Light City/Nordic Roasts in Corretto

[gobears]

Hello All,

Greetings from Northern California.
Just built up a Corretto roaster with great help from examples on this site.
Very much appreciate the great spirit of sharing on this site, its a great resource.

I have an Oster BM converted with insulation, temperature probes, closed cover, etc...
I am roasting for Aeropress or pour over, targeting Kenyans and Ethiopians.
I have done 4 roasts so far in the Full City range (just before 2nd crack).

Now, I would like to do some very light City or Nordic roasts.

I have seen profiles for other machines finishing in as little as 9:30, and as long as 21:00.

Any experiences here about what the best profile is for Corretto roasters when targeting City or Nordic blends (finishing just after 1st crack)?

[flynnaus]

Any experiences here about what the best profile is for Corretto roasters when targeting City or Nordic blends (finishing just after 1st crack)?

Welcome.

Until now, I had never heard of Nordic Roasts. Having googled it, i can see that you can view roast profile graphs. Other CSers who roast for filter brews using a corretto may have their own tips.

You can try to reproduce profiles but don't expect to get the same results as the person who provided the profile on your kit. I've tried following other profiles and the results were not as good as i hoped but i tried, adopted and modified techniques to get something that works for my tastes using my roaster. There are many variables such as the bean type, the batch size and the roaster. Ambient temperature and rest times (before you start consuming) will also have an effect. Then there's you and your tastes.

Be aware that short roast times may mean the beans don't develop enough; the outside of the bean looks roasted but the heat hasn't fully penetrated to the centre. Don't be afraid to go for roast profiles well beyonf first crack. I prefer to take my roasts to the beginning of second crack or just before even for Aeropress brews.

[DesigningByCoffee]

Hi gobears
As flynnaus has mentioned, it is lightening the roast (to reduce the caramelisation that overpowers filter type brews) while still maintaining the correct bean development that is the issue.
FWIW in aeropress I've had the greatest success by either increasing the ramp to first crack slightly and then taking through to just on 2C, or maintaining a standard ramp to 1C then slowing the ramp towards 2C but dropping around 5-6° earlier than usual - this has worked for different beans. Some origins also seem to behave differently. Kenyan beans roasted to my standard espresso profile seem to go great in the AP as-is, while Ethiopians all seem to like the earlier drop method.
Best practice will be to have a go! Couple of options above. I'd buy a bag of something, try 2-3 different profiles on the same day - see what they taste like in different brew methods. Pretty cheap education really!

Cheers Matt

[Steve82]

Hi gobears,

Faster / lighter roasts that are fully developed and sweet can be done on corretto. You just need to think outside the square and not get stuck in any " ideals '. The biggest being that to evenly and fully develop faster / lighter roasts you will need to significantly reduce your batch sizes, so that you can keep your overall environment temp under control so as not to scorch. You will need to have very good control over your airflow and heat application.

I suggest you do a google search using " Creative Nickname Roast and Learn 2014 "
HEAPS of information on light filter roasts using bread machine / heat gun.

[gobears]

Wow, that google search was gold for me, thanks so much!
I've been searching the past few days and not turning up much, until now.

Exactly what I was looking for!

For others, let me give a little background information.
I really like East African, especially Kenyan coffee, roasted light, to keep fruity, floral and high but well balanced and integrated acid, with low roast tones.
I bought several pounds of high grade Kenyan coffee to learn on.

I have read quite alot about Rao's roasting philosophy, versus the slow start / fast finish style popularized by Tim Wendelboe and others.
My plan is to experiment with both of these methods, but I had no idea where to start applying other's profiles to a Corretto roaster.
Now, I have a really great starting point, by following "Creative Nicknames" profiles.

He is using only 200g, but I would prefer to go with 450g (1 pound). I guess for now I can try 225g, and get twice as many experiments, and maybe eventually apply what i learned to larger charge sizes in future.

Can't wait to get started...!

Cheers,
~Bruce

[gobears]

Quick update... I have executed one roast in the "Nordic style", and it turned out reasonably well.
It a Kenya Nyeri AA from Sweet Marias.

I followed a profile from [creative nickname] that was posted as an attempt for SS / FF Nordic style, and which cupped better than alternative roasting methods.

Furthermore, I learned from:

http://nordicbaristacup.com/wp-conte...y_Cropster.pdf

that Tim Wendelboe and Kaffa (2 of the top Norwegian roasters), and running very low temperature profiles.

In fact, I am thoroughly confused by the roast profiles.
There are a total of 4 roast profiles in the presentation, and 2 finish at "normal" higher temperatures, and the 2 I mentioned above finish at much lower than normal temperatures.

Of course, I figured maybe its the thermocouple placement, or an anomaly, but that seems unlikely to happen to 2 of the top roasters in Norway at the same time, in a forum they are presenting to their peers.

So, a question I have for more seasoned roasters is, how did TW & KAFFA both achieve such low 1st crack temperatures compared to the norm?
It seems it would be nearly impossible to me. Keep in mind, these roasts were cupped and these 2 were deemed superior.

[Steve82]

Quick update... I have executed one roast in the "Nordic style", and it turned out reasonably well.
It a Kenya Nyeri AA from Sweet Marias.

I followed a profile from [creative nickname] that was posted as an attempt for SS / FF Nordic style, and which cupped better than alternative roasting methods.

Furthermore, I learned from:

http://nordicbaristacup.com/wp-conte...y_Cropster.pdf

that Tim Wendelboe and Kaffa (2 of the top Norwegian roasters), and running very low temperature profiles.

In fact, I am thoroughly confused by the roast profiles.
There are a total of 4 roast profiles in the presentation, and 2 finish at "normal" higher temperatures, and the 2 I mentioned above finish at much lower than normal temperatures.

Of course, I figured maybe its the thermocouple placement, or an anomaly, but that seems unlikely to happen to 2 of the top roasters in Norway at the same time, in a forum they are presenting to their peers.

So, a question I have for more seasoned roasters is, how did TW & KAFFA both achieve such low 1st crack temperatures compared to the norm?
It seems it would be nearly impossible to me. Keep in mind, these roasts were cupped and these 2 were deemed superior.

I think you have answered your own question. Definitely probe placement, in relation to the batch sizes they are roasting. Probes may only be reading mostly air temps with small loads.

[gobears]

Yeah, I agree that probe placement plays a role.
However, air temps would be expected to be higher, right?
And, I found out elsewhere that Wendelboe stated his temperature probe is closer to the drum, which would explain a little lower reading.
But 70 degrees fahrentheit is still a huge difference in FC.
And, Kaffa has similar low temperature, and it seems unlikely to me that they both have equally poor placement of thermocouples resulting in low BT.

Also, I checked the loads and Wendelboe is roasting 12kg in a 15kg machine, and Kaffa is roasting 30kg loads in a 45kg machine.

[gobears]

Perhaps temperature probe placement & load explain the difference.
Then let me rephrase the question into 2 specific directions:

(1) Can % convection vs. % conduction play a big role in FC temperature, and could a high convection % roaster achieve a lower FC temperature comparatively? We know Wendelboe uses very high air flow.

(2) Can the "pressure cooker" effect play a significant role to enable a lower FC temperature, and how would one maximize pressure cooker effect? I am thinking that a flat ROR would extend the "pressure cooker" effect over a wider range of time, effectively "pressure" cooking the interior of the bean longer, accelerating the roast.

[Steve82]

What you say about TW probe placement on HIS roaster is true. Watch his video on roasting on Youtube if you have not already.

Convection / conduction / radiation ratios and how they work with airflow is different machine to machine. Its not enough to have first crack temps that far out wack.

In the end it does not matter as long as the data is consistent, it just makes it harder to share that data with other roasters.

[gobears]

Ok, now I am sold on the idea of the temperature probe placement, and the fact that its the only way to explain these huge temperature differentials.
Thank You!

[gobears]

Thanks everyone for the great feedback, I am on the right track now.

Did 4 more roasts on my Corretto, all of the roastmasters Panama Natural Process Elida.
The roast times were 7min to 11min, and first crack was consistent on all of them, around 204 celsius.

All of them were good, but the shorter roasts were best.
I tried a profile by creative_nickname at another site, and that one worked best of all, and it was indeed a Nordic Profile.

My next coffee roasts will be some Guatemalans, which I expect to be better with more development.

By far, my biggest challenge with the Corretto roaster is getting control of the RoR curve.
It will be stable at the level I want, then suddenly start to go down, then go down rapidly.
Its rather unstable. I had to increase heat during roast, while heat should be generally decreasing with time.

Any ideas or experience about how to stabilize RoR in a Corretto?

[DesigningByCoffee]

The beans are endothermic before first crack - they take in heat. If you set a start temp and left it, you would get a substantially decreasing profile.
So to achieve a perfectly flat RoR you would need to steadily ramp the temperature up to 1C (and in quite large amounts - think tipping or scorching results most likely).
I use a reasonably slow and gentle ramping roast compared to many - but even then I have a slowly declining profile. I actually log my gun settings on the profile - if you look at the red line down the bottom you'll see where I changed gun temps in steps - the number is my bosch gun temp setting. Then you can compare to the profile curve…
If I want an overall faster roast time (for certain beans or roast styles) - I increase the initial temp by 10 or 20° after turning point - then each step I increase by the same 40° on the gun.

Works for me

Cheers Matt

[gobears]

If I want an overall faster roast time (for certain beans or roast styles) - I increase the initial temp by 10 or 20° after turning point - then each step I increase by the same 40° on the gun.

Hi Matt,

That is helpful feedback, thanks! I can see in your attached photo, that the heat is ramped up periodically in steps, and then ramped down after first crack. When you are ramping up, you are doing so every 30sec to 1min, roughly. What do you use to track when to bump the heat? Do you watch RoR, and when it starts to flag too much, bump it up? Or some other means? I was watching RoR, and when it started to fall too much, I would adjust heat up. However, there was a lag, and the ROR often dropped too low, then overshot due to the heat increasing too much in response. I think I can cure the overshoot by learning how to ramp up my heat more smoothly, but I am pretty sure I will struggle with the undershoots of RoR unless I can have better anticipation.

Looks like you have this very well figured out!

BTW, I insulated my bread pan with one layer of pipe insulation, but the R rating is only 1. Maybe I need to put more insulation to dampen the heat losses too.

[DesigningByCoffee]

Hi gobears
Over time and playing I've just developed a system of ramping 40° on the gun at 75°, 100°, 125° and then 150° log point, measured on the probe (that's a basic system - not taking into account the 'dip' between 145° and 160° which is a whole other story ).
After 1C is a little more involved, but basically I drop by 30° increments every 5° from 1C (say 202° - then 207°, 213° etc) and thereafter to the drop at 223°
With my Bosch gun this system gives almost identical flavour results for a 750g batch with gun set on fan speed 3, or a 350g batch on fan speed 2. Slightly faster roast/profile with the smaller batch when logged by 3-4mins - but actually needs to be funny enough! I find a 17-18mins works well for the small batch, while 22-23 mins is best for the bigger.
Strange but true!