@bradm yes indeed, Scott Roast 1 is the undrinkable one 
Scott, Glad you got a drinkable roast. I think monsooned coffee is a challenging one to start with.
2 Likes
@wngsprd thanks for your thumbs-up. I’d heard MM was a bit tricky to roast and i like a challenge. Now, it’s drinkable though not in any way amazing. I have got some Kenyan and Panama beans to roast next. Any experience with those?
@wngsprd @bradm STOP PRESS! Guys, you’re not going to believe this! So, I kept all three roasts in an airtight container, opening the lid every day to allow to degas. And this morning I was going to throw out the “Scott Roast 1” - this is the one that tasted horrible at cupping +24hrs after roasting. so I ground 18g for an Aeropress and … blimey! that’s a pleasant cup of coffee! What’s going on? It seems to have developed itself over the past week into a bright, mellow coffee. I’m looking at the curve again Any comments?
Seems like “Never give up!” is correct. I notice a big improvement with some coffees as a day or even a week goes by. If I’m going to drink a roast within the first 24 hours, I grind the beans and let them sit out for a couple of hours before the pour over.
Glad to hear your roast isn’t a throw-away. Some of mine are good immediately, but lately I’ve been trying to let 'em rest for a week. Makes you wonder how good one of the ones that you drank up right away could have become.
Hi, I am also a new owner of the Bullet R1 V2. As every new owner very excited, even can barely sleep at night (or is it too much coffee!?). For my first roast I picked a Colombian Supremo (size 18) and followed to the letter the recipe “A Beginner’s Roast Recipe” in the article “How to Begin Roasting Great Coffee on the Bullet R1 V2” (light roast) on the aillio website or owner manual (I think it’s there too). I got great success, clean cup, well balanced (fruit-acidity), good coffee taste, amazing smell. That was 2 days ago and I cannot post the graph as I couldn’t connect to roastime at that time (driver problem!). I reapeted the recipe with a Brazilian with the same success (slightly less developed). Now the problem is fixed and I will experiment using the roastime software. I agree with wngsprd that a Moonsoon Malabar is not an easy roast (whatever the roaster you have). I say that based on my 4 years experience roasting with a Gene cafe.
Anyway, take what I am going to write lightly as I have virtually no experience yet with the Bullet. Looking at your graph though and based on my experience with the Gene, your RoR is too high at first crack and more importantly does not decline constantly enough during the yellow stage. You say the first roast is not good and I am not that surprised given your RoR rise just before the first crack (ie right after 8:30). With the Gene cafe, I got excellent roast with RoR between 6 and 9 at first crack. My readings on success roasts with the bullet showed RoR at first crack around that also. To be explored I guess.
My guess is that you put the fan to F3 too late in the early stage. My understanding (and I might be wrong) is that we should not increase the fan after 120c and before the first crack. After 120c (near 3 min in your curve), increasing the fan will have a convection effect, potentially increasing the RoR or minimally preventing it to decrease. when I look to your first graph, that is what happen when your turn the fan to F3 near the minute 6. I have read in the manual that above f4 it has a cooling effect instead of convection effect. That probably explain the too fast RoR decline when you hit f4 around 8 min. I have read that a good time to increase fan speed (from f3 to f4 for example) would be 20 to 30 seconds before first crack. that would have been right after minute 9 in your first graph. However, if you modify the parameters at the early stage it will certainly affect the time of the FC. So I guess that is what we have to experiment (ie the right recipe for a specific bean and environment). I also read that it is preferable and easier to control the RoR with decreasing the power heat than the fan. I will try to put F2 to F3 at 120c and not touch it just before the first crack, so I guess I will have to experiment before being able to do that.
So all that is theoretical right now (from my readings) so take that for what it worth. However, I am pretty sure that the problem with your curves here are within the maillard stage (ie from yellow to FC). Today I will try the recommended recipe but this time with the roastime software 9to see if the curves correspond to what I taste). Sorry if my english is not top notch as this is my second language.
1 Like
My 4 years experience with the Gene cafe roaster told me that a success roast taste always good right after the roast and improve with time. and a great roast taste amazing right away. I found that some beans are even better after 4-5 days, most after 7 days and some reach their peak at 14+ days. When I visited the Starbuck R&D facility in Seatle in 2016 the master roaster said they let the bean degas for 7 days before distributing. Not that I am a fan myself of starbuck coffee (otherwise I would not roast myself!) but they are still professionals with many years of experience. I have had mediocre roasts over the years turning into acceptable roast after few days.
Hi Again,
From theory to practice. I roasted a Colombian following the suggested recipe in the owner manual without trying to adjust anything (ie not looking at the curve to make adjustments). You can see the curve here.
It gave a decent cup but to me it is clearly underdeveloped. I roasted the same kind of beans in my Gene Cafe and it is soooo good compared to what I got with the Bullet. I noticed that the development time is quite high relative to the 20-25% everyone is talking about. (with the gene I always got within this range without even trying to). To me this batch tastes coffee with underdeveloped notes. So not totally to throw away but I cannot say it gives me much pleasure. When I crush a bean it smells good (but not close to what the bean smells with the Gene). So I guess if I would have stopped to get 25% development (instead of 37%) the coffee would have been undrinkable. I do not know what I have done wrong as the RoR does not look so bad. what do you think? If I raise the charge temp, the first crack comes much too early but the bean is a bit more developed (but too dark for my taste).
I also tried with many 200g charges (at different charge temperature) and it is impossible to get a good reading out of it. (and I throw away all batches because they were too green for me; I do not mind fruit and acidity but it has to taste coffee to not just fruit and acidity like many people like). So I guess 350g is really the minimum. I ordered cheap (6$/pound) but good beans from Brazil to experiment further.
@kafcafe Hi, Hello and wow! Thanks v much for taking the time to give me such a thorough reply. Appreciate it. So. Here is my annotated roast curve with your notes added.
If i roast a la Aillio manual, we can see where the curve adjustments would be.
You explain the effect of F3 and F4 in the curve - cool.
And you describe well the ROR at First Crack - hadn’t known that before.
I’ll also make a roast following the Aillio manual as you did.
Which brings me on to my next questions for us: How do we plan a roast recipe with this knowledge?
- What decides the Charge Temp?
- What decides the End Temp (Bean Probe or IBTS)?
- How steep is the declining ROR curve (as an angle)?
- Should the curve ever be horizontal at any time, or is that an indication of baking?
2 Likes
@kafcafe Hi, So here’s what I get from reading your Colombian curve:
- Charge temp seems high. How about 195-205 and go for a slower convection-roast?
- The ROR is steep between 2:44 to 4:24, then when P5 adjustment gets felt it looks to recover to a shallower curve. Perhaps your bean missed some early development in this phase?
- D9 is fast for 350g. I go for D8. Found a calculation of beans tumbling in a drum and the recommended speed equates to D8 on our Bullet.
- Can you put the Yellow Point and First Crack times on your curve pls? These can be added in the settings in RT3.
- Colombians are known to have bright acid and fruit at lighter roast. I like full-bodied coffees with lingering chocolate and caramel flavours, so i’d go long into the development and drop before second crack. Btw, I found this link, which explains the nuance of the Colombian bean: https://www.sabores.co.za/2019/07/22/guide-to-colombian-coffee/
- 350g would be my min batch size. I’ve read of some folks doing 200g batches. It looks like another different technique because of the smaller bean mass and different exothermic conditions. I’m accepting it’s going to cost me a bit of waste to begin with to learn how to roast 350-700g batches! Where do you buy your cheap Brazils from at $6? Give me their number!
Anyways, hope I haven’t confused everything for us…
3 Likes
Hi my undersranding is that
- The charge temp depends of the qty of beens used, their density and % moisture and the profile we try to achieve. I have read also that the altitude you are roasting might have an influence too. I guess it is the knowledge experience will eventually give us.
- I think the end temperature tells us the type of roasts (given probes are precise). I think once you know the bean, using IBTS or the bean probe isto your choice and maybe i would use the one that is the more constant with a specific bean. Anyway people with more experience might have another (better) explanation.
- I am pretty sure that the RoR diminishing after RoR reach a peak is more important than the rate at which it diminishes per say. However if the decline is too steep, you will stall the batch (ie reaching a RoR of 0 before the fc or before the beans have time to develop). That being said i believe more into my taste buds than into curves. I just noticed after 4 years of roasting with the Gene cafe that everytime i would observe a constant declining of the RoR the coffee would taste very superior, top notch! With the Gene i noted the temp each minutes which brings me to your point 4.
- Other might have a different opinion but I looked into my 4 years log book roasting with the Gene, and all the top notch roasts had declining RoR until the end. Of course the RoR with the Gene is not an almost instant RoR like with Roastime, But taken at each minute RoR of top roasts always declined with an end RoR of 1 or 2 (Gene is not precuse eniugh to know). looking at your picture minute per minute the RoR at minute 4 is higher than at minute 5, 5 higher than 6 but 6 not really higher than 7 and 7 not higher than 8. When i would get that with the Gene it would give a dull cup whatever the color of the bean.
Charles
Thank you for your observations on the last post. I agree that 220c is a high charge (i think) but i just followed the owner manual recipe blindly without trying to adjust anything (same for the D9 + in the manual they say to put D9 for the first 10 roasts but i agree D8 is probably better). Today i did not roast but did cupping of all the roasts i did for the last two days. It turned out that the decent cup improved a lot in 24h. To me it is still not enough developed to my taste (as it seems to be your taste too) but at least it tastes coffee and the mouthfeel is richer after 24h. I suspect it would turned better as the crushed beans under my thumb had a good smell right after roasting.
Next time when i roast i will put the yellow mark, in fact i succeded to roast a very good cup with 200g of the columbian, powerful coffee mouthfeel , vanilla, hazelnuts, sweet caramel finish, 0 biterness, the kind i was able to do before the bullet. I am so happy! I will post it tomorrow. I charged it at 160 (or was it 165) cannot see it now on my phone. Frankly if i can replicate that i will be very happy! It was the last batch of the day, so the drum was at 160 but after many roasts at higher temp charge. You will see a constant decline of the RoR. charging at lower temp gave me a slightly longer yellowing time to fc (than the missed batches). However the ratio development is higher than 25%. Not sure of this ratio importance.
I live in Canada and among others i am buying from this place for the last years. The eagle brazil espresso is 6.12$/pound for 2 pounds. This is not the best brazilian i have had (supplier sometimes had small farm beans from minas gerais that are amazing but more around 12-13$/pound):
https://www.pre-umber.ca/products/brazil-eagle-espresso
Charles
So I looked back at the curve for the first roast and I agree that the RoR is probably too steep from 2:44 to 4:24 and I think that is what prevented to have a significant decline just before the fc, from min 6 to 7. So it is not really declining between 6 and 7 but after fc it is generally declining. Not sure how to “fix” that. Maybe by turning the fan to f4 instead of turning the heat to p5 would trigger a less steep decline…Anyway, the general trend is downward and I think this is why the roast still taste good (but not as great as I would like).
In roast time the yellow point and fc points are showing but not in roast world. However in roast world you have this bar above the graph that indirectly tells you the YP and FC.
Anyway here is a screenshot from roastime (is there a way to directly cut and paste the graph in an other way I did? I did a print screen + paint to get it here:
Charles
1 Like
Thanks @bradm, I guess that’s how I interpret things like heat transfer! Also, appreciate the link to seasoning beans tho I’m not sure they’ll deliver to UK.
Btw, I’m posting an update on the rested Mo soon beans now too…
@kafcafe Hey Charles, I am ordering some Columbian and would like to try your 200g profile if that’s ok? Can you post it up soon?
I can edit my roast profile in beta.Roast.World (see photo) that then allows me to adjust the yellowing and FC markers in the time boxes. Try that and see if they show on your profile.
I’ll add new photos of my Monsoon beans in a minute too - got some interesting updates to share!
So This is the image of the first 200g batch success with the columbian supremo (got few scortch, 6 of them on 200g not too bad) but taste is excellent, with slight bitterness but at least it tastes good strong coffee):
this is the image of the playback, I stopped it before the other one and the taste is even better with less scortch ?(3 only and the beans are not as dark) (my wife thought that was really good and comparable to what I was able to do with the Gene):
I also tried the same recipe doing playback (playback works very well!) with Peruvian but it was less developed (however the bean smell very good so I am going to wait some degassing before judging). I think I have a basic recipe that works for dark roast (or at least darker profile), not sure if I really heard the 2nd crack and will try to improve from there. Also I notice that the color of the beans are not always a sign or success/failure. I started a post on smell.
here are the beans of the columbian playback (just tasted it again and I am very happy!):
The playback recipe (which should be the same as the original “colombie dernier”). Let me know if it works for you. I think you can access those under my pseudo Kafcafe.
Roast World
@kafcafe Charles, interesting curves you have there and the beans look quite dark IMO. I prefer a lighted roast, though am finding that the dark roasts seem to mature well during resting. Mine now have oils sweating out that is bringing a dark chocolate kinda jammy fruit bitterness to the cup - I’m more pleased with these results now!
Here are the pictures of them now: you can see some oil spots where my thumb is.
1 Like