The Golden Bean 2013 - Coffee Roaster Competition

[chokkidog]

I never cooled in the Gene, ................ emergency stop, oven glove and into the colander/fan cooler.

I'm pretty sure that the beans would continue to roast, if you left them in the Gene (?). Others are more qualified than I to comment tho'.
Barry....... what do you do?

[trickydicky2]

I would follow the advice of Barry and Chris.
When I had a Gene it was; preheat to 150°C then 230°C until 1st crack then 225°C til finish.
By setting the temp at 220°C from 240°C I think the thermostat would actually fall below 220°C before it started to reheat but your bean mass
might continue to lose even more heat before stabilising. (????)

I would look at that tho' as your possible cause of baked notes.

Thanks for the suggestion. I think I understand the logic here - In terms of RoR my current profile, rather than just reducing the RoR slightly to extend the roast, is possibly sending it into reverse. By backing off the temp sooner but more gradually it will create a "soft landing" with less risk of baked flavours

Another way of looking at it might be that with the current profile what I may actually have is a 14 minute roast with 2 minutes of poor quality cooling in the roaster

[chokkidog]

- Preheat empty roaster 160C/5m. (While I weigh out green beans, write notes, etc.)
- Load roaster, set timer to 20m
- 160C/4m
- 245C/~4m (until beans are orange)
- 220C/~2m (until beans are tan)
- 240C/~4m (until about 1 min into 1st crack)
- 220C/~2m (until desired roast colour achieved)

I would follow the advice of Barry and Chris.
When I had a Gene it was; preheat to 150°C then 230°C until 1st crack then 225°C til finish.
By setting the temp at 220°C from 240°C I think the thermostat would actually fall below 220°C before it started to reheat but your bean mass
might continue to lose even more heat before stabilising. (????)

I would look at that tho' as your possible cause of baked notes.

[Barry O'Speedwagon]

Hi Barry,

I use a different strategy with my Gene: 225 deg C until first and then 225 deg C until 2nd for a roast time of 16 (ish) minutes on a full 300g load. 238 deg is kept for robusta. I agree that the pre-heat to 150 deg C is worthwhile.

Cheers

Chris

Yes, thanks Chris. I think you were kind enough to suggest this (or something quite similar..might have been 228 reduced to 226?) when I bought the roaster from you last year.....and it certainly works very well for me most of the time. I did find sometimes though that some beans didn't quite get into rolling FC as strongly as I thought they might......partic. during winter up here (and I suspect there may be some generic variation across individual Gene Cafes). I have reverted to your method for that Tanzanian Machare, because it just gets away from otherwise.

[trickydicky2]

I am getting a lot out of this discussion, thanks. (Pleased I asked the question about baked flavours!). Have to admit I did not know what RoR meant in chokkidog's posts. Looked it up and found this interesting article: https://www.cropster.org/focus-on-the-rate-of-rise/

My challenge now, not knowing the bean temp, is to somehow infer the RoR (based on a sixth sense I guess) AND get ahead of the curve by adjusting the set temp by just the right amount and just enough ahead of time... That's what evenings and weekends are for!

[deegee]

Thanks chokkidog. My reference to sweaty saddles was mostly in jest anyway, but I can relate to the harsh & metallic.

The rest is interesting, and while not very applicable for me, it may well be of use to those who use drum roasters.

I have been following Matt's posts with interest, and have tried, with limited success, to get a "Seattle dip" into some of my recent roasts, though I'm still working on how best to achieve that with my gear. I may have to install a PID to get it right.

However, I do have pretty good control of the times from 1st to 2nd and yes, it does involve watching and managing the RoR carefully and using a bit of anticipation/prediction to get the timing right.

[TC]

I also use a Gene, tricky. My present approach based on non-SHB is to preheat to 150degrees (takes no time). Whack the beans in, set temp (actually thermostat limit) to 236 degrees, put plenty of time on timer, hit go. When first crack has definitely started (err on the side of waiting) then drop the temp by 3 degrees, wait 20 seconds and if there's still lots of happy cracking action, drop by a further 2. Second crack can be anywhere from 2 1/2 minutes (the demon Tanz Machare) or 3 1/2 minutes later....I try to pull just before second crack for most beans. For 250g of beans, I hit FC at about 10 minutes 30 in current climate. For hard beans I set the temperature at 238, but everything else the same.

Hi Barry,

I use a different strategy with my Gene: 228 deg C until first and then 225 deg C until 2nd for a roast time of 16 (ish) minutes on a full 300g load. 238 deg is kept for robusta. I agree that the pre-heat to 150 deg C is worthwhile.

Cheers

Chris

[Barry O'Speedwagon]

- 240C/~4m (until about 1 min into 1st crack)
- 220C/~2m (until desired roast colour achieved)

Therein lies a potential problem Tricky. Dropping the temp by 20 degrees might be a bit severe. Remember (AFAIK) the temperature setting is thermostat...it's not a direct measure of the heat that you are applying to the beans, so when you drop the temp to 220 the Gene will coast until the average temp across the 2 sensors drops to that level (someone correct me if I'm wrong). Maybe have a go at gradually reducing temp, and not by quite so much.

[chokkidog]

The saddle reference is more a physical sensation, relating to where and how it feels in your mouth.
Sweat salty, acrid, metallic, harsh and dirty relate to flavours. It was late.........

The part of the workshop where I stumbled and fell was about airflow management and astringent notes in the roast.
The theory being that if you have full airflow from 1st to, (or up to) 2nd crack, then you are at risk of imparting those types of flavour defects.

What wasn't covered is how different roasters and drum types behave in regards to airflow and astringency.
The workshop was run by a person who had only ever used cast iron drums, not carbon or stainless steel or any other material.
So when I went home, I started restricting the airflow..............thought I was doing the right thing!

So began the month of baked roasts..........now that's an oxymoron!! You either bake or you roast, can't do both!

My HG is carbon steel and although I restrict airflow (for a 10°C temp increase) during the first part of 1st crack, it is 100% open from 140°C
and again from the end of the 10°C period, (which is normally 1.20-1.40. minutes long, depending on °RoR) until the end of the roast.
Adopting this method put an end to baked roasts.

The first commercial roaster that I used was a 20kg cast iron drum roaster, the flue damper was only moved from 50% open to 25% open.
I regularly visit a friend who uses a 12kg Probat, also cast iron, his flue damper is never fully open either.

My roasting road trip started with a Gene, I used a technique similar to Barry O' ..... baking was never an issue.
Managing the stretch from 1st to 2nd crack was key to a nice clean roast.

If you look at Matt's excellent work with the corretto roasts you'll see that it is possible to slow the °RoR during the Maillard Reaction
and not adversely affect the roast........ quite the opposite in fact.

I eventually found that it is possible to bake a roast two different ways.
When I do my next baked one, I will post the two different profiles that have the same outcome.
This is the best way to explain the process without getting so wordy.

Sidewayss (post#112) highlights a really important dynamic. Learn to anticipate, rather than react, so that you are always in control.

[trickydicky2]

I also use a Gene...

Wow! Thanks Barry. My current profile is typically this:
- Preheat empty roaster 160C/5m. (While I weigh out green beans, write notes, etc.)
- Load roaster, set timer to 20m
- 160C/4m
- 245C/~4m (until beans are orange)
- 220C/~2m (until beans are tan)
- 240C/~4m (until about 1 min into 1st crack)
- 220C/~2m (until desired roast colour achieved)

Overall time ends up between 15-17 mins depending on the bean and on roast level. When doing back to back roasts I recondition at 160 for 5 mins so that all roasts start at the same roaster temp

For the GB entries I used a different profile where the thermostat was increased from 150 to 240 in gradual increments up to first crack then backed off to 220 for last couple of minutes. The new profile seems to be a bit better (brighter and more aromatic), but at the moment I'm not sure I am recognising the baked flavour so it's hard to comment on whether that's improved

I tried the new profile firstly based on seeing a number of comments around the web on having an initial cool "drying phase" in a Gene roaster, then getting a tip from another roaster to make sure bean is heated as quickly as possible from green to orange roast colour, then to ease back a little to allow development time from orange to tan, then crank up to get through first crack, and finally ease off a little to stretch the roast between 1st and 2nd crack. There's a lot of guesswork going on here! I will give your profile a go to see how it compares. I like that it's a lot simpler than mine!

I would be much happier if I could connect the roast profiles back to some form of theory of roasting at a chemical level... what are the positive chemical changes we are trying to enhance and the negative ones we are trying to suppress. Is there a digestible resource for this out there...?

[Barry O'Speedwagon]

Thanks for all the replies on the baked flavour question. Some clear ideas to work with around adjustment of roast profile approaching and following first crack

With a Gene you don't get to know the bean temp. However I can see the temp set point and the actual air temp leaving the roast chamber. What you do get with the Gene is a great view of the roasting process - colour development and amount of chaff being produced, plus of course the aroma of the exhaust gases
1

I also use a Gene, tricky. My present approach based on non-SHB is to preheat to 150degrees (takes no time). Whack the beans in, set temp (actually thermostat limit) to 236 degrees, put plenty of time on timer, hit go. When first crack has definitely started (err on the side of waiting) then drop the temp by 3 degrees, wait 20 seconds and if there's still lots of happy cracking action, drop by a further 2. Second crack can be anywhere from 2 1/2 minutes (the demon Tanz Machare) or 3 1/2 minutes later....I try to pull just before second crack for most beans. For 250g of beans, I hit FC at about 10 minutes 30 in current climate. For hard beans I set the temperature at 238, but everything else the same.

[kwantfm]

chokkidog,

Great set of posts thanks. I've found that acidity is dependent on both degree of roast as well as speed of roast. You don't find that a slower roast to the same final bean temp is less acidic?

T

[deegee]

I went through a period, following a 'roasting workshop', where my roasting seriously fell off the rails. As I was doing what I had recently been taught, I didn't know what was going on. The 'new' technique was supposed to improve the coffee!

One night, when the house was filled with this stench from the grinder and I was almost in tears of despair, I decided to take some samples and go
cupping with some industry people I respect and who I thought (hoped) could identify the problem and help me resolve it. I mean ........... my coffee stank, it was awful. Well, I got sorted pretty quickly after that. The advice and information was spot on accurate and I turned things around in 1 roast, when I got back home.

A very interesting & informative post chokkidog. I would like more details of the part above i.e. what was the new technique you got from the workshop, and what was the advice that allowed you to turn it around and get back on track ??.

At one stage of my learning curve I became overly concerned with stretching my overall roast times, and the time between 1C and 2C. I baked a couple of batches of beans trying to do that. I can't say I remember the taste of "sweaty saddles", but they were quite ugly in the cup.

Fortunately one of your previous posts and some info on Sweet Maria's put a stop to that nonsense, and got me back on course. But I still have lots of learning curve left, so I'm interested in what pushed you off the rails, and what got you back on again.

Thanks, deegee.

[trickydicky2]

Thanks for all the replies on the baked flavour question. Some clear ideas to work with around adjustment of roast profile approaching and following first crack

With a Gene you don't get to know the bean temp. However I can see the temp set point and the actual air temp leaving the roast chamber. What you do get with the Gene is a great view of the roasting process - colour development and amount of chaff being produced, plus of course the aroma of the exhaust gases

Maybe I should try a few roasts based on ideas above AND a few doing the exact opposite to see if I can reduce/increase the mysterious baked flavours

sidewayss - What coffee did you score the 54.5 with. My best was a blend in the espresso class:
'Moloko'
Brazil Serra Los Crioulos COE
Kenya Gachami AA
Panama Santa Teresa La loma lot 91

[sidewayss]

Trickydicky,

My take is you have backed off too much at first crack to compensate. Suggest you progressively start backing off heat in steps as first crack approaches so that you get a smooth tapering off on the roasting curve, leading to a steady but slower rise per minute, that way you are in control and not in a panic with arms flapping.
You want to back off certainly to create a gap between first and second crack, but you want to still introduce heat during the backing off.
The trick is anticipation and planning ahead, that will come from experience, trial and error as well as taking lots of notes in regards to time and heat input on your device.
Most of us have been in your shoes and learnt from our own mistakes as well as gaining advice from others.

My score was 54.5. This year's feedback was more positive than last year's.