HomeRoast Digest


Topic: My fastest roast ever (14 msgs / 525 lines)
1) From: Ken Mary
Since I appear to be the only person interested in fast roasting, I 
considered not posting this info. But this report may inspire others to
investigate. My intent is not to see how fast beans will turn brown, but how
quickly a roast can be completed while maintaining a "good" cup.
A brief review of where I came from: I started out slow profile roasting
about 3 1/2 years ago. While the coffee was "good", it was unexciting. I too
frequently found a flavor that was unpleasantly tangy or sour. I found that
speeding up the period just before first crack removed the tanginess and
drastically improved the cup. So logic forced me to speed up the entire
roast curve and again found a huge improvement. Luck came upon me in the
form a of a popper that, unmodified, completed a roast in 3 minutes. That
single roast was the best tasting coffee to that point in time. But my
inexperience (and ruining a good Kenya) made me pursue slower profiles
again.
I recently completed a new fast roaster. The results are so good that I have
now shelved all of my slow roast equipment including the mini-drum.
Following excellent roasts of Sumatra and a few others including Nicaragua
Pacamara, I threw all caution to the winds. I chose Ethiopian Harar decaf
and it finished in 115 seconds (yes, that is less than 2 minutes) to city+.
My best roast ever of any decaf. I even broke my rule of slow roasts for
Ethiopian beans. There is a lot more work to be done. There is also the
potential to use this roaster's fast heatup combined with a slower more
controlled finish. These fast roasts are uneven, bean to bean, but only 10%
of sampled beans show nonuniformity from surface to center. Part of the
reason for the good results may be a melange effect. So far, I have not
found a "peak" in the curve of roast time vs cup quality. The faster the
roast, when finished before divots occur and quickly cooled, the better the
cup.
I had previously just about given up on trying new roast methods. I was
happy to just have freshly roasted coffee, and too busy with summer
recreation to bother with this list or a.c. Now I feel renewed. Especially
gratifying to me are the new list members doing work with poppers, and
discussion getting back to roasting.
--

2) From: miKe mcKoffee
Wow, 115sec roast! IRRC that's in line with what some big commercial air
roasters do for maximum batch per hour yield. Actually I found your post
very interesting. Enough so I'm going to do some roast experimenting. I'm
thinking going to 1st onset as fast as possible and slamming the breaks for
my usual 3min development finish. (not stalling but continued gradual temp
increase from 1st to end of roast) I "may" also try a totally as fast as
possible roast shooting for 2min. That'll really be interesting and
challenging to accomplish, the fastest I've done in the past was a 5min
total time to French (trying to emulate a stock FR with my Rosto setup)
Kona Konnaisseur miKe mcKoffee
URL to Rosto mods, FrankenFormer etc.http://mdmint.home.comcast.net/coffee/Rosto_mod.htm

3) From: Lesley Albjerg
Ken,
 
I used to fast roast.  I have not done it in years.  I think it may be time to revisit.  If memory serves me right, my quickest roasts were about 2.5 min.  Are you willing to share your technique?  I could really see an advantage for decafs, it that they have been abused by the process of getting the caf. out.
 
Les
 
Ken Mary  wrote:
Since I appear to be the only person interested in fast roasting, I 
considered not posting this info. But this report may inspire others to
investigate. My intent is not to see how fast beans will turn brown, but how
quickly a roast can be completed while maintaining a "good" cup.
A brief review of where I came from: I started out slow profile roasting
about 3 1/2 years ago. While the coffee was "good", it was unexciting. I too
frequently found a flavor that was unpleasantly tangy or sour. I found that
speeding up the period just before first crack removed the tanginess and
drastically improved the cup. So logic forced me to speed up the entire
roast curve and again found a huge improvement. Luck came upon me in the
form a of a popper that, unmodified, completed a roast in 3 minutes. That
single roast was the best tasting coffee to that point in time. But my
inexperience (and ruining a good Kenya) made me pursue slower profiles
again.
I recently completed a new fast roaster. The results are so good that I have
now shelved all of my slow roast equipment including the mini-drum.
Following excellent roasts of Sumatra and a few others including Nicaragua
Pacamara, I threw all caution to the winds. I chose Ethiopian Harar decaf
and it finished in 115 seconds (yes, that is less than 2 minutes) to city+.
My best roast ever of any decaf. I even broke my rule of slow roasts for
Ethiopian beans. There is a lot more work to be done. There is also the
potential to use this roaster's fast heatup combined with a slower more
controlled finish. These fast roasts are uneven, bean to bean, but only 10%
of sampled beans show nonuniformity from surface to center. Part of the
reason for the good results may be a melange effect. So far, I have not
found a "peak" in the curve of roast time vs cup quality. The faster the
roast, when finished before divots occur and quickly cooled, the better the
cup.
I had previously just about given up on trying new roast methods. I was
happy to just have freshly roasted coffee, and too busy with summer
recreation to bother with this list or a.c. Now I feel renewed. Especially
gratifying to me are the new list members doing work with poppers, and
discussion getting back to roasting.
--

4) From: Edward Spiegel
At 10:12 AM -0400 7/23/04, Ken Mary wrote:
<Snip>
Hi Ken,
Great food for thought. A few questions. Do you tend to prefer roasts in City to Full City range? Have you had satisfying dark roasts with the fast method?
My reason for experimenting with slowing down my roasts (my initial roasts in the Toastmaster were about 3 - 4 minutes) is that my preferences run to Full City+ or Vienna roasts, and I was finding that with super fast roasts that the roast could quickly get away from me.
After doing some informal cuppings with a friend, I am interested in expanding my tasting palette. So, I will try some fast roasting to City+ and explore that whole taste sensation.
Your comment about stopping the roast before the divots form, makes me wonder if  there isn't a point where the heat build-up in the beans is so much that they essentially skip through several stages of 'development' (for lack of a better word) if one takes a fast  roast too far. But if one stops short of that point (for Cith to City+++) perhaps that rate of heat build-up is just right. (Or maybe I am just a theorizing fool).
Would that be consistent with your experience?
What popper do you use? Any mods? Any techniques other than dropping the beans in the hopper and letting it rip?
Thx,
Edward

5) From: Bob Yellin
Hi Ken,
<Snip>
What did you do to "complete" the roaster?
Can you provide some temperature/time data?
<Snip>
cooled, the better the
<Snip>
Have you changed your mind about (extra) slow, controlled cooling? I =
recall
doing a test per your suggestion back in February, and finding that slow =
cooling
improved the results in the cup. Have you abandoned that?
Interesting that you should bring up this topic at his time. I have =
recently
found that certain beans do indeed benefit from fast roasting, especially
certain high grown, hard beans. Particularly when they are roasted fast =
before
1st crack. This idea is hardly original with me. I had read about it in a=
 Willem
Boot article in Roast Magazine and decided to check it out for myself. I =
have
found a significant difference in quality (much better) for certain beans=
 when
they are roasted faster than what is considered "normal".
I went back a couple of years and looked at my notes and compared roast =
times to
cup quality and found that there is a definite correlation but that it is=
 also
very bean-dependent. However, I have never tried 2-minute roasts. But, to=
 check
this out, I took one particular bean and roasted it numerous times to the=
 same
roast level, for various lengths of time, starting at 4:00 minutes and =
taking it
through 14:00 minutes. I found (for this bean type) that the most =
enjoyable cup
was produced by a 7-minute roast. That shattered the 12 to 13-minute =
"holy
grail" for me, but I'm still in the midst of these trials for other beans=
 so I
didn't want to come here and state that "the king has no clothes" just =
yet. For
that bean, 4 minutes did not produce a tastier cup. The bean did not have=
 time
to develop it's fruity content even though it was evenly roasted and at =
the same
roast level as the 7-minute roast. It wasn't bad but it wasn't as good.
But as I said, I haven't tried "ultra-fast" so I'd be curious if you have
temp/time info. Maybe the "melange effect" is at play here. And thanks =
for
bringing this up!
Bob Yellin

6) From: HeatGunRoast
Heatgun roasting allows very fast roasts, especially in the 3-6 oz batch range.  Of
course, these are not the lightning-fast roasts reported here, but they can come in
at close to 4 minutes.  I don't like roasting that quick.  Makes me too tense. 
#:o)From time to time, I've experimented with overall roast times and with fast
ramps for one or another phase of the roast. I've only reported these results in the
most general terms because I'm not confident of my cupping "accuracy."  I await
others (well, a few others) who will give fast heatgun roasts and variable ramps a
try. A short(2.5 minutes) ramp to first produces, IMO, "no difference" from 6-8
minute first for most single varieties.  Since I'm usually roasting a pre-blend, I
stay with the longer period to get beans even before first.  I notice a consistently
less developed cup if the time spent between mid-first c. and start of 2nd falls
below 3 minutes.  
Martin
=====
Martin
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!http://promotions.yahoo.com/new_mail

7) From: Craig Wichner
Hi Bob,
I've been roasting and lurking for a few months, and have enjoyed this forum
very much.  There are some great characters here, and the wisdom you all
have shared has put me years ahead on my learning curve. 
<Snip>
beans 
<Snip>
I am still experimenting with beans & roasting profiles in the 5 - 10 min
range, and am very curious about which specific beans you found worked well
on faster roasts, and which ones you enjoyed best slow roasted.  
<Snip>
I'm also interested in more of your thoughts around the correlation
(processing method, etc?).
Thanks!
Craig Wichner
(I-Roasting in SF)

8) From: Bob Yellin
<Snip>
forum
<Snip>
Hi Craig,
I hope you take advantage of, and enjoy the rich experience you can have =
in this
group now that you've "come out". ;-)
<Snip>
certain
<Snip>
min
<Snip>
well
<Snip>
Well, as I said, I'm still in the process of gathering info on my roasts =
so I
don't want to state facts until I'm sure of them myself, and even then, =
what I
find will probably only pertain to my specific roaster, conditions, etc.
But, for example, Ethiopian Yirgacheffe benefited greatly from a short =
(7-9
minutes) total roast time. I get it to 1st crack in about 4 minutes and =
then
spend another 3 taking it to a state where, by it's looks, smell, and
temperature, it's finished. The bean, at that point is medium brown, and =
looks
very slightly mottled or wrinkled with a strip of white chaff still in =
the
center fold of the bean. At that point, it's most aromatic and has the =
most
varietal character in the cup without being too light-bodied and sour. It=
 loses
aromatics and develops chocolaty notes as the days progress until, after =
a
couple of weeks it's about dead. But what a joy until then! My beans =
don't
usually last that long. 
I'd have to write a book to give you all the beans I've tried because =
it's not
enough to just give the bean name - you would need to know the profile =
and the
rest of the conditions as well. And finally these beans were roasted on a=
 PID'd
Sirocco roaster, a roaster which was discontinued in the 80's. It may be =
the
same for all air roasters but I don't know. I can tell you for sure, that=
 when I
roast Yirg or any other bean on my Hottop, for example, the results are
altogether different and I then have to begin again, trying to optimize =
for that
roaster. It's the same with my other roasters as well (p1s and I-Roast).
<Snip>
bean-dependent.
<Snip>
There's no real "process". For a couple of years now, for nearly every =
roast, I
have taken notes about various time/temperature "landmarks" such as when =
the
bean turns the color of peanuts, when first crack begins, when second =
crack
begins, at what time/temperature I stop the roast, etc. By the way, the =
point at
which the roast is terminated is critical and I do it manually by judging
appearance, smelling the roast and watching the bean temperature. Then I =
take
careful cupping notes on many of these roasts and note which ones gave =
the most
excellent (or very poor) results in the cup.
So lately I have looked over my records for the past couple of years and =
listed
all the roasts and the ending times and looked for correlations between =
cup
quality and times (and found them). Now I'm going back to some of these =
beans
and verifying the length of time to get the best quality. That's it. Hope=
 that's
what you were asking.
Bob
<Snip>
recall
<Snip>
recently
<Snip>
especially
<Snip>
 it
<Snip>
for
<Snip>
for
<Snip>
"normal".
<Snip>
that
<Snip>
roasts.
<Snip>
numerous
<Snip>
4:00
<Snip>
type)
<Snip>
shattered
<Snip>
these
<Snip>
a
<Snip>
even
<Snip>
have
<Snip>
for
<Snip>
unsvbscribes) go to =http://sweetmarias.com/maillistinfo.html#personalsettings

9) From: Ken Mary
<Snip>
I have never tried a heat gun roast, but that may be a solution to
super-fast roasting provided the mixing can keep up with the heat transfer.
<Snip>
consistently
<Snip>
falls
<Snip>
I do not blend beans. But I imagine fast roasting will do poorly on
pre-blends designed for espresso.
--

10) From: Ken Mary
<Snip>
This is an interesting point, I find some "spiciness" in the fast roasts
quickly cooled, but less than that provided by a 4 to 5 minute profile with
slow cooling. The fast roasts are quickly cooled so they do not run away. It
should be possible to quench a fast roast, then cool slowly, but this may
take experimentation to get it right.
<Snip>
same
<Snip>
I should say that the fast roast cup profile is different from slower, there
is much more tip of the tongue impact from the brightness. And at the city
or city+ stage, there seems to be more varietal character than the same
stage of longer profiles. The fruitiness seems less developed or masked. The
one Yirg roast so far did not taste overly acidy or citrusy, but it may have
gone closer to full city. Body seems less developed. But I prefer the
overall cup profile of the fastest roasts for brewed coffee.
I still have different beans to try. My previous thinking was that these
fast roasts were best on well prepped Colombians, and not recommended for
Yemens or Ethiopians. This could be the high vs low grown effect, or the
single estate vs pooled lot effect.
--

11) From: Ken Mary
<Snip>
method?
With such fast roasts, anything beyond the first few snaps of second gives a
burned taste. Even at the first snap of second, most beans begin to have the
same taste, since the varietal character is reduced.
<Snip>
You may want to try the coast method. At the correct time determined by
experiment, usually after first but before second starts, shut off the
roaster and let the residual heat carry the roast into second, then begin
cooling. I have allowed a few to proceed a full 2 minutes into second with
no burned taste. This method seems to optimize the fruity and caramelly
flavors.
--

12) From: Ken Mary
<Snip>
It is fairly easy to do in a lightweight popper with little roaster mass to
heat up, and about 70 grams of beans. But the heater may have to be modded
to reach 1400 to 1500 watts.
In many of my longer profiles, I found a better result by preheating the
popper. And as you say, the fast heatup can be part of a slower profile.
--

13) From: Ken Mary
<Snip>
the caf. out.
I use a lightweight popper with a shortened Presto heater giving about 1400
watts. The fan is driven with a 25v transformer full speed with no dimmer.
Bean weight is about 70 grams or 100cc. The roast is stopped using a 50 or
55 second time after the very first pop. This usually gives a city+ finish.
The first divot used to be my heat off signal, but too many roasts were
overdone for my taste.
--

14) From: Edward Spiegel
At 9:19 AM -0400 7/24/04, Ken Mary wrote:
<Snip>
When you say that "Even at the first snap of second, most beans begin to have the same taste, since the varietal character is reduced" I am assuming that you are talking about with fast roasting. With slower roasts I have found that there is a lot of varietal character there when the roasts are done well even at Vienna. That is why I have been putting so much time into getting my 'slow down techniques' together. For me  (I realize that this is simply a subjective matter of preference), there is a marriage of body and varietal flavor that I have tasted in some full city to vienna roasts that are (to my taste) my preference. Beyond that, I would agree that varietal characteristics disappear.
That's my .02
--Edward


HomeRoast Digest