IKAWA inlet profile calculator - Page 2

#11: Post by **drgary** » April 20th, 2023, 1:56 pm

On my drum roaster, it's typical to see a slight environmental temp dip just before first crack becomes audible. This may have to do with moisture reduction on the bean surface so that the beans are more readily absorbing heat just prior to 1C. I did a brief search of the site for more about this but couldn't immediately come up with it.

April 20th, 2023, 4:30 pm

mgrayson wrote:The RoR signature suggests it was earlier.

That is one of the most distinctive "signatures" I've seen. It's timing also happens to coincide with one of the setpoints, so I can't rule out it being a reaction to the slope change at that point. Nonetheless, this illustrates a potential benefit to setting an inlet profile that follows a smooth curve: it makes deviations from that curve easier to spot - beneficial for Pro owners with an ET sensor to gain additional insights into their roast chemistry, consider me jealous.

Question for Pro owners: when you rotate the doser to load the greens, does the fan stop, slow down, keep going?

And yet another Pro move: you pre-heated before loading. That will put significantly more heat energy into the greens early in the roast cycle, which I think is why FC happened so much earlier. That's the same profile I've been using to reach just-shy of second crack, so that added heat may be enough to push it into SC.

I haven't added a heat input correction (yet) to the FC threshold, so how you approach FC will affect the timing. Despite that, I'm hoping the spreadsheet can adapt, so I tried tweaking the FC threshold down just a bit, to 248.8 then lowered the curvature to 5, which pushed out FC by about 30 seconds - presuming pre-heat temperature, batch size, and greens stay the same. No pressure, but if you were hoping for a shorter development time:

April 20th, 2023, 4:55 pm

drgary wrote:On my drum roaster, it's typical to see a slight environmental temp dip just before first crack becomes audible. This may have to do with moisture reduction on the bean surface so that the beans are more readily absorbing heat just prior to 1C. I did a brief search of the site for more about this but couldn't immediately come up with it.

Observing a dip is interesting, as I thought the chemistry behind first crack was exothermic, which I had figured would cause the temperature to rise. Maybe something else is going on...

On many runs where I've measured BT, I've observed a slope change after first crack. I speculate it's because the specific heat capacity of the beans drops after first crack, causing the BT to rise faster. I think this is more apparent with larger batch sizes. For example, there's a noticable slope change in mgrayson's earlier post:

#14: Post by **mgrayson** » April 20th, 2023, 6:54 pm

The fan on the pro stops when you rotate the doser.

I tasted both roasts - one dropping 30 seconds earlier and the one shown, both left to degass for an hour after grinding. The lighter one had 2 seconds of no taste, followed by the characteristic Polar Ex comfort blend. The darker one hit from the get-go and was stronger throughout. Both were very good with milk, and this is my new favorite roast for that blend. I might try to get FC to hit a bit later, but what a starting point!

#15: Post by **drgary** » April 20th, 2023, 10:31 pm

@Gary

As long as the beans are absorbing heat from the hotter air around them, they aren't exothermic. Their being labeled as such is a misperception derived from earlier practices of describing roasting changes at first crack and beyond.

#16: Post by **drgary** » April 21st, 2023, 8:12 pm

Gary,

Here's a note on your first post. When taking a roast into 2C, the taste can benefit from increasing the ROR between 1C and 2C. You develop the darker roast flavors on the outside of the bean while preserving some of the desirable flavors developed short of 2C.

April 22nd, 2023, 7:36 am

I've always thought that it would just take too long to get to 2nd crack with decreasing RoR. Yet this seems to be the definition of "flick", which is supposed to irrevocably destroy a roast. How do they differ?

#18: Post by **drgary** » April 22nd, 2023, 10:09 am

Hi Matt:

Rao's writing about the crash and flick effect applies mainly to light and medium roasts. Rather than a flick, think of rising ROR toward 2C as more gradual. This quote from Neal Wilson was my inspiration for thinking this way.

N3Roaster wrote:I'd start by removing the constraint on declining rate and suggest not being afraid of cranking up the heat moving out of the range between first and second crack. Especially if you're planning to go substantially beyond the start of 2C, taking that part of the roast faster lets you get to the higher temperatures where the chemical changes you want out of a dark roast start while not spending so much time getting there that you completely destroy desirable flavors developed earlier in the roast.

The most popular roasting class I've been teaching lately has been one that explores lots of different profile shapes across a range of roasts, and my experience with that within the dark roast category is that what I've just described is preferred for dark roasts compared with the same coffee roasted to the same ending temperature following a strictly declining rate by almost all students in the class regardless of the coffee used for the exercise. If, after getting to that end point you want to try again to compare with what you've been trying to do, the experience from this different approach should give you a better sense of how much heat you need for that coffee in your machine going in and you can plot out a declining rate plan based on that which should make it easier to avoid premature stalling. Having that plan written down (or loaded into your data logger) will let you see where specifically you're deviating from the plan (if you're still having trouble), which makes it easier to troubleshoot those specific problematic points.

Here it is in the referenced thread: How to roast darker

#19: Post by **mgrayson** » April 22nd, 2023, 12:31 pm

drgary wrote:Hi Matt:

Rao's writing about the crash and flick effect applies mainly to light and medium roasts. Rather than a flick, think of rising ROR toward 2C as more gradual. This quote from Neal Wilson was my inspiration for thinking this way.

Here it is in the referenced thread: How to roast darker

Thank you. That is a great relief and I will go back to roasting darker than medium with more optimism!

#20: Post by **SquidMan** » April 24th, 2023, 12:12 pm

Since I had some of the same Klatch beans around I gave it a go with all the same parameters using curve 7,9 and 65g batch.

Couldn't hear first crack for any of the roasts, maybe a pop or two 35-40 seconds after estimated FC.
Accidentally knocked the usb cable off first roast which interrupted the logging so I did a back to back using curve 9

Visually I can't really tell the difference, on closer inspection at OCD levels I can see a tab bit more occurrences of tipping and minor defects on the Curve 7 batch.