that this roasting sh** is complex, and for every commandment, there are probably some interesting exceptions out there.