keno wrote:
Here are my observations and a few questions:

• I found it interesting that no mention at all was made in the book about airflow and the effect that this can have on heat transfer. This seems particularly strange to me given that he roasts primarily on a Loring roaster which is a recirculation type roaster that involves more airflow. Perhaps he has not varied the airflow to eliminate it as a parameter for his roasting/cupping tests but it is still an important factor that seems like it should be discussed.

The Loring has its fan linked to the burner, meaning you can't turn the heat off and increase the fan speed. It doesn't specifically mean "more" airflow, because you could easily take an entirely different roaster and crank up its blower and perhaps have higher airspeeds. But it is a unique roaster in that it's essentially all convective, with the only significant heat transfer via radiation and conduction happening from bean strike. He's roasted extensively on all their models, being the company's certified trainer, but I'm not sure he could easily translate airflow from those systems to those that don't act the same. He does however happen to have the same (but older model) sample 1 kilo gas roaster as I do, which he does a lot of testing and sampling off of.

keno wrote:

• For his roasting experiments all differences are given in time differences from basleline but no absolute times are given for baseline. I realize this will vary from roaster to roaster but this makes it hard for me to interpret and apply his principles. I have no ideas whether my times are on the low or high side of his baseline. Nor are any overall roast times provided. I'd like to see a chart with the actual profiles for each of the roasts he tested.

I think the gist of his "manifesto" was to get an idea out and develop it. I hope he develops this process described, myself. It's a data hungry topic, and I take my hat off to anyone who can organize it and still keep it easy to understand. He'll no doubt take input like yours and others so he can expand on the topic he's trying to teach. He'll either have a blog or website that will bear updates of the above.

keno wrote:

• Are his principles at odds with Rao's? A constantly declining RoR would seem to favor a shorter ramp time and longer development time compared to a longer ramp and moderate development time. But then again it's hard to say because Hoos does not provide the curves or times.

Good question. I think he's found his own way that works for him and has worked well on his blind cupping analysis. I personally don't jump on bandwagons enthusiastically. Whether Rao and Hoos agree (and I'm sure they do on a lot) might be immaterial if the individual roaster applying one or the others techniques don't happen to give them the results their seeking.

. . . and we can all afford a Loring . . . .
I kinda don't see how his (well done and documented) findings apply to those of us with perhaps entirely different heat transfer systems. I'm still learning the capabilities and limitations of my Diedrich. And Diedrich does encourage a longer ramp time, but as to airflow, it seems these are totally different animals .

Received my signed copy, and spent couple hours for a good read. I really like the concepts of "modulating of flavors" in this book, and it is a reminiscent of Jim's Espresso 101: How to Adjust Dose and Grind Setting by Taste . It provides a guideline to fine-tune flavors by adjusting roasting profiles in a relative term.

Don't worry about the Loring. I believe he should have tested his observations in various drum roasters and the general trends should hold true. Granted there are physical limitations of various types of roasters, but luckily we home roasters will need to master only one machine, which is the one sitting in our home.

Received my 97 out of 100 copy today...Reading now.

More like professional "continuing education," than a beginner's tutorial or a recipe book. The more you know going in, the more "baselines" with which you're familiar, the more you'll get out of it. That adjusting Ramp and Development in ten second units gives predictable results provides the roaster a lot of power to plan the roast in a universe where "roasting on purpose" is everything.

This is a plot of a roast I did today. It's the first time I tweaked a profile ala Hoos, a profile I'd been using for a specific bean, Nicaragua Los Congos Pacamara Lot B, successfully. by increasing Ramp, Development and total roast time. I also went for a lower drop temp, because this roast and its "control" companion roast are intended for brew only and I wanted to see what if Hoos's suggestions could help me heighten the desirable fruits while maintaining or improving body.

(By the way, every micro-lot of the Los Congos Pacamaras, purchased from Bodhi Leaf, I've roasted gave 90+ results.

Anyway:



Notes:

• I've been roasting the Los Congos in my usual medium-slow/fast/medium profile; with a 6:15 Drying; 3:45 Ramp; and 2:45 Development to 420F, for an early FC finish;
• Hoos defined EOD -- Yellowing/Hay -- occurred at 5:00, 280F. I noted EOD on the plot at 300F out of habit;
• Inteval targets are within 6sec of the proposed profile. Proposed total time was 12:20. Drop temp was 3F less than proposed, because I knew when 25% hit and panicked;
• Development RoR slowed too quickly and fought me all the way to Drop. Had it cooperated it would have had a steadily decreasing RoR. Despite the long Development time, finish (Drop temp, aroma, and appearance) is C+. Like old age, Hoos's standard of +/- 10sec is not for sissies;
• In case it needs saying, Roast 1 this morning was the usual profile, but with a similar C+ finish (drop 416F) and finally,
• We'll see. The proof of the profile is in the cupping. Sadly, I don't think I can get to it til Sunday.

Rob Hoo's blog should help flesh out the book, making it less of a lecture and more of a practicum. I'm looking forward to it.

Rich

Dan and I are very excited to invite Robert Hoos to this thread to directly participate with the forum members, to answer questions about the book, to expound on whatever he'd like involving roasting. Feel free to add to the dialog with feedback, questions or comments in general. Let's all be welcoming and open to some good solid discussions!

Robert, I'll start by asking you to share with us what your thoughts on what translating thermometry and data across a wide audience is like. I know you roast on smaller (1-2 pound) sample gas roasters and all the Loring's. Since it wasn't directly mentioned in your book, it's an easy question to start with.

Have at it

Hi Everyone!

Dan, Tom, thank you both for giving me the opportunity to geek out with everyone on Home Barista about coffee and coffee roasting. And to everyone on Home Barista, thank you for the chance to have a conversation with you about roasting! I have been looking forward to talking about these things with all of you and hope that I can contribute as much as I am sure I can learn from your community.

I will have to be a little shorter tonight but will gladly expound on it tomorrow (I'm already winding down for the evening). I will gladly start by answering the question posed by Tom and hope to provide insights to questions already posed in this thread.

Thanks again for the opportunity and I will chat with you all tomorrow morning PST!

Rob

Good morning,

All right, let's dive right in.

what your thoughts on what translating thermometry and data across a wide audience is like. I know you roast on smaller (1-2 pound) sample gas roasters and all the Loring's.

The fact that I typically don't talk in terms of temperature is no accident.As many of you on HB are aware, different types, thicknesses, placements of thermocouples will read incredibly differently even in the same roasting system, let alone in a different roasting machine. We find ourselves in the situation where talking numbers doesn't really get us anywhere except into a mess of more confusion.

What I have noticed though, both in doing this project and in just working in the industry is that, regardless of the type of machine, and to a great extent the type of heat application, the coffee is going through the same set of physical and chemical reactions.

So, for example, when I am playing with product development on a coffee on my small Ambex-style sample roaster and I need to translate it to the Loring, do so by noting the times of physical changes and chemical changes, and then using my familiarity with where they tend to happen on the Loring, I create a roast plan to match and follow it. The trickiest part (and the part that requires a lot of work to match) is the matching of the final temperature. Since thermocouples don't match up, I keep some of the beans from the other roast near-by and then match the final expansion and color of the coffee. When doing something like this I will then cup the two looking for differences in order to dial it in even more (especially with regard to its drop temperature). Frequently I will then go to my friend's lab and use their Color Track to see how close my senses were at evaluating.

All of this has led me to the conclusion that roast profile is king.

The length of time that we allow coffee to chemically react in each phase, the over-all length of the roast, the final degree of caramelization/pyrolysis, all of these things are what make the coffee taste the way it does. To give one final example to this point, I often am asked to match someone's current profile on my Loring (I like to jokingly call it "wire-tapping"). They send me some of their green, as detailed of profile data from their roaster as they can, and a roasted sample from their roaster. By applying what we just talked about I have been able to get industry professionals (from some highly respected companies) to not be able to tell a difference on a blind cupping table.

So, basically that way a long, and rambly way to say that for me, even working on different equipment (a Quest, Probatino, etc. etc.) you can achieve similar results (if not the same results... I just can't claim same because I don't have a mass spec) if you match the over-all time and duration for physical and chemical changes in the roast.

Hopefully a little later this morning I can get to some of the questions listed prior and keep the conversation going. But for now, time to go to work . Have a wonderful morning. Also, hopefully this all makes sense...tired brain this morning.

Rob Hoos

robhoos wrote: The length of time that we allow coffee to chemically react in each phase, the over-all length of the roast, the final degree of caramelization/pyrolysis, all of these things are what make the coffee taste the way it does.

Welcome.

Do you mind if I ask what you see as the different phases, and what characteristics mark each one?

keno wrote:The gist of can be summed up, I think, fairly simply (like Rao's book in terms of his three commandments) in the following two principles:

• Longer ramp time from yellow to FC (or MAI time as Hoos calls it) promotes body and complexity

• Development is a balancing act, too little development leads to bitter and vegetal flavors, while too much development time leads to bland flavor with low acidity

There are a few other points that are elaborated on but these are the two that seem to be firmly based on his roasting experiments and cupping tests in which MAI time and development time are varied with a few different coffees.

Hi Ken,

I wanted to take a few minutes to talk with you on HB about your questions. But I also noticed that you live in Portland OR. If you have the chance you should totally stop by the roastery and come talk to me for a bit. If I haven't met you then I would greatly enjoy that, and if I have I am sorry I don't remember it (I'm much better with faces than names). Bottom line is come by M-F 7am-3:30pm and lets have a coffee together!

Your summary of the book is a decent one. My largest goal in developing the book was to try and break down the roast into measurable segments that could be independently affected and adapted to create nuanced (and sometimes blatant) differences in the way the coffee tasted. When going through this process I basically ended up with 5 zones of control that I could change relatively independently in order to dial a coffee in to the flavor profile I was hoping to get. 1.) Time from the beginning of roast to the beginning of chemical reaction (ie color change), 2.) From the beginning of chemical reaction to the beginning of first crack, 3.) From the beginning of first crack to the end of the roast, 4.) The entire length of the roast, 5.) The degree of caramelization and pyrolysis. Then (through research and experimentation) I determined how changing those one way or another (as independently as possible) affected the cup. The ramp from yellow to fc, and the time from fc to drop were documented with specific experiments, and the others are referred to through less experiments (though they are pooled from my roasting and cupping experiences as well, just not documented in the same way).

• I found it interesting that no mention at all was made in the book about airflow and the effect that this can have on heat transfer. This seems particularly strange to me given that he roasts primarily on a Loring roaster which is a recirculation type roaster that involves more airflow. Perhaps he has not varied the airflow to eliminate it as a parameter for his roasting/cupping tests but it is still an important factor that seems like it should be discussed.

Tom's response is correct. Because of the nature of the Loring, airflow and burner are necessarily tied together (but that is a discussion for another time ). The reason I didn't talk to much about air-flow isn't because I don't vary it when given the opportunity (in fact, the test roasts done on the Ambex-like sample roaster I have I use the airflow control in every roast... and in other experiments not noted in the book I tried to see how it affected flavor); rather it is because I wanted to give the industry and home roasters a book with general principles that apply across all roaster manufacturers. Some people don't have air-flow controls, but they can still control their profile. In roaster's that I have used that utilize separate air-flow controls, their affect has more to do with heat transfer and profile shape than it does the flavor of cup, and at the end of the day, flavor attributes that come from different air-flows can be explained through profile changes they cause in my opinion. I know this may be a controversial stance, but in my experience roasting on tons of different systems, and doing flavor profile matching across those systems, airflow affects heat transfer which affects profile which affects flavor.

• For his roasting experiments all differences are given in time differences from basleline but no absolute times are given for baseline. I realize this will vary from roaster to roaster but this makes it hard for me to interpret and apply his principles. I have no ideas whether my times are on the low or high side of his baseline. Nor are any overall roast times provided. I'd like to see a chart with the actual profiles for each of the roasts he tested.

This is a fair criticism. The reason I gave no times, only variations, is because I don't want to tell anyone what the "best way to roast" is, or "how a coffee will taste its best." When it comes to taste, there is (in general) no right or wrong, just preference. I hoped to show people how to take where they are at and wiggle their profile into something they would like better. My concern is that people will take what I have and just do that instead of making something that is truly their own. So, for the good folks here at HB, here is a general light roast Brazil profile... take it as you will.
Beginning 360 F 50.2lb Charge, Turn Around 145F @ 0:45, Pale white 300F @ 4:00, Yellowing and Hay-like 320F @ 5:00, Brown and Bready 340F @ 6:00, FC 398F @ 9:20, drop 414F @ 11:40. Though I have the data on those roasts specifically, I think this should get you a general ball-park idea of where I am coming from.

• Regarding his cupping methodology he states that the tests were blind which is good, but I believe they are based on his taste testing only. It seems like it would help greatly to replicate these findings with others.

You are correct, the data presented in the book is only based of my cupping data. Though I have reproduced this with various people in the industry and found that we are in agreement. Since I am presenting it as "one roaster's manifesto" I only including my perspective. I can tell you more about cupping and the industry in person too .

• Are his principles at odds with Rao's? A constantly declining RoR would seem to favor a shorter ramp time and longer development time compared to a longer ramp and moderate development time. But then again it's hard to say because Hoos does not provide the curves or times.

I don't consider myself at odd with Mr. Rao, I've never met the man (though I have heard nothing but nice things and would love to chat with him). But I have read his book and am familiar with the concepts presented there-in. I think what he is trying to do with the constantly declining RoR is not a bad thing, but I also think that it is somewhat dependent on the type of thermocouple and equipment you are working on. For example, if you have a really sensitive thermocouple, you will read endothermic flash (during first crack all the water vapor coming out into the environment) will cause the thermocouple to read a much lower RoR than the bean mass is experiencing. So while I can agree than in general my RoR is typically declining, I would not say that it is prescriptive. Additionally, there have been some coffees where I can say my RoR has not always have been declining because of an intentional choice on my part with the flavor development, and they have been awesome.

Overall, I found the book a worthwhile, if quick read for $25, but I'm still not sure yet how to put this into practice, except perhaps to try slowing down my ramp time. I always learn a lot from everyone on HB so I'm looking foward to following the discussion.

I am glad you found it to be worthwhile! One of my greatest concerns when putting this book together was simply that it would be helpful and enjoyable to people, so I am glad that it was even somewhat helpful. Additionally come on down to Nossa sometime and let's nerd out on coffee together. I hope my responses to your questions were somewhat helpful. Additionally, I deeply appreciate how thorough you were in responding to the book.

Until we meet in person,

Rob Hoos