Given the results of the recent competition, it seems that this method has shown to be a capable roasting device. With several topics detailing profiling/modding/details of other roasting devices, I'd like to open this topic to any and all interested parties to discuss the methodology of HG/BM roasting.

Kupe - looking at you and your strong showing to lead the way!

-s.

I use a HG/BM some of the time, and P1 some of the time.

I'd be interested hearing details about how others have their HG/BM system set up-- insulated or not? location and type of temperature probes? bean-stirrer shape, design?

My setup is: uninsulated - original bread machine housing. A 'lid' made from thin aluminum sheet, with a hole drilled to fit the snout of the heatgun. My temp probe is near the bottom of the pan, I've experimented with different locations and different types of probes. bean-stirrer is thin aluminum bar stock.

I'm a little embarrassed because my setup is about as basic as they come. I don't even have a stand for the heat gun and hold it myself the whole time! I don't have much time right now, but I'll try to take some pictures and talk about how I go about roasting when I get a chance.

I'm a HG/BM roaster.

+BM rewired for continuous agitation
+No insulation
+Leave lid attached, remove sight glass, form aluminum foil "gasket" for heatgun to fit in
+Roast with lid closed
+Drilled 1" vent hole at lower, rear of breadmaker for chaff/smoke to exit
+Heatgun tip hovers just inside the lid, about 3 inches above beans after they've expanded
+Use a triac to control the heatgun for easy variable on-the-fly heat changes
+Stainless thermocouple probe just above agitator, 1" from outer wall, fully in bean mass
+Tilt breadmaker 15 degrees back toward TC
+Preheat to ~300F for a few minutes (sometimes I don't if I want to slow down the drying phase)
+1lb to 1.25lb standard charge of beans
+Roast in unheated, attached garage, with door partially open, and roasting works well even with ambient temps well below freezing

This setup gives great control over heat. Very easy to adjust drying times, ramps, stretching between first and second crack.

CHAD

Apologies to kupe - didn't mean to throw a spotlight on you.

I started with HG/BM based on info found on another forum and based my setup on seedlings, who posted above. Current setup is:

• BM rewired for continuous agitation. Original stirring paddle.
• Fiberglass insulation between basket & outer housing
• Heat controlled by plugging HG into router speed control
• Concave metal lid w/sightglass removed. Hose clamp on HG snout, foil wrapped around hose clamp
• Piece of pantyhose covering HG intake to prevent chaff inhalation. Fringe benefit is that it looks swank.
• Type J TC drilled through outer housing & basket, monitoring at ~1" up from bottom for BT
• Type K TC drilled through lid, monitoring ~3" down for ET
• Typical Load: 350g
• Typical Drop-in temps: 400F measured at BT

I've tested 100g-500g loads and drop-in temps from room temp to 450F, and found 350g with 400F BT sensor to be the best balance of quantity and responsiveness for my setup. At 350g:

• BT turnaround at ~235F at 1:00-1:30
• "Yellowing" ~285F BT at 3:00-5:00, dependent upon heat input
• 1C ~360F BT starting at 6:00-8:00, dependent upon heat input
• 1C duration up 90-150sec, dependent upon heat input
  

I still need cobble together and post the data output from my competition roasts, but I'm finding that, regardless of bean, I can get a 350g load to behave in a fairly consistent manner - turnaround at 1:00, yellowing ~285F at 4:00, 1C starting at 7:00-7:15, lasting up to 2 minutes, stopping at 3-4 minutes after beginning of 1C to produce FC roasts.

One continued source of frustration has been my own ignorance of how to take accurate measurements, how to correlate the measurements, and how to affect change. I'm not even sure that the ET probe is worthwhile, but would rather have too much data than not enough. I'm using the TCs as much as I'm using my eyes, ears and nose to tweak roasts, but there is hope.

I've found one single exercise to be immensely useful as a way of learning both how to use my roaster and how to train my palate. (By no means do I make any claim about the completeness or correctness of this exercise, but it has proven to be a great tool for me, so I'm sharing it. Take from it what you wish, and feel free to provide corrections/updates):
Necessary equipment: notepad, paper, timer with resolution of minutes & seconds

1. Pick one bean, preferably one from a known source that has been cupped by an accepted authority (that is, whomever you believe to have sufficiently-developed abilities with respect to cupping) and pick one load
2. Given some basic guidelines (turn at 1:00, yellowing at 4:00, 1C at 7:00 in my case), try roasting a batch to the end of 1C and stopping there, modulating your heat so that 1C lasts roughly 90 seconds. If you can't meet those guidelines, change your load until you can. Record your results. Also helpful - what are you smelling at these various points?
3. Keep roasting batches until you can determine and follow that profile, within reason (+/- 15 sec.)
4. MOST IMPORTANTLY, cup your roasts. Don't know how? CoffeeGeek, CoffeeCuppers.com, and even Home-Barista give you a starting point. Try cupping 3 at a time. Compare your tasting notes those from item 1. Are you tasting the same, similar, or completely different notes than what is listed? Are you tasting consistent notes across batches?
5. Keep doing this until you can consistently taste similar notes across all of your cupping batches for a given tasting session.

Congratulations. You've just learned how to do a consistent roast. The goal here isn't great tasting coffee, it's consistent results. Once you can repeat your results, you can start tweaking. "Great tasting" is completely subjective. "Consistent" isn't.

-s.

Haha, that's fine, I don't mind. I just really didn't expect to place well and end up explaining my poor methods. I guess this is a good reason to make some long-needed improvements. Here are the only modifications I've made:

• Moved BM's internal thermometer to prevent it from detecting excessive heat and shutting down.
• Removed glass window so that I could put the heat gun through. I don't use any foil or gasket of any sort as I always have plenty of heat.

That's it. Both your and Seedling's setups sound impressive and would make it much easier to be consistent, so I plan on taking what I can from your configurations. I do a lot that can produce inconsistency:

• Holding the heat gun manually rather than with a stand in a static position (I rotate it against the agitation of the beans the whole time. I figured this would assist with obtaining an even roast which seems to be the case, but it is an extra variable that could be eliminated while saving a lot of manual labor).
• No thermometer of any sort.
• Stirring paddle not wired for continuous agitation (I really need to do this). It runs at a slow speed for the first three minutes, but picks up to full for the rest of the roast with plenty of extra time on the cycle. I always have the lid up for the beginning phase and keep the heat gun pretty far away and moving to prevent scorching. Sometimes I won't add the beans for a couple of minutes, but for my contest entry, I did the full 3 minutes of slow agitation. I normally open the lid again at first crack so I can finesse the roast a bit and blow out all the chaff.

Normally I'll just go by sight, sound, smell, and time from there, and that's why my first roast of the Haile Selassie got away from me into vienna territory. 1st and 2nd crack ran together worse than any coffee I've roasted before and it was hard to tell by color in the night with my cheap little headlamp. That was also a tasty roast, but I much preferred it at full city. I do keep my load consistent at 450g and manage similar times to yours, Sherman. 7:00-9:00 minutes for first crack, definitely. Haile Selassie's first crack for me was just short of 9 minutes and I stopped at somewhere in the 11:00 minute range. I've done a roast of it since that ran about 1:30 slower overall for a similar FC roast level, but I don't feel it was quite as good as what I submitted for the contest. Still tasty, though. But that really shows how important consistency is. I want to be able to reproduce what I submitted or improve on it every time I roast. Now that I know for sure that HG/BM is a very respectable roasting method on any terms, I feel it's worth investing into some more. Looks like I need to order some equipment and get modding.

Congrats Kupe,

High praise from Tom of S.M.'s. Check out his video and commentary here:

http://www.youtube.com/user/sweetmarias

Your roast is discussed in Part #2 of 'Macro Mayhem'

-bill

Hahaha, wow...I really am stunned. I guess Tom doesn't know what we all used to roast though. I wonder if he would be surprised how many of them were from HG/BM users? Very cool to see all the roasts up close like that. It's interesting that he mentioned culling. Now that I think about it, I probably removed a few quakers post-roast, but I always do that so I didn't really think about it.

Also, now I finally know how to pronounce "Haile Selassie".

kupe wrote:

• Holding the heat gun manually rather than with a stand in a static position (I rotate it against the agitation of the beans the whole time. I figured this would assist with obtaining an even roast which seems to be the case, but it is an extra variable that could be eliminated while saving a lot of manual labor).
• No thermometer of any sort.
• Stirring paddle not wired for continuous agitation (I really need to do this).

FWIW - seedlings was my primary source for information regarding the re-wiring bit. If you need specific info, we can go into detail here, or you can reference the thread at homeroasters.org. Mods & seedlings, if there are any issues regarding cross-posting, I'd be happy to remove this link.

Regarding thermometers, there are 4 options that I'm aware of:

1. Super-cheap (<$20): analog oven thermometer - Pros: cheap. Cons: only monitors one temp, slow to respond, not necessarily accurate or significant (depending on placement). can buy multiples, but accuracy/responsiveness issue persists.
2. Semi-cheap (~$20): digital oven thermometer - Pros: cheap, faster response than analog Cons: same as analog. can buy multiples, but accuracy/responsiveness issue persists.
3. Slightly spendy ($30-100): eBay-ed thermocouple thermometers - Pros: accurate, placement is up to you. Cons: only 1 input. Usually requires drilling of some sort. Not that it'd affect your roasts, but if you're not handy/confident with a drill, this could be an issue
4. Spendy (>$100): eBayed/retail thermocouple data loggers - Pros: same as others, usually includes multiple inputs, multiple TC type support, plus whiz-bang data logging and download options. Cons: lightens your wallet considerably.

kupe wrote:Both your and Seedling's setups sound impressive...

I can't speak for seedlings, but can share that my setup looks a whole lot less impressive than it sounds. I'm roasting in an unheated, standalone garage, with everything situated on some Ikea shelving that was screwed into the studs. Wintertime roasting means donning an old coat and hat to keep warm. High tech stuff, to be sure.

Good luck with the modding, and I hope you'll share the results.

-s.

Huh...I don't know how I've never run across that site before. Thanks so much! I'm pretty short on time lately, but I'll try to do the mod soon and will ask questions if I have them. As for the thermometer, I think the 2nd category is what I'll go with for now. I've been looking at the one SM carries for a while at the bottom of this page:

http://www.sweetmarias.com/prod.cupping-brewing.php

I figure that would do the trick. I think I'd be able to handle the drilling alright. And as long as people don't mind my rambly non-expertness, I'll be glad to share plenty.