Announcement

Collapse
No announcement yet.

Roasting slow-fast profile in the KKTO

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

  • readeral
    replied
    Originally posted by ArnhemR View Post
    fair points readeral, you are aiming for the opposite flavour profile to my preference so i dont have anything to offer you there!

    I am surprised you cant hear FC clearly - thats never been an issue with my KKTO

    I sort of get the feeling though that you are complicating the process somewhat, I have never felt that controlling the air low was necessary, although I did experiment with a speed controller on the fan but found I couldnt detect any difference.

    I do think bigger batches are easier to control on the KKTO, I would suggest you try the biggest batch that makes sense with your usage. If you can roast as much as 600g green, aiming for 500g roasted (roughly), then I would try preheating to around 180deg and drop in the beans and leave on full heat until just before FC, start burping to slow the BT RoR to around 5 deg/m and finish when you like the look of the beans. Alternatively control the RoR approaching FC by turning the thermostat right down.

    What software are you using to record profiles?

    I would be surprised if it were not possible to roast on the KKTO to suit your flavour profile without major mods.

    Its worth hitting up Paul (the inventor of the KKTO ) for his thoughts, he is always happy to share his vast knowledge and experience.
    It depends on the beans. Some are clear, some are impossible. I work hard to make sure I can hear it, but my thermocouples are a good indicator of when I should be listening out and if it should have finished.

    I'll have a crack at your 600g 180 degree preheat and full on for the rest of the roast suggestion with some of these african beans. That's close to what I did when I first built the KKTO for my first few months of roasting in it, and is similar to my approach for softer beans - but keeping an eye on the upper temperatures of the system.

    I'm using Artisan to record my profiles - prior to that I was using RoastLogger, which I actually preferred, but I have an Arduino with a TC4 as my logger and controller of element, and at the time I was rigging it all together Artisan was going to be more straight-forward. Now they're much for muchness, and I might move back to RoastLogger at some point. It has much better smoothing, and doesn't poll the Arduino for data points, but takes all the raw data and processes it itself.

    Paul and I have had a number of conversations over the past 12 months, mostly in terms of build, but some in terms of roasting approach. A lot of our conversations have been documented elsewhere on this forum. The 'set and forget' might be great for most, but what I'm really after is to be able to manipulate my roast, control the 3 phases of the roast to enhance or suppress bean characteristics, much like Matt and Mal do with their Corretto setups. It could just be that the KKTO isn't agile enough for it, but I'm giving it a go! I like a lot of the KKTO package, but I'm not convinced there isn't an opportunity for improvements (a la my experiments in the past). It's just a question of whether I have the time, tools and tenacity to do it.

    I'd love to be able to hit my preferred flavour profile with a simple approach - but over the past 12 months, it hasn't yet come up the goods.

    I probably sound really ungrateful for the conversation and advice - I'm not, I do appreciate all of the input everyone is putting in. It would just be exhausting to outline everything that I've tried in my roaster thus far, and all of the conversations I've had on and offline with various people.

    Leave a comment:


  • readeral
    replied
    Originally posted by MrJack View Post
    I'll return to the rest of you post a bit later when not on my phone, but I just wanted to ask - where the evaporative cooling concerns have come from? i.e. what do you think is going on, and why do you think it is of particular concern in your roaster?
    Well, Steve mentioned it up in number 12, but I've read up about it with regards to 'the stall' when slow-cooking meat, and that's assisted by increasing airflow. I'm creating a humid environment which is stopping moisture being taken from the bean surface, and so getting in the way of my heat transfer. I think. Possibly. My post says "likely". Again, I'm not particularly hung up on this being 'the thing', it's just (possibly!) a factor, and will probably fix itself with other changes.

    Leave a comment:


  • readeral
    replied
    Excellent, thanks Mal. Glad she's been writing - a mate of mine went on some workshops with her and thought she was brilliant (I actually have a copy of that equilibrium graph on my PC - but not with her heat transfer details). Alas, a lot of what he took away (beyond this excellent kind of content), was specific to his Has Garanti.

    It's good to read the distinction of natural and forced convection, and how it needs to carry the bulk of the roast. Gives me confidence in the lower charge temps, but also my need for improved convection.

    Leave a comment:


  • Dimal
    replied
    Hello again Al...

    An Aussie source of some really good info about coffee roasting comes from highly respected Roaster Extraordinaire Anne Cooper, who currently has been running a series of articles in the Cafe Culture Magazine. All of their articles can be read online or the magazine can be subscribed.

    Anyway, thought that this episode might have some information of interest for you, starting on page #32...
    https://issuu.com/cafeculture/docs/c...22766/37516812

    Mal.

    Leave a comment:


  • MrJack
    replied
    Originally posted by readeral View Post
    Couple this with the likely evaporative cooling due to poor exhaust/air flow, and we've got ourselves a complex problem.

    I'm sure I've got some science wrong - feel free to correct me, but in the context of solving this problem of effectively executing a slow-fast profile in a KKTO.
    I'll return to the rest of you post a bit later when not on my phone, but I just wanted to ask - where the evaporative cooling concerns have come from? i.e. what do you think is going on, and why do you think it is of particular concern in your roaster?

    Leave a comment:


  • ArnhemR
    replied
    fair points readeral, you are aiming for the opposite flavour profile to my preference so i dont have anything to offer you there!

    I am surprised you cant hear FC clearly - thats never been an issue with my KKTO

    I sort of get the feeling though that you are complicating the process somewhat, I have never felt that controlling the air low was necessary, although I did experiment with a speed controller on the fan but found I couldnt detect any difference.

    I do think bigger batches are easier to control on the KKTO, I would suggest you try the biggest batch that makes sense with your usage. If you can roast as much as 600g green, aiming for 500g roasted (roughly), then I would try preheating to around 180deg and drop in the beans and leave on full heat until just before FC, start burping to slow the BT RoR to around 5 deg/m and finish when you like the look of the beans. Alternatively control the RoR approaching FC by turning the thermostat right down.

    What software are you using to record profiles?

    I would be surprised if it were not possible to roast on the KKTO to suit your flavour profile without major mods.

    Its worth hitting up Paul (the inventor of the KKTO ) for his thoughts, he is always happy to share his vast knowledge and experience.

    Leave a comment:


  • readeral
    replied
    Hi ArnhemR - it's not suited to the KKTO, you're right, but alas I own a KKTO and it's coffee is not suited to my tastes! So I'm seeing what I can do to modify the design or approach to make it do what I want. One day I'll fail in this pursuit and get a more suitable roaster for my liking :P But in the meantime I'm a student, living in an apartment, with no income of my own so... this is the extent to which my wife will let me go. All I want is those delicious fruity flavours that I know exist in my (especially African) greens. But maybe I'll have to stick to nuttiness and cocoa.

    My turbo oven already has (almost) as much control of it as I can wield. I can already control it's power level (albeit through pulsing rather than PWM) from always on through to always off. Rather than let it go full boar, I usually use my PID to set a maximum environment temperature. This isn't ideal, as really I should have the thing on full boar most of the time, and have some way of controlling the airflow to manipulate my environment temps. Corretto owners do this with lids on their bread pans, KKTO owners like yourself typically do it with burping. I found that KKTO approach too difficult to repeat but maybe I should give it another shot.

    I'm well aware that my thermocouples will behave uniquely - most of the concern about TCs in this thread is that they have drifted, so what value they have had recently is diminishing - until I fix the problem. Whether I like it or not, TCs are my best guide with my KKTO. With a small bean mass and a large amount of mechanical noise (don't know about you, but my Turbo Oven has got to the stage of rattling and requires retightening regularly) 1C isn't very discernable, and telling steam and smoke apart with such a sealed system is difficult.

    Most of the people I have read that are happy with KKTOs are, like yourself, much further north than I am. Makes me wonder whether there is a distinct advantage for the KKTO user in a warmer ambient climate. I had much less finicky roasts in summer (Again well aware that all roasters require adjustments for winter conditions).

    On the design front, I would posit that a solution would be to actually build in an exhaust (with fan). Either a tube straight down from false floor through the base, or diagonally from false floor to wall of roaster - which will draw the warm convective air across the beans with much more velocity, pull out chaff, draw out steam and smoke and make airflow a controllable dimension of the roast without more mods to my Turbo Oven. Then I could have the turbo oven always on. But the idea of more Stainless Steel drilling is not at all appealing, and further engineering decisions like diameter of tubes would make it a risky move. Still - I might do that some time as an experiment.

    Leave a comment:


  • ArnhemR
    replied
    I think the profile you are trying to achieve is not particularly suited to the KKTO, all my profiles have a slightly decreasing RoR to FC, it starts around 11deg/m and at first crack is around 6.5deg/m, then I continue to control a slowly falling RoR from FC to SC - from 6.5deg/m down to around 3.5deg/m at the end of the roast. I aim for a Rao style profile with 20-25% of the time to FC for the phase between FC & finishing.

    I roast 1.2kg batches in my KKTO, i bypass the thermo couple so the heat is always on, i lift the TO off to 'burp' when I need to control RoR.

    I actually found over the years that bigger batches are easier to manage in the KKTO - as long as you bypass the cycling of the element.

    Temps and thermocouples are something I use relatively - they do vary and its more important to observe whats happening (FC, SC, etc) than relying on what you think should be happening at a given temp. Different beans vary a lot too. I dont bother any more with ET, I think its more important in drum type roasters, I couldnt find any correlation in the KKTO.

    My load temp is 250 deg.

    Leave a comment:


  • readeral
    replied
    MrJack, most of this I understood from many sources I've read, thanks for a concise summary. I might be using the term 'momentum' unhelpfully, but I suppose I have meant 'ongoing heat transfer' especially as the roast has progressed. It's a minor point in a broader discussion that essentially is determining: 'effective heat transfer to the bean, and through the bean is essential' and as the whole roast is progressing very slowly, due to poor transfer, the beans are subject to high temperature for a long time. Couple this with the likely evaporative cooling due to poor exhaust/air flow, and we've got ourselves a complex problem.

    The salient comment you made is "The rate of convective heat transfer depends primarily on the difference in temperature between the air and the bean, and on the velocity of the air." What I'm finding is that although this is a 'convection oven' unit, I'm getting very little convective transfer of heat given how it's now being used. The challenge I'm trying to solve is how to improve energy transfer to and through the bean for a successful roast within reasonable parameters. The success of the suggestion by kwantfm of lowering the batch is contingent on good roaster design to have good, direct convection (and, lesser so conduction/radiation), which I personally feel it does not have. I obviously have no instruments to actually measure this, and chaff flying around the roaster seems to contradict my hunch, but I think the problem may lie in the issue of air pressure (which, if I have understood correctly, means with negative pressure outside the system will produce air flow through the roaster) - rather than just air movement.

    The reason - I suspect, and honestly all of this is just hunch given I am a musician not an engineer - that lowering the batch will not help much anyway is that: given the size of the roaster, the amount of energy lost to the increasingly exposed metal surfaces of the roaster (to be conducted outward to the ambient air), will offset the desired gains from a lower batch size. Creating better convection directly to the beans, rather than to the surfaces of the roaster, should help a lot more.

    I'm sure I've got some science wrong - feel free to correct me, but in the context of solving this problem of effectively executing a slow-fast profile in a KKTO.

    Leave a comment:


  • MrJack
    replied
    Originally posted by readeral View Post
    I'm assuming that by delaying my ramp up, I can get better momentum into first crack for a more rigorous crack, as well as avoiding those high environment temperature negatives, damaged beans from earlier high heat input and hopefully carrying it into post-FC development a lot better.
    There are some interesting observations in this thread, and clearly a lot of collective roasting experience, but there also appear to be some misconceptions about the fundamentals of heat transfer.

    Forget about 'momentum' - it's a misleading analogy. What you are describing is a product of the temperature gradient in the bean (i.e. that the middle of the bean is at a lower temperature that the outer surface).

    If you start with a cold bean and pass hot air over it, heat (energy) is transferred from the air to the bean surface. The rate of convective heat transfer depends primarily on the difference in temperature between the air and the bean, and on the velocity of the air.

    Meanwhile, heat is being conducted away from the surface of the bean towards the centre. It takes time for this to happen (and from memory the dynamics are primarily a function of conductivity and heat capacity).

    If conduction through the bean is rapid relative to convection then there won't be a significant temperature gradient (i.e. if heat is transferred away from the surface nearly as fast as is is transferred to it, then the temperature at the inside of the bean will be close to that of the surface).

    If, on the other hand, the rate of convection is very high relative to conduction (through the bean) then the surface temperature will rise faster than the centre temperature (resulting in a temperature gradient).

    This effect is amplified when the air temperature is increasing (because the surface temperature will rise with it faster than the centre).

    Now, first crack occurs once a particular temperature is reached. The slower the temperature rise of the bean surface in the lead up to FC, the closer the centre of the bean will be to FC temperature - perhaps resulting in the observed "tighter" first crack.

    This of course is somewhat simplified - chemical reactions and physical changes in the bean as it expands complicate the situation beyond simple heat transfer - but it is still useful for picturing what is happening.


    A picture is worth a thousand words: https://youtu.be/NwpJDV5vzm8
    Last edited by MrJack; 20 August 2016, 03:21 AM.

    Leave a comment:


  • Yelta
    replied
    Originally posted by readeral View Post
    I think I'll be looking at converting this into a "glorified breadmaker" and go with a heat gun, and ditch my electronic control via my roasting software. I suspect it's the only way I'm going to get the kind of airflow that I really want. Just a bummer that so many seemed satisfied with the KKTO over the years. Maybe I have much higher expectations!
    Have never used a KKTO so cant comment on the device, however I have been using a Coretto for over 10 years with great success, my setup really is quite basic, Breville single loaf bread maker, Bosch variable speed heat gun, DMM with temp probe and timing with wrist watch.

    I roast 725 grams weekly, over cooked one batch slightly in the early days, since then have not had a failure, perhaps I'm not as fussy or as scientific as some, however i do know and enjoy good coffee and consider my efforts as well above average.

    The one disadvantage with a Coretto is uncontrolled chaff, I'm sure this could be controlled, not a problem for me as I roast in a shed and set up a pedestal fan to blow the chaff into the garden.

    Not attempting to denigrate the KKTO, simply explaining a setup that works very well for me.

    Leave a comment:


  • readeral
    replied
    Originally posted by kwantfm View Post
    So the problem is that despite getting ET close to the limit (I find that ET greater than about 270 C causes ashiness and scorching) that you still can't get enough energy into the system. How is airflow controlled in a KKTO? If I had the same problems on a drum roaster I'd increase airflow and raise gas to get same ET but higher system energy.
    Also I may use "ET" incorrectly, as I'm referring to the general environment temperature, rather than an exhaust temp. That's the other downside with this whole setup is not being able to create negative air pressure with an exhaust to draw the hot air through. It (generally) is fairly static air. The definition of baking. Given turbo ovens are... ovens... that's what it's really designed to do!

    Leave a comment:


  • readeral
    replied
    Originally posted by kwantfm View Post
    Will be very interested to see whether your fan tweaking experiments yield results. Good luck!
    Thanks! I'll give it a go.

    Leave a comment:


  • readeral
    replied
    I think I'll be looking at converting this into a "glorified breadmaker" and go with a heat gun, and ditch my electronic control via my roasting software. I suspect it's the only way I'm going to get the kind of airflow that I really want. Just a bummer that so many seemed satisfied with the KKTO over the years. Maybe I have much higher expectations!

    Leave a comment:


  • kwantfm
    replied
    Will be very interested to see whether your fan tweaking experiments yield results. Good luck!

    Leave a comment:

Working...
X