I'm trying to calibrate my approach on a Skywalker V2, the previous batch did taste good but I dont remember the data. What's your typical total roast time for a washed Ethiopian Yirgacheffe, and what's the highest RoR you're comfortable seeing early in the roast? Mine finished in just under 7 minutes, so I'm wondering if I should be backing off the heat earlier. The beans look and smell good though 😋
I've been trying to understand consistency between roasts, comparing roast history and slowly learning how different coffees behave. Hopefully this book will help fill in a few of the gaps. For those who've read The Artisan Roaster, what chapter or concept had the biggest impact on the way you roast? Happy to get any books recommend on modulation etc.
I've been reading quite a bit about roast modulation and consistency over the last few weeks, so before putting more coffee through the roaster this week I started comparing a few simulated roast profiles.
The funny thing is... I'm not actually sure how I should be reading them yet.
I can compare roast curves, RoR, burner changes and phase distributions, but I'm still trying to figure out which differences are actually meaningful and which ones I should ignore.
When you're comparing multiple roasts of the same coffee, what do you look at first?
I'm trying to become more systematic rather than changing one variable and hoping for the best. Any advice from more experienced roasters would be much appreciated. ☕
I'm still learning and I think I've been marking FC too early.
If you hear 1-3 isolated pops, do you mark FCSTART immediately, or do you wait until the crack becomes consistent? I recently logged a roast with:
Dry End: 3:51 FCSTART: 5:40 Drop: 7:51
Which resulted in a 27% development ratio. The beans look more like a City/City+ roast, so now I'm wondering whether I marked first crack too early and artificially inflated the development time.
What's your rule of thumb for calling FCSTART?
#1 Brazil Santos - burnt it 😅 7 days later the burnt taste is mostly gone and it tastes surprisingly similar to some of the commercial coffees I used to drink before getting into specialty coffee.
#2 Colombia - tasted it after 2 days because apparently I have no patience. Need to redo this one. My family deserves better 😂
#3 Ethiopian Yirgacheffe - best roast so far. First one that genuinely made me think, “okay, maybe I’m starting to understand this.”
#4 Ethiopian Djimmah - smells amazing, but the roast came out a bit uneven. Some beans are lighter, some darker. I tasted it this morning and it was still quite acidic, so I’m curious to see how it develops over the next few days.
A few lessons I’ve learned so far:
* Don’t panic.
* The graph is useful, but the cup tells the truth.
* Every machine has its own personality.
* Let the beans degas before judging the roast.
* Change one thing at a time or you’ll have no idea what actually improved.
Roasting feels a bit like learning to swim. Every time I think I’ve figured something out, I discover three more things I don’t understand 😂
What’s the biggest lesson from your first few roasts? And do you remember your very first roast? What was it like?
#1 Brazil Santos - burnt it 😅
7 days later the burnt taste is mostly gone and it tastes a bit like the commercial robusta-style coffee from the supermarket.
#2 Colombia - tasted it after 2 days because apparently I have no patience.
Need to redo this one. My family deserves better 😂
#3 Ethiopian Yirgacheffe - best roast so far - I'm proud of it!!
First one that genuinely tasted like something I’d happily buy from a trusted specialty roastery.
#4 Ethiopian Djimmah - smells amazing but uneven roast some beans are lighter and some darker. I tasted it this morning and it does taste quiet acidic - I might need to invest into PID espresso machine to pull those light roasts though
Trying very hard not to repeat mistake #2.
Lessons:
Don’t panic.
The graph is not the coffee, cup tells the truth about your roast.
Every machine lies differently. skywalker v2 hardware has it's personality
Let the beans degas. Too much CO2 will taste bitter if you pull your shot after 1 day of roasting
AI, Artisan and roastline.ai can help, but the cup still gets the final vote.
To me roasting feels a bit like learning to swim. Every time I think I’ve figured it out, I discover three more things I don’t understand 😂
What’s the biggest lesson from your first few roasts? And do you remember your very first roast? What was it like? What machine do you use and how you deal with it?
Probably my nicest roast so far and, more importantly, the weight loss finally looks realistic 😅
400g → 347g (~13.25%)
One thing I’m noticing though is that the beans aren’t roasting completely evenly. Some are noticeably lighter, some darker. I saw the same thing on my Colombian roast as well.
I’m running drum speed around 60-75% and this was a 400g batch. Funny enough, the most visually consistent roast I’ve had so far was my very first 200g Brazil Santos… which also happened to be the batch where I slightly burnt some beans. So apparently my choices are consistency or not setting the coffee on fire 😂 I’ve been comparing the roast logs in Roastline and what’s confusing me is that the graphs keep looking better while the bean consistency doesn’t seem to be improving at the same rate.
I’m wondering:
There’s also still quite a bit of chaff attached to some of the beans, although maybe that’s normal because I dropped fairly light.
Overall I feel like I’m slowly getting the hang of this machine. Every roast teaches me something new, usually right after it teaches me that I don’t know what I’m doing ☕🔥
Took a lot of advice from you guys after my first slightly burnt Brazil Santos 😅 Less burner and airflow definitely seems to work better on this machine.
This time I went with 300g instead of 200g and it felt much easier to control. I also tried lowering drum speed to 60% initially but bumped it back to 75% because sampling was a bit awkward.
One thing that confused me is I started with 300g and ended up with around 225g, which is about 25% weight loss. That seems crazy high so maybe I’ve measured something wrong.
I also logged charge a bit late in Roastline so the roast was actually longer than the chart shows. Looking back I can see a RoR spike during Maillard where I increased the burner because I thought I was losing momentum… probably just me panicking again 😂
Overall though, the beans look much better than roast #1 and the smell is on another level compared to the Brazil Santos.
Curious what you guys think about the chart and whether that weight loss sounds remotely realistic.
I used Davinci Resolve for this with Arri Film Look
Blackmagic Film Gen 5 → ARRI LogC4 LUT
Usually looks wrong: odd contrast, shifted skin tones, strange saturation, because the input colour space/gamma does not match.
PYXIS 12K → CST/colour-managed transform → ARRI-like grade
Better approach. In DaVinci Resolve, use Blackmagic Design Film Gen 5 as input, transform to a working space like DaVinci Wide Gamut/Intermediate, then build an ARRI-style look: softer highlight rolloff, lower digital sharpness, warmer skin bias, gentle halation/grain, and controlled saturation.
PYXIS 12K can get close aesthetically, especially because it has high resolution, RAW, and wide dynamic range, but LogC4 itself is not a magic ARRI look. The ARRI look is sensor response + colour science + highlight behaviour + lensing + exposure + grade.
A bit burnt, but it smells good! Started with 200g ended up with ~170g. I think it was scorched....
First of all, thanks to this community.
I’ve spent the last few weeks reading posts here, watching videos, and seeing so many people start their own roasting journey. It finally pushed me to stop researching and actually buy a roaster.
I recently received a Skywalker V2 (or ITOP V2, depending on where you’re based), and after what felt like a very long wait, it’s finally sitting on my desk ready for its first proper roast.
I’ve attached a photo of the initial setup and preheat.
I’ve got around 4kg of green coffee ready to learn on:
So fortunately I have enough coffee available to make plenty of mistakes.
Nothing too ambitious.
I’d like to learn how to consistently roast around:
and gradually improve my understanding of:
Tomorrow will be my first proper roast, so we’ll see how much theory survives contact with reality.
Looking forward to learning from everyone here and sharing some results soon.
☕🔥 Wish me luck. I have a feeling I’ve just started a very interesting new hobby.
First of all, thanks to this community.
I’ve spent the last few weeks reading posts here, watching videos, and seeing so many people start their own roasting journey. It finally pushed me to stop researching and actually buy a roaster.
I recently received a Skywalker V2 (or ITOP V2, depending on where you’re based), and after what felt like a very long wait, it’s finally sitting on my desk ready for its first proper roast.
I’ve attached a photo of the initial setup and preheat.
A week ago I couldn’t have told you what first crack was.
Now I’m trying to understand:
It’s amazing how deep the rabbit hole goes once you start learning.
One thing that surprised me was looking at roast curves.
I uploaded a roast log and compared it against a profile I was studying, and the differences were much larger than I expected.
The profile I was looking at appeared very smooth and predictable, while the actual roast data looked much more dynamic.
As a complete beginner, it made me realize how much there is to learn about interpreting what’s happening during a roast rather than just looking at a graph.
I’ve got around 4kg of green coffee ready to learn on:
So fortunately I have enough coffee available to make plenty of mistakes.
A big shout-out to RoasterKat and a few other creators whose content convinced me to stop overthinking and actually start roasting.
Nothing too ambitious.
I’d like to learn how to consistently roast around:
and gradually improve my understanding of:
Tomorrow will be my first proper roast, so we’ll see how much theory survives contact with reality.
Looking forward to learning from everyone here and sharing some results soon.
☕🔥 Wish me luck. I have a feeling I’ve just started a very interesting new hobby.
First of all, thanks to this community.
I’ve spent the last few weeks reading through posts here, watching videos, and seeing so many people start their own roasting journey. It finally pushed me to stop researching and actually buy a roaster.
I recently received a Skywalker V2 (or ITOP V2, depending on where you’re based). The box arrived from Hong Kong and while the advertised price looked attractive, I quickly learned that wasn’t the final cost.
So if you’re in the UK and considering one, factor that into your budget.
I’ve attached a photo of the initial setup and preheat.
I’ve been playing around with Roastline.ai because I’m not at the stage where I can justify Cropster yet.
I uploaded a roast log into the software and compared the actual roast against a simulated profile.
The simulation showed a very smooth and controlled process. However, when I compared it to a Colombian roast log from a Skywalker V2, the curves looked quite different from what I expected.
The preheat phase looked reasonable, but after that I started wondering:
As someone who only recently learned what drying, Maillard, development time, and first crack actually mean, it was a little intimidating.
At the moment, I have around 4kg of green coffee ready to roast:
So there’s plenty of room for mistakes and learning.
A week ago I couldn’t have told you what first crack was. Now I’m reading roast curves and trying to understand what they’re telling me.
One thing that really caught my attention was the cost of green beans.
I recently visited a speciality cafe and noticed an Ethiopian Yirgacheffe green beans selling for around £17 for 250g.
Meanwhile, I bought 1kg of green Ethiopian Yirgacheffe for roughly £17.
At first glance, roasting almost feels like a cheat code.
Obviously, the more I learn, the more I realise retail pricing includes far more than the cost of beans:
And of course there’s roast loss, where you lose around 12–18% of bean weight during roasting.
Still, it was one of those moments where I understood why so many people get hooked on home roasting. Even if my coffee ends up being only “pretty good” for a while, the economics seem surprisingly attractive compared to buying specialty coffee every week.
A big shout-out to RoasterKat on Youtube. Her videos were one of the reasons I finally stopped endlessly researching and actually ordered a machine.
Nothing fancy.
For now I’d like to learn how to consistently roast around:
and gradually understand:
Tomorrow will be my first proper roast, so we’ll see how much theory survives contact with reality.
Looking forward to learning from everyone here and sharing some results soon.
☕🔥 Wish me luck. I have a feeling this might become an expensive new obsession. 😅