Roasting Profiles for Espresso?

[miKe mcKoffee]

Generally speaking my espresso profile targeted for straight shots runs 2 to 4 minutes slower than various SO roast profiles for "coffee". (Even though coffee usually means espresso brewed Americano or cappuccino.) Currently I've stuck with pre-roast espresso blending rather than more advanced and or time consuming multiple roast multiple profile post roast blending.

The espresso roast profile I've been using for a while now is:
200f @ 1 minute (Initially hit the heater hard to 150f @ ~25sec then back way off)
250f 2
275f 3
300f 4
320f 5 with continued straight 20f/min ramp to
400f 9 minutes (my usual start of 1st point, sometimes a few earlier)
410f 10 with continued straight 10f/min ramp to
440f 13 then slowing ramp
445f 14
450f 15 end of roast ~FC touch of 2nd, very few initial droplets on very few beans (with pre-roast blend)
and in roaster cooling
200f 2 minutes
100f 4 minutes

At one time used a 30 second to 1 minute final temp hold without slowed final ramp before ending the roast as I saw Stumptown do with their Hairbender roasting. Tried the very slow final 2 minute stage and liked the results just as well so have stuck with it. Though if my final stage runs a bit fast I'll still add a bit of a final temp hold before cooling.

This is Caffe' Rosto roasting with my desired very slow bean movement rather than fast like typical fluid bed roasting. Temps measured with temp probe about 1/3 down into bean depth green about middle as they expand, 1/2# batches.

FWIW this profile is very similar to the one I developed to tame the acrid tendency of Kenya Mika noticed by Jim during that particular 4 way roast exchange we did back Summer '03. Those roast exchanges we did were fun and great learning experiences!

[mattg]

What type of profiles are you trying to achieve by dropping the temperature toward the end? My attempts at running a lower temperatures for the final 3-4 minutes have ended up with a profile that muted most of the flavors and tasted almost baked.

[IMAWriter]

Generally speaking my espresso profile targeted for straight shots runs 2 to 4 minutes slower than various SO roast profiles for "coffee". (Even though coffee usually means espresso brewed Americano or cappuccino.) Currently I've stuck with pre-roast espresso blending rather than more advanced and or time consuming multiple roast multiple profile post roast blending.

and in roaster cooling
200f 2 minutes
100f 4 minutes

[snipped]

hi...just wondering if that cool down cycle of yours is a bit long...I roast via CO/UFO combo...I dump into colander, and have room temp beans in less than 2 minutes...for a 1 lb roast....
I commend you on your attention to your roast...I would give this hobby up if i had to go through as many temperature machinations as you!
I've read the Rosto is a fine way to roast....almost bought one, but I go through 1.5 #'s of espresso and 2#'s of vac/drip coffee in a week in my home/studio...thus the combo serves my needs...
Back on topic...I bet Jim could give us a hint as to which beans...say A Daterra Brazil, Harar (dry processed), Monsooned Malabar, Yemen, Sulawesi, etc could be roasted in combo!...what say you, Mr. Jim?

[another_jim]

I'm sort of fixed on

0 - 1 min to 250F (basically top speed)
1 - 4 min to 300F (drying phase, an absolute must)
4 - 7 min to 380F (top speed again, the longer you're here, the worse it tastes)
7 - 11/12 min to the end, 11 for brewing, 12 for espresso (balance fast vividness and slower sweetness)

But I'm now experimenting with an extra step, slowing down to 10f per minute through the first crack, and speeding to 15f/min afterwards. I'm hoping this will produce a better tradeoff of vivid and sweet. It'll take a few months before I can decide.

[Rainman]

7 - 11/12 min to the end, 11 for brewing, 12 for espresso (balance fast vividness and slower sweetness)

Jim- what temp do you shoot for here?

[another_jim]

Jim- what temp do you shoot for here?

Depends. 425F for cupping (5F over the end of the 1st crack); 440F for espresso tasting (5F before the first pops of the second); then to whatever roast level I think will work for that bean and prep method.

As a general note: these profiles are machine specific and have to be calibrated to each machine's crack ranges as measured on each machine's thermometry. My crack range is: first crack from 395F to 420F, second from 445F to 455F.

[miKe mcKoffee]

But I'm now experimenting with an extra step, slowing down to 10f per minute through the first crack, and speeding to 15f/min afterwards. I'm hoping this will produce a better tradeoff of vivid and sweet. It'll take a few months before I can decide.

Lately I've been thinking along similar lines. Currently getting great sweetness but perhaps lacking in acidity definition. Ideal might be doing two separate roasts of my straight shot blends, one faster one slower like with your excellent WTC blend and roast of a few years ago.

Alternately try reducing my final stage by a minute speeding up the end. Thinking of comparing three finishes: One as is profile above, another going from 440 to 450f end of roast in 1 min instead of 2 and a third going from 430f to 450f 1min instead of current total 3 minute. I'm going to do this in the next couple days and send 1/4# samples of each to Abe to compare. (plus greens of the blend to roast his method since these will be Rosto air close but no cigar roasted )

[miKe mcKoffee]

I'm sort of fixed on

0 - 1 min to 250F (basically top speed)

Top speed of course is relatively to ones setup! If I went top speed with the Rosto (~132v to heater, ~144v to fan) and my 1/2# batches for a minute it'd shoot past 300f. For whatever arbitrary reason I've chosen 200f as my 1min mark, 250f two minute mark.

1 - 4 min to 300F (drying phase, an absolute must)

Same target temp @ 4min

4 - 7 min to 380F (top speed again, the longer you're here, the worse it tastes)

Here again top speed with a qualifier you've mentioned in the past, top speed while maintaining even bean color change. With my over sized slow moving 1/2# Rosto batches faster than 4 minutes to start of 1st from 300f usually results in too uneven going into 1st. But agree too slow will/can result in a dull roast.

7 - 11/12 min to the end, 11 for brewing, 12 for espresso (balance fast vividness and slower sweetness)

But I'm now experimenting with an extra step, slowing down to 10f per minute through the first crack, and speeding to 15f/min afterwards. I'm hoping this will produce a better tradeoff of vivid and sweet. It'll take a few months before I can decide.

I've come to the preliminary conclusion my current ~6min start of 1st to end of roast too long. Not baked, but not enough acidity definition or vividness as you describe I believe.

Did a roast of my espresso blend today changing two stage parameters from posted above: increased 300f to 400f ramp rate for 4min total time (was 5 min, 4min is actually my norm for most non-espresso roasts already), decreased start of 1st to end of roast from 6 min to 5 min using straight 10f/min ramp 400f to 450f end of roast. Likely won't test resulting shots 'til Sunday young 3&1/2 days rest then will adjust the three test roasts to split and send half of each to Abe next week accordingly. Of course could have should have done two roasts today each with only one changed stage but where's the challenge in figuring out the results from that!

Mapping today's roast profile realized I haven't done any serious profile experimenting since Dec '05, date on my current profiles sheet! Man time flies.

[another_jim]

Did a roast of my espresso blend today changing two stage parameters from posted above: increased 300f to 400f ramp rate for 4min total time (was 5 min, 4min is actually my norm for most non-espresso roasts already), decreased start of 1st to end of roast from 6 min to 5 min using straight 10f/min ramp 400f to 450f end of roast. Likely won't test resulting shots 'til Sunday young 3&1/2 days rest then will adjust the three test roasts to split and send half of each to Abe next week accordingly. Of course could have should have done two roasts today each with only one changed stage but where's the challenge in figuring out the results from that!

I'd be interested to know how it goes. My experiment yielded a too bright taste, but terrific aromas (unfortunately nothing like the taste, a bait and switch roast if I've ever tasted one). Getting the taste match the aroma will be my next job.

[cpl593h]

Did a roast of my espresso blend today changing two stage parameters from posted above: increased 300f to 400f ramp rate for 4min total time (was 5 min, 4min is actually my norm for most non-espresso roasts already), decreased start of 1st to end of roast from 6 min to 5 min using straight 10f/min ramp 400f to 450f end of roast. Likely won't test resulting shots 'til Sunday young 3&1/2 days rest then will adjust the three test roasts to split and send half of each to Abe next week accordingly. Of course could have should have done two roasts today each with only one changed stage but where's the challenge in figuring out the results from that!

I did the exact same thing with some Daterra Yellow Bourbon and it was wonderful. It wasn't so great with anything else, but it did the trick for the Brazil.

[miKe mcKoffee]

I'd be interested to know how it goes. My experiment yielded a too bright taste, but terrific aromas (unfortunately nothing like the taste, a bait and switch roast if I've ever tasted one). Getting the taste match the aroma will be my next job.

Just pulled the first shot of this roast profile at 4 days (and 30 minutes) rest (mason jar vac sealed directly post roast cooling as is my roast rest and storage norm). All I can say so far is WOW! Definitely seems much better top end definition but not lacking anywhere else. I'd say perfectly balanced but that's just going too darned far! Besides, then wouldn't leave room for improvement, and there's always room for improvement. But dang, one of the best shots of anything I've had in a very long time. Of course, didn't roast any using the previous profile at the same time to compare so taste memory comes into play. Tomorrow or Tuesday plan to roast 3 or 4 slightly different profiles for direct comparisons. (half of each also being sent to Abe to compare)

Jim, IIRC this profile was a minute longer start of 1st to end of roast than yours, might try lengthening a hair.

I'll post the detailed profiles after comparisons done by myself and Abe.

[cafeIKE]

I don't want to rain on the parade, but we must consider the variables.

For example, last Friday, I roasted a batch and after two days it was, well, terrible. The dry aroma was quite nice, but the cup not so. Weedy would be the best descriptor. I haven't roasted this bean since the summer, when it was excellent.

I left the remainder of the batch in the cupboard in a vented jar until this morning, a full 9 days later. The average temperature is probably about 60°F in the cupboard. When I opened the jar, the aroma was fair, but faint. Nothing exciting.

In the cup, it was astounding. The weediness was completely gone. Mocha notes were outstanding. A great balanced taste with a back end that wouldn't quit. Y_U_M

Normally, I'd expect a 9 day old roast to be ready for the bin, especially in the summer when the average temperature is about 75°F.

Which leads me to speculate whether there is "a roast".

There maybe a roast in a roaster for a bean / blend in a condition rested for a time ground on a grinder pulled on a machine by a barista on a day in a season.

Whether said roast would transfer across the street, town, county or planet maybe open to discussion.

What say ye to this, stout fellows?

[miKe mcKoffee]

Which leads me to speculate whether there is "a roast".

There maybe a roast in a roaster for a bean / blend in a condition rested for a time ground on a grinder pulled on a machine by a barista on a day in a season.

Whether said roast would transfer across the street, town, county or planet maybe open to discussion.

What say ye to this, stout fellows?

Of course this is no one single best roast profile, who ever said or implied there was?

For a specific bean/blend from my experience and that of other's there is little or no reason to believe a specific profile is not transportable with very duplicatable results, including across roasting platforms.

[Arto]

Until now my standard procedure has been to increase the temp up till 260 ° C chamber temp (around 15 min), and stop the roast when desired roast-level has been reached (around 20 min).

Works OK, but I will experiment more... Shorter times has been to acid/non-sweet to me

[miKe mcKoffee]

Well the pressure's on! Getting ready to roast three 1/2# batches of my Winter Solstice Blend attempting to (relatively) accurately vary the start of 1st to end of roast profiles while maintaining the same earlier stage profile.

FWIW just in case you're wondering the blend consists of Aged Sumatra DP 17.4%, Monsooned Malabar 8.7%, Brazil Cerrado DP 19.7%, Brazil Cerrado CD 19.7%, Kenya Gethumbwini 17.4%, Panama Eleta 17.4%

Profile will be as follows with temps measured in slow moving Caffe' Rosto bean mass:
150f @ 30 sec
200f @ 1 min
250f @ 2
275f @ 3
300f @ 4
325f @ 5
350f @ 6
375f @ 7
400f @ 8 min (anticipated start of 1st)

Profile1 (current excellent batch with level ramp)
410f @ 9
420f @ 10
430f @ 11
440f @ 12
450f @ 13 end of roast tickling 2nd

Profile2 slow fast
406.7f @ 9
414.3f @ 10
420f @ 11
435f @ 12
450f @ 13 end of roast

Profile3 fast slow
415f @ 9
430f @ 10
440f @ 11
445f @ 12
450f @ 13 end of roast

Approx' cooling in Rosto down to
275f @ 14
200f @ 15
150f @ 16
100f @ 17

1/4# of each finished roast will be sent to Abe valve bag sealed shipped USPS Priority today, I'll keep the remainder of each batch. Originally had challenged Abe to simply try what I felt was a quality "close but no cigar" espresso air roast. But now imposing three different profiles on him. So you may or may not get more than one person's review of whatever differences may result.

[another_jim]

It'll be interesting to see how the different profiles turn out. I'm using a 50/50 Idido-Cenaproc mix done on the new profile, stopped quite light, 7F before the first pops of the second, at City+, for an espresso dosing experiment. So far it's working as planned -- sweetly drinkable shots no matter how I abuse the blend.

[miKe mcKoffee]

...done on the new profile...

Would you mind posting the specifics for your new profile. I know many may find it boring and or useless, but I'm a glutton for punishment.

It will indeed be interesting so see what differences are apparent with those three different finish profiles I did today. But must wait at least three days preferably four ideally five to try them out!

I do know prolile1 has been stellar days 3 through 6. Haven't messed with shot temp variation much keeping primarily around competition ~200f temp. Or more accurately using HX flash-n-go intra-shot profile with definite HX hump pulling peak temps ~202f with main shot fairly level ~200f. Only tried as high as ~203f and as low as ~200f HX hump start of shot temps, all good but different of course. The 2nd shot pulled this afternoon with lower temp ~200f HX hump which would be ~198f main shot surprised me in the amount of sweet brightness maintained with no hint of sour.

[another_jim]

I'm sort of fixed on

0 - 1 min to 250F (basically top speed)
1 - 4 min to 300F (drying phase, an absolute must)
4 - 7 min to 380F (top speed again, the longer you're here, the worse it tastes)
7 - 11/12 min to the end, 11 for brewing, 12 for espresso (balance fast vividness and slower sweetness)

But I'm now experimenting with an extra step, slowing down to 10f per minute through the first crack, and speeding to 15f/min afterwards. I'm hoping this will produce a better tradeoff of vivid and sweet. It'll take a few months before I can decide.

Final stage is split in two:

7 - 10 min to 415F
10 - 11/12 to final temp, 1 minute to 420 to 430 for cupping, 2 minutes to 435 to 450 for espresso.

For calibration purposes: 1st crack 395F to 410F; 2nd crack from 442F to 452F

[miKe mcKoffee]

Final stage is split in two:

7 - 10 min to 415F
10 - 11/12 to final temp, 1 minute to 420 to 430 for cupping, 2 minutes to 435 to 450 for espresso.

For calibration purposes: 1st crack 395F to 410F; 2nd crack from 442F to 452F

This is where I'm confused. If 7-10min to 415f from start of 1st @395f, a 20f temp rise, and experimenting with 10f/min ramp doesn't fit the 3 minute 7-10 minutes!? Also if I understand correctly, you're using a total time of 5 min from start of 1st to end of roast for espresso roasts? (Which happens to be what I just shortened mine down to from 6 min with good results.)

[cpl593h]

Jim - is this a preblend? I have both of those in my stash, I'd love to see if I get similar results.