Having already been through the usual steps - having moved from an old Phillips "Espresso" machine (which had been accumulating dust for years) to a Rancilio Silvia three years ago (after coming back from a trip to Italy) and then to my actual LaSpaziale S1 Vivaldi II; from a Braun burr grinder to the Rancilio Rocky to my actual Baratza Vario - I then wondered about buying a Hottop-B roaster (A case for home roasting?), which I did buy a month ago (before the price increase) (Hottop KN-8828B-2).

I've since bought and skimmed through Home Coffee Roasting, Revised, Updated Edition: Romance and Revival, read many interesting threads from the Home Roasting section of Home-Barista and other sites.

In the end, I realized that many excellent information and insights are often hidden inside forum threads, making it difficult for a home roasting newbie like me to get to it in one step.

Hence, I've decided to document my own experience, hoping this will be useful to home roasting newcomers. Feel free to comment and complete the information with your own experience and knowledge!

As I'm the only espresso drinker at home and aiming for quality rather than quantity (I cannot, yet, be considered a caffeine addict), and as I want to use artisan roasts all along to compare to my own roasts, I will probably do only 2-3, maybe 4 roasts per month. A good roasting log will therefore be of crucial importance to me.

References I've read: (will be updated as I find new ones)

• Roasting guidelines, my attempt at summarizing roast profiling. (added in Feb. 2012)

General coffee roasting knowledge

• A case for home roasting?, or why I roast. (added in Feb. 2012)
• "Home Coffee Roasting, Revised, Updated Edition: Romance and Revival", by Kenneth Davids. A great introductory reading, but it lacks the latest trends/science, such as roast profiling.
• HB: How to Home Roast
• Sweet Maria's An Updated Visual Guide to the Roast Process
• HB: Coffee Roasting Defects Pictorial
• Sweet Maria's Library
• How best to maximize "fruit bombs"?
• iNeedCoffee.com: nice but basic information only
• HB: Post by Jim Schulman: How to recognize a City and a City+ roast.
• Roasting by Smell
• Fresh roasted coffee smell is off - help!!
• HB: Post by Ken Fox on recommended duration between 1st and 2nd crack.
• BootCoffee.com: Good general roasting knowledge/how-to.
• CoffeeResearch.org: Good information on everything coffee
• HB: Two posts from Jim Schulman on Maillard/caramels/distillates temperatures and on rate of rise & proportion of time spent in each phase of the roast (warmup, ramp to first, and roast finish)
• A one page roasting summary from George Howell's Terroir Coffee
• Sweet Maria's Roast Profiling
• Sweet Maria's Using Your Senses to Determine Degree of Roast articles: Smell, Touch, Taste, Sight, Sound, all summarized in Use All Five Senses To Determine Degree of Roast
• Correlation between brightness(acidity):sweetness and ramp:finish, from jammin (added in April 2012)
• Jammin advice on MET, espresso profiling and some Hottop advice (added in April 2012)
• Ken Fox (Hopefully) Useful Home Roasting Tips (added in April 2012)

Temperature management and measurement

• Choosing thermometry equipment for home roasting
• HB: Some recommendation on rate of rise
• HB: Charging at a High ET without Tipping/Scorching/Charring? (probe locations, loading temp, etc.)

Hottop specific references

• HB: A solid, catch-all, Hottop espresso roast profile?
• Installing ET and BT probes in a Hottop (added in Feb. 2012)
• Hottop USA user guides
• Hottop USA Roast Profiling

Blending information

• Sweet Maria's Blending Basics (with Espresso Blend section) (added in April 2012)
• An Aficionado's Guide to Espresso Blending
• HB: Simple Espresso Blends, post from Sherman

Cupping and tasting

• SCAA Cupping Protocol (PDF) (added in April 2012)
• Cup of Excellence Cupping Form (PDF) (Sweet Maria's) (added in April 2012)
• SCAA Coffee Taster's Flavor Wheel: SCAA online store, Google image search
• HB: Roasting 101s: Understanding degree of roast
• Aroma selector for cupping
• Jim Schulman on a cupping protocol
• Sweet Maria's Teaching to Taste (added in Feb. 2012)

Legend of terms used: (UPDATED)
Ref. Sweet Maria's Dictionary
1C: First crack.
2C: Second crack.
EoR: End of roast.
RoR: Rate of rise.

Batch #1, June 2011

Preparing for the roast

I've bought 12 lbs of greens from Sweet Maria's. Understanding their warning, I ordered a 8 lbs sampler and a 4 lbs sampler:

Sweet Maria's wrote:Please note that if you choose the Espresso sampler - we send some of our Sweet Maria's blends, and some single origins considered good for espresso. The selection in this option is more narrow, tailored for demands of espresso brewing and the 4lb espresso sampler will be mostly/all espresso blends.

I therefore expected to have many bags of the same coffee and be able to roast the same blend more than only 1-2 times. Sweet Maria's was rather kind to me and I have only two blends with two identical bags. For my next order (in a few months...), I will take the advice from Jim Schulman and others and order only a few different beans, to get my method together and improve it between batches. I'll take this first green beans order as a quick world tour to begin with.

So... here I am, alone in the house, Hottop-B under the range hood, 250g of Sweet Maria Espresso Monkey blend and an annotated color printed copy of Sweet Maria's Visual Guide to the roast process, with bean temperature (roast levels) mapped to the approximate display temperature on the Hottop.

The goal for this blend: the beginning stages of 2C, for a Light Vienna roast.

As this was my first time experience, I chose to use the Hottop auto profile and to let it go without changing parameters (fan, heater), while deciding to push the Eject button at my own convenience. I thought this would allow me more time to concentrate on the roast itself (smell, colour, smoke) instead of playing frantically with the buttons.

The roast


Hottop-B automated profile. Temperatures are from the Hottop display. Until I get a datalogger, the ET curve is to be taken with a grain of salt. It simply gives an idea of the temperature rise but sudden change of indicated temperature mostly comes from manual sampling.

Notes:
5:00 285°F Beans are turning greener.
7:30 325°F Turning yellow (as far as I can tell).
8:30 345°F Going to tan.
9:00 350°F First pops of 1C. Tan.
9:30 355°F Toasted bread smell.
10:00 360°F Lot of smoke. Turning to light brown.
11:00 380°F Light brown. Toasted pop-corn smell.
11:45 390°F Beginning of 1C. Brown.
12:30 405°F Rolling 1C.
13:00 413°F End of 1C.
14:00 428°F First snaps of 2C. Beans ejected at programmed temp.

About 2:30 min of drying time.
Abourt 2:15 from 1C to EOR.

I was able to follow the roast, easily identifying the so called drying phase, 1C (pop-corn) and 2C (rice crispies). Sounds are easy to hear with the Hottop low noise level but the pops covered a large period of time. This may be due to the use of a blend instead of a single origin (different beans/crop roasting differently?). At one point, I am pretty sure I heard both 1C pops and 2C snaps mixed together. This is probably worsened as the delay between 1st and 2C was too short (1-2 min).

In the end, the roaster ejected the beans by itself (temperature reached) a fraction of a second before I pressed the Eject button.

I used two metal colanders to clean off the chaff from the beans, outside the house, after the roaster cooling cycle ended.

How it went and lessons learned

Using the integrated profile was a dumb approach, with the fan/power settings changing at timings that didn't seem to match the roast in progress. Next time I will manually control the roast settings.

The bean colour was uniform over its surface, seemed dark to me, with the chaff remaining in the crease also brown. Bean surface was smooth, probably around Vienna / Light French. A few days after the roast, oil appeared on the surface of the beans, confirming that I probably roasted too dark.

The result in the cup (espresso) was around what I can get from the grocery store beans. Drinkable, with essentially roast flavours and not much else (ie. certainly not adequate to keep me in home roasting!)

PS. Sorry, didn't take a picture of that batch.

Batch #2, June 2011

Preparing for the roast

As I wanted to stay away from the integrated auto profile, I looked for a good starting point. There are already different threads on the subject, many well documented, with temperature graphs, for example A solid, catch-all, Hottop espresso roast profile? and Hottop-B Espresso: A Fast, Low ET, Profile from coffee.me). I also found in the middle of the Hottop user guide their own profile recommendation: http://www.hottopusa.com/profile.html. It sounded right, inline with what I read and linked bean temp to the shown temperature on the Hottop display.

The profile meant that I was to drop the heater power during the drying phase, push it back to full power just before the beginning of 1C, then lower it to surf the roast and prolong the delay between 1C and 2C.

I stayed with the Sweet Maria's Espresso Monkey blend. Same batch size, 250g. Roaster set to 20:00 min., max temperature. Beginning with full power and no fan.

As I wanted to avoid getting, again, to a Vienna roast (roasting out all the origin flavours), I aimed for a lighter roast this time, ending it with the first snaps of 2C.

The roast


0:00 250°F Drop-in.
4:00 320°F Wet hay-grass smell. Bright green colour.
6:00 ___°F Yellow.
7:00 340°F Tan colour.
7:30 Smoke coming out the loading chute.
9:00 355°F Light brown. Theoretically approaching 1C. Lowering heating power.
9:45 360°F Brown. Still lowering heating power.
11:30 375°F Toasted smell. First snap. Smoke coming out.
12:00 377°F Where the hell is 1C. Panic.
12:15 385°F Beginning of 1C. Turning power back down.
13:00 390°F Rolling 1C.
14:00 400°F End of 1C.
15:00 402°F First snap of 2C.
16:00 404°F 2C going on.
16:10 405°F Ejecting.

About 3:15 min of drying time.
About 4:15 min between 1C and EOR.

Everything seemed to go well until I reached 350°F, the displayed temperature indicated as the moment to drop the heater power. Mistake. While I was expecting the first pops of 1C, I got a complete silence... I was stalling my roast... I pushed power back to 100% until I heard the first pops but this was probably too late... With first crack under way, I again lower the heater, but too late again, with the temperature rising hastily to 2C. As soon as I was sure I really recognized snaps, I ejected the roast.

How it went and lessons learned

Following the roast and adjusting the profile at the same time was a stressful experience for the newbie I am but went well (from this perspective alone).

The roast looked better than last time, with crease chaff still of lighter colour than the rest of the bean. Beans kept some surface texture, unlike the first batch. Some oil spots appeared after about five days, indicating I was probably in the Full City+/light Vienna range.

All was well, until I grinded some and tasted it: completely bland coffee. Very smooth, no acidity, no bitterness, not too much roast flavours, no recognisable fine flavours... no nothing. Probably a very good average cafeteria coffee... Meaningless would be a good epithet.

Searching HB with the term "bland" under Home-Roasting lead me to this post from endlesscycles and this one from Ken Fox. Both gave me the two part answer: too much time spend reaching 1C and too little time between 1C and 2C. I cooked my coffee.

Next time, I probably need to wait until the first pops of 1C before I rapidly drop the heater power and keep the power low until 2C, maybe also using the fan at full power if needed between 1C and 2C to help lower the temperature rise.


Batch #2. Sweet Maria's Espresso Monkey blend. Picture taken about 6 days post-roast.

Batch #3, July 2011

Preparing for the roast

Having learned my lessons about how to cook coffee and the great taste you get when you do so, I aimed this time to keep the power high until I was sure 1C was near (without previous logs and bean temperature probe, this mean waiting for the first snaps of 1C) and to try keeping the delay between 1C and 2C as long as I could (low power and high fan).

Sadly, all was left of my two bags of Sweet Maria's Espresso monkey blend was not enough to do a third 250g. I shuffled the coffee bags, peaking at Sweet Maria's website, and decided for the Espresso Workshop #17 "Ethiopiques".

Goal: [quote="Sweet Maria's]This blend works well at Full City. I would not take it to 2C unless you feel it is too bright at FC roast.[/quote]

I again used the Hottop recommended profile, this time adjusting the thresholds:
250°F Drop-in.
290°F Reduce power to 70%.
320°F Reduce power to 50%.
In the middle of the drying phase (judged by colour), fan at 50%.
At the end of the drying phase (colour = tan), power back to 100%.
At the first pop of 1C, reduce power to 50% and raise fan to 75%.
At the very first snap of 2C, hit Eject.

The roast


0:00 255°F Drop-in.
1:00 265°F. Colour remain unchanged.
2:00 290°F Colour turning greener. (H7)
3:30 320°F (H5)
5:30 330°F Yellow.
6:30 340°F Turning tan. (H10)
7:00 ___°F Starting to get smoke. Toasted smell.
8:00 ___°F Turning to light brown. (F3)
9:00 ___°F Light brown. (F2)
10:00 367°F Turning brown. (H7)
10:30 375°F First snaps of 1C. (H5, F3)
11:00 ___°F 1C beginning. (F4)
11:30 388°F Rolling 1C.
12:00 393°F 1C slowing down.
12:30 ___°F Last pops of 1C.
13:30 400°F
14:00 403°F (H7, F3)
14:30 404°F First snaps of 2C. Hit Eject immediately.

About 4:00 min of drying time.
About 3:30 min between 1C and EOR.

As the roast began, I saw two brown beans in the batch... two roasted beans from the previous batch, probably stuck in the drum when it emptied. I hoped for the best; that they would not convey carbonized flavors to the rest of the beans.

How it went and lessons learned

The roasted beans look really nice to me. Light brown. Chaff in crease stayed light. Beans show a nice texture/colour contrast (a marbled look). Probably a City+/Full City roast.

A few hours after the roast, their smell was very interesting: pepper and cacao notes. Twenty-four hours later, they show a very nice smell: chocolate, subtle roasting aromas. I dared to bite into a bean and the taste was also very good.

I'll wait another day or two, pull a shot and report back.

This roast could really confirm my motivation/interest to home roast...

PS. I already realize the limitation of relying on the environmental temperature (the Hottop displayed temp) rather than having direct access to bean temp. As our computers are desktops, I would probably need a two probe temperature data logger (not having the computer next to the roaster during the roast). No perfect solution seems to exist... only pricey solutions... I will probably wait a bit before committing to buy one.


Batch #3. Sweet Maria's Espresso Workshop #17: "Ethiopiques". Picture taken 2 days post-roast.

UPDATE (12 July): I'm finishing this batch. The beans staid dry, confirming a City+/Full City roast. The chocolate aroma dissipated after a few days. The cup result is very satisfying and I could compare this roast to what I use to buy from good but not top tier roasters. OTOH, this blend is indeed intense, with pungent dark chocolate aromas. Hence, I'm happy with the roast but I may not be a fan of all Ethiopian blends. I didn't take the time to try the coffee at different brew temperatures... (used 93°C).

This is a great post, very useful for someone that's just starting to home roast. Both the list of good links, and the record of your different attempts. I'm looking forward to seeing how the latest roast tastes, and what changes you'll be making on your next batch.

allon wrote:As you have noticed, the beans don't behave linearly;

There are stages within the roast where they need more heat to keep the same rate of rise, and others where they need less heat. As the beans approach 1st crack, this is probably the most prominent of this sort of step, where the beans need more heat. I think of it as a hill...you don't want to shove too much heat early in the roast, or you'll race to 1st crack and jump over the hill, and you'll be heading full tilt towards 2nd crack - you won't be able to slam on the brakes. But if you approach the hill and don't pour on the juice, you won't make it over the hump and into 1st crack. It can be a tricky balance, especially when you are running close to the capacity of the roaster.

This is what I understood for my preparatory reading.

Electric drum roaster also have a lot of inertia, probably unlike gas roaster. Anyone who used a classical electric range vs a halogen electric range, or the latter vs a gas range will understand this easily. On my 2nd roast, I in fact put two feet on the gas to avoid staling the roast, which resulted in going full speed through 2nd crack, even if I cut the heater as first crack began.

Having a good log from roast to roast would probably ease this, as would having a bean temperature probe (to know when to cut the heater before actually reaching 1st crack).

allon wrote:Welcome to home roasting! I love the variety that is possible. You sound like you're on the right track.

Thanks!

allon wrote:How long do you rest the beans before pulling shots? It can be hard with home roasting to keep the pipeline full, and even harder to adjust technique when you aren't even tasting what you did for a week.

I used Sweet Maria's recommendations: 3 days post roast for Espresso Monkey (batch 1 and 2). As they don't provide any for the Workshop blend "Ethiopiques" (batch 3), I'll have to see for myself.

I still need to figure out if I should leave them in an open container or if putting them (after proper cooling) into closed (but not sealed) Mason jars would be worse/same/best.

I also use deep freezing to preserve roasted coffee and will use it if needed for my home roasts (http://www.home-barista.com/store-coffee-in-freezer.html).

allon wrote:You also didn't mention what shot parameters you used; I tend to pull Espresso Monkey on the cooler end of the spectrum (longer HX flush on my machine - I don't measure temps).

Since I've got the Vivaldi (three months now), I found that most coffee are best at 93°C (199°F) (not Scace calibrated and not sure my vendor did his job on this). I have not specifically tried different temperatures for the Monkey (taste comparison test). The first batch was better at a lower temp (92°C) as it seemed over roasted. I didn't bother to change the temp for my second batch: I thought that bland is bland and that I could not do much to change this. Cappuccinos and Mokaccinos it has been.

TrlstanC wrote:This is a great post, very useful for someone that's just starting to home roast. Both the list of good links, and the record of your different attempts. I'm looking forward to seeing how the latest roast tastes, and what changes you'll be making on your next batch.

I'm glad you find it useful; this is exactly why I wanted to start this thread.

Just keep in mind I will roast only every one or two weeks... but they say patience is a virtue...

Buying glasses...?

I have to admit that the urge to buy thermocouples/data logger has taken me, as I realise that what is needed to use temperature as a roast progress indicator is really the bean temperature (BT). Else, I can only use indirect indicators: color, sound, smell, smoke and past experience as guidelines to know where I am in the roasting process and when to end it.

UPDATE: I've started another thread in hope to get answers to a few questions I still have, and my research might be useful to others: Choosing thermometry equipment for home roasting.

Adding a bmp to your roaster you will never regret no matter which method you use. I preferred putting it through the rear wall but you can't loose if you use Randy's method and can always change your mind if you want later. It is my opinion that becoming familiar with all your senses when it comes to roasting is very important as well and should not be replaced with raw data. Some of my worst roasts where after I added a bmp since I stopped paying attention to what I knew and tried to roast by data, once I used both data and my eyes, nose, and sound I was getting great roasts. I still don't use a data logger, maybe one day I will but for now when I taste a roast and it tastes off I know what to change after many, many roasts, experience is king.

I will search for photos I used for bmp placement in rear wall. You want it low enough to be covered by beans when using as little as 150g.

Batch #4, July 2011

Preparing for the roast

Today I am roasting again a batch of Sweet Maria's Espresso Workshop #17 "Ethiopiques".

As I didn't screw up the last batch, I want to follow about the same profile, but as this blend is said to have bright fruit notes, something I couldn't notice in my last batch, I will adapt my profile with this comment from Ed Bourgeois on Sweet Maria's forums: How best to maximize "fruit bombs"?:

farmroast wrote:Generally
1. Check Tom notes for best roast level and any other additional notes,
2. don't over preheat
3. don't over dry
4. don't go too slowly during ramp from 300f to first crack
5. ramp down the rate of bean temp. increase from about 30 secs before start of 1st til suggested finish. So that time from start of 1st to dump is 3 1/2 -5 1/2 mins. I'd shoot for 4 1/2 for brewed and 5-5 1/2 for espresso.
cheers,
Ed

and the following post by Thompson Owen:

Sweet Maria's wrote:The reply is very good - don't drop the coffee in to a "too hot" environment. If you want to maximize the highest tones of fruit, I recommend NOT going for a really slow finish, where you would start to drop the temperature just a bit before 1st crack (just a little, not so much as to stall) and then slowly drop temp. as 1st finishes to have a nice slow progression to your finish temperature, which I would say should be 430 or so on a roaster where 1st crack starts around 400-405. Now normally what I describe above is a great technique to develop body, sweetness and complexity, but to keep higher fruit tones, I sometimes think that having a nice slow, controlled first crack, and finishing the roast without dropping heat will give that to the cup. Of course, as always, I disprove my own theories constantly, so it comes down to cupping all the time. Cup it, re-roast it. Cup it, re-roast it. Repeat!

To summarize the changes:
- Don't over preheat. Instead of the 250°F I used before, I will try 200°F (default drop-in on the Hottop is at 160°F).
- Turn power to max when I feel the drying phase is about to end (somewhere between yellow and turning tan)
- Reduce power a bit more during 1C (<50%) and use a higher fan setting, to prolong 1C and aim (as far as I can!) for 5 1/2 min. between beginning of 1C and EOR.

Chosen profile:
200°F Drop-in.
290°F Reduce power to 70%.
320°F Reduce power to 50%.
In the middle of the drying phase (judged by colour), fan at 50%.
Before the end of the drying phase (color yellow to tan), power back to 100%.
At the first pops of 1C, reduce power to 40% and raise fan to 100%.
Getting out of 1C, take fan back to 75%.
At the very first snap of 2C, hit Eject.

Roaster has been emptied from jammed beans this time...

The roast

Note: This batch weighs only 209g.

0:00 200°F Drop-in. (H10, F0)
1:00 230°F No color change.
2:00 255°F Greener.
3:00 285°F Turning bright green. (H7)
4:00 310°F Turning yellowish.
4:30 320°F (H5)
5:00 325°F Yellow, turning to tan. (H10; F2)
5:30 340°F Fully tan.
6:00 345°F Turning to light brown.
7:00 355°F Constant smoke.
7:30 365°F Turning to brown.
8:00 370°F Brown.
8:30 380°F First pops of 1C. (H4, F4). Also removing loading chute cover to further slow down the exothermic reaction.
9:00 385°F Rolling 1C.
10:00 393°F End of 1C. Put cover back on. (still F4)
11:00 399°F (F3)
12:00 First snaps of 2C. Ejecting.

About 3:00 min of drying time.
About 3:30 between beginning of 1C and EOR.
209g pre-roast. 170g post roast. -19%.

How it went and lessons learned

Hottop recommends a 250g load, but I remembered today (d'oh!) that 1 lbs (one bag from Sweet Maria's) ≠ 2x 250g... I will standardize my future batches on 225g instead of 250g, so that I can do two identical batches out of 1 lbs. This batch probably was roasting faster than usual.

I think I was able to go through the drying phase just a bit faster, but maybe not as much as I would have liked. We'll see later if I was able to preserve bright fruit aromas.

I went though 1C as fast as usual, while I was trying to go slower. Without BT and past experience with the same beans/weight, I dropped the power when I heard the first pops, which was too late. I chose to remove the chute cover to let the heat out (quite successfully), but this was not enough to slow down 1C (duration of 1:30). First snaps of 2C came 1:30 after that.

Hopefully, by using a new standard batch weight and my logs, I will eventually be able to turn the heater down before 1C without risking stalling the roast. Of course, the final solution would be thermometry with the bean temperature reading. Still need to thing about it.

The roasted beans look really nice. Light brown. Chaff in crease stayed light. Beans show a nice texture/colour contrast (a marbled look). [Correction:] A Full City+ roast.

Same thing as the last batch: after the roast, pepper and cacao aromas.

I'll have to see what it tastes like... (better of worse?).


Batch #4. Sweet Maria's Espresso Workshop #17: "Ethiopiques". Picture taken right after the roast.