Buyers Remorse - Kaleido M10 Astringent Roasts - Page 2

Discuss roast levels and profiles for espresso, equipment for roasting coffee.
User avatar
yakster
Supporter ♡
Posts: 7344
Joined: 15 years ago

#11: Post by yakster »

Your charge temps seem awfully low, I agree with Gary that you may be under developing the inside of the bean because you compensate with a low charge temp by applying a lot of heat to get the bean to first crack. Does the roast color of the inside of the bean (or ground) seem a lot lighter than the outside?

I would recommend a higher charge temp and you may even consider a soak for naturals to help allow the inside of the bean to catch up to the outside.

Apples and Oranges, I know, but for my Bullet I typically see C1 (First Crack) at 211 C and I charge at the following temps. As I roasted along and got more experience, I raised my charge temps. My roast typically progresses by gradually decreasing the power as I approach first crack to get the momentum that I prefer.
285 C for 700 grams
280 C for 600 grams
270 C for 500 grams
265 C for 456 grams (1 #)

The Liberica Batangas coffee I picked up from the Philippians was an exception, my roasts would run away from me if I charged as high as I normally do, I found that charging at 255 C for 500 grams worked well.
-Chris

LMWDP # 272

GDM528
Posts: 852
Joined: 2 years ago

#12: Post by GDM528 »

yakster wrote: a soak for naturals
Just for natural processed greens, and not for washed? And to clarify, the soak you refer to is at the beginning of the roast?

User avatar
yakster
Supporter ♡
Posts: 7344
Joined: 15 years ago

#13: Post by yakster »

I would read up on the soak more from others, I rarely if ever use this technique, but I believe I've read it's used more with naturals. I'd try a higher charge by itself first, then consider trying adding a soak.
-Chris

LMWDP # 272

Rytopa
Posts: 228
Joined: 7 years ago

#14: Post by Rytopa »

A few tips for the Kaleido,

Charge higher, use lower % power thru out..i am generally using 70% - 80% starting heat, Highest AIrflow constant throughout. .. I am using 2-3% power cuts after 160c for every 5c, very small adjustments until 180c, @ 180C 5% power cut very 5c, at around 200c depending on the ROR , u may need to do 8% power cuts.

IR roasters like the Kaleido are quite different compared with normal roasters, they tip easily, leading to bitterness and astrigency, so the trick is to use higher charge, but lower peak % to get the roast going because we are using lower % to begin with, power adjustment cannot be drastic, but very subtle, slowly lowering the power thruout the roast, at around 180c is when bigger cuts happen.

User avatar
Ypuh
Posts: 312
Joined: 3 years ago

#15: Post by Ypuh »

I'm no expert by any means, but I got the Kaleido M10 for about 2 years now and a little over 100~ roasts. Not every roast is perfect, but despite some tests, I never binned any beans and drank or shared them all. There's only a few that didn't come out at my taste, the Monsooned Malabar being one of them (I guess they require a different approach to roasting that I haven't managed to figure out yet).

Here are some (randomly selected) of my roast profiles.

1. Brazil Sorocobana - 450-1200m elv. - Natural


2. Indonesia Java Sunda Tilu - 1400-1700m - Washed


3. Indonesia Mandheling Sumatra - 750-1500m - Washed


Depending on the hardness of a bean and batch size (usually around 650-850g) I might play with lower power at the start. Usually anywhere between 70-100% and tapering down based on smell/bean color.

4. Colombia Supremo - 1500-1650m - Washed
I don't want a Decent

User avatar
drgary
Team HB
Posts: 14392
Joined: 14 years ago

#16: Post by drgary »

Looking at the four profiles above, you're coming into first crack with too much momentum. This creates a plateau just before first crack and a crash that can lose sweetness. Where there's a flick after the crash you may find some roasty or baked flavors.

My roaster is different, but I corrected the vulnerability to that plateau by charging a bit hotter and backing off the heat earlier in the roast, well ahead of first crack.
Gary
LMWDP#308

What I WOULD do for a good cup of coffee!

Agtron70 (original poster)
Posts: 15
Joined: 2 months ago

#17: Post by Agtron70 (original poster) »

M10 Astringency Update:

Be aware I am a green novice. Please interpret my experiences with that filter. Also the typos make it real.

Thank you all for your insightful and constructive advice.

I have taken action based on your comments.

I have experimented with increased charge temperature, gradually creeping it up over many more roasts - unfortunately roasts headed for the bin.

I started with a charge temp (based on my BT reading) of 160C. Over trials, I increased the charge temperature to 209/210C. Please keep in mind that I believe my BT is producing readings are that I believe are 20C too low. I believe the BT sensor is 20C too low because, for a bean well known to me, I observe a Kaleido M10 FC at 178C for that bean vs an observed FC of 198C on my pair of Kaffelogic Nano 7 roasters. Therefore I believe my displayed charge temperature which is based on my BT temperature of 160C is actually 180C. Similarly when the charge temp displays 210C, it is actually 230C.

I am aware there is a symbolic temperature adjustment that can be made in Artisan under "CONFIG=>Device, symb ET/BT tab", however for my purposes it adds more confusion when PID roasting, which is the subject of a different post.

So for any BT, SV, Charge temperature I assume they are displayed 20C too low. It regrettably adds an opportunity for confusion and data misinterpretation in a situation which requires data interpretation and analysis for remediation. I have sent a note to the Kaleido manufacturer, the entity I purchased the unit from, re: BT temp inaccuracy display. I have not gotten a response at this point.

So to recap the trials, the charge temp experimentation has ranged from 160/180C (displayed/actual) to 210-ish/230-ishC (displayed/actual).

I am PID roasting. I create a roast profile in Artisan designer, I load that profile into the Artisan background, I heat the roaster, I begin the roast by dropping the beans from the hopper into the drum. This change in temp of the BT is correctly interpreted as the start of the roast by Artisan/Kaleido and the roast is temperature adjusted automatically for the entirety of the roast and I observe the roast temp/time graph in Artisan as the roast progresses.

I'd give that experience a "B +/-" grade. Factoring in the cost for Artisan - ZERO - I'd give it an "A" grade - it delivers way more than the price I paid. Overall, in absolute terms it's much more of a secondary school science experiment than I would like. It can be a struggle to get Artisan Designer to generate the desired roast curve. I appreciate Apple - lick-your-palm-friendly - and this isn't it. Fundamentally roasters are stoves for coffee beans - I don't sweat heating a pizza in my oven - I'd like roasting great coffee to be the same level of stress. As an aside, considering that Artisan is the roast control system, I think it is amazing that roaster manufacturers don't invest more in the development of Artisan to bring to a more friendly and capable state. Artisan is currently amazing freeware that needs to be next level and probably fee based. Reminder - I know nothing.

Back to business. During the roasting process, the PID control liberally juices the heat up to 100% to get and keep the roast on track with the roast profile - that maybe the nature of PID control - I am not a PID guy. It defiantly does a pretty good job of keeping the roast in line with the background profile after a startup period. As stated before the heat is redlined for some time, and the delta BT/RoR isn't always pretty.

Here a roast that produces great cups from the Kaffelogic Nano 7. What a wonderful roaster - it is a very Apple-ish in end user experience - it works - no science experiment -and I don't sweat the experience - I just wish it had a bigger capacity - If It had more capacity I would never have wandered into Artisan/Kaleidoland.:


After many failures this is the best PID roast I have been able to produce on the Kaleido M10:



From the taste of the two roasts - The Kaffelogic Nano 7 roast and the Kaleido M10 roast - , it is apparent that the roasts use the same bean. The difference is the Kaffelogic roast is clean and bright, while the Kaleido has a "darker more done" flavor (you're not dealing with a "Q Certified" taster here) and a subtile but very noticeable tongue drying astringent after taste. Still an unsatisfying experience. Even more so considering the effort and expense.

Interestingly the following M10 roast, similar to the one above with a 210C-ish charge, but for a longer roast period, produced unbearable astringency. Same charge temp - longer roast - much worse result.:


There was an excellent observation made by Rytopa where Rytopa comments that the infrared heating of the Kaleido design results in different experience/result - paraphrasing here - and has a tendency for tipping and astringency creation.

I think that may have something to do with root cause of the issue (that and my inexperience).

Further on the design of the Kaleido as a factor in astringency creation - is that Kaleido changed the design of the roaster. I have a new design - a Gen 2. What changed? It was actually a pretty big change. Kaleido made the new design more "sealed" and more efficient. Still using infrared heat generation from Gen 1 to Gen 2, Gen 2 just has less waste, which I assume results in a different experience for the beans under heat. In a Gen 1 Kaleido M10 roaster, the little side wings can be flipped up, and you can look at the infrared heating tubes - they are kind of in the open. In a Gen 2 the infrared tubes are not visible because they are contained in the sealed system. For identification purposes, visually, the Gen 2 has a little three lobed twist knob to drop the beans into the hopper, where as the Gen 1 has a "bent lever" that moves vertically.

In my research leading up to my purchase decision I considered the deign change. There are quite a few vids on the old design - very few if any on the new design. I may have rolled the dice and lost on this one. I thought more efficient would be better. Maybe it's just an astringency factory instead.

So the experiment continues. But its not going to continue for long.

I have burned through in bad roasts - no pun intended - about shy of 1/10 the cost of the Kaleido M10 roaster. Its getting old and at the risk of sounding like a broken record - expensive.

I look froward to reading your comments, but I only plan to post an update if I can master this.

Thanks again for your advice and comments.

Agtron70 (original poster)
Posts: 15
Joined: 2 months ago

#18: Post by Agtron70 (original poster) »

Actually heres the correct Kaffelogic Roast for comparison purposes:

PaulTheRoaster
Posts: 263
Joined: 18 years ago

#19: Post by PaulTheRoaster »

Back to business. During the roasting process, the PID control liberally juices the heat up to 100% to get and keep the roast on track with the roast profile - that maybe the nature of PID control - I am not a PID guy. It defiantly does a pretty good job of keeping the roast in line with the background profile after a startup period. As stated before the heat is redlined for some time, and the delta BT/RoR isn't always pretty.
Do you need to use PID control? Most drum roasters use power instead (and wouldn't oscillate from 100% to 30% and back). That could possibly be the issue.

User avatar
drgary
Team HB
Posts: 14392
Joined: 14 years ago

#20: Post by drgary »

Without time at the moment to consider driving the roaster with Artisan and a PID (have you thought of reaching out to Kaleido and showing them your curves? They may suggest a change to PID settings), I generally charge my propane North Roaster with an environmental temp of 215C. This after at least preheating the roaster for 20 minutes. The greens measured with the bean temp probe start a lot lower, but I'm paying less attention to that number (for my roaster, which has lots of metal mass. It weighs about 250 lbs and has a 1 Kg capacity. Upon charge, I'm also heating the initial drying stage pretty hard. By charging this hot and adding lots of heat early but short of tipping or scorching, I have enough energy in the beans to be backing off of heat from there.

Here's a nice, recent profile, where I'm running air at 100% to distribute heat pretty evenly through the perforated roasting drum. The roast in this profile came out sweet and juicy. It's a Yemen and all of those coffees are dry processed. The stepped reading at the bottom shows me starting with high heat and lowering it throughout the roast without changing the air setting. This works well for my roaster, which is very different from yours. By comparison, your roast looks pretty rough, especially with the dip in bean temperature rate of rise early on. Uneven ROR like that will tend to develop the coffee unevenly.



A fluid bed roaster like a Kaffelogic or my IKAWA Home drives a lot more air through the beans than on my drum roaster.
Gary
LMWDP#308

What I WOULD do for a good cup of coffee!