Raost

[benjaminfleon]

I don't personally own a moisture meter, but have access to a couple from different roasters I work with. Whenever I buy green, I measure the moisture and then dose all of it in vacuum sealed bags. Then I open them and rehydrate between 8-12 hours before roasting, according to Feran's protocol.

I'm glad to know the samples you cupped were nice; I've found it's a pretty good benchmark profile for a number of different beans. Right now I'm roasting a washed Pink Bourbon from Colombia, and it's been super juicy and sweet.

[benjaminfleon]

Raost v.2

I went further into a few things described by Rao in his most recent book (Coffee Roasting Best Practices), and a couple of podcasts. Having learned a few things from several logs and adjusting the original profile to a couple of different profiles, I came up with this new version.

It's basically the original profile with these modifications:
— slightly lower preheat power, as most coffees I tried went above the curve at the start (I think this should be adjusted primarily based on bean size, and not so much based on moisture content, unless you're rehydrating; bigger beans, higher preheat power).
— power profiling zone to keep power constant during FC, as well as constant fan speed during that same time; as to minimize flicks or crashes.
— power profiling zone of -5 from 6:15 to 6:30, just before 12% DTR, to account for the tendency of a flick around 16% DTR; and another one of -2 from 6:30 to 8:00 to avoid flick around 22% DTR

Raost v2.kpro

(3.15 KiB) Downloaded 355 times

I've been able to roast up to higher temperatures with this approach, without the roastiness I used to get at them (i.e, at 222-224 ºC). From what I understand, the flicks that I was getting in the original profile were leading me to roast under 220, and miss out on a little more texture and body.

Keep in mind, this is suited for my specific preference of roast level (which currently sits between 1.9 and 2.0).

I'll post a few logs soon!

*EDIT: I uploaded the wrong file. The one that's up now is the correct one.

[benjaminfleon]

Slow Raost

Slow Raost.kpro

(3.33 KiB) Downloaded 320 times

This profile is one of the two profiles I've been playing with a mods from the original. It's basically a slower version, that tries to get the same approximate %'s for the different phases, but renders a slightly more developed flavor and heavier body.

The other one I'm working on is a faster version, that does the same, but aiming for a more nordic-style clarity and bright acidity. That one's not ready yet, as I'm still working on controlling the flicks and crashes in the RoR.

For reference, Raost v2, the version I posted on the last post, hits FC around 5:20, and finishes just before 8:00, for a 32% DTR at 222ºC (2.0). Slow Raost hits FC around 6:50, and finishes around 9:30, at the same 222ºC (2.0).

Here's a snap of the last roast curve using this profile, and the log beneath it

log0258.png (122.84 KiB) Viewed 5662 times

log0258.klog

(72.1 KiB) Downloaded 257 times

I'm still working on smoothing out the RoR curve, but I'm pretty happy with the results so far. Tomorrow I'll cup a bunch of samples, so I hope I can report back soon on how taste is going.

[benjaminfleon]

Raost v3 & Fast Raost

Hey everyone,

So after a bunch of roasting using different profiles, I've come closer to a "less is more" approach. I'm still trying to apply Rao's ideas to these profiles, which is why I'm keeping the names; although these two profiles are meant for very different kinds of roasts.

So, first, on the v3 of Raost, I fixed the curve in order to have a higher RoR at the end, as I was having issues with such a flat curve after FC in the past versions. As for the Fast Raost, I'm going for a faster roast (duh), similar to what I would call a "nordic" style —clarity and acidity over texture and complexity—, but with the same intention of keeping a steadily declining RoR curve with plenty of development.

Both profiles work great with 100 gram batches, or 80 gram rehydrated batches (see first post), but of course you'll see a difference in development, which I tend to fix with end temp. I guess you could see them respectively as modern espresso and filter roasts, while Slow Raost could be equated to a more traditional espresso roast (I'm also using this approach for this profile's new version, but haven't been able to log anything with it yet).

For both profiles, I use a power profiling zone around FC and a slightly higher than recommended min desired rate of rise setting, which I attempt to compensate with a negative boost between approximately 12 and 16% DTR. I've found this helps to give enough room for the PID to do most of the work, and it's more versatile than relying too much on power profiling.

Also, I'm using a similar approach to Steve's CDFP (constantly declining fan profile) on all profiles, but I'm keeping the speed constant to coincide with the power profiling during FC.

Raost v3.kpro

(1.99 KiB) Downloaded 338 times

Raost v3 profile.png (116.39 KiB) Viewed 5564 times

Raost v3 fan.png (43.58 KiB) Viewed 5564 times

Fast Raost.kpro

(2.09 KiB) Downloaded 287 times

Fast Raost profile.png (107.92 KiB) Viewed 5564 times

Fast Raost fan.png (42.67 KiB) Viewed 5564 times

Let me know what you think!

[Damian]

I had a crack at it, didn't work well for me

1.jpeg (147.88 KiB) Viewed 5525 times

[benjaminfleon]

I had a crack at it, didn't work well for me
1.jpeg

Hey Damian,

I'm sorry to hear that. Although I'm curious about the version you tried, cause the power curve seems to have some power profiling after FC. I was having similar issues with some beans with higher exothermic behaviour on past versions. Could you upload the log and provide info about the coffee?

By looking at just this, I'd recommend trying the last version (v3) if it isn't; adjusting the duration of the constant power profiling during FC to start and end 5-10 secs sooner; or adjusting the min desired rate of rise setting to slightly higher value to avoid that crash after 6 min.

[mr. bean]

What bean did you roast Damian? And what version of the profile?

[Barry O'Speedwagon]

Can report that the Raost Sidamo profile works very well for the Ethiopia Shakiso Estate formerly available through CS. I've been using 110g doses (as that is my standard approach) and doesn't seem to present an issue (for me).

[Damian]

Sorry, I thought I replied, must not have
Roast V2 Yemen Mocka Ismaili

Thanks for the suggestions, I'm not going to persist with this sorry. I profile quite differently, but wanted to give it a shot.
Anyway, Kudos for sharing

[mr. bean]

I know this is off topic, but out of curiosity are you still roasting with D-roast or something similar Damien?