Philip P. Ide

Author, programmer, science enthusiast, half-wit.
Life is sweet. Have you tasted it lately?

User Tools

Site Tools


Coffee Blog: 2022-07-24

Update: Profile H tested, see below

I've been creating roasting profiles, and recently changed from a numeric system to an alphabetic. I don't know why, perhaps the numbers were making me depressed at how many times I'd done it and still not reached a particularly good roast other by accident.

Anyway, I decided to do two roasts this weekend. One I called 'Profile G' and the other 'Profile H'. Here's the first one:

Profile G

Profile G
Load 200g
Bean Columbian Bucaramanga
Pre-heat 200°C
Stage 1 (to yellow) Drying 200°C 6:00
Stage 2 (to 1st crack) Caramelising 245°C 10:50
Stage 3 (to 2nd crack) Development 225°C 12:50
End 12:50

Firstly, I should say that things didn't go to plan right from the start. My Gene Café has a maximum load of 250g or to the maximum load line, whichever is reached first. I knew I was going to roast 400g, and for some reason - me not thinking mainly - I measured out 400g. So although the recipe called for 200g, I used 400g.

I have previously taken 200g recipes (made by other people) and tried them and they worked. Not brilliant (every machine and location is different), but they worked. When I tried them with 250g loads, they failed. Badly. Strangely, this recipe worked, after a fashion. The main thing is, it didn't break the roaster and there were no fires. I think the machine did an admirable job of reaching the temperatures asked of it - but the result was some of the beans were burnt. I had to spend 30 minutes after picking out the burnt and excessively dark beans.

As usual, I made a drink in a French Press. Doing this so soon after roasting is likely to leave a slightly sour and a burnt taste, so you have to ignore those flavours and see what else is there. Over the next seven days, the sour and especially the burnt taste will diminish as the CO2 outgasses. To my surprise, the coffee had plenty of acid flavours. I was onto something!

Other than that, I'm going to ignore the first roast because of the excessive load which I won't repeat in case it damages my machine - and I don't want to spend thirty minutes of my life picking out burnt beans. But the fact it somehow worked, after a fashion, is interesting.

Profile H

Profile H
Load 200g
Bean Columbian Bucaramanga
Pre-heat 200°C
Stage 1 (to yellow) Drying 200°C 6:00
Stage 2 (to 1st crack) Caramelising 245°C 10:00
Stage 3 Development 225°C 13:00
Stage 4 (to 2nd crack) Development 235°C 16:00
End 16:00

In this recipe, because I used half the amount of beans they reached the target stages quicker (and more evenly). Dropping the temperature to 225°C in the third stage had much more of a dramatic effect for the same reason, so after three minutes in stage 3 I pumped the temperature up to 235°C to bring the beans to 2nd crack at 17 minutes.

The first stage, to yellow, happens in a nice 6 mins. This is a perfectly respectable time.

The second stage should be 60% of the first stage (30% overall), equating to 3.6 mins (216 secs). The actual timing was 4 mins (240 secs). That's actually a good approximation. Next time, if I pump the temperature up to 250, I might hit that 216 second mark more accurately.

The third stage should be 40% of the first stage (20% overall), which should equate to 2.4mins (144 secs). As you can see, it was taking too long, so after 3 mins I pumped it up to 235°C and creating a 4th stage in the process. Combined, the last two stages took 6 mins.

I made myself a drink in a French Press. This was much better! Despite the overly long development phase, it still had some nice flavours.

Based on these results, this is what Profile I is going to look like:

Profile I
Load 200g
Pre-heat 200°C
Stage 1 (to yellow) Drying 200°C
Stage 2 (to 1st crack) Caramelising 250°C
Stage 3 Development 240°C

That should give me some leeway on the last stage to raise or lower the temperature to bring it to 144 secs, but that tweak will occur on Profile J. However, the Columbian I've got left is El Carmen. I roasted El Carmen before, and it's a lovely coffee, but because it's different I must be prepared to adjust the recipes accordingly.


It's been a week since I roasted Profile H, so it was with a bit of excitement and a bit of trepidation, I made a double-shot of espresso in my Rok EspressoGC. It came out with a lovely head of crema, which is both expected and important. The crema is an emulsion of the oils from the coffee and carbon-dioxide (which is present in roasted coffee in large quantities). The crema is - or should be - where all the bitterness lives. The coffee itself was smooth to taste, and full of flavour. I am very pleased with this roast. The flavour could be improved, especially for French Press or Americano, but this profile is a solid place to start tweaking.

This website uses cookies. By using the website, you agree with storing cookies on your computer. Also you acknowledge that you have read and understand our Privacy Policy. If you do not agree leave the website.More information about cookies
blog/coffee_blog/2022-07-24.txt · Last modified: 2022/07/29 08:19 by Phil Ide

Except where otherwise noted, content on this wiki is licensed under the following license: Copyright © Phil Ide
Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki