Roast and Learn Together - May 2015 - Page 2

#11: Post by **another_jim** » April 27th, 2015, 6:50 pm

jalpert wrote:One last follow-up if you don't mind: I recently read the Rob Hoos manifesto, and one takeaway there was that the length (in absolute terms) of the maillard phase is important. Theoretically the longer the maillard phase (in minutes), the more insoluble large molecules form, which leads to greater body and mouth-feel.

That's a good observation; I'll need to give that manifesto a close reading.

A 15 minute roast that simply is a stretched profile of an 8 minute roast can be indistinguishable in the cup if the heat transfer to the beans was slowed down proportionately, so that all the chemical reactions are slowed down. On the other hand, if the heat transfer to the bean remains the same, Rob Hoos is right, more of the sugars and amino acids will be turned into Maillard compounds and fewer will be available for caramelization. I'm not sure which compounds have more mouthfeel, but the Maillard compounds will have a complex savory notes that the caramels lack.

This is why the description of this coffee attracted my attention. I was interested in seeing how much the people who are participating can play with the maillard flavors.

SJM wrote:I am looking forward to seeing how using the solid drum/fast motor (as opposed to the perforated drum/slow motor) on that HUKY informs the results ))Susan

Does the solid set up have more ventilation? Usually solid drum roasters are ventilated, while perforated ones can get by with convection currents off the heat source.

#12: Post by **SJM** » April 27th, 2015, 7:03 pm

Here is the diagram that Hank posted for us when we were trying (are still trying) to clarify how the solid drum HUKY and the perforated drum HUKY differ. Both configurations are supplied with the same fan.

#13: Post by **jalpert** » April 27th, 2015, 8:21 pm

another_jim wrote:Does the solid set up have more ventilation? Usually solid drum roasters are ventilated, while perforated ones can get by with convection currents off the heat source.

As Susan points out, both drum types are ventilated on the Huky. However, the fan probably has a greater effect at the same CFM with the solid drum, just based on Bernoulli.

With the perforated drum, increasing the fan up to a point actually pushes up ET, with a subsequent BT RoR drift upward. After that point, increasing fan speed lowers ET.

I would guess that the solid drum has more immediate feedback upon changing the fan setting. Huky roasters on hukyforum often report this makes the solid drum easier to control as you get more immediate feedback.

another_jim wrote:A 15 minute roast that simply is a stretched profile of an 8 minute roast can be indistinguishable in the cup if the heat transfer to the beans was slowed down proportionately, so that all the chemical reactions are slowed down.

This is fairly mind-blowing for me; I guess the more experienced guys take this for granted as you've seen this firsthand and I'm just making it up as I go along. I had always assumed that lengthening the browning phase in terms of seconds had a good deal of effect on what compounds ended up in the bean. I'm not terribly familiar with maillard chemistry (and honestly I don't remember too much of my chem 101 either) - but I had figured that all you had to do was get a food to the activation energy for the maillard reactions, and then it would more or less proceed from there, throwing off first order chemicals, which then react and form second order, third order, and so on to create whatever orgy of brown stuff happens to come out. That is interesting that the progression of the reactions are dependent on dT of the substrate. I'll have to do more reading on maillard reactions.

#14: Post by **another_jim** » April 27th, 2015, 8:51 pm

Thanks for he lesson on the Huky.

Way back when the Hottop first came out, they were set up to do 20 minute roasts. We did some blind tests using the same profile with an air roaster, and the air roasts tasted flatter. The same shape done in about 13 to 15 minutes cupped the same. (The original Hottop was slower than the current models, and tasted a little flatter, but still pretty good). These results are old, before we were controlling the ETs, so there's a good possibility that there's lots of factors we didn't know about.

#15: Post by **SJM** » April 29th, 2015, 2:54 pm

Using the HUKY with the solid drum, FC was definite but very early on at 370F (I don't usually expect it until about 387), which made the time from 'end of dry' (don't shoot me I still call it that) to FC very short.

#16: Post by **stinkyonion** » April 29th, 2015, 10:31 pm

It would help for us newbies if the acronyms were used a little less. Most roasting posts I have to read at least 3 times before I can understand most of what is said.

#17: Post by **another_jim** » May 2nd, 2015, 3:53 pm

stinkyonion wrote:It would help for us newbies if the acronyms were used a little less. Most roasting posts I have to read at least 3 times before I can understand most of what is said.

It would be helpful for us oldies as well.

#18: Post by **another_jim** » May 2nd, 2015, 4:36 pm

I've done and cupped a pair of roasts, one slow stat/fast finish, the other fast start/slow finish. My roaster is an electronics free zone, so I recorded the profiles using the dial thermometers and stop watch, and transcribed them into R (the stats language formerly known as S). Here's the graph:

The slow start roast took 14:40 minutes and ended at 212.5 C (414.5F). The fast roast to 10:35 minutes and ended at the same temperature. The temperature corresponds to a few degrees after the end of the first crack on my roaster. In the graph, the fast start profile is offset by 4 minutes, so that they end at the same spot of the graph. This allows people to see that the overall heat input is roughly the same (which is why fast start/slow finish roasts should run faster than slow start/fast finish roasts).

Both roasts were done on the Quest, at 110 gram lots. The slow start roast had environmental temperatures starting at roughly 200C (400F) and ending at 265C (510F), using minimal airflow; the fast start roast ran pretty constant at 250C (480F), using maximal airflow.

The color of the whole and ground beans were identical to visual inspection.

I cupped the beans for the first time 20 hours post roast.

As usual, the taste when the cup was warm and the wet aroma were very much the same, whereas the dry aroma and cool taste were slightly distinct. I was able to correctly guess which roast was which was which based on the taste, but the difference was not at all dramatic. People who do not cup blind exaggerate these differences, and since they dominate the on-line coffee discussions, readers think the difference between two competent roast profiles is far larger than they actually are. The big differences are due to roasting mistakes, not competent profiling preferences.

Both cups tasted roasty and only mildly acidic. The fast start cup was dominated by bitter/sweet caramels, with only a hint of savoriness in the finish. The slow start cup was more balanced, with stronger savory and acidic notes to the taste. So the slow start cup had about two points more on the flavor and finish scoring. However, the dry aroma was won by the fast start roast, which had a strong and inviting caramel aroma, whereas the slow start roast was laid back. Overall, I scored the slow start roast at 88 and the fast start roast at 87. I'll do a more picky cupping and espresso shots on Monday.

The savory flavors in this coffee are malt like, reminiscent of whisky; so stretching out the roast up to the first crack does help as far as my taste is concerned. If the savory flavors are more like beef broth, as in some Kenyas, it is better to stick with a faster start, since a hint of that is cool, but more gets weird.

#19: Post by **Andy** » May 2nd, 2015, 6:57 pm

I roasted 3 batches on 4/28 and tasted this morning. Like jalpert, I had a difficult time detecting first crack. I think I heard a single pop in each roast, so I estimated the start and duration of 1c based on past roasts. All 3 were roasted to about the same finish temp, 420F. First roast finished at 12:00; second at 9:30. The third was a double roast. the first phase stopped at 340F in 4:30; beans were then cooled to ambient temp and re-roasted in 10:30 to 420F.

I brewed them all at the same time this morning -- 4 day rest -- using Melitta cones. I neglected to note the fragrance and I never get a satisfactory impression smelling the aroma in a paper cone for some reason. I will revisit these soon to fill in those blanks. All seemed equally well balanced as far as acidity, roastiness and fruitiness. It was a blind tasting for my wife and she picked the second roast as sweeter than the others. I agreed, but it wasn't pronounced enough that I would have trusted my non-blind evaluation without her confirmation.

One of my favorite things about the RL&T threads, and this forum in general, is that I am prompted to try coffees that I probably would not get around to otherwise, simply because my volume of consumption is low. This is an example of a very nice coffee that I would not have encountered otherwise.

May 3rd, 2015, 2:18 pm

I'm looking forward to profiling this coffee today and tomorrow. It's a fascinating bean to approach, since it's usually "wrong" to hit a pulp natural with a lot of high heat early on, yet on the flip side, it's a peaberry which are notorious for not taking heat as well as normal coffee. I think I might actually order more of it, and profile them separately, some on the Quest, some on the gasser. I like the airflow properties of my gas drum more.