Behmor 1600 Coffee Roaster - 1st Look

[miKe mcKoffee]

Note: This is the final pre-production factory production test run unit.

I just wanted to toss this out. I posted this over on CG also. I currently have on testing loan one of the pre-production models of the Behmor 1600 and have been test driving about a week.

With bean weight, 5 profiles and length of time adjustments (the profile length changes if you change in initial length of roast as it is based on % not minutes) there are a LOT of permutations to test out, but if people have some specific questions they want answered now that one is available to test, please ask them.

Just thought I'd mention a second sighting has occurred. Here's cut & paste post I made on SM List earlier this evening...

Damn, I'm impressed early with what a MMAAP $299 (Mfg. Minimum Allowed Advertised Price) 1# roaster can do. Yeah, mine arrived this afternoon. During PNWG V had way too much other stuff to do to give it much attention. Obviously haven't had much time with it, like less than 3 hours. Did a empty burn in then 1# Generic Guat test roast to City+, 18 minute roast using P2, a profile with heater 100% about the 1st 2/3 time then drops to 70% end stage. Or more accurately this profile has 70% heater followed by a short 10% of time at 100% heat very end. (Pound of Guat Joe had left, given to him and stated to be less than stellar.) Decided to roast in the kitchen with hood off. No visible smoke, the bare faintest roast odor. After cooling had kicked in waited about a minute and opened the door for a faster cooler, now smoke came out (but not that much) and I kicked on the twin cyclone 700cfm exhaust hood. Cool to touch in about 4 minutes from end of roast. Roast a bit uneven but not severely so. Total of two beans stuck in the mesh drum. Chaff removal very good. Of course, opening the door during cooling does allow some chaff to escape the roaster, mini shop vac on the ready.

Ok, since this morning I'd roasted two half pounders WP Yirg for a frozen greens roast test (frozen and defrosted before roasting plus usually vac sealed room temp greens) decided to roast a half of the same Yirg' in the Behmor. Used a different profile than the Guat. Three stages each stage increasing heat. Has five built in profiles, all modifiable in time
percentages middle and end stages. The "profiles" themselves cannot be changed, yet still has good flexibility. More accurately a given stages heater power percentage cannot be changed, how long at that stage can. Decided to leave door closed for entire cooling, outer door moderately warm 5 minute mark, virtually room temp end of 10 minute cooling cycle. Drum loading and removal easy and straight forward IMO. Oh, no gloves required. Unless you want to remove the drum at end of roast to manually speed cool of course. But I'm going to do all my testing cooling in Behmor.

Preliminary indications IMO this is gonna be one hell of a 1# sub $300 off the shelf roaster. Expect them to hit the market end of August first part of September per talking with Joe today. More will be revealed. Hopefully Tom decides to carry them...

If there's any negative so far it's that I can't smell the roast progression nearly as well as from the Rosto exhaust. Dang afterburner doing it's job!

Oh, IMO the inside light does a good job illuminating the roast for visual. Will want to keep the door window clean.

May do an early 3 way cupping of the Yirg's tomorrow morning.

[Cathi]

This IS good news. I'll be following this one closely. More, more, more! (Please )

[miKe mcKoffee]

First not sure if Dan would prefer these posts in the New home roaster to debut at SCAA Long Beach CG thread, but since A) I spend more time here and B) that thread was primarily speculations rather than observations I'll continue here unless requested otherwise. (edit: Dan okie-dokied keeping it here)

I'll get the potential major problem out of the way from the git go. Namely this, some may find it too easy to roast too much too fast! Especially when playing with a new toy. Glad I picked up a used 13cuft freezer targeted for dedicated coffee storage just last weekend. While was intended for vac'd greens looks like some vac'd roasted may be there right quickly.

The preliminary pre-release documentation includes the following:

The Behmor 1600 is the world's first and only one-pound capacity home coffee roaster designed strictly for roasting coffee. When used with high quality beans, The Behmor 1600 roasts in virtually every application with no visible smoke up to 2nd crack*. The Behmor 1600 is the only roaster in the world manufactured with a variable speed motor to facilitate proper bean rotation, plus allow for higher more efficient cooling and chaff removal. In addition, the Behmor 1600 variable-speed fan controls allowing for optimum functionality.

The Behmor 1600 will, in the overwhelming majority of cases, take up to a single pound of green coffee into 2nd crack* without visible smoke being emitted. For Full City+ roasts, we recommend you reduce the quantity of the roast to ½ pound. The Behmor 1600 is not intended nor meant to roast coffee to levels known as Vienna, French, and Italian or darker.

Taking close note of the admonition of doing smaller 1/2# batches for darker roasts I of course proceeded to start today's trials doing a slightly over spec' sized 18oz batch to FC maybe FC+. Chose the most heat aggressive profile with roast time set for 20 minutes, this profile basically 100% heater full time. 1st started ~15minute mark, few early 2nd's 19:30, hit cooling @20 as 2nd just really starting, last slow 2nd's ~1:30 into cooling. (Did not take it into rolling 2nd, I never do:-) A wee bit of smoke was noticeable as cooling started, to be expected. Decided to go for total door closed cooling. Had added 30 seconds for max 20:30 but killed at 20 anyway anticipating 2nd continuing first part of cooling. FWIW this was of a very mixed Arabica espresso blend. Was a blend of mine comprised of 25% split between aged Sumatra & Monsoon Malabar low end, 40% 2 different Brazil mid base notes, 35% WP Yirg' and Guat' bourbon. But this batch even more mixed! Only 11.1oz left of my blend so topped if off with Tom's Monkey Blend for 18oz. FWIW I hadn't intended (as in didn't think of it) to shoot for a post roast pound but will later. 18oz yielded 14.9oz, 17.2% weight loss. Next time I'll do 19.2 oz for post roast full pound test.

Initial impression gorgeous FC+ roast with nary a drop of oil, but not quite exactly were I'd wanted it. Had targeted FC but missed slightly with the continued bit of roasting with door closed cooling, will just be a matter of learning the roaster. Very even roast, especially considering the extreme pre-roast blend. Definitely looking forward to how the shots pull in a few days. Ok waited a whole hour and pulled a shot. Actually not bad at all yet way to young to judge.

Ok, time for a really serious test. Next roasted a 1/2# Kona Zuma Estate. Chose P5, the gentlest profile, initial time set for 14 minutes. Added 30 sec additional time during 1st and ended roast 14:15 as 1st thoroughly finished but well before 2nd, about City+. (Why I didn't record 1st crack start times the next two roasts is beyond me. ) Now I've roasted literally hundreds of Kona batches the past going on 7 years. My norm is to post roast munch a few beans and have been able to fairly well tell how well it'll be in the cup from the munch. Based on the munch it's looking to be an excellent balanced smooth and sweet with a good acidic dance Kona roast.

Next tossed in a 1/2# of India Mysore Nuggets Extra Bold again using P5 set 15. During 1st added 30 seconds targeting FC and let cooling kick in 15:30. Again in roaster closed door. This time a hair lighter than wanted, should have either added another 15 seconds or used a more aggressive profile. But the post roast munch sure was good so maybe not!

Was going to roast some Rwanda Bourbon next using a more aggressive profile but decided what with roasting 3 half pounders of Yirg' plus a pound of Guat' yesterday plus today's roasts I'd best quit playing.

So far nothing has changed my early opinion the Behmor 1600 is going to be a winner. Sure it would be nice to have things like bean mass and roast environment temp monitoring/display, tryer, full user programmable profiles including automatic end of roast by bean temp option, no drum removal bean dump. Then again commercial roasters with that type of control cost many thousands just for the control systems not ~$300! And while there are very good control systems for home roasters being built (and sometimes sold) even those control systems cost way more than the Behmor 1600. OTOH in the hands of the geekier home roasters this puppy might well turn into a totally automated fully profilable roaster. But it's got pretty good roast flexibility just the way it is.

(Edit) First corrected Kona roast time above. Also forgot to mention that while P2 profile is "100% heater" applied full time that really doesn't mean the heater will be on 100% of the time. It means it'll be on 100% heat power until max designed roast temp is attained then toggle on and off maintaining max roast temp.

[miKe mcKoffee]

Dan gently chided me

Try to restrain your enthusiasm, HB prides itself on dispassionate evaluations.

Ok, yes I'm more than somewhat surprised and delighted at the feature set and and preliminary performance results for such an affordable off the shelf 1# max (plus a bit more ) roaster.

That said I don't particular like the way it looks reminding me of a kitchen appliance small microwave/large toaster oven, 'taint no stylin' SF1! I'll add it seems more solidly made like a good microwave than a toaster oven regardless the looks. Pic's and videos to follow.

What can I say. Been home roasting going on 7 years via variable boost voltage controlled Rosto and this is the first off the shelf roaster to really get my attention. No it's not in nearly the same profile control league as Jeffrey's CCR modified HotTop or Mike's (just plain) PID P1 Uber Poppers or Jim's or others, but not in the same price league either plus larger batch size as CCR HotTop and no hacking electronic knowledge required for "some" decent indoor roasting control. Considering it's batch capacity, (limited) profile capability and price point the Behmor 1600 could be a serious HotTop/Gene Cafe killer (read forcing price reductions). Don't see how it'll do anything but improve the home roast community's options.

But I'll attempt to keep future ramblings strictly to observed details and facts and post-roast tastings!

[dBndbit]

Thanks for the great first-hand info. So far from what you've said, I'm most concerned that the long roast times for 1# batches will miss the flavor peaks on many beans. So I'm eager to hear any flavor shoot-out info you might have. Also, since we're totally ignorant of all the details on this machine... you mentioned "max roast temp". Is that adjustable on the front panel? I mean other than cranking up the voltage with a transformer?

Jim

[luca]

Dan gently chided me

Try to restrain your enthusiasm, HB prides itself on dispassionate evaluations.

Yes, I was thinking the same thing. I can't remember how many reviews I have read where people said that something was the "best" and promptly ended up changing their minds several months later.

Cheers,

Luca

[miKe mcKoffee]

Thanks for the great first-hand info.

You're welcome.

So far from what you've said, I'm most concerned that the long roast times for 1# batches will miss the flavor peaks on many beans. So I'm eager to hear any flavor shoot-out info you might have.

Long compared to what? IMO while the roasts times with max rated load of stated 18 to 23 minutes may seem long, but compared to what? Roasters like Diedrichs when operated at their max capacity roast in these time ranges, the HotTops advertised roast times for 9oz batch are 17 to 23 minutes. Compared to a fluid bed roaster sure. However the beans heat differently in a drum and for one take longer to come up to actually roasting temps from room temp versus fluid bed. And different drum roasters heat differently. With the Behmor the drum environment is not pre-heated before bean load like many drum roasters including the HotTop. So a portion of the total roast time of the Behmor doesn't really count in a time comparison.

The truth of course is only one place, in the cup. The challenge for flavor shoot-outs is multifold. Can't afford to roast a whole bunch of 1# batches just for taste testing and tasting would be limited to myself!

That said if some would like to compare a Behmor roast to their roasts I'm willing to do it for them. They'd ship me a pound of greens plus a check or MO for $4.60 to cover USPS Priority US shipping back to them. I'd roast and ship same day. If wanting more than a pound roasted say to compare 1 & 1/2# batches of the same or multiple greens or same green different profiles etc, shipping would be $8.95 US up to about 12#. (Canada or Mexico would be $9 for 1#, $23 up to 12#. All other countries $11/$37.) I'd also request $0.50 per batch roasted to defray cost of valve bag for the roast(s). Sorry but finances are tight and this isn't a sponsored review with a bunch of coffee provided for tests!

Also, since we're totally ignorant of all the details on this machine... you mentioned "max roast temp". Is that adjustable on the front panel? I mean other than cranking up the voltage with a transformer?
Jim

I've been requested by Joe not to give out design detail specifics do to competitor snooping etc. but think this is ok to answer. The maximum roast chamber temp is fixed (safety design feature) and equates to ~482f. Cranking up the voltage would not raise the maximum temp, the heaters would simply be toggled off more. Until/unless hacking the Behmor of course.

See also http://www.behmor.com/

[miKe mcKoffee]

Yes, I was thinking the same thing. I can't remember how many reviews I have read where people said that something was the "best" and promptly ended up changing their minds several months later.

Cheers,

Luca

I don't disagree. For the record note I've used terms like preliminary results, preliminary indications, impressed early, initial impression, early opinion etc. Never said it's the best home roaster imaginable, even enumerated a few things I wish it had.

[dBndbit]

I'm relatively new to roasting, so I refrain from claiming any fixed opinions. But from what I've read and what I've experienced, many beans loose flavor when the roast time to 2nd crack goes out much beyond 15 minutes. And pre-heating a machine is usually an easy way to minimize time to first crack.

I understand and sympathize with the heartbreak of sacrificing a bunch of good beans. But this is for science! So I'll gladly help out. I've got some pretty good Kenyan PB in the stash. They roast a little faster than normal beans due to their size, but we can mentally adjust for that. PM me with your address and I'll send a bag your way.

Here's what I propose... I'll trust your taste buds. I'll send a couple of pounds to cover at least three roasts: 1#, 1/2#, and 1/4#. You keep the coffee and just post your flavor analysis, or just comments on any differences in the roasts. Brew it however your taste buds would be most sensitive. For roasting, pick the same (fastest=hottest?) profile on the Behmor. Pre-heat the Behmor in some predictable way for each roast. Roast all three exactly to the beginning of 2nd crack. (first crack of second crack?) Normally I wouldn't roast these beans quite that far, but we need an exact comparison point. Please keep a record of times for beginning of first crack and second crack. And please, if possible, also pull the beans out of the roaster for an immediate cool down. Is this an OK plan? The results would be really interesting, I think.

But this asks a lot of time and fiddling, so if your life is tightly scheduled I understand why this kind of testing is unreasonable.

Regardless of the testing, I look forward to reading all your comments.

[miKe mcKoffee]

I'm relatively new to roasting, so I refrain from claiming any fixed opinions. But from what I've read and what I've experienced, many beans loose flavor when the roast time to 2nd crack goes out much beyond 15 minutes. And pre-heating a machine is usually an easy way to minimize time to first crack.

I understand and sympathize with the heartbreak of sacrificing a bunch of good beans. But this is for science! So I'll gladly help out. I've got some pretty good Kenyan PB in the stash. They roast a little faster than normal beans due to their size, but we can mentally adjust for that. PM me with your address and I'll send a bag your way.

Here's what I propose... I'll trust your taste buds. I'll send a couple of pounds to cover at least three roasts: 1#, 1/2#, and 1/4#. You keep the coffee and just post your flavor analysis, or just comments on any differences in the roasts. Brew it however your taste buds would be most sensitive. For roasting, pick the same (fastest=hottest?) profile on the Behmor. Pre-heat the Behmor in some predictable way for each roast. Roast all three exactly to the beginning of 2nd crack. (first crack of second crack?) Normally I wouldn't roast these beans quite that far, but we need an exact comparison point. Please keep a record of times for beginning of first crack and second crack. And please, if possible, also pull the beans out of the roaster for an immediate cool down. Is this an OK plan? The results would be really interesting, I think.

But this asks a lot of time and fiddling, so if your life is tightly scheduled I understand why this kind of testing is unreasonable.

Regardless of the testing, I look forward to reading all your comments.

I believe would/could be more beneficial to have roasts tasted by those familiar with the bean compared to their other roast method(s). Plus more tasters would provide greater feedback. Might also be possible to get some advanced cuppers (like Jim) to agree to have roasts shipped to them. But sorry to say I'm not in a financial position to pay for a bunch of shipping.

Cooling wise no, I'm not willing nor want to use extreme put on the gloves yank out the drum cooling methods. First using the documented modified method of opening door a minute or two after cooling kicks in cools amply fast enough IMO (and that of many professional roasters) plus want to see how total closed door cooling affects the cup. I will be doing some roasts of same bean same profile etc. comparing the those two methods. My understanding the total closed door cooling is a by design ease of use low cost compromise solution for the novice who wants it clean and simple and in the kitchen.

[miKe mcKoffee]

Roasted another couple 1# batches today using two different profiles, both to FC+. Don't have time to post specifics about the roasts at the moment but wanted to post a bit about the oft discussed in roaster long cooling cycle times. By default the cooling is a 13 minute timed cycle for 1# roast. But does that mean it takes 13 minutes to cool the beans? And if it actually does how does it affect the cup? I'll not attempt hazard an answer for the 2nd question yet. One thing I observed yesterday was that the afterburner stays on for the first 3 minutes of the cooling cycle! Egads me thought, since it's located inside the roast chamber that can only make cooling worse! Granted it's located in the top towards the front positioned as far as possible from the rotating bean mass but still the heat is generated inside the roast chamber. So the 2nd roast of the day I broke out the K TC and decided to see what cooling temps actually where. (Actually I started by attempting to monitor chamber temp during the roast but in a rush chose a terrible position bottom right front which turned out to be over 200f low actually bean roasting temp based on when element shot off during 100% stage. Think it happened to also be near fan inlet to chamber...) Anyway after taking the roast into cooling going for best possible in roaster cooling by immediately opening the door. Fished the TC to top just off center curling down to just above bean mass height. (Couldn't feed directly center do to the after burner being there.) Started recording the roast chamber cooling temps at 3 minute in mark. (Next time will start recording sooner)

With ambient temp 75f opening door beginning of cooling FC+ roast:
3min 190f, 4min 161f, 5min 140f, 6min 122f. Recorded down to 9min 96f but really meaningless at this point IMO. Maybe not great but not nearly as terrible as some may have feared. And IMO at the upper end of acceptable since getting down below 200f in a timely manner most important. A way suggested to me for speeding up in roaster open door cool could be directing an outside fan at the chamber. Should work if desiring faster cooling without removing hot objects. If removed the (hot) chaff collector, which would partially impede outside blowing air from entering, would work even better. Or of course simply removing the (hot) drum and fully different cooling method. But removing the chaff collector and or drum with air blowing would also lead to a messier proposition with chaff blowing around. Even simply opening the door during cooling does allow a small bit of chaff to escape.

I've come to believe there may have been some fundamental error in designing the roast profiles. Possibly caused mainly because they were suggested by professional roasters! Huh? A regular commercial drum uses a profile based on beans being loaded into a preheated drum/chamber not starting roast from ambient roast environment like the Behmor! The problem is primarily with 1# roasts, actually pretty good times for 1/2# batches. Specifically starting 3 of the 5 profiles at 70% heater power then raising in two later steps. (Going into more discussion than I'd planned...) For example a 1# roast with P1 100% all the time profile FC+ roast took 18:30 with couple early 1st's ~14, 1st in earnest 16:45 and first early 2nd's ~18. Maybe a bit fast end stage but fine for accentuating brightness. OTOH P4 profile uses 70% heater power first 30% of time, 85% - 30%, 100% - 40%. This profile took 24:30 for FC+ with early 1st's ~20min, 1st in earnest ~20:45, early 2nd's 24, 2nd kicking in 24:15 ended roast 24:30 (max allowed time) with open door cooling last trailing 2nd ~40sec into cooling. The profile I thought I'd use most, P2, has 100% power first 60%, 70% - 30%, 100% last 10% of time. I've ended up not using it much yet, feeling the 70% (targeting ~374f environment) is dropping to low with the air flow for bean momentum to carry. My gut tells me closer to 80% (~410f environment target) would be better. But I need to play with P2 some more to confirm and with good TC placement.

Unfortunately while the three profiles with 70% first stage seem ok for 1/2# batches, seem a bit slow first stage for 1# batches. What might work better for 1# max batches could be initial 100% short stage to get it going then drop to ramping up. FWIW the base profiles are identical relative to percentage of roast time for any chosen batch size, only the total roast time varies.

This dang thing is so close even for a roast control freak like me. Yet I do believe most newer home roasters would be delighted and even many not so new. And I'm not giving up trying to make it work reasonably well for 1# batches (without hacking) yet! Now simply hacking to manually control the heaters with a variac, or automatically with PID or computer and I've little doubt this pupply could really sing.

And for my reality check. Designing an electric roaster to perform even half way decently with 1# batch on a 15A circuit is no small feat, and virtually smoke free and for $300 at that.

[miKe mcKoffee]

I've hit a wall and am going to back up and punt. This will likely be my last Behmor 1600 post for a spell. Basically going to start over attempting to be methodical and dispassionate. Last night realized I really need a bunch, a whole bunch, of disposal greens to really get to know the 5 profiles and their potential variations before doing any real roasts for cupping. Tom graciously agreed to ship me 40# of some Columbia for cost of shipping that he

liked the coffee okay, but felt it had a hint of travel damage (ie it steamed in port or on the ship)

and the broker didn't want back.

Few (or more) parting comments.

Concerning P2 and it's 70% heater relating to ~374f roasting temp. Not what I observed today! First I did an empty P2 run with TC mounted upper right inside chaff collector which equates to about 1" below top of drum, centered between afterburner and side wall. Max initial stage (100% heater) peaked highest point ~500f, 70% stage ran around ~410f. That looks much more like what would be needed to ease through 2nd so decided to go for another pound test roast. Recap default P2 has 100% heater first 60% roast time, 70% 30% time, 100% last 10% roast time. The max time possible this profile is 22.5 minutes.

Played with the numbers and chose 21 minutes for start time which would give 12.6 minutes 100%, 6.3 minutes 70% and 100% last 2.1 minutes. Chose this because previous 100% up to 1st had it starting ~13 minute mark. (Would have liked S2 shorter ~5.5 minutes but not possible without also shortening S1 so had to compromise. Also decided to add time at the end by starting roast and immediately adding 1.5 minutes to max 22.5 yielding last stage 3.6 minutes at 100%.
Running modified P2 22.5 minutes:
S1 12.6 min 100% heater
S2 6.3 min 70% heater
S3 3.6 min 100% heater
Ambient 70f, no load 120.4v, 118.1v start of roast, 117.2v when afterburner and fan kicked on. (Diidn't catch the time, was watching time versus temp and writing it down.)
Times listed counting down from 22.5
21 215f
20 230
19 323
18 356
17 380
16 400
15 412 (afterburner and fan kicked soon after causing)
14 383
13 420
12 440
11 453 (main heater shuts off for first time)
10 440
9:30 439
9 426
8:30 401
8 395
7:30 398
7 401
6:30 405
6 407
5:30 411
5 413
4:30 416
4 418 (thought I heard a very few early 1st's)
3:30 420
3 422 (not hearing 1st cracks?)
2:30 430
2 447 (still quiet)
1:30 450 (thought I heard real 1st kicking in, but was mistaken, it was 2nd!)
1 457
0:30 450
0 463 (uh oh, that was definitely 2nd as fan kicked to high a more smoke than I'd ever see before from the Behmor leaked out)

13 minute cooling cycle, first 2 minutes left door closed still thinking 2nd was 1st
12:30 366f (still slowly cracking)
12 340 (very intermittent cracks)
11:30 320 (no cracks heard)
11 304 (opened door)
10:30 269
10 240 (3 minutes into cooling with door closed first 2 minutes, also afterburner turns off now)
9 193
8 160
7 135
6 118 (quit recording cooling environment temps and let it run to end)

No question was 2nd crack I was hearing, Light Vienna or a hair darker, fairly dark brown with very light sheen.

Big thing I learned (or rather was vividly reminded of) easing through first like that can silence 1st crack!
Second thing Behmor quite capable of darker roasts full pound. Intentionally left the door closed at first to let it keep roasting longer first part of cooling stage, cooling would have been faster immediately opening the door. (Which will be documented later)

Based on this roast result a FC would have been attained ~20 minute this profile. Some may say too long, I'll say nothing until cupping some samples these times this roaster! And interestingly a well respected international cupper and greens supplier says this about the HotTop on his site:

The roast profile of the long (17-23 minute) roast cycle reproduces the results of shop drum roaster more than any other home roast model. This especially suits folks who roast for espresso extraction, but in cupping the HotTops lighter roasts, I am impressed with the aromas and delicate balance of the cup quality.

Another thing I learned (again) is this sure would be easier with a data logging digital thermometer!

Now to wait for forty pounds of steamed Columbia greens to arrive for serious testing to begin. (Anyone have a datalogger they want to loan me?)

[Alchemist]

You hit the same wall I have in regards to beans. The permutations are so large, that in practicality there is no good way for one person to test the Behmor. It is really going to take it making it to market and having people post and pool data and experiences. I stopped roasting for tasting a dozen roasts or so ago and have been roasting a throw away bean just to learn and get a feel for the profiles.

I see you hit the same items I did in a couple regards to observations. Yeah, Joe got these profiles from professional roasters and profiles of bean mass temperatures. A bit different from ambient roaster temperatures IMO. I recall Joe mentioned P2 was from a professional drum roaster, and that makes sense. Hit high, reduce heat, raise to finish (sometimes).

I too have found that 1 lb is possible, but IMO it is better suited for 8-12 oz roasts, at least on the later profiles. One thing you may not have noticed about the design (that you data does show). When the afteburner kicks in, there is an ambient temperature drop -- just like you would want in a drum roast, just not "programmed". So even P1 that is 100% is not 100% if you catch my drift. It is 100%, then some lesser amount at about 40% of the roast time in as the afterburner (and draw fan) kick in. Another item I noticed was that the after burner does contribute to the roast profile, both as above in dropping the temperature, and again a little later from radiative heat. I inadvertantly tested this once by having the afterburner unplugged during a p1 roast. It didn't hit 1st until 21 minutes.

The p2 profile adjustment I like best so far is to set it to maximum time (when doing a pound) so the first stage is maximized for heat soak. I then pull the end time immediately back to 20 minutes for timekeeping purposes since an adjustment after the roast starts simply takes time off the final stage.

I have also found TC placement to be REALLY tricky and that is going to be the crux for comparing data. Even though this is not where I have been placing mine, I might suggest placing it where the roaster's TC is located. If you have not found it, it is in the back right, in front of the heaters. It is actually mounted on the other side of the wall, inside the control and wiring space on the right. With it's placement, I have noticed there is a lag time because it is not directly reading the roasting environment, but the wall temperature through the wall. Note, that is how the profiles are based, so I don't see a major issue with that. Just a note. This also means that wall needs to remain clean or roast profiles may start to drift (remember that in the alpen roast?).

I remain undecided about the roast times (length that is). I have started one item that I like. If you start the roaster empty (doesn't matter what profile) for 1.25 minutes, you can have a consistent preheat. Not as good as a hot preheat, but something. That time gets the interior to about 200 F which is below the safety set that is programmed in (you can't start the roaster over 130 C). Basically start a roast, count 1.25 minutes, cool/off, load and start. It helps with giving a slightly more robust 1st as opposed to the "sneaker" 1st you (and I) have experienced. Even so, I have yet to get a really robust 1st crack with a pound. The are almost all sneakers with a merge to 2nd. Not a complain really, merely and observation. At 8 oz and 12 oz most 1st cracks are more distinct. I have yet to play with 1/4 lb at all? I wonder if anyone will?

[miKe mcKoffee]

You hit the same wall I have in regards to beans. The permutations are so large, that in practicality there is no good way for one person to test the Behmor.

Sure there is. Just have a few hundred pounds of greens and multiple Behmor's and and unlimited time. Wait, you did say

in practicality

It is really going to take it making it to market and having people post and pool data and experiences.

Absolutely agree. Same as it's been for all other home roasting appliances/methods. (Professionals and commercial roasters too for that matter.)

I too have found that 1 lb is possible, but IMO it is better suited for 8-12 oz roasts, at least on the later profiles.

Basically agree and realize we view this differently. Just not an issue to me that a 1# max batch size rated roaster may not perform at it's optimal level at it's max batch size. Right or wrong I keep thinking analogous to a $9000 7# max batch size roaster performing best with 4 to 5# batches while at 7# max batch little profile control and longer roast times. (Like a Diedrich IR3.)

One thing you may not have noticed about the design (that you data does show).
When the afteburner kicks in, there is an ambient temperature drop -- just like you would want in a drum roast, just not "programmed". So even P1 that is 100% is not 100% if you catch my drift. It is 100%, then some lesser amount at about 40% of the roast time in as the afterburner (and draw fan) kick in. Another item I noticed was that the after burner does contribute to the roast profile, both as above in dropping the temperature, and again a little later from radiative heat.

Yes I'd noticed that but failed to directly address it. When the afterburner and draw fan kick in initially there's a drop in temp while the afterburner heats up to temp and the draw fan is cycling air out. Later during a 100% run the maximum measured environmental temp appears to be greater than prior to afterburner/draw fan. Here's a 12 minute no bean run. I did not pre-heat, but rather pre-warmed simulating back to back batches. (Heated a couple minutes then fully cooled before next "batch run".)
12 minute 100% P1 no bean temps:
12 90f (start of roast environment before run following pre-warm)
11:30 103
11 163
10:30 241
10 306
9:30 352
9 389
8:30 417
8 441
7:30 462
7 481 (afterburner/draw fan just kicking in, coincidentally heater turned off virtually simultaneously)
6:30 472 (heater kicks back on shortly after)
6 457
5:30 479
5 498 (heater kicked off right at 498, hit high of 502)
4:30 488 (heater back on a bit later)
4 465
3:30 485 (heater off 3:15)
3 487 (heat on 2:42)
2:30 457
2 492 (heat off 1:42)
1:30 490
1 460 (heat on right at 1:00)
0:30 455
0 482 (kick to cooling, did not record further temps.)

I inadvertantly tested this once by having the afterburner unplugged during a p1 roast. It didn't hit 1st until 21 minutes.

A different effect and benefit of having the afterburner manually controlled would be turning it off at the start of cooling cycle during in roaster cooling for a bit faster cooling. This of course would defeat the "smoke-less" feature and hence is not part of the stock design for in the house usage. Ideally the afterburner would be external to the roast chamber so it didn't add heat to the roast environment during cooling. This would also mean different overall design spec's including heater/draw fan relationship since afterburner wouldn't be compensating for draw fan during roasting.

The p2 profile adjustment I like best so far is to set it to maximum time (when doing a pound) so the first stage is maximized for heat soak. I then pull the end time immediately back to 20 minutes for timekeeping purposes since an adjustment after the roast starts simply takes time off the final stage.

That (or similar) would likely work well targeting anything from Cinnamon to FC rather than the Light Vienna I achieved P2 yesterday!

I have also found TC placement to be REALLY tricky and that is going to be the crux for comparing data.

Always is any two roasters!

Even though this is not where I have been placing mine, I might suggest placing it where the roaster's TC is located. If you have not found it, it is in the back right, in front of the heaters. It is actually mounted on the other side of the wall, inside the control and wiring space on the right. With it's placement, I have noticed there is a lag time because it is not directly reading the roasting environment, but the wall temperature through the wall. Note, that is how the profiles are based, so I don't see a major issue with that.

TC placement wise simply fished over the top and looped back into the chaff collector seems to work fairly well as "temporary" location. I believe (but haven't confirmed) no bean temps read higher than with batch since the beans block the direct radiant heat path between the TC and the heaters.

This also means that wall needs to remain clean or roast profiles may start to drift (remember that in the alpen roast?).

Yes, that's one of the purposes of the suggested no bean burn every 5 batches. Another is to burn off any oils that may accumulate in the exhaust path.

I have started one item that I like. If you start the roaster empty (doesn't matter what profile) for 1.25 minutes, you can have a consistent preheat. Not as good as a hot preheat, but something. That time gets the interior to about 200 F which is below the safety set that is programmed in (you can't start the roaster over 130 C). Basically start a roast, count 1.25 minutes, cool/off, load and start. It helps with giving a slightly more robust 1st as opposed to the "sneaker" 1st you (and I) have experienced.

Though you mentioned it via email (yesterday?) haven't implemented it yet. Makes sense especially for 1# batches. FWIW just tested pre-heat 1:30, works too. From idle. Then tested 1:30 simulating back to back batches and it was too hot so wouldn't start until cooling further. (Had drum out chaff collector in for pre-heat test portion so could simulate having drum loaded and ready.) Suspected and confirmed really need to go by pre-heat temp not fixed pre-heat time for repeatability. (duh)

I have yet to play with 1/4 lb at all? I wonder if anyone will?

If this was a formal official review probably should. Though no I haven't and it's not high on my test priorities.

(So much for not posting for a spell, or it was just a very weak spell. )

[Alchemist]

Following a note from Joe, I wanted to point something out. I currently have two roasters. One is a pre-production model that I took a roasted into a rolling 2nd and caused a thermal shutdown of the system. It currently doesn't roast right due to that. The roasts now take too long and unless otherwise stated, none of my observations are based on that roaster after the incident. The other roaster that was bounced around by FedEx that Joe sent me. He did care for the way it roasted. Noting that, I find that it matches the first one pretty well, based on my roasting notes. Keep that in mind that the following data MAY not represent the exact conditions Joe expected.

That all said, I had the best roast off the Behmor 1600 so far this morning. I used my own espresso blend of 1/3 Ghimbi, 1/3 harrar, 1/3 Sidamo, all DP. I roasted 9 oz (because 3 x 3 oz was easier) on an 18 min P2 profile. I preheated 1.25 minutes to 150 F initial temperature.

Here's the roaster temps with the TC at the inside point of the roaster's TC (i.e. on the back right wall).

18 150 F
16 278
15 315
14 345
13 365
12:30 368 (heard the heating elements click off)
12 340 (beans are yellowing)
11:30 (heater on)
11 365
10:20 350 (draw fan and afterburner came on)
10 340
9 375
8 390 (pop?)
7:30 400 (yes - 1st - most robust I have heard so far) (at 7.2 minute the lower portion of the profile started - PERFECT timing)
6 340 (1 st is over)
5 345
4:20 350 (2nd starting)
4:10 360 (EOR)
---- cooling
13 350 (opened door)
12 240
11:30 200
11 175
10 140
9 110
8 98 stopped cooling

I really liked this roast. Oddly, the finished roast smelled of buttered popcorn

I want to slowly inch up the weight and starting time so that 1st and the drop in temperature continue to coincide.

[miKe mcKoffee]

That all said, I had the best roast off the Behmor 1600 so far...

(Playing devils advocate) Best roast? As compared to ? You've cupped this roast? I infer you mean by best roast attaining your desired temperature profile for a roast to early 2nd in 13:50 time for end of roast. Without comparing same blend in the cup roasted a different profile, or same/similar profile with larger batch longer time... We're all possibly guilty of preconceived notions!

Here's the roaster temps with the TC at the inside point of the roaster's TC (i.e. on the back right wall).
18 150 F
16 278
15 315
14 345
13 365
12:30 368 (heard the heating elements click off)
12 340 (beans are yellowing)
11:30 (heater on)
11 365
10:20 350 (draw fan and afterburner came on)
10 340
9 375
8 390 (pop?)
7:30 400 (yes - 1st - most robust I have heard so far) (at 7.2 minute the lower portion of the profile started - PERFECT timing)
6 340 (1 st is over)
5 345
4:20 350 (2nd starting)
4:10 360 (EOR)
---- cooling
13 350 (opened door)
12 240
11:30 200
11 175
10 140
9 110
8 98 stopped cooling

Defintely appears your TC placement reading substantially lower than mine, and IIRC much lower than design spec' target temps. (100% power heater of max ~482f. Your highest reading during roast 400f while mine ~60f higher, my first heater ~100f higher) Roast from 6 to 4:20 minutes reading ~350f yet 2nd kicking in, maybe enough momentum or maybe low temp reading from closer to actual "roast environment" (or whack tc/meter). Not that it matters except in relation to comparing. How 'bout running a no bean P1 12min and record temps for direct comparison to what I ran earlier today. If you do it remember I started it from pre-warmed simulating back to back roast and cool but not pre-heated.

I really liked this roast. Oddly, the finished roast smelled of buttered popcorn

I could understand it smelling like buttered toast since some feel the Behmor looks like a toaster oven but it doesn't look like a popcorn popper! Ok, microwaved popcorn, looks more like a small nuker than a toaster oven to me.

[Alchemist]

(Playing devils advocate) Best roast? As compared to ? You've cupped this roast? I infer you mean by best roast attaining your desired temperature profile for a roast to early 2nd in 13:50 time for end of roast. Without comparing same blend in the cup roasted a different profile, or same/similar profile with larger batch longer time... We're all possibly guilty of preconceived notions!

Yes, I should have been way more explicit. To expand a bit, that blend mentioned is the same one I brought to the espresso jam and what I roast weekly. I know it very well, fresh out of the roaster to 10 days old. In one form or another (different crops of the same origins and prep {DP always}) I have been using this blend for about 3 years. It behaved as I expected from my drum roaster, albeit a little less even than I am used to. Strong 1st crack, good gap in cracks, and a coast into 2nd. Except for the buttered popcorn smell, it also had the fruity brightness and earthiness I am used to fresh from the roaster. I have not cupped it, just tasted it as an espresso pull. I don't really care for this blend this fresh (1 hour) but I know it's flavor and that flavor was there. That is what I meant

Is that better? Thanks for keeping me from getting to lazy.

[miKe mcKoffee]

Yes, I should have been way more explicit. To expand a bit, that blend mentioned is the same one I brought to the espresso jam and what I roast weekly. I know it very well, fresh out of the roaster to 10 days old. In one form or another (different crops of the same origins and prep {DP always}) I have been using this blend for about 3 years. It behaved as I expected from my drum roaster, albeit a little less even than I am used to. Strong 1st crack, good gap in cracks, and a coast into 2nd. Except for the buttered popcorn smell, it also had the fruity brightness and earthiness I am used to fresh from the roaster. I have not cupped it, just tasted it as an espresso pull. I don't really care for this blend this fresh (1 hour) but I know it's flavor and that flavor was there. That is what I meant

Is that better? Thanks for keeping me from getting to lazy.

Ah, so you did "cup" it already, as a shot pulled in a demi' cup! I understand having a good idea how a known to you blend pulled at 1 hour post roast will turn out when properly rested. (FWIW I recognized the blend you posted as your long time standby but doubted the majority of readers would know so pushed for more detail. )

Also FWIW I've not meant to imply your opinion or suspicion that Behmor ~20min pounder roasts times will flatten roasts is right or wrong, just not willing to let suspicions or assumptions preclude taste confirmations.

Preliminarily I found that even 1/2# roast can be flattenrd with wrong Behmor profile. (No real surprise, same roaster same time to same finish degree but different profile ramps will yield different cups.) This evening Debi & I French Press pseudo cupped 4 day mason jar vac' rested WP Yirg' roasts comparing my 12:30 Rosto profiled roast with Behmor P5 15 plus 0:30 added late in roast for 15:30. Roasts appeared comparable whole bean and ground, 'bout LFC just before 2nd. We both found Rosto cup livelier, definitely better top end. The Rosto batch also had better aroma and fragrance. Going to re-cup tomorrow and/or Sunday to confirm. Have a sneaking suspicion myself that your focus on tweaking P2 may at least often yield the best results (for my tastes) too. Back during PNWG V it was the only profile that really caught my eye when I first saw the 5 profile graphs. But I let the stated 70% heater equating to 374f lead me astray, until finding out via actual TC in chamber is not that low! (My TC placement anyway.) With the Behmor tstat outside the chamber it may well equate to 374f. Regardless way too soon to make any conclusions regarding any of the profiles any batch size. This one cupping in no way equates to the Behmor not being capable of an excellent roast of this WP Yirg'. Heck I've been Rosto profile roasting with variable control of both heat and fan for 5 years so knew how to roast the Yirg' with the Rosto to make it sing. Only had the Behmor for 4 days and the Yirg' roast was my second ever Behmor roast! Not exactly a fair comparison. Oh, and didn't mean to imply the Behmor Yirg' was bad, just not as good.

And what the heck, it's 1am! I think I'll quit rambling and hit the hay.

[miKe mcKoffee]

I've hit a wall and am going to back up and punt. This will likely be my last Behmor 1600 post for a spell. Basically going to start over attempting to be methodical and dispassionate. Last night realized I really need a bunch, a whole bunch, of disposal greens to really get to know the 5 profiles and their potential variations before doing any real roasts for cupping. Tom graciously agreed to ship me 40# of some Columbia for cost of shipping that he

liked the coffee okay, but felt it had a hint of travel damage (ie it steamed in port or on the ship)

and the broker didn't want back.

See, I'm not posting and waiting for the 40# test beans to arrive before doing anymore roasting. Except for posts since then and the roast I just did.

Decided to do another 1/2# roast of the same WP Yirg' applying what I'd learned so far and Alchemist's experiences. So reviewed the profile I'd used with the Rosto since it's cup clearly was better (based on the one cupping and haven't re-cupped to verify results ). The Rosto roasts in brief was fast to 200f 1min, 300f @ 4min, straight ramp 300 to 400f @ 8min then hitting the heater brakes, 400f to ~5f before 2nd @12:30 @ my 450f. Chose P2 13min which would give S1 7.8min (actual) 100% heater (initally screwed up and thought it would be 9.6min using 70% {S2 heater ?} of 13 instead of 60% S1 time), S2 3.9min 70% heater 30% time, S1 1.3min 100% planning to immediately open door cool end of roast. Decided to not pre-heat but rather continue with pre-warm. Before loading heated @100% heater for 2min then cooling. Pre-warm sequence temp peaked 360f, cooled to 100f and proceeded to roast.
P2(suppose to be) 13 min 1/2# WP Yirg'
13 - 93
12:30 - 103
12 - 144
11:30 - 194
11 - 242
10:30 - 280
10 - 308
9:30 - 331
9 - 351
8:30 - 368
8 - 380
7:30 - 405
7 - 418
6:30 - 435
6 - 448
5:30 - 458
5 - 467
4:30 - 474 heater kicks off
4 - 467 heater on
3:30 - 457 (was thinking S2 should start soon and heater should go 70%)
3 - 468 1st starts, not just one or two but starts and almost immediately fairly steady
2:30 - 475 heater off back 2:40 (why has it not gone to 70% heater, look at panel it's on P1 OOPS!)
2 - 455 heater on (re-thinking end of roast)
1:30 - 453
1 - 464 heater off
0:30 - 458 end of 1st
0:15 ended roast, since goofed on using P2 for gentler 1st via 70% heat decided to use closed door cooling first minute

13
12 - 324f opened door
11 - 251
10 - 191
9 - 157
8 - 132
7 - 116 ....

Let's see. In haste screwed up the roast planning P2 calculations, 13 min total time would have yielded 7.8 min S1 rather than desired 9.6 min. yielded from mistake calculation. IF I'd selected P2 13min S2 70% would likely have been too soon. As it turned out running P1 13min time gave solid 1st kicking in @ 3 (30 seconds after I'd planned to reduce heater to 70% was spot on) 1st ending about 0:30 (2:30 long ) then letting it go to 0:15 before ending roast with 1min closed door roast/cool coast may have worked very well for acidity based on very nice and lively post roast smell and bean munch! Quite probably livelier than the Rosto roast which I'd intentially used a gentlier through first to bring out more body versus acidity but not so slow flatting the roast.

Next time for this Yirg' 1/2# batch I'll try P2 with 15:45 or 16 min initial time which WILL yield 9.45 or 9.6 min 100% heater first stage. Maybe use 15:45 with pre-heat and 16 with pre-warm and compare. Tempted to go for it now but do only have 5 more pounds of this stuff, and it definitely ain't a garbage bean. Though do also have 8# more total two other WP Yirg's, both of which I think slightly superior to this one. )

Ok, I'll go do something else like pull some shots instead of playing with the Behmor some more.

[dgrove]

Hi Mike,

Got a testing request for the Behmor for to add in the queue whereever appropriate - if you find it worthy. A problem that many have with some of the roasters is a dearth of heating power due to low voltage. What I'm curious to know is how this roaster can perform on lower-than-normal voltage. You appear to have all the necessary pieces for performing this - a way to measure the voltage and a variac to alter it. Perhaps you could (at some point in time) run few batches at "low" voltage to see how performance is affected.

This may be a good thing to try down the line a bit, after you've some of that 40 lbs of Colombian under your belt.

Thanks!
Doug