No announcement yet.

Behmor Plus - Roasting Approaches

  • Filter
  • Time
  • Show
Clear All
new posts

  • Behmor Plus - Roasting Approaches

    I've read through some of the threads on Behmor plus in here and one thing I've noticed is a bit of a lack of thorough discussion on the manual mode, and techniques people are using to roast different varieties of beans. This panel update, going from the old 1600 to the plus, has been revolutionary for me, and wanted to share some of the things I've picked up, but also was hoping for others to throw a few ideas around to see other techniques being used. Anyway here goes:

    I write ET generically for your measure of environment temperature, whether it be the Behmor thermistor or a thermocouple somewhere in the chamber.

    I've been mainly focusing on trying to get Ethiopian DPs and Brazil pulped naturals down pat. I prefer roasts which are balanced in acidity, body and sweetness. I typically do 200g batches.

    Ethiopian - dry processed:
    P1-auto mode until FC. I hit D for double drum speed at FC. As FC really starts to get going, I hit P1-manual to cut the power, keep an eye on the ET to make sure not dropping. This gives me a good loud first crack, which I'm always searching for on the Behmor. FC for me typically occurs at around 9 to 10 mins for a 200g charge. FC typically runs at around 1:30 duration. After it fades I hit the beans for another 45secs to a minute at half heat (P3-manual) until end of roast (cool). Typically my ET is creeping up at this stage, although I have a suspicion I should hit them a bit harder, maybe P4-manual, and for a shorter time.

    Brazils - pulped naturals
    P4-manual mode (75% power) until bread baking smell, then P5-manual (100% power) until FC. Aim here is to go a bit easy early on to avoid tipping, but once drying phase is over get some momentum up before hitting FC. Once I hit FC I drop heat to P1-manual, hit D, when I hear the first few consistent pops of FC. I drop heat earlier then I do with the Eth DPs, because I've noticed the Brazils take off like a rocket if you heat too much during early FC. I let FC fizzle out, which is around ~1:30 mins duration, then add back P2-manual 30 seconds, P3-manual 30 seconds, and P4-manual 30 seconds, listening intently for any sounds of SC (and watching for smoke). At some point in this ramp I hit cool. One issue I have here is I can't seem to just butt up against SC (ie a few snaps and nothing else). Typically if I hear one snap of SC during the roast, I hear 15-20 seconds of SC snaps during the cool, and my roast is a bit darker then I'd like. I'd be keen to hear how others drive the Behmor to achieve this outcome - a roast delivering just a couple of snaps of SC.

    General tips I’ve learnt along the way:
    - Don’t add much (any?) heat during the start of FC or your roast will go off to the stratosphere
    - For almost all roasts I hit D at FC to kickstart the exothermic, but I use slower drum speed for the beginning of the roast. Keeping it on double drum speed the whole roast led to too ‘bright’ tasting roasts.
    - If you want to notch down a power level, say 100 to 75, the key sequence is P5, P1, P4. The elements have fixed cycle times which can lag this power change, P1 ensures power is cut straight away.
    - Before hitting cool, hit P1 beforehand to turn off the heating element. This stops that extra little kick cool gives your roast, if the element is on.

    Appreciate any input or discussion this post generates….

  • #2
    A very detailed review pb200 - should generate some great Behmor roasting discussion! Sounds like the new control panel is giving some great manual profiling options…
    Cheers Matt


    • #3
      Great write-up pb200 and thanks for sharing your tips!
      My experience with manual mode somewhat mirrors yours. Although, I haven't had a chance to play with drum speeds or cycle times yet.
      Here's a recent snapshot of my general approach:

      Guatemala Huehuetenango
      300g charge
      400g | P1 | Start
      Straight into manual mode 100% power (P5) to start of rolling first crack.
      Drop to 0% power (P1) for approx 20 seconds.
      Back up to 50% power (P3) for approx 2 minutes.
      Hit cool just before second crack.
      First few snaps of second crack follow. After around 20 seconds, I usually open door and speed up cooling by removing chaff tray along with a blast with some cool air.

      Completely unnecessary, but I have installed 2 bead probes to help diagnose and repeat roasts. One inside wall LHS and the other in the exhaust vent. Attached profile shows the above roast in action.

      Attached Files
      Last edited by moreCoffee; 25 May 2015, 08:41 PM. Reason: small correction


      • #4
        Thanks pcf1978, this is good info. I see one key difference with your heat application and mine is you hit them a bit earlier, after dropping power at FC. Is this because your ET probes tell you the temp starts dropping at this point (FC+20 seconds) ?
        Interesting that 50% is enough to get you to second in a reasonable timeframe, and this is a hard bean. Makes me think I should revise my ramp strategy between first and second for the Brazils.

        Maybe 50% power is all you really need between FC and SC...


        • #5
          Originally posted by pyrmontboy200 View Post
          Thanks pcf1978, this is good info. I see one key difference with your heat application and mine is you hit them a bit earlier, after dropping power at FC. Is this because your ET probes tell you the temp starts dropping at this point (FC+20 seconds) ?
          Interesting that 50% is enough to get you to second in a reasonable timeframe, and this is a hard bean. Makes me think I should revise my ramp strategy between first and second for the Brazils.

          Maybe 50% power is all you really need between FC and SC...
          I've found if I drop to 0% for any longer than 20-30 seconds the exhaust temp and rate-of-rise starts to flat line and then you need to hit with a lot more power to reach 2C in the same amount of time, which can give you less control approaching 2C.

          Some beans are far more aggressive at FC and can carry on to 2C very quickly, in this case I try and extend the time at 0% for a few extra seconds to provide better separation.

          Obviously without a probe in the bean mass it's hard to know exactly what the bean is doing amongst all this.

          If you decide to experiment, I'd be very interested in learn how you go!


          • #6
            Behmor Plus - Roasting Approaches

            Ok so motivated by pcf's pointers above I have performed some repeat roasts on the same bean, Brazil Toffee Cerrado, experimenting, among other things with the duration time of the power cut post FC. My daughter keeps tearing up and drawing all over my roast notes, so I will log them here for my own benefit and anyone else who may be interested.

            The first thing to note is that I am now able to control the entry into SC a lot more predictably, simply by using P3 manual after the power drop, as suggested above by pcf. The first thing I noticed with roast #1 below is this offish acidity in the background, and a slight lack of body given I had just touched SC. I'm no great cupper, and probably have the palate of a pig's ass, but I was able to tell this by comparing against roast #2 and also checking with my wife (unprompted!) who has a better palate than me. Researching this a bit further my hypothesis was a bit of a stall in the roast, and the post here Let's Talk About Roasting - Matt Perger has got me thinking about paying a lot more attention to the roast post FC to SC.

            Roast #1 -200g - Double drum speed at FC
            P4 manual
            P5 manual @ 6:50
            FC (consistent pops) @ 10:35
            P1 @ 10:48
            P3 @ 11:30
            cool @ 12:52

            One or two snaps of SC, could've missed them completely.
            Notes: brown sugar, toffee like sweetness, some ashiness, slight background acidity which tastes out of place with this bean.

            With the hypothesis of a bit of a stall in mind, I decided to shorten the power cut in Roast #2. Regarding the ashiness I thought this was due to overdrying, so I cranked up the drum speed the whole roast. This turned out to be a mistake, I think it made the problem worse.

            Roast #2 - 200g

            Double drum speed the whole way
            P4- manual start
            P5 @ 5:00
            FC (consistent popping) @ 11:20
            P1 manual @ 11:29
            P3 manual @ 11:58
            Cool @ 13:00
            Several snaps of SC on cool, lasted for 5 seconds or so. Remove chaff tray and opened door on cool


            Gobfuls of brown sugar, toffee like sweetness, little acidity, great body, but a little more ashiness than above.

            Next roast I will be trying to remove the ashiness, hope to report back on this success soon.
            Last edited by pyrmontboy200; 15 June 2015, 05:44 PM.


            • #7
              All in all both the roasts above were very acceptable, they are incredibly sweet and true to their marketing name Toffee Cerrado. Lingering in the background of each cup was a little ashiness which I'm determined to get rid of. Some beans in the roast above exhibited centre line scorching. With the hypothesis of too much heat after end of drying to FC as causing this ashiness, I set about a new roast where I reduced heat post drying and made the time
              back up before drying. This turned out to be a big mistake :

              P5 manual until the 5:00 mark
              P4 manual til FC @ 12:00.
              P1 manual @ 12:05. P3 manual @ 12:25
              Cool @ 14:30

              The first thing I noticed with taking 75% power into first was a really flacid first crack, which never really got going, probably because I cut the heat too early.

              There was no evidence of centre line charring, but still some internal scorching and ashiness. Cupping this I got that baked/under developed taste that I know so well. A week later and the beans still do not smell.

              Conclusion, I should follow my taste buds and stick to the basic profiles above, with some slight tweaks on the drying phase, which I cover in the next post.


              • #8
                More research on this forum tells me to go slower in the dry phase for Brazils, so I took the basic outline of my successful roasts above, and reduced the heat early on leading to the following profile:

                Roast run entirely in manual mode:

                2 minute preheat on P1.

                P4 to 8 minutes
                P5 til FC @10:35, rolling at 11:15,
                hit P1,D @ 11:15
                P3 @ 11:45
                cool @ 12:30

                A few snaps of SC in the cool for maybe 10 seconds, it was trying to work up to consistent snaps, but never got there. Happy enough with that....

                On inspection the roast looks pretty good. Some 'internal' tipping, not visible from the outside, so maybe still too much heat too early? Will have to experiment.

                A preliminary cupping today reveals a very balanced cup that has good caramelisation, and little to no bitter tones. As my cupping is not the best I will wait for the ultimate acid test, a brew in the V60 at work this week before I declare success.


                • #9
                  Brewed this roast yesterday and today in my v60, and it is a far superior cup to any of the previous roasts above. Full bodied, and sweet, very little acidity. It is a welcome change from some of the brighter Ethiopians I've been drinking.

                  So I think I am converging on a Behmor 'profile' for this bean, and dare I say a template or starting point profile for all Brazil naturals. Tasting such a big effect in the cup of applying less heat during drying, I am going to keep experimenting, I want to push this point a little further, so next roast I will use the P5 - auto profile until FC, then to manual mode and a similar stretch as above. The theory here being to drop the early heat even more, and see what effect this has on the final roast.


                  • #10
                    With the theory that p5 is the most suitable auto profile for my, and perhaps all, Brazilian naturally processed beans I proceeded with the following 200g roast:
                    P5 auto til fc
                    P1 manual for 30 seconds, at rolling FC, turn on double drum speed
                    P3 manual til the start of SC

                    FC hit at 13:00, rolling at 13:25. I powered back up to 50% at 13:55, and at 15:55 I saw a puff of blue smoke, and hit cool. I heard a couple of snaps of SC during start of cool.

                    This turned out a superior roast to the attempts above, far less ashiness if any at all, and got all the tasting notes with ease. It stood up well in the v60, the only pickle being as espresso this weekend I could detect a slight bitterness. This may be down to user error on the machine, however i suspect this particular bean should not touch SC at all, so next roast i will try and pull about 15-30 seconds earlier than the above, and see where that leads me. The theory here is this bean is so ridiculously sweet that it turns to charcoal a lot quicker, i.e a double edged sword.

                    So there we are, P5 auto on the Behmor plus use of the manual functions to power cut and slowly stretch to SC, seems to work a treat for this bean. Next bean I am attacking is a dry processed Ethiopian.


                    • #11
                      Thanks for all the info on your roasting techniques. I have ordered a Behmor and will use your experience to give me a reasonable start with my roasting adventure. If you have any other info/advice, I am all ears.



                      • #12
                        So with slow and low being the general guide for Brazil naturals, I've gone back to the Yirgacheffe Koke natural, which I typically hit hard and roast fast. I took 200g with the following plan:

                        P1 auto til FC, p1 manual for 30 seconds, then P3 til SC start or just before.

                        So after error 7ing the first batch while daydreaming, the second batch hit FC start @ 8:53. 9:11 rolling and I could tell these bad boys had some momentum so I hit P1 for 40 seconds until 9:53. P3 manual til cool @ 11:51, no SC but good development and pretty sure it was about 10 seconds away.
                        A pretty standard 4:4:2 profile if you allow some time for the Behmor to heat up at the start. Cupped it and got balanced acidity, big berry and rose water aroma, some sweetness giving way to dryness which I'm still unsure whether I like. Big body, and flecks of oil on the beans two days post roast confirmed this adequate development. A week later drinking this at work in my v60 was really enjoyable. The dryness which I had reservations about, seemed to have abated quite a bit.

                        All in all a successful roast, and I will probably only tweak the profile above at the margins. Next I will apply this same guide to another Yirgacheffe I have lying around, and see where that leads me. I will continue to report these results back here.

                        Ps the thought occurred to me that when the Behmor connected panel is released, a lot of information in posts like this will be redundant. It will simply be a matter of sharing a profile that works for a particular bean on the cloud. Clearly you will have to be in a similar temperature zone to the person sharing...
                        Last edited by pyrmontboy200; 2 August 2015, 03:12 PM. Reason: Clarification


                        • #13
                          Tasting this Ethiopian again today after a bit more resting, I've decided that the dryness discussed above is a roasting defect. It is not an astringent taste but rather a component of the acidity which bothers me, I'm calling it dryness because although there is only a hint, it is there with every sip and reminds me of a dry wine. So I still have work to do on this bean! Theories to be tested are as follows:
                          - underdeveloped, I.e although the beans show all the signs of proper development, right buried down inside the bean is not as well roasted as the outer
                          - I've been keeping the power cranked at full bore until FC consistent pops, then slamming on the brakes with p1 manual for 30 or so seconds. So I'm conscious of a bit of an over correction here which leads to a stall. I wonder if a more gentle slowdown is called for here, I plan to experiment with a taper down in power as the first outliers go off.


                          • #14
                            OK I am going to dust off my Behmor 100 and get back into it. thanks for the motivation, and the details on roasting profiles.


                            • #15
                              I did another roast of the Yirgacheffe Koke Natural on the 7th, this time 300g with the following approach. I was interested to see what difference upping the batch size would have.

                              P1-auto 400g til FC start @ 11:05
                              P4-manual +D @ 11:13 (tapering down to 75% power heading into FC)
                              P1-manual @ 11:27 (roll FC)
                              P3-manual @ 12:07
                              14:10 cool with 1 or 2 snaps of SC.

                              Drinking now as pourover and it tastes much the same as my last batch, the dryness encountered last time is less prominent however. It's a good brew and all the fruitiness and sweetness is still there, however there is slightly less body than I remember the last batch having. Might need to rest it a bit more maybe... The jury is still out on the taper off of power going into FC, but I suspect it isn't going to be a big saviour. I realise how unscientific a lot of my tests of hypotheses are btw!

                              I will be trying out another Yirg I have lying around in the Behmor and report results soon.
                              Attached Files