Oh my, what a great coffee. Prepped KONE/Chemex: 60g/finished quart of coffee. City plus roast. Think I dropped at 418* on our system.

Picked up peach on the grind. Nose was 18" from the grounds and an overwhelming, compelling peach hit me.

Peach in the cup as well. Chewy, earthy peach. Makes me salivate. Sweet caramel/noughat with a high note from the tartaric <?> acidity. Get some cocoa (espresso grind size) notes that appear in the finish and are predominant in a lingering after-taste.

My profile is at the shop. Will post it this week.

Tom Owen gives it an 87.2. I think he underrated this one. I give it an 89.75.

OK I'm here to roast and learn. It's a little intimidating since everyone seems to be trying to fine tune their roasts to bring out that last point or two in the score. My aim is a little different- I'm just trying to roast a coffee that is "decent", and I am having trouble.

A quick intro: I roast on a modified Quest M3. It's the newer model with the mirror imaged controls. The drum is painted black, and the outside shell is insulated; I've changed the fan control to be like the old Quest, where I can shut off fan power completely. EricS thermocouple placement.

I did 5 roasts with this little Tanzania, using several different profiles just to see what kind of approach would work for me. The best of the lot could be described as "acceptable", but not really what I'd call good. I'm sorry my vocabulary is not very sophisticated, so as well as I can describe the coffee is to say it's kind of dry in both aroma and taste. The coffee has adequate body, and decent acidity. The aroma is kind of lifeless, with only faint caramel/cocoa. BTW talking about brewed coffee here, v60. Overall, it's pretty drinkable, but I'd like to take it to the next level.

Here is the roast data:


120g batch, 14.4% weight loss.

Just in case it's not obvious what you're looking at, the bold lines represent this roast, which was the best of the 5. The faded lines are the previous roast, which I judged to be the worst. As some others have said, this bean doesn't seem to make much first crack noise, so my 1C indication on this graph is based on one of the few cracks I heard, so the 31% development time (which depends on the 1C time) is really not a reliable indicator of anything in my opinion.

Help. What should I do differently?

Thanks,
Rob

Have you tried a longer roast, say 12 minutes? I think hitting 14.4% weight loss in under 9 minutes might be a bit aggressive on the Quest. Your Quest mirrors mine, except I've long since took the black off the drum. I think (and this is entirely speculation, but based on a lot of experience with the Quest) that it doesn't like pulp natural, natural processed and peaberry coffees, particularly in combination. Not that you can't work around it, but I strongly believe despite our best hopes, that the Quest is still largely a conductive roaster. The fan does not behave like a traditional drum roaster drawing hot air thru the drum. I think what often happens (especially on small 120-ish gram charges and a black drum) is you are more prone to develop the outside of the coffee quicker than say, what you'd get with a traditional drum roaster.

This wouldn't help a bean that seems to need a nice amount of Maillard development to bring out those wonderful caramels and gooey chocolate notes (which really need to be optimized to make this bean shine, since you're not preserving any snappy sparkling acidity). You get a darker bean surface and a lower degree of development internally.

Pulp naturals, and naturals have to be handled differently, their higher degree of sugar and amino acid makeup on the surface can easily take on acrid off notes if developed too fast. Especially in a roast that behaves more conductively than most. This alone would be the reason I'd try to stretch out the roast with slightly lower heat and more time.

Case in point, this same coffee roasted in my 1 kilo gas drum was done very fast, under 9:30 if I recall correctly and had only a 13.6 or 13.8% weight loss. The bean expansion was great, the cup fantastic, and certainly not dry or lifeless. I credit a proper convective flow of hot air allowing me to thoroughly develop the bean internals and bring out those complex aromatics and caramel notes without overdeveloping the surface of the bean.

Try any of these suggesions:

1) Use the exact same profile but increase your charge weight to 180 grams and see what that gives you

2) Same charge weight and a lower charge temp, try to stretch out the roast.

3) Consider clipping your post 1C phase a bit, incase you're overdeveloping the bean surface and losing some precious aromatics.


All of these suggestions only apply to the Quest, given it's particularities. And note, the only reason I painted the drum black originally was so I can charge a larger batch, and early apprehension about burning out the elements, which isn't really much of a problem anymore. With a 120g charge, and a black drum, it's almost like high tech wok cooking. If I ever modify my drum to perforate the rear and reroute the airflow, I might consider toying with a black drum again, but my strong belief is it's already too much of a conductive roaster, I don't want to make it more so.

Boldjava wrote:
Peach in the cup as well. Chewy, earthy peach. Makes me salivate. Sweet caramel/nougat with a high note from the tartaric <?> acidity. Get some cocoa (espresso grind size) notes that appear in the finish and are predominant in a lingering after-taste...

One test I have for coffees is they are good, right to the bottom, when cooler.

Today, I had a to-go cup and didn't finish it. Came out of church an hour later and it remained balanced, without any sharpness or edginess. Good fine, bro' Thompson.

TomC wrote:Have you tried a longer roast, say 12 minutes? I think hitting 14.4% weight loss in under 9 minutes might be a bit aggressive on the Quest.

I did a couple of longer roasts, but they were notably less successful. One 11 minute roast saw 15.4% weight loss. I can't figure out how to keep the moisture in with this bean, 14.4% is the lowest loss I saw, and that was with a very short roast.

...Case in point, this same coffee roasted in my 1 kilo gas drum was done very fast, under 9:30 if I recall correctly and had only a 13.6 or 13.8% weight loss. The bean expansion was great, the cup fantastic, and certainly not dry or lifeless. I credit a proper convective flow of hot air allowing me to thoroughly develop the bean internals and bring out those complex aromatics and caramel notes without overdeveloping the surface of the bean.

Maybe I need a little more airflow. I've been keeping it minimal.

1) Use the exact same profile but increase your charge weight to 180 grams and see what that gives you
2) Same charge weight and a lower charge temp, try to stretch out the roast.
3) Consider clipping your post 1C phase a bit, incase you're overdeveloping the bean surface and losing some precious aromatics.

Thanks for the suggestions! I'll report back later in the week when I've had a chance to try them.

Rob

robmatic wrote:I did a couple of longer roasts, but they were notably less successful. One 11 minute roast saw 15.4% weight loss. I can't figure out how to keep the moisture in with this bean, 14.4% is the lowest loss I saw, and that was with a very short roast..

Rob

This comes as no surprise The best cure is to increase your charge weight and otherwise keep your parameters the same. That will make a significant difference and you won't be chasing your tail if you make several other changes and try to figure out what's working.

I'll say it again, unless you want to roast larger (>250g) batches, a black painted drum will not help you much. Your higher degree of weight loss per batch matches my own findings.

TomC wrote:This comes as no surprise The best cure is to increase your charge weight and otherwise keep your parameters the same. That will make a significant difference and you won't be chasing your tail if you make several other changes and try to figure out what's working.

Funny thing is that I used to do larger batches, and have been gradually reducing the size in my quest for the right combination of variables.

I'll say it again, unless you want to roast larger (>250g) batches, a black painted drum will not help you much. Your higher degree of weight loss per batch matches my own findings.

Well that was the reason that I painted the drum in the first place, but I was never able to get good results. And that's how I came back to smaller batches.

Anyway thanks again for the suggestions. I will continue to experiment.

Rob

Following Jim's lead, I did a blind comparison of two roasts, one SSFF and one FSSF. Detailed results are posted below, but the short version is that both my wife and I preferred the FSSF approach with this coffee, and I gave it a few extra points in blind scoring. For my tasting, it had clearer acidity, a more satisfying middle, and a bit more sweetness. For the below, the tastify plots represent blind tasting, while the written comments relate to all the brews I did with the coffees, most of which were unblinded.

---

Roasting Info - FSSF:

Bean: Tanzania Pulp Natural Igamba Mpitu PB
Roaster: USRC Sample Roaster
Charge Mass: 350g
Charge Temp: 395F
Dry/Ramp/Development: 3/3:20/2
FC-start temp: 317F
Finish Temp: 390F
Overall Roast Time: 8:25
Moisture Loss: 14%

Profile Plot:



[BT=Red, Exhaust=Green, Gas=Blue, Fan=Yellow]

Cupping Notes:

Rest: 3 days
Brewer: Cupping
Grinder: Mini-Bunnzilla
Water: 202F, 200g
Coffee: 10g

Tastify Plot:



Overall Impression: My best brews from this profile remind me of a really good Black Forest Cake -- sweet cherry acidity, cream, chocolate cake, and a moderate, pleasant body. I don't get too many savory notes from this profile, just a really pleasant, balanced cup of coffee that reminds me favorably of an excellent offering from Yemen.

---

Roasting Info - SSFF:

Bean: Tanzania Pulp Natural Igamba Mpitu PB
Roaster: USRC Sample Roaster
Charge Mass: 350g
Charge Temp: 300F
Dry/Ramp/Development: 5/2:20/1:30
FC-start temp: 362F
Finish Temp: 392F
Overall Roast Time: 8:50
Moisture Loss: 14%

Profile Plot:



[BT=Red, Exhaust=Green, Gas=Blue, Fan=Yellow]

Cupping Notes:

Rest: 3 days
Brewer: Cupping
Grinder: Mini-Bunnzilla
Water: 202F, 200g
Coffee: 10g

Tastify Plot:



Overall Impression:This was still a great cup of coffee, but it fell behind the other just a bit in terms of sweetness. It had a less complex middle and a bit of astringency in the aftertaste. Still lots of pleasing cherry acidity and chocolate notes, but not quite up there with the best cups I got from the first roast.

I've been playing with the roasts a bit, and working on a faster roast (FCs @ 7:30, EOR @ 10 @ 408F). I noticed what many have noticed; fruity, less savory. The fruit was tart, but not quite citrusy, kind of hard for me to identify. I felt like it was almost "not quite ripe" fruit, except that carries a bit more of a negative connotation than I felt like it deserved. This morning, I decided to pull a ristretto (19 g in, VST 18g basket, 20g out at 31 sec) and WOW!!! what a great change! I'd been missing a bit of sweetness, and by shortening the shot pull, it heightened the pop of acidity AND added that touch of sweetness that I was missing. I'm still playing with the roasts (it IS RLT, after all), but I was surprised at what a difference the pull made with this coffee.