www.klatchroasting.com: USBC champion, voted 2009 'best micro-roaster'

Interesting professional roasters' discussion

Discuss roast levels and profiles for espresso, equipment for roasting coffee.

Link to "Interesting professional roasters' discussion"by Marshall on Tue Aug 12, 2008 9:27 pm

You can't post there, but you can read an interesting thread by respected professional roasters answering a professor's questions about roast profiling and the effect in the cup: http://forum.coffeed.com/viewtopic.php?f=16&t=2286.
Marshall
Los Angeles
User avatar
Marshall
 
Posts: 1132
Joined: May 13, 2005
Location: Los Angeles, California

Link to "Interesting professional roasters' discussion"by RapidCoffee on Tue Aug 12, 2008 9:43 pm

Probably the same "Bear" who initiated an interesting thread on roast profiles on CG.
John
User avatar
RapidCoffee
 
Posts: 1928
Joined: Dec 11, 2005
Location: Rapid City, SD
www.chriscoffee.com: quality & service, second to none
www.chriscoffee.com: quality & service, second to none

Link to "Interesting professional roasters' discussion"by Marshall on Tue Aug 12, 2008 9:59 pm

RapidCoffee wrote:Probably the same "Bear" who initiated an interesting thread on roast profiles on CG.

Certainly. The difference is who is answering.
Marshall
Los Angeles
User avatar
Marshall
 
Posts: 1132
Joined: May 13, 2005
Location: Los Angeles, California

Link to "Interesting professional roasters' discussion"by farmroast on Tue Aug 12, 2008 11:16 pm

Marshall
Thanks for the heads up and link, BUT I won't hold my breath waiting for much in details cause it just never seems to happen in public. There have been a few similar threads on RG during the past few years that I have been watching that have abruptly stopped after scratching at the surface along with the general wisdoms. Overall for specialty coffee it would be a good idea but for competitive professionals it doesn't make much sense to do so.
Hoping they prove me wrong,
farm
Ed Bourgeois
LMWDP # 167
http://coffee-roasting.blogspot.com/
It's about the dirt.
User avatar
farmroast
 
Posts: 559
Joined: Jan 01, 2007
Location: Amherst,MA.

Link to "Interesting professional roasters' discussion"by another_jim on Tue Aug 12, 2008 11:54 pm

One of the reasons I left Coffeed -- every topic turns into ten posts by the usual suspects talking up their "craft experience," rather than answering the question, as if it were an advertising bulletin board from the 15th century. By the time that's done, everyone's eyes have glazed over in boredom and anything interesting posted to the thread is lost. If these usual suspects bothered to pick up any of the works on coffee research and perhaps a food science text to help decode what they are reading, they would be able to give a fairly precise answer to the question, since a lot of the required information is well known.

There are three basic things happening when you roast coffee: 1) water evaporates, 2) organic acids and aromatics break down or are boiled off, and 3) water, sugars, and amino acids combine in a chain of chemical reactions collectively called the Maillard reaction.

  • Water Evaporating This happens from the moment the beans are heated up to the first crack, when the remaining free water escapes. As the beans exceed 300F, a steam wave moves out of the bean from its center. This wave also starts the Maillard reactions (see below). Below 300F bean temperature, not much is happening except water evaporating. If the bean is too moist going above this, the bitter organic acids will not break down as quickly (see next section). If too much water evaporates, the Maillard reactions are starved, and the roast will be dominated by dry distillates. The drop-in temperatures or early bean heating should be adjusted to achieve this balance
  • Organic Acids and Aromatics Breaking Down or Boiling Off Roasts have to go to partway into the first crack, since this is where the chlorogenic acids that make unroasted coffee intensely bitter finish their breakdown. The smaller acid molecules and aromatics, which are responsible for the fruity aromas and acidic tastes start breaking down around around here. The longer the beans stay at temperatures above about 390F, the lower the acidity of the result. For instance, it is said that Italian espresso roasters stall the beans around the first crack to reduce acidity without reaching temperatures that caramelize off the sugars. Too much of this is a roasting flaw called baking, which overly flattens the flavor of the coffee.
  • Maillard Reaction Chains This is where the complexity of coffee is created, since these reaction chains are hugely complicated. However, there are some overall guidelines. The early Maillard reaction, from 300F to the 1st crack, creates nutty, toasty, and woody flavors. At higher temperatures, sugars stop reacting with amino acids and start caramelizing on their own, creating caramel, vanilla and chocolate flavors. Finally, also at higher temperatures, and when the water required for early Maillard reactions and caramelization runs low, the Strecker degradation changes the compounds created earlier in the roast to dry distillates: smoky, spicy and peaty flavors. The simplest lesson here is that these processes compete for water, so that taking longer in the ramp to the first will and less time thereafter will favor the woody, toasty, nutty flavors and reduce the caramel ones. Also, if the roast is to get very dark, the only way to avoid overwhelming distillate flavors is to dry the beans less (go as fast as possible earlier in th roast), so the Strecker degradation creating these flavors is controlled.

There is also the inconvenient fact that beans are 3 dimensional objects. Roasts faster than about 6 minutes will have significant differences between the inside and outside of the bean, and the inside can be under roasted; slower roasts will be more even. However, the slower a roast, the more the aromatics are cooked off.

So, after making sure one has dried the bean to the right degee in the roast below 300F, and balanced the flavor developments that takes place below and above 400F, one should roast as fast as possible.
User avatar
another_jim
 
Posts: 4529
Joined: May 05, 2005
Location: Chicago

Link to "Interesting professional roasters' discussion"by farmroast on Wed Aug 13, 2008 1:33 am

What is done or must be considered to bring out desirable nuttiness? Tweaking my 2nd leg has really brought out the sweetness in my roast thanks to the ET thread and your suggestions.
Ed
Ed Bourgeois
LMWDP # 167
http://coffee-roasting.blogspot.com/
It's about the dirt.
User avatar
farmroast
 
Posts: 559
Joined: Jan 01, 2007
Location: Amherst,MA.

Link to "Interesting professional roasters' discussion"by another_jim on Wed Aug 13, 2008 10:58 am

Sweetness and nuttiness are in competition, since the production of nutty flavors requires sugars and amino acids. The faster you go from 300 to first crack, the fewer the nutty flavors, and the sweeter the result. Your best bet is reducing the sub-300 warm up a bit, extending the leg to and through the 1st crack, and speeding up the end final leg. My first try would be keeping the overall length the same, taking a minute off the warm up and end, and adding two in the middle. This is the way I roast the Bolivian Cenaproc; for it, I'm looking for a dominant marzipan flavor, i.e. sweet and almond.
User avatar
another_jim
 
Posts: 4529
Joined: May 05, 2005
Location: Chicago

Link to "Interesting professional roasters' discussion"by farmroast on Wed Aug 13, 2008 11:44 am

Damn, It's all starting to make sense!!! So the first trade off of sacrificing a bit of sweetness before first and a bit of fruitiness after first will gain in the nuttiness(when appropriate with the right bean). Going too long in the mid leg will continue to lose sweetness and head towards toastedness and then to woodiness? Generally seems that Pacamaras have this potential too, being careful because available sugars are less?
Ed
Ed Bourgeois
LMWDP # 167
http://coffee-roasting.blogspot.com/
It's about the dirt.
User avatar
farmroast
 
Posts: 559
Joined: Jan 01, 2007
Location: Amherst,MA.

Link to "Interesting professional roasters' discussion"by another_jim on Wed Aug 13, 2008 2:46 pm

I find Pacamaras are tough to get right. The easiest is just to blast through the stretch to the first crack, go for sweet, and rely on the Marogagype genes to give you some of the characteristic wood in any case. But if you are looking for the signature taste, and the bean is high quality with lots of sugars, you can afford to convert some of those into flavor components -- the pre-first woody ones in this case, with a really light roast ending level to keep the Caturra style acids around. But truth is on my last two Pacamaras, by the time I was halfway happy with the result, I had roasted through the entire stash.
User avatar
another_jim
 
Posts: 4529
Joined: May 05, 2005
Location: Chicago

Link to "Interesting professional roasters' discussion"by DigMe on Thu Aug 14, 2008 12:38 am

Excellent explanation, Jim.

brad
DigMe
 
Posts: 262
Joined: Oct 14, 2006
Location: Waco

Link to "Interesting professional roasters' discussion"by John P on Fri Aug 15, 2008 11:53 pm

Jim,

when roasting Pacamara, charge your drum at 215-225F, and have a more gradual roast--about 6 degrees per minute on your bean temp-- until you hit first crack (375... give or take a few degrees) increase to about 9-10 degrees per minute (bean temp) after hitting first crack... level out your temp after about two minutes ....and you should finish in about 30 seconds to 1 1/2 minutes max after that.

Pacamara and Margogype are not as dense as your typical SHB and if you roast at the same rate as your typical roast the inside of the bean will overheat. This will kill a lot of the flavors.
John Piquet
Salt Lake City, UT
caffedbolla.com
John P
 
Posts: 77
Joined: Jul 14, 2006
Location: Salt Lake City

Link to "Interesting professional roasters' discussion"by another_jim on Sat Aug 16, 2008 3:52 pm

Thanks, an "ultra-drum," slow start, fast finish roast. I'll give this a try.
User avatar
another_jim
 
Posts: 4529
Joined: May 05, 2005
Location: Chicago

Link to "Interesting professional roasters' discussion"by jason_casale on Sat Aug 16, 2008 4:26 pm

Jim,
this is interesting information about half of it I all ready knew.
I understand that polyphenols are breaking down in the stage before 300 as well as moisture.
My profile for the verona espresso that you have tried used this type of profile.

Turn around at 200 degrees 1:30 seconds into roast. 250 at 4 minutes.
300 at 7 to 8 minutes. 1st crack 370 375 at 11 to 12 minutes before reaching 1st crack I dropped the heat to low at 365 about 30 seconds to a minute before 1st crack would begin.
I would leave it on low for another 2 minutes until 1st crack would finish.
This would produce a slow controlled 1st crack that usually would not go above 400.
The gas would be on low a total of 2:30 to 3 minutes a little before 1st crack all the way thru first crack.
This technically would be referring to the Italian style of roast slowing down the roast to stretch out the sugar chain as much as possible.
I have found that most controlled 1st cracks take a total of 2 minutes from start to finish.
If you leave the gas on low past 1st crack you start to lose bean temp heat and stall the roast baking the coffee.
At the end of 1st crack I was some where in the 400 degree range.
14 to 15 minutes in roast at this point.
I would slowly bump the heat in increments 1/4 turn up on the gas wait 1 minute then a quarter turn up on the gas every 30 seconds for not more than 1 1 minute interval and 3 30 second intervals. (Medium to medium high heat on the roaster)
This put me at 425 430 a little past the beginning of second crack roast end.
medium roast dry beans out of the roaster.
Second crack would begin at 415 420 or so.
The reason for the slow bump on the heat at the end was 2 fold for me.
I wanted to mellow out the acidity quit a bit for northern italian espresso.
Second If I ramped up the heat to fast in the end of the roast sometimes I would get scorching tipping.
A note about 1st crack and second crack if your heat is to aggressive in either stage you to have very loud fast hard cracking that can cause what are known as blow holes or divots.
The other reason for the slow controlled heat bump right before second crack.

Most people who tried my verona blend where amazed that such a light roast had that much body and was not astringently sour but very balanced. This was accomplished using the temp drop before 1st crack and thru 1st and controlled temp slow ramp up after 1st thru second to roast end.

I would generally use this same theory for other coffees sometimes just quicker for strictly hard bean. 9 minute 1st crack 10 minute 1st crack for medium soft beans 11 to 12 minute 1st for espresso or soft bean.
generally it would work very well. I read alot about roast techniques from willem boots ruling the roast articles all very good.

Thank you Thank you for putting this information together it will greatly continue to help me as I continue to roast everyday on our companies diedrich ir3 a complete time temp profile roaster which I love.
Thanks Jason
jason_casale
 
Posts: 62
Joined: Oct 19, 2005
Location: Mesa Arizona

Link to "Interesting professional roasters' discussion"by John P on Sat Aug 16, 2008 9:38 pm

Jim,

yes... your breakdown on the roasting process is simple and to the point.
A wealth of information hidden inside there.
John Piquet
Salt Lake City, UT
caffedbolla.com
John P
 
Posts: 77
Joined: Jul 14, 2006
Location: Salt Lake City

Link to "Interesting professional roasters' discussion"by another_jim on Sat Aug 16, 2008 10:59 pm

Thanks, but this information is readily available in the literature, and fairly easily confirmed in its rough outlines.

I don't think there's enough to tell you how to roast a specific bean up front, but it might reduce the number of roasts needed to get to the best profile.
User avatar
another_jim
 
Posts: 4529
Joined: May 05, 2005
Location: Chicago

Link to "Interesting professional roasters' discussion"by Viridian on Tue Aug 19, 2008 12:18 am

I think that it is kind of funny that Bear, in the article referenced in the first post, refers to a set of experiments that Jeffrey Pawlan conducted. I remember that and have written to Jeffrey to ask him about the results of those experiments. He assured me that when he had worked his way through the sheaf of results he would publish his findings.

"Dear Erik,

I carried two huge binders of data weighing about 30lbs around the SCAA
conference showing it to various people. I had done some statistical analysis
and graphing of this. Then I had a couple of meetings with the statistician who
was helping me in his spare time. He wants to change the method of analysis. I
must wait for him to be available. When we are done, it will be published.

Regards,
Jeffrey Pawlan"

http://www.pawlan.com/RoastProfileResearchSession2.html

I don't doubt that he is a very busy man.
Viridian
 
Posts: 6
Joined: Aug 18, 2008
Location: Eugene, Oregon, USA

Link to "Interesting professional roasters' discussion"by Viridian on Tue Aug 19, 2008 5:46 pm

another_jim wrote:Thanks, but this information is readily available in the literature, and fairly easily confirmed in its rough outlines.

I don't think there's enough to tell you how to roast a specific bean up front, but it might reduce the number of roasts needed to get to the best profile.


Hi Jim,
Might I ask what literature? I think it is time I delved into it. Would Illy's book be a good place to start?

Thanks!
Erik
Viridian
 
Posts: 6
Joined: Aug 18, 2008
Location: Eugene, Oregon, USA

Link to "Interesting professional roasters' discussion"by another_jim on Tue Aug 19, 2008 11:02 pm

The chapter in Illy is hard going, but has some good stuff. Clarke and Vitzthum's edited volume explains the 1990s switchover by the large roasters from fast 3 minute to more craft roast like 8 to 10 minute ones. The best intro I've found is Schenker's roasting dissertation. The full dissertation is 26megs, so it's best to save the PDF locally.
User avatar
another_jim
 
Posts: 4529
Joined: May 05, 2005
Location: Chicago

Link to "Interesting professional roasters' discussion"by Viridian on Wed Aug 20, 2008 5:57 pm

Thanks Jim, I'll do my best to wade through it.
Viridian
 
Posts: 6
Joined: Aug 18, 2008
Location: Eugene, Oregon, USA

Link to "Interesting professional roasters' discussion"by Frost on Sat Aug 23, 2008 5:01 pm

Thank you Jim for all the knowledge and information you bring on this subject. I've been working through the Schenker paper. Plenty to sort and study, but I am most perplexed why he choose to do his lab roasts (HTST,LTLT) as Isothermic roasts. I'm guessing he was not a 'home roaster' at the time or he would have developed more 'empirically optimized temperature/time profiles' as is done in commercial/industrial roasting. I think this is unfortunate, but there is still alot of good work here.
Frost
 
Posts: 124
Joined: Aug 23, 2008
Location: Nevada City, Ca

Next

Return to Home Roasting