noah wrote:
Also, can a roast go too quickly through the 300-400 stage?

Yes, you can BURN the beans during this phase. I've had a couple of batches of beans where I really pushed the flame height on my drum during this phase, and the coffee tasted "burned." Jim S. happened to visit me recently when I produced such a result (with a batch of Wondo Bonko). We learn from our successes as well as our failures, if we pay attention. And it wasn't obvious to me that this was the cause of the burned batch, until I gave it a lot more thought during a subsequent roast session. Ever since I have been more careful not to overdo it during this phase, just to get to first crack at some arbitrary number of minutes. I'd rather be a minute "late" if I get behind, then to push the heat up so much and so fast (just to make the time landmarks I shoot for), since being "late" might flatten the coffee out a bit, but it won't make it undrinkable. Burning it will make you toss the coffee in the trash, unless you are like a few friends of mine, who can and do drink anything . . . .

noah wrote:While the exact progression of compounds may not be able to be mapped, cant we generalize along the lines of, for example, from 400-430, you have the best chance of maximizing acidic fruitiness, from 440 on, chocolates, etc. The general principle of 300-400 gives bready, nutty, and 400 to second crack gives chocolates is well known. But cant we further break this down, or is it so bean dependent that it is a fools errand?

First, you'd have to decide on whose "400-430" you were talking about, since any of these arbitrary numbers will mean very different things depending on your roasting equipment and where your temperature sensors are located. I think it is possible to think too much about this sort of thing. Most people will learn more about roasting if they keep things relatively simple. Initially, it is most important to learn how to identify the cracks, and to become comfortable with the changing appearances of coffee beans as they roast. As one becomes more accustomed to using a particular piece of roasting equipment, one will learn to correlate the measured temps (wherever they are being measured) with what is going on in the beans in the roast cycle. At this point, your "400-430" will hopefully mean something to you, even if they are just abstract numbers to me. Finally, you will achieve some sort of routine (to which Jim alludes in his own personal roasting approach) with which you are comfortable. You can then take that routine and apply it to different beans.

As an example, I roasted 3 coffees this afternoon, a total of 6 batches starting out with a pound of green for each. The coffees I roasted were Ethiopian Worka, Ethiopian Wondo Bonko, and a Bolivian Cenaproc I bought from the Coop 6 months ago. All the roasts took approximately the same total time (roughly 15 minutes), however the Cenaproc was roasted much darker, in the same total time. I know from experience with my roaster and where I have my probe situated in the drum, that 2nd crack will start within 1/2 a degree F from 435F. I roasted the two Ethiopians to 433, and the Cenaproc to 439. I did this because I have roasted these coffees before and have learned from prior experience with them. Although I generally don't like to roast into 2nd crack, I have learned that the Cenaproc is simply unsuitable as a single origin for espresso at a lighter roast level -- it needs 20-30 seconds in 2nd crack to be usable. Even then, if I had it to do over again, I'd probably not have bought the coffee, but it does serve a purpose as a counterpoint to all the Africans I drink most of the time.

To get back to your second paragraph that I quoted, you could try to apply the same logic to choosing wines, but I think you will find it similarly useless, from a practical standpoint. You would probably be much better off finding some flavors that you like in coffee, then trying to achieve them by the beans you choose to purchase, and the results you can easily get by varying simple parameters such as overall roast time, length of time between the cracks, and the final temperature (point in the roast cycle) that you roast to. Molecular approaches, although intellectually satisfying, seldom will get you anywhere with coffee, whereas relentless experimentation with simple stuff will usually point you in the right direction.

ken

another_jim wrote:If one wants to try to get a general idea of, say, the effect of a slow/medium/fast profile for drying, first ramp and second ramp, that's nine roasts right there. Then add light, medium, and dark roasts, along with a appropriate variety of coffees, and you are in the order of comparing around 100 roasts.

Thats why I ask all of my questions. Hoping that some of these comparisons have already been tested and general notions established.

Ken Fox wrote:Molecular approaches, although intellectually satisfying, seldom will get you anywhere with coffee, whereas relentless experimentation with simple stuff will usually point you in the right direction.

Yes. I think that you hint to an issue that pops up in people (like myself admittedly) trying to make their coffee taste like something it simply wont. Also, almost all of the beans I roast end up serving double duty, being used for espresso and brewing. I am beginning to discover, albeit stubbornly, that I simply cannot have a one roast fits both profile (at least not for most beans).

Thanks for your help guys!

Most of my roasts, roasted to roughly 45 agtron in 12-14 minutes, reach the first crack at about 360 degrees and are dropped between 397 and 403 degrees. Four to five minute drying phase, four minutes between the first and second crack. I only mention these number to point out how wildly bean temperature varies depending on the placement of the bean probe. I'd be cautious about putting too much emphasis on bean temp without strongly considering the other variables.

Fullsack wrote:Most of my roasts, roasted to roughly 45 agtron in 12-14 minutes, reach the first crack at about 360 degrees and are dropped between 397 and 403 degrees. Four to five minute drying phase, four minutes between the first and second crack. I only mention these number to point out how wildly bean temperature varies depending on the placement of the bean probe. I'd be cautious putting too much emphasis on bean temp without strongly considering the other variables.

Measurements like this are extremely useful within the confines of your own roaster in your own location if you have the probe located in a logical and consistent place. This is a highly qualified sentence and bears rereading a few times to understand that all of this is relative to a particular situation. The way that I roast, I hit first crack around 423F. This is obviously a very fallacious reading, but that is what my digital thermometer reads when I hit first crack, and is due to the type of probe I'm using and where it is mounted, plus the fact that at this point in the roast, the bean temp is obviously very different than the probe temp. At this probe temp I cut the heat input dramatically and over the next 4-5 minutes the beans "catch up" to where the probe is, and by the end of the roast I believe the probe is measuring something very close to actual bean temp.

This level of certainty (whether or not I am right) is based upon more than 5 years experience with my setup, correlating what is going on in the beans (by sight and sound) with the probe, and taking into consideration huge changes I've made in my roasting approach over this same period of time. If someone else were to play around with my own equipment and input heat at a different rate at different times, then all bets would be off and I'd no longer have confidence in what I was measuring unless I had a chance to watch this person several times using this equipment with which I am very familiar.

Most of us do not have "dream" equipment, and although what I have approaches this more than what most home roasters have, my roaster still has significant limitations one has to overcome in order to use it to best effect. So none of this is all that easy, but if you pay attention to what you are doing, you will ultimately get the hang of it and will learn how to adjust those things you can adjust, in order to get the results that you hope to get (or at least close, I hope).

ken

another_jim wrote:Stalling the roast in the mid-300s is a way for baking out the acids and creating predominantly toasty flavors. By flattening the acids out, the brew will taste relatively sweeter.

It is vague and highly general, also it uses robusta beans, but I noticed this reference in Schenker:

from Schenker's "Invenstigations wrote:Coffee samples were roasted on a laboratory scale roaster at different temperatures for a constant length of time of 5 min. resulting in products of various degrees of roast. The authors stated from cup testing that aroma formation starts around 170C, when a peanut-like roast note can be percieved. At 180-190C coffee-like flavor arose, whereas the "real" flavor of roasted coffee only appeared at 220-230C. After passing this point, the flavor was judged to be slightly over-roasted (240C) and typically over-roasted (250-260C).

While this certainly does not validate "whiskers" comments, it is interesting to note that the nut like flavor was targeted in this example to have developed in the period prior to 350F. I only post this, because, prior to all the feedback, after reading his post, I was intrigued and tried an adapted profile which would have reflected the principles said poster was advocating. I used Mexico Organic Chiapas Proish Coop greens from Sweet Marias, which has been excessively nutty in the roasts I did where I allowed an even ramp taking four minutes from 300-400F. I tried a profile where, after I hit 300, I flew to 350 as fast as I could (under 1 minute) and then took 2.5 minutes of a slow ramp to 375, then cranked it to hit first crack in another 30 seconds. The results surprised me: The nutty flavors were almost unnoticeable. This is not to say much of anything as this coffee is not one with which to do a lot of experimenting with as its flavors are so mild and subtle, so I cannot say that this profile had a positive or negative effect on the total profile at this time, but the muted nuttiness is a surprise.

noah wrote:I was intrigued and tried an adapted profile which would have reflected the principles said poster was advocating. I used Mexico Organic Chiapas Proish Coop greens from Sweet Marias, which has been excessively nutty in the roasts I did where I allowed an even ramp taking four minutes from 300-400F. I tried a profile where, after I hit 300, I flew to 350 as fast as I could (under 1 minute) and then took 2.5 minutes of a slow ramp to 375, then cranked it to hit first crack in another 30 seconds. The results surprised me: The nutty flavors were almost unnoticeable. This is not to say much of anything as this coffee is not one with which to do a lot of experimenting with as its flavors are so mild and subtle, so I cannot say that this profile had a positive or negative effect on the total profile at this time, but the muted nuttiness is a surprise.

Just a general thought worth considering.

Although I am no fan of the in-your-face blends produced by the marquee roasters, most of which are designed to be dosed in the range of 18-23g, I'd be the first to acknowledge that these same roasters are sourcing some of the best single origins on the planet, either directly or in cooperative buying groups.

If you look at what these top roasters are doing to make a name for themselves, you will see that they are not doing it with their roasting prowess. What I mean by this is that although they are competent roasters, their claims to fame is that they are sourcing the top coffees and using their skills at blending in order to produce roasted coffee that is regarded by most as among the best in the world.

You do occasionally read promotional materials from roasters who claim to be god's gift to roasting, employing meaningless terminology such as "slow roasted," whatever that means. Generally, the roasters who tout their roasting skills are unknown and are not regarded as being among the leaders in the industry. It is not uncommon to hear fans of Italian roasters say that what separates the Italians from everyone else is their roasting skills. These same people are unable to define what they mean by this, implying that it is some sort of secret process that is handed down through the DNA. They are also unable to explain why it is that these same Italian roasters buy some of the cheapest and most uninteresting coffees on the planet, in what appears to most impartial observers as being financially motivated. If these Italian roasters have an exceptional skill, it is that they can buy very cheap coffees and combine them in a way that the end product is drinkable. Notice that I said "drinkable," not "exceptional."

Roasting properly is a skill and it does take time to develop it. There is nothing secret about the process however, and the difficulty of doing it well would be of the same degree of difficulty as learning to bake bread.

There is a reason for the emphasis on the quality of the raw materials. That reason is that roasted coffee is at its core an agricultural product in the same sense that wine is. When it comes to assessing the percentage of end product quality that comes from the various stages of its development, the overwhelmingly most important factor is the bean quality in the case of coffee beans, and the grape quality in the case of the wine. The winemaker can make things worse, and maybe in marginal situations can improve things a tiny bit, but the care and feeding of the grapevines and the selection of the grapes on the sorting table will determine 90%+ of the quality of the finished product (e.g., the wine).

It is very much the same with coffee. By the time the coffee ends up in the 132lb sack headed towards the roasting plant, most of the potential end product quality has already been determined. Yes, you can surely diminish what came to you in the form of green beans, but your options for actually improving upon them will be fairly limited. If you look at what the high end roasters are doing, they are not doing very much special in the roasting plant, and they are concentrating their efforts on visiting the various origins in order to develop relationships with the best bean growers.

If what you want to do is to improve the quality of the coffee you drink, then of course it will help to learn the basics of roasting, especially what not to do that can destroy the value in the green beans you have purchased. Most of the improvement that you can hope to get in the cup will come from buying the best beans, rather than from modifying various roasting parameters in an attempt to improve the potential that is inherent in the beans themselves.

ken

It is always nice to have things put into a greater perspective. Ken, I wonder, what is your opinion on the on-line green bean marketplace in general for home-roasters? Do we have access to the same caliber of beans that the esteemed roasters do? (care to do any endorsements?)

Without middle-men or large sourcing outfits to send us samples to roast, cup, and then buy, we are at a distinct disadvantage in that all we can do is buy, then try. So in this sense, the ability to "tweak" roasts, and highlight, and or diminish certain aspects of beans is naturally of greater practical importance to us. We have already spent the money and are stuck with the beans. If I had 20 different roasted samples in front of me at a cupping table, all roasted with a basically generic profile, and a handful were outstanding, it only makes sense that I would buy those beans and "just roast" them. But, necessity breed ingenuity, as they say, and much like the historical development of all cuisine, the advancements come from trying to stretch what you have, or improve that which is not quite to ones liking. That the bigger roasters have the luxury of not needed to worry about such annoyances is very much to our advantage in that it empowers us to be the innovators, which I think has already occurred to a large extent. But it just seems a bit too easy on ourselves to be content to stop at such vagueries when there are so many of us doing our own roasting and such a good forum of information exchange like this forum for data collection, eventually from which new principles and generalities can be extrapolated.