Behmor 1600 Coffee Roaster - 1st Look - Page 2

[Alchemist]

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.

I actually have that capability also. How low of a voltage were you thinking? I know from discussions with Joe that it was spec'ed out at "less than optimal" so it would not have that classic Achilles' heel, but I don't recall how low the voltage testing went. Joe?

I wouldn't think much lower than 100 V.

[tahoejoe]

Doug,

Obviously voltage supply into someone's home is something totally outside my control but something we foresaw and adjusted as best we could.

Given I've run perhaps a thousand tests in my home I can probably answer this better than anyone. Every test roast I perform in every location I check the voltage with a kill a watt meter.

My home is almost 40 years old in San Diego. The house has the typical wiring of that period, copper clad aluminum. The voltage in the mornings has a start up of 117-118 but can descend to 112 or lower in the afternoons depending upon the time of year. Summer usage is the most affected. Example the other day it dropped to 108..and it pushed the roast out to it's maximum allowed time.

The system is designed that for every profile and associated weight there is a maximum allowed roast time. This was not something we chose to do but rather were forced by the OSHA appv'd test house to do.

Now I've laid out the parameters on to answering your question...

As with any home applaince that heats up foods etc the roaster can be affected to some degree by a low voltage feed.

The unit was designed around the idea 116 was the avg in a home. No matter what we chose it would be wrong but 116 was an average voltage I found from a survey I conducted of about 200 people. If we had set the baseline at 110 or 108 we would likely run into a sizeable number of burned out systems. One well known roaster did the 108 and has suffered very large field failures and set the example as to why one must be careful in it's chosen baseline. Point is the 116 was weighed carefully as were all my decisions on the roaster.

What I've found is the area that could be most affected is the 1/2 pound roast.. I say could because in my hundreds (about 400+) of 1/2 pound tests I've only failed to reach desirable roast times because of voltage in maybe 3 cases. All of them hot summer days ..afternoon/late afternoon time.

I can't ever recall running out of time on full pound maybe because the heat of the greater mass of beans. Times were pushed out but never in my home using any variac type device did I ever run out of time. This isn't to say it couldn't happen.

In the case on a pound and a home that consistently has low V..they'll possibly need to reduce roast weight by a few ounces. No way around it or use a variac etc..

One caveat is to try various outlets or roast early or late in a day. I was doing a demo at Ronco(industrial setting) and one outlet sat at 108 to start yet adjacent on another wall maybe 8 ft away the start up was 118v and as pointed out above, time of day in my home affects the available voltage.

Now back to half pounders.. The beauty of the system is the profiles are set to recognize the time entered when started and when started it sets itself..

Let's say we do a 1/2 pound and it max's out at 15 min using P4 (stepped profile), yet you were unable to get to 2nd crack. In this case instead of reducing weight go to the higher weight program.

In other words , put 1/2 pound in the cylinder then hit 1, P4 and the timer should be reading 18:00... here is the kicker or rather beauty of our design..before starting the system now subtract (- button) time down to say 15:30... Then press start... The system will react to the P4, 15:30 and function as such.

If one follows the instructions of reducing times, never leaving the roaster unattended etc. no issue other than reheating elements are not set to activate on percentages of time as are the profiles, but rather at set times. In other words on a 1/2 pound setting they come on around 5 minutes into the roast whereas on a full pound they don't come on until around 7. The remote possibility exists a very very small amount of smoke could leak out of the exhaust channel because it may get into to the channel using basic laws of physics..smoke rises... If this were to happen it would be minimal at best.

Hopefully this thoroughly answers your question.

Joe

[dgrove]

Hi Joe,

Thanks very much for the detailed reply. Can't say I completely followed all you said, as I'm not fully familiar with the
the controls of your roaster. However, I think it's great that you've done such a thorough job on designing/testing this roaster and believe that it can do a reasonable job even under lower voltage conditions. I actually don't know the typical voltage is in my house, but I know that it's not very high. I've got a 102 yr old house that still has a few circuits that are partially knob-and-tube... I've been roasting with a caffe rosto for about 2 years now, and even after purchasing a variac and partially covering the exhaust with foil I *still* can't get a full city + without a decent ambient temp. I've grown tired of that and also very much would like a larger roasting capacity. I've been following the coffeegeek thread and also various mentions on the sweetmarias roasting list, and am very excited about this roaster. But since I've had such a tough time with my current roaster I just had to find out what the "worst case" performance was for your roaster.

I have no problem whatsoever with how you've designed the roaster, I simply thought it would be nice to find out how it fared under less than optimal conditions. After all, even though I now have a variac it would be nice if it didn't *have* to be used.

I'm very much awaiting the release of the roaster. I've completely given up on my rosto and haven't roasted any coffee in several months. I don't yet know how I'll justify the purchase to my wife, but will cross that bridge later...

Doug

[tahoejoe]

doug,

If you have the time..go to the behmor website... http://www.behmor.com

Contact page and send me an email.. w/ phone I'll gladly call you and explain anything and everything I possibly can...

Thanks
Joe

[prof_stack]

The Behmor website lists the roast size as 1 pound to 1/4 pound. I was under the impression that smallest roast size was 1/2 pound. When did the 1600 get spec'd for the 1/4 pound and how will it affect the roasting times?

Thanks.

[tahoejoe]

Actually it has always been spec'd for 1/4, it could be either I've not really pushed that aspect, which is entirely possible or it hasn't been the focus of questions so I've not reminded people of the potential for doing a 1/4...

The roast times were based upon my home... ergo actual volatge floating apps... 1/4 ~8 min.. 1/2 ~12... 1lb~ 18. This is on high heat P1(fastest profile)...

[miKe mcKoffee]

Sorry for the no recent updates, been busy plus waiting for the 40# test beans to arrive.

Some information in the pre-release manual is available on the Behmor website, like the 5 profile curves, but much is not yet up there so in hindsight am cut & pasting some sections of the manual.
Note: Manual still undergoing revisions before final release.

Control Panel:
1) Roast Weights: Weight of green coffee beans to be roasted. When pressed, all of them will turn the system on and specify a program for other components, such as program times. The weight buttons also act as a default setting, allowing the user to "reset" the program at anytime before pressing start.
2) Programs/Times: System roast times directly associated with a weight pressed. Example: Pressing ¼, then pressing A will result in the display showing 8:30. Pressing 1, then pressing C will result in the display showing 21:30. See Tips and Maintenance for more complete information related to these examples.
3) Time Increments: Add or subtract time to/from the roast programs prior to start or during roasting. This can be done at any time during the roast program, up to and including one second remaining on the timer. Pressing "+" adds 15 seconds, "-" subtracts 10 seconds.
4) Profiles: These buttons are used to adjust how heat is applied to the beans. By adjusting how the heat is applied, you can alter some coffee characteristics of any given type of bean, such as acidity. P1 is the highest heat application, thus fastest roast time. P5 is the lowest overall heat application, thus slowest roast time. See Diagram of Profiles for details on heat applications.
5) Start: Starts the roasting cycle.
6) Cool: Allows the user to stop a roast at any time and begin cooling the beans, rather than waiting for time to expire for default cooling.
7) Light: Light switch for viewing beans in either the roasting or cooling cycles.
8) Stop: When pressed stops all system functions. DO NOT press Stop during the roast cycle except in an emergency situation (such as fire). Pressing Stop before the roast cycle is complete could lead to excessive smoke and possibly create a fire hazard.

Tips and Maintenance

1) After five roasts cycles, you must always perform a system self clean. To do so, insert the cylinder (empty), chaff tray, close the front door, press ½ > Start and let the system complete an entire roast and cool cycle. Failure to perform this maintenance procedure will in time affect your roaster's functionality.

2) In general terms, the following is a good guideline for time from start of first heard cracks/snaps to roasts going into the second crack/ being finished.
a) ¼ lb - 1:40-2 minutes
b) ½ lb - 2:20-2:40 minutes
c) 1 lb- 3:30-4 minutes

3) Times may vary due to voltage, but a good understanding of the A, B, C, D (Program /Times) to P1, P2, etc., is as follows:
A to P1
B to P2
C to P3
D to P4 (subtract 30 seconds using)
D to P5 (add 30 seconds using)

4) Associated initial start times of A, B, C, D to weights ¼ , ½ , 1

¼ ½ 1
A 8:30 12:00 18:00
B 9:30 13:00 20:00
C 10:00 14:00 21:30
D 10:30 15:00 23:00


5) In order to prevent possible issues with over roasting, we have built in maximum times associated with each profile. If you should happen to be roasting at P1, the system is designed to prevent you from allotting more than 20.5 minutes roast time. This is a safety feature. The same feature is true for the other profiles but with longer times. See Chart below:

1/4 1/2 1

P1 10 13.5 20.5
P2 10.5 14.5 22.5
P3 11 15.5 23.5
P4 11.5 16.5 24.5
P5 12.5 17.5 25.5

6) For quicker cool down an option does exist, should the user decide take the following measure. After the beans have ceased snapping (approximately 1:30), while keeping the system on cool, simply open the front door. Doing this will facilitate a greater airflow over the beans. The one drawback is small amount of chaff will find it's way past the chaff tray.


DO NOT STOP THE SYSTEM IMMEDIATELY AFTER COMPLETING A ROAST TO PULL THE CYLINDER AND DO A BEAN DUMP FOR ALTERNATE COOLING METHOD. DOING A SYSTEM STOP BEFORE REASONABLE COOLING COULD DAMAGE THE SYSTEM'S COMPONENTS.

7) Adjusting profiles by time additions and subtractions at proper times can be used to slightly alter the pre-programmed profiles. You can shorten or lengthen the middle leg of a profile or end leg of a profile. This can be done by adding or subtracting time prior to starts or just after starting the roast cycle

Example: P5 is 33% of each power application throughout the entire roast. So if you were to set the timer at 6:00 minutes each leg of the profile would last approximately 2:00 each.

However if you were to set the timer for 6:00 minutes to start, then upon starting a roast add 2:00 minutes you will have immediately reduced the middle leg's total percent of total roast to 25% versus the original 33%.

Conversely let us say you want to lengthen the middle leg and shorten the end leg of the roast. Before starting press the timer to read 10:00. The system then reads it is to run approximately 3:20 per leg. Once starting however reduce the time to 8 and you will have the first legs of the roast at 3:20 each but the final leg at 1:20.

If you wish to shorten the roasts middle leg and lengthen the end leg. Reduce times to start, then once started add time back.

If you wish to lengthen the roasts middle leg and reduce the end leg. Add time to start, then once started reduce the time back.

The possible variations number in the hundreds once you understand the roaster and the beans you are roasting.

This should give you a good how idea of the profiles, their intended roast times for different weight batches, and how to manipulate them.

Not ready to give any cup quality opinions other than to say at least for 1/2 to 3/4# batches believe it has great potential. And will even say the shots from the full pound batch 20/40/40 Aged Sumatra/Brazil YB/Misty Valley Yirg' where I'd used a too slow/low early heat profile resulting in 24:30 total roast time plus used first 3 minutes closed door cooling (mistaking 1st for 2nd!) turned out much better than expected. As expected with noticeably flattened acidity though not as severe as I'd thought it be. Actually somewhat pleasantly surprised.

[miKe mcKoffee]

Actually it has always been spec'd for 1/4, it could be either I've not really pushed that aspect, which is entirely possible or it hasn't been the focus of questions so I've not reminded people of the potential for doing a 1/4...

The roast times were based upon my home... ergo actual volatge floating apps... 1/4 ~8 min.. 1/2 ~12... 1lb~ 18. This is on high heat P1(fastest profile)...

I'd add that this doesn't mean it is limited to roasting exactly 1/4, 1/2 or 1# batches. As indicated in the manual what the 1/4, 1/2 and 1# buttons do is set a time associated with a profile. Prior to starting a roast you can subtract time down to 10 seconds, or increase to the maximum for the specific profile. This means you can adjust the time for a profile you want to use for the size batch size. Smaller batch subtract time, larger batch add time.

On a different note just finished a slight mod' I'd planned on doing but was going to wait until further testing completed. Since test beans ETA today and UPS usually doesn't arrive until late afternoon decided to do it now since I had some time. Specifically split wiring the main heaters for independent voltage input. Wired through a toggle switch so can use "stock" configuration or separate voltage input to heaters. Funny thing was right after returning from the appliance repair store for some high temp insulated 14 gauge wire this morning UPS arrived! Did the heater mod' anyway.

[miKe mcKoffee]

Columbia Test One How fast can 1# roast at room temp with pre-heat sequence (stock configuration, no heater boost).

Note: Moved TC to left drum mount bracket so easily duplicated and don't have to remove it to remove chaff collector or drum yet is temporary. Ran tests and determined pre-heat to 160f, cool about 10 sec. until elements no longer glow and then can start a batch. (Temps reading much lower than front top of chaff collector.) Pre-heat by time will have to be worked out for idle versus back to back batches etc.

1# Columbia subject bean, P1 (max heat 100% time), 18 minutes (default), kitchen ambient 69f, no load line voltage 119.2v. Loaded greens in drum, with chaff collector in pre-heated to 160f and cooled 10 seconds, inserted drum in roaster.
18 - 148f
17:30 - 150
17 - 161
16:30 - 174
16 - 187
15:30 - 200
15 - 213
14:30 - 226
14 - 238
13:30 - 250
13 - 262
12:30 - 273
12 - 283
11:30 - 293
11 - 300 - heater first off
10:30 - 294 heater is back on, drawfan & afterburner turned on
10 - 290
9:30 - 296
9 - 304
8:30 - 309
8 - 314
7:30 - 318
7 - 322
6:30 - 326
6 - 331
5:30 - 325
5 - 334 heater off back between 5:30 & 5, early 1st's
4:30 - 328 heater on 4:41, 1st more pronounced
4 - 334 1st picking up
3:30 - 339 solid 1st going
3 - 339 heat off 3:20
2:30 - 330 heat on
2 - 335 end of 1st
1:30 - 340 heat off
1 - 340 heat on, early 2nd 1:15
0:30 335 2nd just barely starting, ended roast
Initial Closed Door Cooling (in kitchen, so kept closed to lesson smoke)
13 minute cooling cycle
12:30 - 334 2nd going
12 - 326 2nd about dead
11:30 - 315 Opened Door
10:30 - 263
10 - 235
9:30 - 213
9 - 197
8:30 - 186 ....

There was no visible smoke during roast or cooling until opening door. As I opened the door I also now turned on the 700cfm hood. Some but not much brief visible smoke opening the door. Very nice and even looking medium dark brown no oil 1# FC batch in 17:30 (or 19 if counting first minute of cooling with door closed) roast. Post roast much was better then expected for one of Tom's reject beans. With just a touch of sweaty gym socks in the after-taste!

(EDIT) Note: no visible roast does not equal no roast smell! I (and the manual) advice using with good ventilation or yes your house will smell like roasted coffee regardless if the smoke detector goes off or not. However, I didn't want the noise of the hood blowers to interfere with roast observations.

Next up, what'll it do with 10% boost to heater until start of 1st then backing off heat a bit (or not), will it work well or cause uneven roast? Or maybe just P2 stock configuration. Yeah, probably should document all 5 profiles for 1# room temp stock before going for any manual heater control. But not this evening...

[miKe mcKoffee]

Saturday's 1/2# batch WP Yirg' P1 12:45 roast time 1 minute close door cooling cup, my usual vacuum sealed mason jar rested. Made a cappuccino with it yesterday (4 days rest), first cup of the morning Americano today. (Americano my usual first cup of the day) In a nutshell excellent. I believe on par with any Rosto profiled roast done with this bean though did not do a Rosto batch this time for direct comparison. The question remains can this good of cup of this bean be achieved with 1# batch. More will be revealed!

[miKe mcKoffee]

Columbia Tests Two & Three

How slow will 1# roast with low 110v no load, P1 18min, 68 ambient, pre-heat 160/cool 10 sec.

18 - 133 (voltage load 106v)
17:30 - 135
17 - 140
16:30 - 158
16 - 158
15:30 - 169
15 - 179
14:30 - 190
14 - 200
13:30 - 210
13 - 221
12:30 - 230
12 - 240
11:30 - 248
11 - 257
10:30 - 264 fan/afterburner kick on, now 104v
10 - 261
9:30 - 262
9 - 269
8:30 - 273
8 - 277
7:30 - 279
7 - 282
6:30 - 284
6 - 287
5:30 - 289
5 - 291
4:30 - 294
4 - 296
3:30 - 298
3 - 301
2:30 - 303
2 - 306
1:30 - 308
1 - 310
0:30 - 313 add 2:30 (20:30 max for P1)
0/2:30 - 316 early 1st right after 0:30, becoming slow steady
2 - 318
1:30 - 320
1 - 322 still slow 1st
0:30 - 323
0 - 325 mandatory end of roast cool - Heater never turned off
13 - 325 closed door cooling
12:30 - 320
12- 310
11:30 - 300 opened door
11 - 281
10:30 - 259
10 - 236
9:30 - 214
9 - 194
8:30 - 182 ...

City maybe City+ best it could do with substantially low voltage, maximum 1# batch size, hottest profile, longest time possible.

How well will reducing the batch size to compensate for low voltage actually work?

3/4# P1 18 min ambient now 70 same 110v no load same 160f pre-heat cool 10 sec.

18 - 126 (load 106v)
17:30 - 130
17 - 138
16:30 - 148
16 - 160
15:30 - 172
15 - 183
14:30 - 195
14 - 208
13:30 - 219
13 - 231
12:30 - 241
12 - 252
11:30 - 261
11 - 271
10:30 - 284 drawfan/afterburner now 104v
10 - 276
9:30 - 277
9 - 286
8:30 - 291
8 - 295
7:30 - 299
7 - 302
6:30 - 305
6 - 309
5:30 - 311
5 - 313
4:30 - 316
4 - 319
3:30 - 321 early 1st 3:15
3 - 325 1st picking up slowly
2:30 - 326 1st slowly constant
2 - 327 1st continues
1;30 - 329 1st slowing
1 - 331 1st ended
0:30 - early 2nd, add 2:30 (max for P1)
0/2:30 - 2nd coming on solid
2 - 376 2nd becoming rapid, ended roast, 18:30 total time, closed door cooling. Again heater never turned off.

13 - same as 2 end of roast
12:30 332 2nd continues, still fairly robust
12 - missed temp, 2nd continues slowly, opened door
11:30 300 2nd has ended
11 - 270
10:30 - 248
9:30 - 201
9 - 188 ...

Nice 18:30 Fc+ roast, med dark brown no oil and a bit surprisingly a couple divots. (Just a bit darker than yesterday's 1# good voltage P1 batch.) With 2 minutes possible roast time remaining obviously could have gone much darker. Clearly reducing batch size from 1# to 3/4# made a big difference with 110v no load voltage. 3/4# batch size enabling dark roasting while 1# only City/City+ at best. I suspect could reduce to 14oz instead of 12oz batch and still get FC+, in longer time than 12oz batch of course, but am not going to run that roast. I believe this test illustrates the voltage/batch size/roast time relationship adequately. Those suffering from bad voltage can learn to fine tune their batch sizes accordingly, or buy a variac.

So much for running all 5 profiles 1# batch good voltage first! (But hey, low voltage was specifically asked about.) Next up 1# P1 boosted heater voltage.

[miKe mcKoffee]

Columbia Test Four

How fast will 1# roast with 132v to main heaters, 120v P1 18min rest of roaster, 70 ambient, pre-heat 160/cool 10 sec.

18 - 128
17:30 - 136
17 - 154
16:30 - 171
16 - 188
15:30 - 206
15 - 223
14:30 - 240
14 - 255
13:30 - 277
13 - 284
12:30 - 287
12 - 310 recalling not much higher temp was stock max good voltage for heater cut out, reduced heater to 120v
11:30 - 318
11 - 326 seeing a bit of smoke escaping, reduced heater further to 110v
10:30 - missed temp, fan/afterburner kick on now.
10 - 328 any visible smoke now gone
9:30 - 328
9 - 331
8:30 - 334
8 - 335
7:30 - 337
7 - 338
6:30 - 324 early 1st
6 - 324 1st becoming rapid, reduced heater futher to 105v
5:30 - 343 reduced heater to 100v
5 - 342
4:30 - 341 1st continuing, slowing
4 - 340
3:30 - 340 end 1st?
3 - 340 1st definitely ended, reduced heater further to 95v
2:30 - 339 early 2nd, ended roast 15:30 roast time, immediate open door cooling

13:00- 339 begin open door cooling, few 2nds
12:30 - 330 2nd ends
12- 288
11:30 - 267
11 - 254
10:30 - 239
10 - 226
9:30 - 215
9 - 199...
7 - 157 ...

Beautiful, even Light Full City 1# roast in 15:30 with well defined ~3 minute 1st crack kicking in ~11:30 minutes into roast, ~1 minute pause end of 1st to early 2nd. Next time will try reducing heater further to 95v early 1st instead 105v then 100v then 95v end of 1st. Obviously changed the plan mid roast, so much for going flat out for how fast it roast a full lb! What can I say, my years of Rosto profile roasting kicked in and decided to go for a potential "good" roast profile instead of just speed roast. I believe this demonstrates some pretty darned good 1# profile roast control potential be it manually or PID etc. Yeah, modders just may have some fun with the Behmor! With faster early ramp would like to see afterburner/drawfan kick in sooner, like around 5 minutes mark for 1/2# settings, but the longest 1/2# profile time is 15 minutes so didn't use one of them. But hey, just other potential mod's to gain full control of afterburner heater and drawfan.

I suppose I should return my focus to testing stock configuration/profiles, but it's gonna be tough.

[miKe mcKoffee]

CT5 (Columbia Test)
1#, P2 20min(B) (As you may or maynot recall P2 S1{stage1} 100% heat first 60% time = 12min, S2 70% heat 30% time = 6min, S3 100% heat 10% time = 2min), ambient 67f, stock confuration @ 120v, 160f pre-heat

20 - 132f
19:30 - 130
19 - 140
18:30 - 152
18 - 165
17:30 - 179
17 - 192
16:30 - 206
16 - 219
15:30 - 232
15 - 244
14:30 - 256
14 - 267
13:30 - 277
13 - 287
12:30 - 296 draw fan/afterburner on now as expected
12 - 292
11:30 - 294
11 - 301
10:30 - 306
10 - 310
9:30 - 314
9 - 318
8:30 - 322
8 - 325 now goes to S2
7:30 - 316
7 - 307
6:30 - 305
6 - 304
5:30 - 304
5 - 305
4:30 - 305 early 1st
4 - 304 1st very slow
3:30 - 305 anemic 1st continues
3 - 305 "
2:30 - 305 "
2- 305 now goes to S3 as slow 1st continues
1:30 - 311 ditto 1st
1 - 317 end of 1st ?
0:30 - 323
0 - 331 early 2nd let go to cooling closed door

13:00 - 331
12:30 - 326 slow 2nd
12- 317 few 2nds
11:30 - 308 2nd about dead opened door
11 - 285 no more 2nd heard
10:30 - 260
10 - 235
9:30 - 210
9 - 197
8:30 - 196 ?
8 - 183
7:30 - 170 ...

Nice even appearing 20 minute LFC with early 2nd right at start of cooling with first 1:30 cooling closed door.
Note: could have added up to 2:30 minutes additional time this profile.

CT6
1#, Modified P2 20min (manual heater controlled 7min 110% 132v, 3min 100% 120v, 4min 70% 84v, 2min 80% 96v, 2min 100% 120v), ambient 68f, usual 160f pre-heat

20 - 141
19:30 - 139
19 - 147
18:30 - 159
18 - 173
17:30 - 188
17 - 204
16:30 - 220
16 - 236
15:30 - 251
15 - 265
14:30 - 279
14 - 291
13:30 - 303
13 - 314 now to 120v
12:30 - 322 drawfan/afterburner on now as expected
12 - 323
11:30 - 327
11 - 331
10:30 - 335
10 - 338 now to 84v
9:30 - 336
9 - 333 early 1st
8:30 - 330 slow 1st
8 - 324 slow somewhat steady 1st
7:30 - 324 "
7 - 320 1st picking up a bit
6:30 - 315 remains moderately slow steady 1st
6 - 315 now to 96v, ditto 1st
5:30 - 314 1st slowing
5 - 313 1st very slow
4:30 - 313 end 1st
4 - 314 now to 120v
3:30 - 317
3 - 325 early 2nd @ 3:10, ended roast @ 3 immediate open door cooling

13:00- 325
12:30 - 314 no 2nds heard since started cooling
12- 283
11:30 - 272
11 - 256
10:30 - 245
10 - 225
9:30 - 213
9 - 199
8:30 - 186 ...

Again nice even appearing 17 minute LFC, 3 minutes overall faster 1# roast same/similar profile. CT6 appears just a hair ligher than CT5, confirmed comparing grinds. Also grind color very comparable to whole bean both. Slight finish roast degree makes sense with CT6 early 2nd's 10 seconds before initial open door cooling versus CT5 early 2nd right at start of initially first 1:30 minute closed door cooling enabling 2nd to slowly coast longer.

Overall pleased with CT6 modified P2 but making a couple tweaks. Next time I'll lengthen 132v 1min to 8min, shorten 120v 1min to 2min, invert 84v/96v times for 84v 2min (preparing environment for 1st) followed by 96v 4min (carrying through 1st). I think this will bring 1st on initially a bit stronger and carry through more even rate better. FWIW this is in preparation of roasting 1# batches for comparison stock P2 versus modified P2 of espresso blend 20/40/40 Aged Sumatra/Brazil YB/Idido Misty Valley Yirg'. In addition to my comparing them, I'll be sending samples to two H-B moderators. (One knows it'll be coming, the other if he reads this knows who he is. If he doesn't read this it'll be a surprise!:-))

(Edit Add) Thinking about the roasts to ship for comparison, maybe should add 1# P2 full closed door cooling in addition to open door cooling, and possibly a 4th 1/2# roast batch, but then closed or open door cooling or 5 batches doing both. Hmmm, this isn't just Tom's reject test beans I'll be roasting! Maybe keep to two roasts of "good" stuff and send various test bean roasts for comparions...

[miKe mcKoffee]

FWIW I highly suspect for 1# batches at least (unmodified) P1 & P2 will be of most use. However will still document P3, P4 & P5 1# runs with their 70% initial heat proliles.

Also an important note of omission, CT5 was run drinking first cup of the morning Cafe Crema Americano LFC WP Yirg' (coarser ground super lungo ~5oz double shot pull topped of with hot water), CT6 second cup of the morning cappuccino normale pulled City+ Kona.

[miKe mcKoffee]

Drinking an excellent first cup of the morning FC+ Sulawesi roasted 9 days 19 hours ago, mason jar vac sealed room temp rest/stored (with opening for brewing re-vac'ing along the way), Cafe Crema Americano. (Very few oil droplets have appeared, when roasted none.) This was a 1# batch run P1 18:30 roast time with total closed door cooling. (2nd full day I had the Behmor. I did not record the temps, hadn't started that yet.) Common wisdom says it should not be this good with anemic cooling yet it does not seem flat or baked at all. Great typical Indo' spice notes I look for in a Sulawesi. I wouldn't call it bright, but rather smooth and balanced and very satisfying. Seems very reminiscent of Sulawesi I used to buy fresh roasted from a local roaster using a Probat. Hate to admit it but believe I like this Sulawesi cup better than most of my Caffe Rosto Sulawesi roasts the past 6+ years. Will definitely have to include full closed door cooling on serious roasts intended for cup evaluations, especially on bright character beans.

[miKe mcKoffee]

I'll add that this 2nd cup of the morning WP Yirg' Caffe Crema Americano is also excellent. Wonderful bright lively smooth cup. (Easily on par with profiled Rosto roasts me thinks.) This was roasted 6 days 18 hours ago (my usual vac mason jar stored...). It is from that 1/2# batch run P1 12:45 roast time with 1 min closed door cooling. (The batch I'd intended to used P2 and goofed.) So even preliminary worst case scenario might be to really accentuate acidity roast smaller 1/2# batch and more open door cooling, but of course need to test 1# with this Yirg', especially using boosted heater!

Now to do a ho-hum record the numbers for P3 1# batch...

[RapidCoffee]

Mike, thanks for your reports on this promising new home roaster. Any chance you could be convinced to present the data in graphical form? It only takes a moment in Excel, and I think it would help in evaluating the measurements. For example, here's your CT5 roast profile:

[miKe mcKoffee]

Mike, thanks for your reports on this promising new home roaster. Any chance you could be convinced to present the data in graphical form? It only takes a moment in Excel, and I think it would help in evaluating the measurements. For example, here's your CT5 roast profile:

I knew somebody was going to ask for graphs! "Only takes a moment" to create graphs in Excel assumes a couple things: 1) have Excel and 2) already know how to quickly create charts in Excel. Ok for 1) I do have full Office Suite. However for 2) I am not proficient with Excel and spent over a half hour messing with a line chart and how the hell do you change the axis so each 30 second temp data point entered is 30 seconds (or 0.5 minutes) instead of 1 minute increment. I'm not wasting any more time on it, don't have the time to waste!

Anyone is more than welcome to convert my data to graphical! I have more test data to key in and post and many many more test roasts to do rather than learning Excel graphing...

[miKe mcKoffee]

CT7

1# P3 (70% heat 20% time, 80% 35%, 100% 45%) 21:30 "C" as recommended (max possible 23:30 time P3 1#), ambient 70f, pre-heat 160f

NOTE: No longer using the cheapo CircuitSpecialists MS6500 digital thermometer & TC, it's wonkers. First didn't really think it's been reading accurately placed on left drum mount bracket but seemed to track fairly closesly with Alchemist's example (he's using same MS6500 IIRC which is why I used it). It did seem realistic temps when simply fished over top of chaff collector but didn't seem to like being wrapped around the bracket. This morning after pre-heating to 160f, removed the chaff collector, about to put in the drum and noticed it was displaying 26f! Messed around with it one second it'd be in the 80's then 20's then 30's then 80's.... Fortunately recalled I have a Fluke 16 multi-meter with K tc so have switched to using it. It's temps seem much more realistic for dead center drum height middle of roaster front to back top to bottom. So previous test roast temps are meaningless for direct temp comparsion to any subsequent tests.

21:30 - 143f S1 70% heat
21 - 144
20:30 - 155
20 - 170
19:30 - 185
19 - 202
18:30 - 218
18 - 232
17:30 - 247
17 - 260 now S2 85% heat
16:30 - 274
16 - 290
15:30 - 304
15 - 319
14:30 - 331
14 - 344 drawfan/afterburner on
13:30 - 345
13 - 347
12:30 - 349
12 - 352
11:30 - 354
11 - 357
10:30 - 359
10 - 363
9:30 - 367 now s1 100% heat
9 - 376
8:30 - 385
8 - 395
7:30 - 403
7 - 412
6:30 - 421
6 - 429
5:30 - 437
5 - 444 early 1st
4:30 - 449 heater off, back on 445
4 - 443 where'd 1st go?
3:30 - 447 1st back weakish strong
3 - 454 1st cont', heat off 459
2:30 - 458 heat on 453
2- 451
1:30 - 453 1st cont' slowing
1 - 459 1st very slow, heat off 462
0:30 - 462 no 1st
0 - 456 early 2nd, let go to cooling (closed door)

13:00- 456
12:30 - 431 no more 2nd
12- 419 opened door
11:30 - 406
11 - 389
10:30 - missed it
10 - 363
9:30 - 331 afterburner off
9 - 304
8:30 - 285
8 - 260
7:30 - 255
7 - 221
6:30 - 204
6 - 189 ...
0 - 81

City+ maybe very Light Full City, even roast.


Fluke 16 P1 100% heat no bean test run
(Manual advises doing 12 minute empty burn every 5 batches so what the heck:-))
12 - 70f
11:30 - 78
11 - 122
10:30 - 186
10 - 249
9:30 - 305
9 - 357
8:30 - 403
8 - 447
7:30 - 485
7 - 522 drawfan/afterburner on, heat off 523
6:30 - 514 heat on 509
6 - 502
5:30 - 513
5 - 529 neat off 543
4:30 - 543 heat on 530
4 - 526
3:30 - 522
3 - 535 heat off 541
2:30 - 537 heat on 529
2- 521
1:30 - 528 heat off 536
1 - 534 heat on 524
0:30 - 518
0 - 523

10 - 523 (1/2# batch selected cooling defaults to 10min instead of 13, closed door entire test)
9:30 - 483
9 - 437
8:30 - 398
8 - 365
7:30 - 334
7 - 310 afterburner off (stays on same first 3 min cooling as 1#)
6:30 - 288
6 - 263
5:30 - 237
5 - 216
4:30 - 198
4 - 183
3:30 - 169
3 - 157
2:30 - 146
2 - 137
1:30 - 129
1 - 121
0:30 - 114
0 - 108

I had no more time for tests this morning. May not for a couple days though may squeeze one or two in Sunday.

[RapidCoffee]

I knew somebody was going to ask for graphs! "Only takes a moment" to create graphs in Excel assumes a couple things: 1) have Excel and 2) already know how to quickly create charts in Excel. Ok for 1) I do have full Office Suite. However for 2) I am not proficient with Excel and spent over a half hour messing with a line chart and how the hell do you change the axis so each 30 second temp data point entered is 30 seconds (or 0.5 minutes) instead of 1 minute increment. I'm not wasting any more time on it, don't have the time to waste!

Anyone is more than welcome to convert my data to graphical! I have more test data to key in and post and many many more test roasts to do rather than learning Excel graphing...

Mike, I don't mean to sound ungrateful. It's great of you to share your observations on the Behmor roaster. But presenting your data in an easy-to-comprehend format is hardly a waste of time. Trust me, you don't have to become an Excel guru to make simple graphs. (Hint: use an X-Y plot instead of a line plot for your data.)

Just shoot me an email or PM me and I'll be glad to send you step-by-step instructions. If you really don't want to do that, then send me the data and I'll graph it for you - preferably just the data points, without added commentary. But I'd rather teach a man to fish...