How to Roast Single Origin Espresso - Page 2

#11: Post by **Ken Fox** » February 18th, 2010, 11:15 pm

another_jim wrote:First: Melange blends only make sense if the two extreme roasts mixed taste better than the intermediate roast. If this is the case, great; but if you want a touch of a great dark roast flavor, why not use 20% of a Sumatra or Sulawesi, where even the cheap ones produce great dark roasts. Guatemala Antigua Bourbons are just about the only bean I'd think would work consistently, year in, year out, as a melange, since they make both lovely dark and ultra light roasts.

Second, great SO coffees are extremely rare; great Brasils are not. All in all, if I ranked all the coffees I drank each year, my "make do" blends would win two years out of three, and there would always be seven to eight in my top ten, even in a super SO year.

Chacun a son goût, or so the expression goes (I prefer a different version I heard recently, e.g. "chacun sa merde," however this is a family website so I'll refrain from any translations

)

The problem with anything like this is that it assumes that everyone has the same taste. Now we know you have good taste, and we know you have spent a lot of time exploring this, but this is, for sure, your taste.

This is why I always try to make it very clear that I am expressing a personal bias, e.g. "I gravitate towards dry processed east African beans," rather than making an assumption that other people share my taste.

I'm about 95% certain that George Howell does not share my taste; after all, he dislikes the natural process beans that are my favorites. I'm unwilling to accept the idea that my taste is crap, but at the same time, given George's dislike of what I like, I'm more than willing to accept that when I express my preferences, they are opinions, an expression of my taste, and probably not a whole lot more than that.

Using wine as an analogy, which is an easy one because the wine newsletter industry is hugely more established and respected than anything in coffee -- I don't share Robert Parker's taste, but at the same time I have an idea of what he likes. Our likes and dislikes are not mutually exclusive. After reading his writings for two decades and tasting many of the same wines as he has, I'm aware there are many points of intersection of our tastes, and when he recommends something that is in an area where I think we have somewhat similar tastes, I stand up and take notice.

ken

February 19th, 2010, 10:03 am

another_jim wrote:First: Melange blends only make sense if the two extreme roasts mixed taste better than the intermediate roast. If this is the case, great; but if you want a touch of a great dark roast flavor, why not use 20% of a Sumatra or Sulawesi, where even the cheap ones produce great dark roasts. Guatemala Antigua Bourbons are just about the only bean I'd think would work consistently, year in, year out, as a melange, since they make both lovely dark and ultra light roasts.

+1 to that. Not only is it unlikely that a single bean will respond equally well at two extremes of the roasting process, but you also have to deal with different sweet spots for grind and temperature. Darker roasts generally require a finer grind and lower temps than lighter roasts. Can you simultaneously optimize flavor extraction of both light and dark roasts of the same bean? Not impossible, but certainly unlikely.

Has anyone ever had great espresso from a melange? I played around with melanges when I started home roasting. IIRC they were OK for non-espresso brewing, but I never had much luck with espresso.

February 19th, 2010, 10:13 am

Two extremes of the roasting process is not at all what I had in mind, nor is it what I've seen pros do when they do it successfully. I'm talking about stopping the roast in one version a minute earlier than the 2nd one. And no, I don't accept that this would result in extraction problems or problems with the use of a grinder. If that was true, then we'd have the same problems with commercial blends, many of which not only have different types of beans in them, but also take those beans to different roast levels, as post-roast blends.

And I'd be astonished if there is a person posting on this thread who has experimented enough with melange blends to make any sort of meaningful comment on their utility, that would be worth more than a guess. In fact, this is a topic that hardly ever gets discussed (at least here) and in private emails with other home roasters I can't recall it coming up at all over the last couple of years.

ken

February 19th, 2010, 10:47 am

Agreed. The smaller the difference in roast, the less these factors will come into play. No roast is perfectly uniform, due to natural variation in the beans.

I look forward to hearing about your upcoming experiments with different roast levels. That will be far more conclusive (although perhaps less entertaining) than arguing about it in the absence of any evidence. I'd offer to participate, but I've pretty much quit home roasting for espresso... partly due to your provocative posts on this subject.

February 19th, 2010, 11:25 am

Hi John,

Of course I don't know whether this approach will work; I haven't done it (intentionally, that is

) in years. A problem with this particular forum is that due to an absence of either professional input, or of very many people who really have confidence in their roasting skills, discussions can be killed by a single dismissive comment even if in reality the (dismissive) comment is a statement of opinion, rather than "provable fact."

So I don't know from personal experience whether this approach has merit, because I haven't tested it since I began to feel comfortable with my roasting skills, and it has not been a topic of frequent conversation. But to me, at least intuitively, I think it merits study. It is certainly a "purer" approach for someone seeking SO espresso, than simply turning an ok but not great SO candidate into a salvage blend with other beans one has lying around.

Of course, that approach, going into "salvage blend" mode also has merit. It just depends on what you want to end up with, and stating opinions as though they were facts does not further our understanding of these options one bit.

ken

February 19th, 2010, 2:38 pm

Seeing this title, I was hoping to gather more information on actual profile considerations to get the most out of a "potential" candidate for an excellent SO Espresso. Let me share some of my recent observations on a journey to "roast at home", some greens...with this pursuit in mind. A fellow HBer and I have been getting together for roasting sessions for over a year now, with still much "work to do". It hasn't been easy as Mike (mivanitsky) lives 150 miles from me and with busy lives, we have to work around schedules.

To further complicate our process...we have changed roasters from electric to propane gas powered. What Mike had produced on his electric HR1, was for many greens, very close to the goal. The limitations we ran into was being able to "ramp up...or down" the heat in a timely fashion during the profile. As any professional chef would not attempt cooking on an electric stove (for the most part...I'm sure some exceptions exist), I decided to get a gas powered roaster, and add higher capacity at the same time. Enough of a background...

To add to the current posts, our recent roasts of a DP Ethiopian Oromia turned out with a better SO espresso from a melange of the two roasts we did that day. It was by accident that one roast had not had the "watlow" controller increased after the drying phase and had an extra minute in the ramp to 1st, both were ended at the same finish temp...I am posting this mostly for "food for thought" in the concept of merely taking one "to a darker roast level" and then blending the two.

I feel that we have come to a point where infinate variables in the roasting process start narrowing down areas that certain varieties of greens need greater attention. Once the equilibrium and drying process is complete it is not just merely timing to first crack that is critical...but how you get there. Not proceeding to quickly or slowly through first, and finishing with attention to the rate of rise in bean mass temp to the desired level of roast.

The content of this thread now makes me wonder if some bean varieties (which have been chosen as good candidates for SO espresso) will require blending at different roast profiles to create a more "balanced" or "complex" result?

February 19th, 2010, 3:39 pm

There's just not much to lose in experimenting with homeroasting. Thankfully they are not business decisions. I learn from my failures and successes. Even when a Pro says that from their experiences they've decided to go a certain way I find I want those experiences too. I like Jim's thoughts of roasting something else to add if a questionable SO comes up short and also relate with Ken's concept of trying to keep an SO pure and maybe a melange might do the trick. When occasions arise I will keep both in mind.

February 19th, 2010, 3:44 pm

Pat & Mike; I'm wondering if your profiles are on the slow side? This may cause a melange of roasts to taste better.

The prime difference between a fast roast and a slow roast is that in the fast roast, the interior of the bean is not as darkly roasted as the outside of the bean. This is because a slow roast is done on a cooler roaster so the roast is even, while a fast roast is done on a hot roaster, so the beans is very done on the outside and not so done on the inside. Think of the difference between grilling and braising meat.

This means the spectrum of flavors is wider on faster roasts. If the roast is too fast and light, the inside of the bean can be grassy while the outside is scorched. This is why even Maxwell House has stopped doing four minute roasts. However, on a gas powered drum roaster, a rolling 2nd crack done in 10 minutes is likely to have both more light roast and more dark roast tastes than a first pops of the second roasted in 15 minutes. Both roasts are within normal parameters, both will grind to roughly the same color, but the fast roast will taste louder.

#19: Post by **Hamilton** » February 19th, 2010, 3:51 pm

another_jim wrote:Another thing to realize is that high end espresso blends are simple. I know the recipe of several high end commercial blends, and they all use 2 to 4 different coffees. the basic idea is to source a sweet, creamy Brazil, and then add it to coffees that would make great SOs if only they were a little more balanced. Think of these coffees as singers who can't quite go a capella, but who do fine with some strings in the background.

This field of "semi-SOs" is a lot wider than the perfect SOs. I order roughly 20 to 30 pounds of good Brasils every year and use them to fill out coffees I tried that are not-quite SO. I roast a wide variety of new coffees, and always roast of Brazil along side. If the coffees are good SOs, I drink them straight; if they need help, they get filled out with the Brasil. If the new coffee makes a great SO or fill out, I order more.

I'm not sure what makes this so complicated or advanced.

That makes a lot of sense to me.
Are there any sort of go-to coffees you fill in espresso blends with? I am about to buy a some more beans from Sweet Marias and I'm not sure what might work with some Brazils for espresso. I think I have heard that some espresso blends use some central Americans along side Brazils in espresso.

February 19th, 2010, 4:03 pm

The Morinhena Formosa Raisin is very nice for this and well named; at a leisurely first pops of the second roasts, it's like adding raisins and brown sugar to the blend.