Could you put up links to some of these? The links you put above also don’t work. (Remember to click on Actions->Share to get a usable link.) I’d be interested to see how the bad roasts compared to the good ones in the roasting curves.
Congrats on your Africans cupping better! Esp that dur Feres - it’s a fav.
@itspaul you can find @omegacoffeecollectiv on RW here Roast World - Cup, grade, and analyze your coffee roasts in depth and possibly match up the roasts with the dates mentioned in the post. Only 15 roasts logged so might be easy to “find”
Nice report! Lots of information there.
It’s surprising that all of your Honduras roasts tasted the same. Looks like you carried too much power into FC in the first roast, but changed this in Hon2 and Hon3, each with progressively more development time and correspondingly more weight loss. Hon3 is still pretty light though. Maybe that bean just doesn’t lend itself to a light roast? Are you thinking of extending Hon3 for another 15-30 seconds? Would be interesting to see if that alone does the trick, or if you need more post-FC temperature too. How’s the chaff level in the roasting chamber when you finish with only F3?
1 Like
The chaff level is not too bad at all and I think this may be due to my lighter roasting style. My latest versions of Honduras I chose to speed up drying, lengthen maillard, slightly lengthen dev time and go for a slightly higher final temp. I think this roast style will produce more of the caramel and toffee sweetness I want to pull from this coffee by lengthening the Maillard phase to 47ish %. I will analyze how the coffee is with this style at 410 drop temp and decide to roast more versions at higher temps or leave it at 410. Will report back on findings and further analysis shortly!
Ha, didn’t notice you already roasted 3 more tries on April 2. Looks like you bumped up the preheat 5C, eliminated F3 entirely, and progressively cut power earlier to achieve exactly the goal you described. If one could drink the curves you’d have nailed it. Any signs of tipping or scorching? Fingers crossed.
Zero signs of tipping or sorching, i did that particular adjustment because I find this bean to be very dense AND hard instead of an Ethiopian natural that may be dense and soft. As such it handled the heat quite well. Currently gassing off and revealing the flavors. It’s already quite fragrant so I’m getting quite excited!!!
My experience with Ethiopian is that they tend to roast a bit “faster” than Yemen even if I’m using the same recipe. I hit the yellowing faster and FC a touch sooner (up to a minute earlier for my 1kg batches). I’ve not really paid much attention to density and hardness of the beans but seems like you’re on to something here. 
1 Like
This thread is my favorite thread on this forum to date. Thanks to all of the participants and contributors. I am learning a lot.
I have a question for @Omega: How are you determining that the bean is hard?
I understand how a person can do a water displacement test and measure density, but I have no idea how to measure hardness of a green coffee seed.
Also, are you eventually going to scale up the size of the roast? I would be very interested in how you end up going about this. This is a common and recurring question on the forum.
Thanks for sharing.
Great Question Bill!
When discussing bean hardness I mean it in a relative manner. The hardness is arbitrary but the point is comparatively some beans are harder than others. Generally I find that the more intense the processing method(crazy anaerobic fermentation, super long natural processes, carbonic macerations, ect) the much less hard the bean. The more traditional the wash the harder the bean and generally the central and south americans can be harder in my opinion than africans. But that is just my experience. I can easily tell this by how easily the bean takes on the heat and how quickly I can progress through all the various stages. Some of my intense naturals roast extremely easily and actually require restraint to roast correctly. Some of my central americans feel like they take the power of the sun itself to roast as quickly as I desire. It may be totally flawed but I consider that to be hardness.
In regards to your batch question. This has been a difficult matter with which I have been wrestling for some time. My plan is to dial in all coffees at 400g batches because I sell 12oz bags of coffee and a 400g batch leaves me with almost exactly 1 bag. Then I will replicate the curves with 800g batches which will provide me with 2 bags per roast. To replicate the 400g curve I will do the following: increase charge temp, run higher power settings for longer, and possibly augment drum speed to allow for faster heat transfer to the beans conductively.
We’ll see how my 800g plan works out but first I want to fully tackle my 400g plan.
Some clarification for you all on my situation. First I want to thank you all for the incredible help you have been on this thread and all the other threads some of you have participated on. I should be clear that I have some of my roasts hidden for a variety of reasons but partially so that you guys can see the roasts I’m talking about and not wade through the junk. It may appear as though I have achieved all of this growth and understanding in 20 roasts. I have just crossed 120 roasts on the bullet at this point so a lot of this learning has been informed by my 100 roasts that were trash, although some of those were replications of a coffee I nailed early on. So if you are early in your roasting journey with the bullet keep pressing forward, it is absolutely worth it. Also I think that when this thread has been completed I will post pictures and descriptions of all of the coffees and the final roast profiles so that others can use that resource. This may prove helpful because some of my coffees will be very lightly roasted and clear, while one or two may be more traditional and developed. Again thank you all for the journey. I hope every one of you brews something tasty today!
1 Like
Thanks for the explanations
So, in regards to the hardness: Decaf coffees beans are typically going to be less hard because the way they are processed and saturated with whatever solvent they are using. This makes sense to me and I get that impression when I roast them. I just hadn’t thought of them in those terms. I hadn’t separated density from hardness in my mind.
Also: Thanks for the insight into why you chose to do 400g batches.
I have been working on trying to come up with some generic, starting point, baseline 500g roast batches to not waste beans and have been exploring some of the variables. Drum speed and other things.
The decaf analogy is the perfect way to think of it!! I only use this terminology to help myself better understand roasting and better be able to communicate why I roast the way I do.
I would definitely say you could 400g-500g and get great batches. Don’t let anyone fool you into thinking this roaster is only good at 600-800g.
If you want a fun experiment try getting a great roast and curve with just the power setting, then try it by using lower power and just fan. This won’t produce great coffee but it will really improve your understanding of how to think about how each of the controls adjusts things and critically how quickly it adjusts them.
1 Like
In several tastings they have proven to be flat regardless of brew method or extraction levels. Interior of the bean still appears a bit green. Taro 1 and 2 from 4/4 seem best but still are not tasting great. Any suggestions of how to analyze the issues and adjust? Or experiment to find a solution?
1 Like
Just my 2 cents, if the interior of the bean still appears a bit green or the cupping appears grassy, to me it seems to indicate that the roasting may have been too fast (the drying phase is too short maybe).
I’m seeing 4 Taro roasts: Taro1 (3/25), Taro1 (4/4), Taro2 (4/4), Taro3 (4/4).
Comparing the Taro1 roasts, looks like you cut power sooner on 4/4, stretching the mid-phase, eliminating a small post-FC crash and maintaining a higher RoR. This allowed you to reach a similar drop temperature with 40 seconds less development time, and reduced weight loss by 0.5%. Did the 4/4 Taro1 have more brightness, in spite of being perhaps underdeveloped? One oddity is that the 4/4 BT began pretty low. Was that the first roast, or some different between batch protocol?
Your Taro2 used 220C preheat, which steepened the overall RoR and got you to FC sooner. You also managed to cut power to P4, so your dev time increased to almost 1:30 but drop temperature decreased slightly. Weight loss back up to 13%. This was a hotter, quicker roast with more development time. But it was still under developed?
Taro3 cut power slightly earlier than Taro2, but for some reason carried a little more energy and allowed you to shorten the roast by 25 seconds with similar development.
One thing to try, if you want to experiment, would be to start with D5. Lower drum speed gets more heat into the beans and minimizes the cooling effect of the fan. If the bean could take that much heat, try the Taro2 or Taro3 profile with D5 until yellow, D7 mid-phase, and D9 around FC? Might get more development of the interior of the bean this way. Watch out for scorching!
Edit: I’ve also seen more uniform development in light roasts by moving to larger batches. Maybe now is the time to try 800?? 
Sorry everyone, life got crazy and I forgot to reply so I’m revisiting this thread now.
Thank you all for all of the help!!!
Since last post I have found that my Taro beans were in fact suffering from age and moisture loss. As such, I used a slower overall profile with :30 more dev time and found success in a richer and slightly less vibrant version of the original.
I have since then dialed in 2 anaerobic natural Colombians that caused me to learn a-lot. The resulting profiles are about 8:35 in length with me marking yellowing around 4:00 and FC around 7:30 with about 1 minute of dev time.
This resulted in a coffee that had beautiful fruity qualities while balancing the line of rich and complex.
If any of these profiles interest you I can send them in a reply
I have also found great success transitioning my 400g roasts to 800g when demand calls for it. I have does this by drastically increasing my preheat temp, removing my :50-1:00 low heat drying phase, and reducing drum speed to 7. This results in a roast that follows my 400g roast curve well and in indistinguishable from a 400g roast on the cupping table.
Through all of this (I’m closing in on 450 roasts now, which is still barely any)I have built a framework that works well for me in achieving my roasting goals. Personally I seek to balance the origin specific qualities of a coffee with it’s potential sweetness and richness. Sometimes this comes at a cost of reducing perceived acidity by 10-25% but for me it produces incredible cups in both filter and espresso applications.
I follow these basic rules for my 400g roasts and I dial all roasts in at 400g to prevent waste as some of my beans are $15 per lb
:30-1:00 drying phase in the beginning at P6. I modulate this depending on if the bean is an exotic process. More exotic the process the longer the drying phase. On the cupping table for my palate this has helped preserve complexity in the final cup.
Increase heat to P8(exotic) or P9 for 1:30 then P8(washed) until 329° which is where I mark yellowing. Not sure if a fixed yellowing is good or not but it works awesome for me.
Decrease heat to P6(exotic) or P7(washed) about :30 after yellowing.
Ride that til 1st crack around 7:30-8
Begin slowly decreasing power and increasing fan speed just before 394° without crashing the roast.
Adjust dev time depending on the final cup characteristics I want but usually have an end temp between 406-412 for my personal palate.
All of these are completely adjustable and I adjust them frequently. While using the above system I constantly try to adjust the roast as little as possible. I have found that for my taste preferences the less I adjust with the roast the better they turn out. Some of my roasts are even just 2 power adjustments and that’s it and it produces incredible results. I often feel as though over adjusting causes the need for more adjustments and that can just create a lackluster bean. For me letting the bean roast how it wants to seems to go quite well.
Through the process of developing this framework I have learned a ton. Please remember this is all descriptive and NOT prescriptive. Don’t just go copying this and assume it will get you what you want. I am roasting for my particular desires and my customer needs which is to highlight sweetness and maintain origin characteristics, while still having some acidity, and being able to use the coffee for filter pour overs as well as 1:3 ratio espresso shots.
Thank you all for you help and wisdom. If you have any desire to see some curves, just ask and I will provide.
Blessings, and remember drink what you love🖤
3 Likes
Blessings to you too @omegacoffeecollectiv and thank you for sharing.
I would be very interested in the profiles you mentioned if you would be happy to share them please, especially the ones for the 400g size batches.
Thanks again
Gaby
Here are the two Anaerobic Naturals from Colombia. Both decently light roasts focusing on higher extraction filters brews trying to emphasize sweetness and richness while still keeping a little acidity in there for fun. They were a little wild to control but ended up tasting real good.
