Delays and the final countdown for the roasting competition


Delays and the final countdown for the roasting competition

Delays and holidays (and GDPR)

Next week Crankhouse will be on vacation. Not a coffee farm trip or a short weekend break but a proper holiday, on my bike for a week in the Picos De Europa with my wife. I've endeavoured to get extra stocks to all my wholesale customers this week. Fingers crossed I've remembered everyone since the worst thing that can happen is that a customer is left without the coffee needed to run their business and serve their customers. They'll have to go elsewhere regardless of the quality or consistency I normally provide, and then what ! Luckily I have a good friend and helper who is prepared for the emergency call to bag up and deliver coffee at short notice. Let's hope his services aren't required.

This does mean that retail orders through the Web store will also be delayed until my return. There is a clear note on the site front-page as well as on the checkout page FYI.

On the topic of GDPR, I've had a couple of emails from annoyed customers who have continued to receive this newsletter who did not opt-in to the GDPR mail-out. That's absolutely my error since I missed a crucial step in segmenting the opt-in respondents. I have hopefully rectified that now but if you're receiving this and did not opt-in then please email me ( and I'll remove you from the mailchimp list that this goes out to. My apologies again to anyone offended.

Roasting competition update

This is a follow-on from a couple of previous posts which covered the roasting competition that I've entered. The 500g final roasted sample needs to be submitted to Falcon (the green coffee sourcing company) by the 11th September for the blind judging on the 14th. I'm out of the country until the 12th which means I need to roast and submit the coffee before I leave on the 3rd. I'll be roasting this final batch this weekend.

I roasted this coffee on 3 separate dates in 3 batches of 3Kg giving 9 total profiles, each separated by 3 or 4 days. I asked some coffee friends if they'd mind helping out and got an affirmative from all ten. Some incredibly knowledgeable and talented coffee professionals gave me their time and expertise to help narrow down a 'good' or 'best' profile for this coffee. Due to the time constraints I wasn't able to send all 9 profiles to all of my helpers. Joel in Norway and Zoltan in Poland only received the initial 3 batches but their responses were both incredibly detailed and constructive.

Of course I want to win this competition, but I also wanted to use it as a mechanism to demonstrate how incredible this specialty coffee community is. I've met all but one of the ten helpers but some only once for a quick hello. They're not on my whatsapp group or my telephone quick-dial, but they all agreed to help and all took the time to cup the roasts and give me their detailed professional feedback. That's quite something. I also wanted to use it as an educational exercise. Seldom do I get to 'play' with a coffee. I'm running a business and testing 27Kg of an expensive 87.5 rated coffee doesn't make any business sense. Selling those 'test' roasts would make sense but what if they're not that good ? For me it has been an investment in my knowledge, and hopefully it's made me a better roaster. 

It's easy to look back and think of how I would do things differently. Common sense would tell me to use the IKAWA Pro sample roaster to roast 60g batches. Identify the 'best' and then simply roast multiple 60g batches with that profile and submit that. But that isn't real. I need to be able to get the best out of a coffee with my production roaster, not a computer controlled hot air blower.

A huge thanks to all my helpers: 

  • Tom @ Crafthouse
  • Joel @ IId Og Bonner
  • Zoltan @ Holistic
  • Freda @ Raw Material
  • Stephen @ Bailles
  • Andrew @ Colonna
  • Chris and Harley @ Outpost
  • Will and Jack @ Roastworks
  • Kane @ Nude
  • William @ It All Started Here

Plus all the local coffee folk in Exeter who continue to donate their time and energies in cupping regularly and learning together, Tony, Lewis, Jo, John and Josh and our newcomer Justine. Anyone welcome every Thursday 6pm at Exploding Bakery.

For the geeky few I've included all 9 profiles below with a brief explanation of how I narrowed it down to the top 3 and how I'm going to approach the final batch this weekend.

Profile1 (ORANGE LABEL): 

Profile 1

Charge: 190C, 1'20sec soak then 70% power, reducing gradually after drying through to FC with a power adjustment to maintain a steady RoR. FC: 7'12", drop: 8'12".

Short 1'00" development and consensus was that this was 'green' and underdeveloped, a little thin with not much body. 

Profile2 (RED LABEL): 


Charge: 190C, 20sec soak then 60% power, reducing gradually after drying through to FC with a power adjustment to maintain a steady RoR. FC: 7'24", drop: 8'36".

1'14" post FC development and some people preferred this as the best of the initial 3 roasts with more sweetness and body but still with a greenness and quick finish.

Profile3 (YELLOW LABEL): 


Charge: 190C, 1'40" sec soak then 70% power, reducing gradually throughout. FC: 7'54", drop: 8'57".

1'03" post FC development and even though this is the slowest roast that came about by the excessively long soak. Therefore the period of power and development is probably quite similar to the previous 2 batches. A majority liked this the least. A little flat and dull.

Based on the cuppings with my regular crew at The Exploding Bakery in Exeter and the feedback from my specialist group of helpers the consensus was that these were all underdeveloped and this coffee needed to be given more heat and more time to allow the sugars and body to take effect. I think all of us struggled with the 87.5 rating that the importer had given this coffee.

For the 2nd round I decided to try and slow the roasts down through the maillard phase by reducing the power a little earlier. The long soak seemed irrelevant.



Charge: 190C, 30" sec soak then 60% power, reducing gradually throughout. FC: 7'33", drop: 9'03", with 1:30 sec post FC (16.6% DTR)

I was very happy with this profile and this was quickly highlighted as the best roast so far. Good fruit and sweetness, body and a nice finish. Kane from Nude Coffee just told me to STOP right there. Numero 4 ! 



Charge: 190C, 10" sec soak then 70% power through drying then a big power reduction trying to get the RoR down and keep it constant (ie not declining) through FC . FC: 7'09", drop: 199C at 8'18", with 1:09 sec post FC (13.9% DTR)

In terms of roasting this was a disaster. You can see the way the RoR curve did the old ski jump trick, up and then rapidly down. Losing all momentum in the roast. I think I just wrote RUBBISH next to this one when I first cupped it and the feedback from the crew confirmed it. Flat and dull with an ashy aftertaste.





Charge: 188C, 30" sec soak then 60% power through drying then a big power reductions and an increase into FC to try and acheieve a smooth RoR. EPIC FAIL. Another ski jump flattening and dropping rapidly. Scott Rao covers this in detail and he reckons this leads to baked coffees. No disagreement from me or the panel. A little lacklustre.

By this point I was beginning to realise that this was not an easy coffee to deal with. This secind batch of 3 (4,5,6) went out to a sub-set of the tasters and Stephen from Baiiles made a useful suggestion. Try and slow the whole thing down not by reducing power but by reducing the initial charge, allowing more time for the necessary reactions to take place for the sweetness and body to develop.



 Charge: 170C, 30" sec soak then 60% power right through drying and continuing that energy through maillard. With the reduced charge I was worried about not having enough energy in the system reduction trying to get the RoR down and keep it constant (ie not declining) through FC . FC: 8'48", drop: 195C at 10'09", with 1:21 sec post FC (13.3% DTR). The slowest overall roast of all 9.

This one made my final cut, partly because it had the best dry aroma. Again I felt I was in control with this roast and avoided the dreaded ski jump drop off. 



Charge: 170C, 20" sec soak then 70% power through drying then dropping off the power to try and extend maillard. FC: 8'15", drop: 195C at 9'36", with 1:21 sec post FC (14.1% DTR). I didn't reduce the power into FC and the result was the flattening then ski jump drop . Once it starts it's too late to stop and I felt once again that I'd lost control of this roast.

On the table this cupped empty. Hollow and flat. 



Charge: 187C, 30" sec soak then 60% power. FC: 7'54", drop: 197C at 9'06", with 1:12 sec post FC (13.2% DTR). This tie I reduced the power into FC and then gave it a little boost after the first few pops to avoid the ski jump. 

Not one of the favourites on the cupping table and this didn't make the final cut. Another one that was a little dull and lacking any vibrancy.

The Final Cut

First impressions are definitely worth listening to but having some understanding of how coffee changes as it rests after roasting also needs to be considered. I cupped the initial 3 batches (that were all considered to be underdeveloped) daily over the follwing two weeks, alongside the newer batches as they were roasted and started resting themselves. For me the first batch No1 started to show some delicate and clean fruit flavours. I put it in my top 3.

No4 seemed to attract the largest number of positive comments from everyone. Juicy and sweet with a creamy body. 4 happens to be my lucky number so it had to be in.

Then No7, a controlled roast and the slowest of all allowing more time for stuff to happen. 

At the end of the day the coffee on the judges table will have my name on it (actually it won't since the judging will be completely blind.. just figuratively) and it's my decision. For me 4 tastes a little heavy, perhaps slightly too much body and not quite as clean as I'd like it to be. For sure it would make a banging espresso but that's not how it will be judged. No1 is a little thin and maybe still a little green but cleaner and it definitely opened up over the testing period.

The 'final' roast profile I'm going to aim for is somewhere in between the two. ie trying to keep some of the clarity of No1 but allow more development to happen but not quite to the point of No4. That's the plan ! Wish me luck.

Just to show you how generous and professional my helpers were here's a copy of the email from Tom and Zoltan..

12 minutes: notes of deep, rum, drying, notes of overdev, hollow
18 minutes: it has become more complex, but still drying and lacks acidity
25 minutes: still drying, dark, no progress
cold: it died completely, no character
12 minutes: more balanced in terms of acidity and sweetness, no signs of overroasting
18 minutes: round, sweet and balanced, slightly drying
25 minutes: the extra acidity disappeared, otherwise still clean
cold: cleaner and more acidic than the red, still shows some notes of origin, hollower than the orange
12 minutes: the sweetest cup, but maybe less clean than the yellow one. Notes of toasted seeds.
18 minutes: even better than the yellow. My favourite cup due to the extra sweetness 
25 minutes: the extra sweetness remained, this is the winner cup
cold: still the best and sweetest cup, kept integrity


Roast development (cracking it open):
4 nice and even possible a little light on the dev
6 good little dark on the inside.
7 nice and even smidge dark in spots.
Dry Aroma:
4 candy sweet
6 a little roasty
7 great balance, sweet.
4 nice and sweet, sugar nugart
6 a little dry, strongest 
7 sweet nice a little spicy 
Taste 9:00
4 little dry, yellow fruit creamy
6  Brazil, nutty? Silky. Not much acidity
7 hoppy, astringent (I think thats how you spell it?) best acidity
Taste 15:00
4 opened up a bit, nice and sweet. Dry fruit, apple. Kinda Guatemala like 
6 once again opened up, Little more juicy. Cream little hints of mid level lime. Nutty.
7 dropped off. A little tasteless. Dry astringent
Taste 20:00
4 stone fruit, kinda nice apricot - prob the best. Kina honeydew? 
6 yeah, apricot again. Juicy. Still a little nutty for my liking. Kina honeydew? 
7 still dry. Me no like the mouthfeel. 
I would say 4, now that its cool it has a nice stone fruit sweetness. A little juicy. A little dry in a nice way, like a high grown Colombian. All are a little nutty/ nugartey. All deffo have this kinda wine like body and mouthfeel.

Update.. the submitted profile (2/9/18)

I roasted the two final batches yesterday, planning for one batch half-way between 1 and 4, and the other half-way between 4 and 7.

I immediately cupped them after cooling and had an idea of my preference but needed some validation. I asked my wife to pop out with me to the roastery today and with Volvic in hand cupped the final two roasts again.

There was a clear winner which will go in the post to Falcon tomorrow. It's a few days ahead of where I'd ideally like to have roasted this coffee but sometimes other things must (and should) take priority. 

Leave a comment

Please note, comments must be approved before they are published