false1001 wrote: Context from the other light roast thread is I was initially worried I was spending too much time in the maillard phase, so the goal here was to drop low, maybe soak, and hit with high heat 2 minutes in to maximize drying time and reduce Maillard percentage to ~40% or less. There are two profiles in the What Is A Light Roast thread where I tried a more traditional profile and I was ending up with weak, bitter, and roasty notes with little or no terroir.

I'd like you to try the advice I gave you, just to see what happens. The peru you posted in that thread is a bit odd that you had FC starting at 190c. You were also charging quite hot - its hard to tell how your sensor equates to my original one but on a light roast I'll charge at 190C 195C drum temp with a 155-160 BT.

I just don't think using full power on a batch with half the capacity of the roaster, makes sense. I might try the heatsoak just for the heck of it, with slightly different adjustments from you.

I actually did follow your advice with the fan, as you can see here: https://roast.world/r/b_BtUz4LGDNoR0zehkQSW, and it's been pretty helpful. However as far as the heat settings go I need to charge around 450F (220ish °C) in order to finish a roast in under 11 minutes using p7 and below... and charging that high means I race through drying in 2-3 minutes.

false1001 wrote:Colombia is at work, i'll take a photo tomorrow. Here's the best Burundi roast. The soak has not resulted in good looking roasts, i'll tell you that.

My Burundi looked similar, and I did not soak. I think I have noticed that Burundi seems splotchier than most. However, My Colombia Aponte that I just roasted was also more uneven than I was expecting, and I took it to the end of first crack then dropped.

winslette wrote:I think I have noticed that Burundi seems splotchier than most. .

+1

winslette wrote:My Burundi looked similar, and I did not soak. I think I have noticed that Burundi seems splotchier than most. However, My Colombia Aponte that I just roasted was also more uneven than I was expecting, and I took it to the end of first crack then dropped.

Sorry for the delay, been rather busy recently. As promised:


A little splotchy, but relatively uniform. No noticeable increase of splotchiness from non-soak roasts.

Off topic, but such a wonderful coffee. I usually bring screw up roasts or lower rated coffees to work since I only have an Encore at the office but this coffee delivers regardless of brew method and grinder.

Another addition to the light roast experiment: https://roast.world/r/3NTzWMhxxiCTE3bf2_0Jr

Probably my most tea-like result to date, but not unpleasant by any means. Very subtle, but still noticeable floral notes with some indistinct berry/citrus fruitiness and simple syrup sweetness. What is interesting is that while both acidity and sweetness are muted, the acidity is drastically more muted, resulting in a sweet but rather flat profile. I shared this roast with some friends and the general consensus was that it wasn't "strong enough", but was very interesting to drink. It initially doesn't taste like much, but really sticks with you. I might try to tweak this so I get slightly more acidity without ruining the general delicacy of the flavors.

Recently acquired a Bullet, rather impulsively, but really having fun with it thus far. Initial roasts went well, I used some of the profiles posted in this thread as a rough template. Here's a rather aggressive light roast of a natural Eth that I'm drinking this morning: https://roast.world/r/8YtGIwgM_Dx8YjF2QnPf~. There's always room for improvement, but as a starting point this one turned out super bright and fruity.

The lightest I have roasted is a drop @ IBTS showing 203. This is very aggressive indeed. Did you make espresso with this roast? or drip?

I tried finding a good Bullet topic to resurrect that would be useful for people who google or site search. A problem with my roasts from the Bullet was that they used to taste good for a few days, then would become hollow and dull. I read someone's post about roasting on the Bullet who had tried Alan's advice of using constant fan and altering heat only. I decided to give this method a go since I was trying to find a solution to my problems. While the original poster used F5 constantly, I found F4 worked best for me. My curves still weren't perfect, but my coffee does not become hollow after a few days anymore. My old fan routine would be to start at something like F2 and go to F3, then F4 at first crack. My time spent in maillard was low compared to time to yellow, which is what I am assuming at this point is the reason my roasts taste better and stay better. The higher constant fan really helped extend my time post yellow. Hoping this serves as a data point for Bullet roasters who are trying to diagnose their roasts. I still have a ways to go, but this has helped a lot.

This is a bit funny since Alan and I who both advocated one fan setting on our drums, are now experimenting with a low start and one adjustment between then and DE. Maybe starting two low was the issue on the Bullet.