Tips on developing the nose


Hello everyone. I have been enjoying the trial and error of the home roaster and have found a lot of posts useful. I was wondering if anyone has some ideas on how to help me with the wet sock aroma I get from some decent Guatemalan beans I bought. I finally found a decent tasting roast for them (around city+) but no matter what I do, the sock smell is there. I actually haven’t been enjoying any of the coffee smells even with IKAWA stock beans and profiles. Help?


My thoughts are that you might live in a wet environment maybe your beans are absorbing the moisture in the air, and/or the beans weren’t dried well before being shipped to the distributors. But tha’ts not something you can’t get around. I think if you slow down during green and then as it gets yellow, rocketing up to…460-480*F maybe gets you though that “wet sock” for me I’ve gotten wet burlap sack and horse, definitely a lot of horse. Not on Guatemalans per se but there tends to be that wet smell if yellow to brown is too slow. I have found that ‘rocketing up’ part to be a good thing in my experience, I’ve seen that the flavor comes out more in the area of burnt flavors in the cup but I’m seeing if those roasty/burnt flavors will dissipate and leave the fruit, caramels,etc after some time of rest.

edit: *through that “wet sock”…


Thanks so much for the reply. I was able to remedy it somewhat by going more info a medium roast. We don’t have much humidity here at all so I don’t think that’s the issue. I do think that maybe I was stopping the roasts too early and some of the recommended profiles by IKAWA never hit C1 which is odd, isn’t it? Maybe they were just under roasted.

Based on what you were saying and your past observations, do your profiles tend to be more like roast 1, 2 or 3 usually? Pictures below.


You’re welcome, I think a key thing is going by color is a little hard to do if the beans don’t fully expand and puff out. So a dark looking medium roast is actually more underdeveloped than a cinnamon roast. It is probably because of that, that you stopped the roast without hearing the 1C.

It is somewhat odd, but other roasters on other machines have noted that going very slow during a roast like Italian roasters do, or American’s who roast Sumatran’s 1C may be inaudible. But for us, if you don’t hear 1C you fudged up. If you hear 1C and 2C together, or if you walk away on a unfamiliar profile and missed when 1C started…all of these things mess up a roast.

I don’t use any of those types of profiles for reasons: charge is too fast and too short, being able to clearly define a space for beans to get bright green and slowly (1-2min) get to yellow. (Steady increments of declining ROR I feel gets a very proper 3rd Wave cup) I think we do not have an accurate way to measure bean temp (even with approximation) to be using a declining/negative slope because a) you cant catch the bean before it crashes b) slows the momentum of the bean, where (at this point) I’d rather use the fan to do that, at the end, but really depends on the whole profile. The fan does not look like it’s high enough especially in the beginning to dry out the bean, and the middle part where the rotation of the beans will most likely stop and then you have the beans in contact with the metal and really hot air and the middle part getting nothing. Also things like charge temp and speed depend on 100% of the bean you are using.

I think you should look at either my Brazil Sitio Profile or Marshal Etheo those should be a good place to start and get a better understanding of roasting while having a blueprint to get past most of the checkpoints. The end goal is to have a profile with characteristics of my A profile, where a firm grasp of the concepts of roasting are understood, familiarity amongst different beans and their relationship to each other in order to properly set segments of the profile for the specific bean (charge, green, yellow, brown, 1C, 2C)

What have I done wrong?

Update: for the most part I generally agree with what I said. Or at least I haven’t seen to much to change. I don’t mind a high charge now but I still have not been able to match the clarity of flavor in my more random looking profiles vs really high charge with a generally sustained spike and high temp profile. Now, as to where those pronounced flavors are…I do not know in general (I start moving along the z axis and then change charge or Millard, etc until I strike gold Trying to Break-down Profiling, Different types of Profies, and Troubleshooting)for new beans of a different variety and location, but once I found it, there generally seems to be the “flavor area” in that location/profile. But this will take more evidence and testing for me to feel comfortable in concluding as “fact”.