Increasing fruit flavor sweetness and body in this roast

I’ve been roasting for a little while on the bullet v2 and I have began to question my methods (once again). I’m striving for the perfect cup and constant improvement and it’s an endless uphill battle. I find that with most beans I can easily get what I’m looking for, but I have a Brazilian pulped natural that is playing hard to get…

I will attach my roast curve along with times and temps. It’s current profile is fruity with good acidity, but lacking in body which means anything past a 6 oz flat white all of the interesting flavor gets lost… It can often taste flat and I’m trying to change my roast curve so that I retain the fruit and the acidity (a little less is okay) and increase the body so that it’s a richer more flavorful cup when made into a latte.

any ideas how I would approach this with this roaster?

Just noting that this graph wasn’t as smooth as usual near the end, but it turned out better than my profile I was following, so I thought it’s best to work from here.

I have been using the bean temp probe as my primary input as this is most similar to commercial roasters, everything is always very clean inside so reading are accurate. I find if I use IBTS the beans come out way darker than the temp suggests which is why I don’t use it… FC started around 194/195 which seems high as it doesn’t give me much room to develop the beans and stay in a medium roast. this batch dropped at 205 degrees and same thing again… fruity, mildly acidic, but lost in milk, this bean shouldn’t be like that but cant figure out why? Am I hitting it with too much heat? not enough heat? too long? too short? Just seems like a rich sweetness is not present, but I know this bean has that potential, I just cant figure it out on this roaster… any help or advise would be appreciated.

Is this 1kg batch?

1 Like

Ah yes sorry 1kg

What’s your ambient temp and relative humidity ? Just curious.

I’ve been elongating my Maillard a bit more lately (up to 40% in some cases), and trying to go harder/faster into FC with anything washed (with good results related to body).

ambient temp 20c and humidity around 40% This is pretty standard and usually doesn’t change much.

Fair enough, but if I go too hard and fast into FC then I’ll blow past a medium roast before the roast is developed enough. its also worth mentioning the bean are all rather small in size…

1 Like

My 2cents - given you’re doing 1kg batches, try keeping at P9 from the start and lengthen how long you stay in P9. Consider also playing with the drum speed. With the bean mass of 1kg I found I had to do things a little differently than a 500g batch. Feel free to take a look at my roasts on RW for 1kg. I’m still messing around a bit with D speed with 1kg especially at the start to make sure the beans really move around more to soak up the heat during that drying phase. You also start with a slightly high PH than I do, which is ok. I did that once or twice but used an F3 at the start of the roast. If you’re roasting to make coffee drinks (as opposed to drip or pour over for black coffee) you may want to extend to a higher drop temp so the cup has more body to deal with the cream/milk additions - that’s really my 2 cents… I don’t take any additions (sugar or cream/milk) in my coffee at all :slight_smile:

It is a learning process unfortunately.

I also track my roast weights to help me understand my roast levels. Up to 13% is my cutoff for light roasts, and some beans and batches reach that quicker than others, so I rely less on absolute temps and more on what the bean says.

I like your preheat and the bean turnaround around 100 is what I go for as well, but I would keep a little more heat on during the roast, up to yellow, reducing to P7 with F3-4 by then. Coast along until first crack, then around 1.5 mins and/or 15-20% dev time. After dropping and cooling, weigh your batch and see if you’re between 11-13%. Adjust your drop point accordingly to achieve it with your next batch. I also start smelling the tryer constantly as it gets close, looking for those sweet smells where the bean starts releasing the oils after FC and they bloom (like parching spices in a pan on the stovetop).

I am hardly an expert - pretty much a newbie. I am not an endless tinkerer and gratefully am not a victim of good taste. I roast to what I drink and the endless subtleties and descriptors of cupping sessions bore me to tears.

I use a variety of vacuum pots to brew and drink my coffee with a sprinkle of sugar and a lot of cream. (I know - that is beyond the pale and I chalk it up to a moral failing on my part that no number of twelve step programs can rectify.)

In order to stand up to these travesties, I try to maximize body - my ideal cup would have the mouth feel of a cup of hot chocolate - mostly by selecting varieties noted for having a heavy body. So far I’ve liked the high elevation Brazilians and Bolivians the best. My daily go to is Brazil Oberon (Coffee Bean Corral). Mostly varieties described as fruity, don’t appeal to me. I lean towards things the cupping mavens describe as chocolatey, nutty, caramely, brown sugary.

I have been following the advice of coffee consultant Morten Munchow, who claims that 80% of coffee flavor is color and 20% is development time. He apparently pays little attention to RoR except to the extent necessary to meet his development goals.

Generally, I roast to medium-light to medium - perhaps slightly darker than the so called Nordic roasts. Since I am not comparing to anything but my own previous roasts, I strictly use IBTS temps, which are remarkably consistent and, I think, tell you what the actual bean surface temp is. Bean probe temps are affected by way too many things unrelated to the actual bean temperature, so I mostly don’t know what to make of them.

I use IBTS temp as a proxy for color and my end temps are generally between 215°C and 218°C. Per Morten’s suggestions, my goals are to reach FC in around 9 minutes, and end temp 3 to 5 mins after FC. Mostly I shoot for 4 mins development.

I have been roasting 500 g batches, as i feel I have better control, but am trying to move on to KG batches, so the following is based on 500 g.

PH 250°C

P8, D9, F2 ----> 4minutes.
P5, D9 F2 ----> 200°C
P3, D9, F2 ----> End

My roaster is indoors so ambient temp is fairly consistent and around 24°C.

This gets me in the ballpark for hitting my milestones for most bean varieties, but may require a bit of fine tuning for some. Obviously, this needs to be tweaked if you’re roasting to a different final temp or figure your development times should be different. My take is that changing fan speed will change your RoR much quicker than changing power.

As always, if you like the way it tastes then you’ve done it right.

-Gray Haertig




Totally agree with your take. Reduce variables, focus on color, keep it short for fruitiness/acidity, go a little longer to reduce acidity. Morten has really simplified roasting.

Just a small comment regarding Morten Munchow’s analysis of the roast cycle since there was some discussion here about increasing the body of the coffee. As mentioned above, according to Morten’s research 80% of the flavor variations are determined by the final color of the bean. 15% is determined by how fast you approach that final color after 1st crack. The other 5% is determined by other things which could include the lengthening of the Maillard phase as advocated by Rob Hoos. So the point of Munchow’s research concerning body is that changes in the roast profile have very little effect on perceived body of the coffee.

I have found that body in a cup increases with a longer development phase (leading into your 1st crack). I’ve also found that flavor falls flat when your RoR flat-lines in the end of the roast. I see your graph has very little change in the RoR in the end. I would try adjusting power and/or fan speed to allow the beans to take on more heat over a longer period at the end of the roast.

I hope this helps!