That's interesting, DJ, and certainly goes against what is becoming the conventional wisdom these days. Have you done a blind, side-by-side comparison of the two approaches? If so, I'm sure we would all appreciate it if you could share the profiles you compared and your tasting notes regarding the differences.

thepilgrimsdream wrote:I had recently spoken to a notable roaster about underdeveloped coffees, he actually recommended increasing airflow for the full roast, charging much lower and increasing the gas/power about 20-30 degrees before first crack and to enter with at about 20-22f/ror, and then tapering off as you enter a rolling first crack. I've cut my development back to 1:00-2:00 because FC happens with such good momentum

Deaton Pigot speaks about the same thing in the Cat and Cloud podcast.

I've been doing this for about 3 weeks and my coffees is much sweet, cleaner and flavorful than ever before.

Honestly I feel like trying to follow Rao and Joe's techniques have led me astray.

This is a slow start fast finish (ssff) vs the fast start slow finish (fssf) of rao and Joe. These are the two basic styles of roasting, with of course tweaks along the way on both sides. Jim Schulman has poste references in that regard in the past. Both approaches are capable of flavorful coffee. Personal preference and roaster specifics may influence your choice, but trying both approaches is very useful. I'm currently doing fssf, but this reminds me to experiment with ssff.

You can do SSFF without increasing heat during ramp, and with either high or low airflow, so the specific package of recommendations DJ is making does seem a bit different. Either way, I'd be interested to see a more detailed write-up of the profiling approach he has become such a fan of, and what he is comparing it to.

thepilgrimsdream wrote:I had recently spoken to a notable roaster about underdeveloped coffees, he actually recommended increasing airflow for the full roast, charging much lower and increasing the gas/power about 20-30 degrees before first crack and to enter with at about 20-22f/ror, and then tapering off as you enter a rolling first crack. I've cut my development back to 1:00-2:00 because FC happens with such good momentum

Deaton Pigot speaks about the same thing in the Cat and Cloud podcast.

I've been doing this for about 3 weeks and my coffees is much sweet, cleaner and flavorful than ever before.

Honestly I feel like trying to follow Rao and Joe's techniques have led me astray.

Question - when you increase gas 20-30 degrees before FC, do you end up having an increase in ROR? This is somewhat similar to a technique that I've only experimented with a couple times on a dry processed coffee but the results were QUITE good. Charge lower, use more air early on, extending dry phase, and then increase heat shortly after 300f, then taper down so the 2nd half of the curve ends up being your normal RAO curve (but it is a faster finish through development). I believe this is more like a typical diedrich curve. I'm very interested to experiment with altering length of dry phase to set up a city/city+ roast vs a full city/full city+ roast. It SEEMS that there are very different strategies you should utilize for both....

How long has your dry phase been for these roasts? And how long was the overall roast?

The more I read, the more techniques and variables seem HIGHLY dependent on the roaster that you're using. On the other hand, I finally ordered and read Rob Hoos' book, its amazing to me how small the window can be to really change the flavor from bad to fair to good to excellent to good to fair to bad. So ... many... variables.

My basic roast has been:

7/3:30/1:30 ish

22f/min 18f/min 15f/min

Yes my ROR increases before entering FC

In the middle phase my ROR drops to 15-17f/min, but around 350f I crank the gas and enter first crack around 20f/min. I taper off the gas slightly and when it is rolling in FC I cut back and finish at about 6-10f/min(gas is still on).

This is only achievable with high airflow


I use this technique on a Quest M3 and Probat 12Kilo

The long drying helps the inter/outer bean be properly developed. I've found Rao/Joe's method (higher charge, low starting airflow) was causing me to get slight and barely visible tipping, possibly too high ROR needed in order enter FC with good momentum to have a robust rolling FC with a constant declining ROR

This allows me to have no tipping and much much more energy entering FC very effectively with a very evenly roast bean inside and out.

I drop 15-25f from start of FC

Fantastic - thanks for the detail.

I somewhat tried what you said in a roast I did last night on the 2nd try of a Colombian bean I just got from SM's...

The SM cupping notes say its supposed to be super sweet. For my first roast I tried a nice steadily declining RoR. I think my phases were something like 5:25/4:40/2:50. I went into FC with a RoR of around 14F and let it gradually taper off. The profile looked great (I'm at work, otherwise I'd post it), I had nearly no dip after FC started (I bumped heat when FC snaps started to compensate the dip, then I let off around 45 seconds to 1 minute after to ensure it doesn't go back up). I cupped it last night, blah, nothing special and really frustrating cause I thought it was going to be good.

So last night was attempt #2, I tried to do what you said aside from a long dry phase (before you made the detailed post). I think I bumped heat at around 360 which on the huky I hit FC at around 395 usually. I entered FC at around 20F (solid momentum); what resulted was a BIG group of snaps at FC, which caused my ROR to really drop quickly. My theory on the drop is that so many of the beans snapping at one time equates to more steam being released and thus a quicker dip. From there I held the ROR fairly steady at 10F, it climbed a little towards the end, but didn't "flick" upward beyond where the natural curve would've been.

At first I was really mad because my overall RoR curve looked pretty flat. My drying phase didn't have as high of a RoR as I intended but thats because I charged a bit higher so the turning point wasn't as low (which greatly affects the ROR in that first stage of the roast). Then I started going through all of the curves of really successful light roasts which I've done on the huky. By successful I mean, sweet, defined fruit notes, acidity kept in check with sweetness. Guess what, they're all pretty similar, they're like my "2nd attempt" and it sounds like they're somewhat similar to yours based on the variance between the average RoR in each phase. Those successful sweet light roasts have had pretty flat looking ROR curves, where I enter into FC with a good amount of energy. I think I (we) may be onto something.

My next attempt is going to be even more like you suggested, intentional long drying phase, shorten the maillard phase and build energy going into FC. This style of roast seems to break "rules" but "rules" be darned, the proof is in the cup and what works for your roaster, period. It seems that shortening the maillard phase and having sufficient energy to carry through finishing phase is very very important.

I really appreciate your suggestions (which seem to validate my early experiences with similar curves)!!!

I hope it works for you! Let me know the results.

I had a thought.

I have had underdeveloped roasts with "enough development time" and properly developed roasts with not "enough development", if we're talking about "the norm". The only difference is the deg/min.

Does 20f/min for 0:30 = 10f/min for 1:00?

I'm starting to think that development time or DTR is completely irrelevant without taking into account f/min average of the Development time.

Looking back, my best roasts have an average of 15-17f/min for the development time, finishing with at least 30-50% gas still on.

Roasts that are closer to 10-13f/min avg. have a lot more breadiness(if properly developed), and less sugars. (Yes with a constantly declining ROR).


Not hard and fast rules, just some questioning and observations I've been making over the past few weeks.