Announcement

Collapse
No announcement yet.

First roasts with data logger

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • First roasts with data logger

    Hello CSers

    I got a data logger a couple of Beanbays ago and did a few roasts with it, nothing else changed in my technique (corretto setup) so I could plot the actual temps of the way I have been roasting for ages.

    I would like your input/critique on the plots or pointers in terms of desirable temp rise rates etc.  Ie so I can setup a "correct" base line to try and follow for future roasts.

    I have a cover over the BM, with the heat gun above the "entry" hole (ie not sealed to cover) and an exit hole at the other end where the hot air/chaff blows out.

    I have been generally roasting as follows:  Turn on heatgun and bread maker and as soon as I hear first crack I raise the gun a couple of cms to reduce the heat.  Once I hear second crack I kill the heat and dump the beans to cool as quickly as possible.  I use a blower vac on suction to cool the beans and they cool very quickly (~15-30 secs).  Coffee has always tasted great.

    I have attached 3 screen dumps of my first data logger roasts:

    Brazil Yellow Bourbon Especial:


    Tanzania Uru Estate:


    Ethiopian Yirgacheffe (smaller batch so heated more quickly - Brazil roast above as background template):


    I am using version 004 of the software.

    Also, when you enter the bean type, weight etc into the box that pops up, where is this data saved/displayed?

    Any tips greatly appreciated.

    Cheers

  • #2
    Re: First roasts with data logger

    I tried searching for some "default" charts to get an idea if mine are in the ball park but have not been able to find anything.

    Would anyone like to comment on these to give me some pointers?

    Cheers

    Comment


    • #3
      Re: First roasts with data logger

      Hi Artman
      While I use the DMM, I havent computerised the process. But from my hand-written notes (see attached) I have recently started getting some stunning results on my rig from the following basic profile… box 1 is temp, the second is the gun height (not in height from the beans - just on my tripod scale)

      The overview:
      Slowly warm the pan, beans in, at max height for the first 2mins while the BM cycle is fussing about
      Lower gun and roast at a consistent input (ie controlled by gun height on my rig) for 6 minutes, then slowly increased the temp (lower the gun 1 gradation at a time) up to 1st crack at 11mins.
      Then back off the temp to stretch out the time to 2nd crack, which comes on at about 16-17 minutes. I also raise the gun a fraction at about 208 degrees - or it can run away pretty fast to the drop at 222 degrees.
      My last few bags using this technique have been quite spectacular - compared to my early attempts ;D

      Hope this helps
      Matt




      Comment


      • #4
        Re: First roasts with data logger

        Originally posted by 5D4E48515D523C0 link=1318770716/0#0 date=1318770716
        I would like your input/critique on the plots or pointers in terms of desirable temp rise rates etc. Ie so I can setup a "correct" base line to try and follow for future roasts.
        The correct profile is the one that you like!

        Sounds a little naff but the profiles will only really mean something to your setup. Slightly different probe placement, ambient temperature and a dozen other things mean that your profiles are only meaningful on your setup.

        Having said that... they look excellent. Nice ramping and taper after 1st crack. They look similar to what I expect on the commercial drum and I would be surprised if you would want to make too many changes to those.

        Originally posted by 5D4E48515D523C0 link=1318770716/0#0 date=1318770716
        I am using version 004 of the software
        Err... whats 004 software?
        I can see you are running Roast Monitor v2.37

        Originally posted by 5D4E48515D523C0 link=1318770716/0#0 date=1318770716
        when you enter the bean type, weight etc into the box that pops up, where is this data saved/displayed?
        You can save as either a CSV (template) or a JPG (image) or both. They are saved in the "profiles" directory under the directory that you have installed Roast Monitor in.
        It shows the path and filename in the images above in the Alerts box.

        Depending on what RML you are also loading it might write into the log file in the Roast Monitor directory too.

        Originally posted by 78594F555B5255525B7E457F535A5A59593C0 link=1318770716/2#2 date=1319425586
        My last few bags using this technique have been quite spectacular - compared to my early attempts
        Nice. Thats exactly what you need to do with all roasting setups. Get yourself a baseline that you can work from.

        Comment


        • #5
          Re: First roasts with data logger

          Whats this data logger software???

          Comment


          • #6
            Re: First roasts with data logger

            Originally posted by 5D4E48515D523C0 link=1318770716/0#0 date=1318770716
            I am using version 004 of the software
            Err... whats 004 software? 
            I can see you are running Roast Monitor v2.37
            Suspect he means the v4 of the RML Program I put up

            GrahamK

            Comment


            • #7
              Re: First roasts with data logger

              Thanks heaps for the input guys, much appreciated.  Just wanted confirmation nothing was obviously out of whack.  Looks like I am doing it OK, good idea re watching the end, it gets smokey very quickly at the end. 

              Derrr - didnt even think about different probe placements to other people will affect the readings (a bit like grinder settings) - just got too excited with my new toy!!  [smiley=cheesy.gif]

              So far I have just been adjusting the height of the gun to vary the temp, it sits at one height to first crack, then I raise it a bit, wait untill second crack and and dump/call as soon as I hear 2C.

              I will try another gun that has a temp adjuster (electronic) so hopefully will make it easier to adjust/follow profiles.

              Sorry Andy, got my terminology wrong, I am using the RM v2.17 as you say, but the version 004 is the RML - this one:
              http://coffeesnobs.com.au/YaBB.pl?num=1304437799/21#21

              and this asks for the bean name, weight etc.  I can find the CSV and JPG it saves, but cant find the bean info etc saved.  Will have another look.

              Cheers guys

              Edit: GrahamK beat me to it explaining version 004

              Comment


              • #8
                Re: First roasts with data logger

                Originally posted by 4143504358474747220 link=1318770716/4#4 date=1319432434
                Whats this data logger software???
                Have a look here:

                http://coffeesnobs.com.au/YaBB.pl?num=1236075603

                and

                http://coffeesnobs.com.au/YaBB.pl?num=1304437799

                If you have the DMM from beanbay you can run this software and it does the plotting and you can save roast profiles etc. Its what I have just started using.

                Cheers

                Comment


                • #9
                  Re: First roasts with data logger

                  Originally posted by 4F5C5A434F402E0 link=1318770716/6#6 date=1319433729
                  I can find the CSV and JPG it saves, but cant find the bean info etc saved.  Will have another look.
                  Bean Info is logged in the Logfile. An example of what is logged is provided in the same thread you got the rml file from.

                  GrahamK

                  Comment


                  • #10
                    Re: First roasts with data logger

                    Ah no I have a different multimeter so I guess it wont work with mine hey! That sucks... would have been good to have that sort of thing to map it out all for you!

                    Comment


                    • #11
                      Re: First roasts with data logger

                      Ahhh... that 004.
                      Didnt even think of that.

                      Originally posted by 3A292F363A355B0 link=1318770716/6#6 date=1319433729
                      I will try another gun that has a temp adjuster (electronic) so hopefully will make it easier to adjust/follow profiles.
                      I doubt that will make much difference. Changing the height is just as good I expect. Like I said, your profiles look great from here.

                      Importantly though... how is the coffee?

                      Comment


                      • #12
                        Re: First roasts with data logger

                        Thanks GrahamK, I had a look for the txt log file, but its not there. Maybe I am doing something wrong? I was full of excitement with the new toy when I last tried it so will have another go in a more relaxed state!!

                        You are right Andy, but I just need to make it easier to adjust the height of my gun, and can easily monitor the effects on the plot.

                        The coffee tastes great!! I was just thinking it might taste even better? Either way, I am very happy and cant believe its taken me soooo long to start roasting. A few friends of mine are a bit mad with me, I gave them some beans and now they are hooked on the flavour of freshly roasted beans!! [smiley=grin.gif]

                        Cheers

                        Comment


                        • #13
                          Re: First roasts with data logger

                          Glad you asked the question Artman. I have been thinking of posting a similar query. Ive been logging my last half dozen roasts, and thinking to myself.. what does a good roast profile actually look like? I hear people say ramp up the temp and taper off to draw out the second crack, but how fast do you ramp to the first, how long do you taper to the second crack?

                          I understand that every set up renders a unique profile based on the variables involved with that particular set up, but there must be a good general curve we can use as a starting point to aim for?

                          I did the same thing as you, just started by logging my normal roasting practice, just to see what I had been doing by ear/eye before I could see the graphic evidence, and the result was.. well a result. But does my profile resemble a GOOD profile? Should I use this as a start point and make adjustments based on this for individual beans, or am I making a critical mistake right from the start which will effect every roast from here on in. Sure, my roast tastes pretty good. But could it be better? If only I had ramped the temp to first crack earlier, or drawn out second crack more??

                          Here is a profile from my first KJM blend. Looks OK to me, but I dont know what I am looking for??

                          I have experimented recently with reducing the heat after first crack & in some cases removing the heat and letting the thermal mass carry the roast through, as I have read others doing but on my log it shows a drop in temp. Would this constitute a stalled roast? Or is there a certain dip in temperature that is acceptable?

                          Oh so many questions.



                          Comment


                          • #14
                            Re: First roasts with data logger

                            Originally posted by 363F33313C66500 link=1318770716/12#12 date=1319467076
                            But does my profile resemble a GOOD profile?
                            That again comes from experience. In my limited roasting experience, I simply experimented until I hit the sweet spot - ie the coffee tasted just as good as any commercial brown. Before that - they just tasted OK. But when I hit the sweet spot - you can still smell the ground aroma 1/2 day later
                            Then with the profile I had recorded, I could then see what Id done, and try to repeat it. In my handwritten log I borrowed a basic profile to get ball park (from GrahamK I think?) but in the end, the biggest shift for me was taking the roast through to the right dump temp - about 222 deg in my set up. Everything before that was under-done & sour.
                            And now that Im into the really drinkable range - I can experiment with time/ramps etc.

                            So what do your roasts taste like so far? You happy with them? At least as good as bought? If youre happy with the results - why change?
                            Matt

                            Comment


                            • #15
                              Re: First roasts with data logger

                              Thanks for the input guys.

                              Originally posted by 7150465C525B5C5B52774C765A53535050350 link=1318770716/13#13 date=1319486286
                              the biggest shift for me was taking the roast through to the right dump temp - about 222 deg in my set up. Everything before that was under-done & sour.
                              Regarding this, on my roasts, I find that if I dont dump as soon as I start hearing second crack, the beans end up roasted very dark. Maybe my temp is not tapering off sufficiently towards the end? Generally speaking, is the final roast colour dependent on the end temp or what is happening temp wise during the roast?

                              Are the cracks releated to a particular temp (for a particular bean)?

                              Originally posted by 7150465C525B5C5B52774C765A53535050350 link=1318770716/13#13 date=1319486286
                              So what do your roasts taste like so far? You happy with them? At least as good as bought? If youre happy with the results - why change?
                              All my roasts have tasted great, (only one was off and it was becuase I stuffed up in the end and didnt dump/cool quick enough, so they were...ummm... a VERY off white colour!! [smiley=cheesy.gif]), and no worse than freshly roasted bought beans, and I reckon sometimes better. Just want to know if/how I can get it better. I guess its a bit like before I tasted a REAL coffee, I thought Gloria Jeans was great!! Not quite to that extent, but you get the idea. Now where is the bucket, dont feel well after hearing the GJ word....

                              Comment

                              Working...
                              X