D-Roast

[mr. bean]

Steve and Damien, could you guys comment a bit more on why you've gone for higher fan speeds in your profiles?
It would be helpful for me (and maybe others) as we consider messing around with your profiles, lowering fan speeds especially towards the end as the heat climbs higher and the power required to deliver high temp air and high speeds gets too much for some.

My understanding on fan speed so far (or at least what I think I know!):
- higher fan speed right at the very start will help get denser beans moving and air flowing evenly through them all
- higher fan speed can act like higher heat in terms of getting more energy into the beans, as long as you can heat the air fast enough so you're not just blowing faster colder air on (I don't know if this is true throughout the whole roast, or more so at the start)
- higher fan speeds later in the roast will result in less smokey flavours by clearing out air the beans are roasting in
- higher air speeds will result in lower actual bean temps for any given temp reading on the KL due to increased influence of air temp Vs bean temp on the probe (so modifications to airspeed will affect things like expected first crack etc...)
- lower air speeds will taste more like a drum roaster, and higher air speeds will taste more like a commercial airbed roaster

[Damian]

https://kaffelogic.com/community/viewt ... ?f=6&t=196

[fnq]

That is a good link , thanks Damien i hadn't seen that thread before.

I have tried D roast, Darkside and a merged D roast with darkside fan features, but at approx 1320-1330 watts i flat line and my roasts drop away from the profile from about the 8 minute mark. I will try a different power line next and decrease size from 100 to 90 gms in.
I mostly roast medium to light medium (3.3 -2.8) after not being able to hit an earlier 4.6 roast effort,, ( but i hadn't truly understood the fan impact- so more things to try now.

For those others power challenged, i have found Classic, RTD, GTTT and ninja low power roast profiles to work for me.

If you (Damien) or others have cupping breakthroughs on lower roast size/ lower power, could you point me to those profiles.

Thanks again:- my appreciation to you clever folks for sharing your work. Cheers Darryl

[mr. bean]

https://kaffelogic.com/community/viewt ... ?f=6&t=196

Thanks Damien! Hadn't seen this. Keen to experiment more with these higher fan speeds now.

I'll put further comments / questions on that thread...

[Geronimo]

Hi
Re: Damians D-Roast with modified fan

Help required please, on what could be going on at the beginning of roast. Comparing to others that have had a crack at running Damian's profile, my roaster doesn't appear to be running on all cylinders as it doesnt take the steep road, it is more of a gradual gradient at the start.

log0135.png (47.01 KiB) Viewed 9188 times

[Steve]

Hi
Re: Damians D-Roast with modified fan

Help required please, on what could be going on at the beginning of roast. Comparing to others that have had a crack at running Damian's profile, my roaster doesn't appear to be running on all cylinders as it doesnt take the steep road, it is more of a gradual gradient at the start.

log0135.png

Im assuming you mean right at the very start where it falls under profile straight away. In which case try increase the "pre heat power" by 50W increments until it starts above profile.

Even so I think it looks fine without adding more power.

[Steve]

Steve and Damien, could you guys comment a bit more on why you've gone for higher fan speeds in your profiles?
It would be helpful for me (and maybe others) as we consider messing around with your profiles, lowering fan speeds especially towards the end as the heat climbs higher and the power required to deliver high temp air and high speeds gets too much for some.

My understanding on fan speed so far (or at least what I think I know!):
- higher fan speed right at the very start will help get denser beans moving and air flowing evenly through them all
- higher fan speed can act like higher heat in terms of getting more energy into the beans, as long as you can heat the air fast enough so you're not just blowing faster colder air on (I don't know if this is true throughout the whole roast, or more so at the start)
- higher fan speeds later in the roast will result in less smokey flavours by clearing out air the beans are roasting in
- higher air speeds will result in lower actual bean temps for any given temp reading on the KL due to increased influence of air temp Vs bean temp on the probe (so modifications to airspeed will affect things like expected first crack etc...)
- lower air speeds will taste more like a drum roaster, and higher air speeds will taste more like a commercial airbed roaster

Yes the higher fan speed at the start is to facilitate good agitation of the mass. Without this some beans are being exposed to much higher temperatures than the rest. Some will get damaged that you can see later, others it can be invisible but present as flavour defects, hence roast defects.

Most Arabica unless its island coffee can take a fair wack of heat at the start as long there is even agitation of the mass or pile. So the higher air flow to a certain point will allow the beans to better utilise the heater power. From what ive observed it would be pretty hard to use too much air flow at the start on the KL. Getting as much heat as possible in the first 2 to 3 mins seems to be the key as long as there is good bubbling fluidisation from start.

I dont think lower fan speeds contribute to smokey flavours on their own, as in smoke hangs around and absorbs into the beans, even on a drum roaster with nearly no air flow thats pretty hard to do. My theory is that most of the off flavours created are from baking, unless one is roasting well into rolling 2nd crack. It is possible to roast up to 2nd crack and have no defected flavours, just darker developed caramels.

Yes the different fan speeds can have minor impacts on fIrst crack temps, something like 2 to 4C which is also within the range of variance found in Arabica from different parts of the world and human perception of when first crack actually starts...so not a problem.

This is a bit of an internet myth that drum roasters taste a certain way and vice versa. It depends on the human operator and how well they know the machine. More often than not many commercial drum roasts are baked and or underdeveloped or charred, ive had fluid bed roasts just as bad or worse. Ive also had some really clean, sweet vibrant drum roasts not all necessarily light roasts either, same with air roasts.

Give my modified NinjaTurtle profiles a try, have a look at the slight changes I made in the fan profiles for the Med - Dark.

https://kaffelogic.com/community/viewto ... ?f=2&t=198

[nrdlnd]

Give my modified NinjaTurtle profiles a try, have a look at the slight changes I made in the fan profiles for the Med - Dark.

Yes I've tried both the original NinjaTurtle and Steves modified Med-Dark with an Indonesian Sumatra Lintong. The original was with L 2.8. With the original I could clearly hear the 1C. I stopped the roast at DTR 25%. The bean looked fine but in my notes I wanted to try L 2.7 next time. I don't remember that the beans had tipping or scorching defects. I have a disclaimer here that I don't remember if I checked the movement of the beans in the beginning of the roast

With Steves modified profile I couldn't hear 1C (I have difficulties as I'm hearing impaired) so I decided to stop the roast around 230 deg C. L became 3.2 one decimal below Steves lowest recommendation. The beans looks well developed and about as dark as I remember they were with the original. I can't compare the batches as I have nothing left of the original batch.

The preheat setting of the original profile is 1100W and the modified 1160. The fan speeds are 1470 and 1575 RPM respectively in the beginning. What is interesting is that the beans heat faster with the original profile. With the modified profile it is a more gradual rise. After 2 min the temperatures are about the same but after 1 min it is 5 deg C higher with the original profile.

Does this matter and what is "better"? I don't really know as I don't have much experience yet. Scott Rao writes in his book about the importance of a high heat in the beginning of the roast and to create a high Delta T (difference) between the outer and the inner temperature of the bean in the beginning as it affects the development later in the roast. A later try to "catch up" by prolonging the roast may lead to baking for example. If Scott Rao is correct then the original profile is better in this respect. I have a disclaimer that it's two weeks between these roastings so there may be other factors affecting the result and it's only two single roasts.

I have another observation and this is also only a comparison between two single roasts but this time with the same profile but different beans with different sizes and they are processed differently. Maybe the best roast I've made for espresso is with a Brazilian Yellow Bourbon natural processed and roasted with Steves DarkSide. It takes heat faster than the modified Ninja Turtle but not as fast as the original Ninja Turtle with the Lintong. After 2 min the roast follows the profile to the end. L was 4.6 and the roast ended at about 10 min. I calculate DTR to about 21% from the ROR.

The other roast is with the same profile but with quite another bean. It's a bigger bean and it's wet processed. It's an Indian Kalledavarapura. After 1 min the temperature is 5 deg lower than with the Brazil and at 2 min 2 deg lower. What's happening here is that at the last 1½ min of the roast doesn't the temperature follow the profile! There doesn't seem to be enough heat available from the roaster. What I wonder here is if the beans had got more heat in the beginning of the roast had this happened then? L was 4.4 and DTR about 24%. A higher preheat, a lower fanspeed in the beginning (but there must be o good movement of the beans) or a smaller batch size could maybe compensate for this?

I had a similar problem when I tried the D-roast with an Indonesian Tana Toraja. I do usually have a batch size of 100g. In this case I had 110g. In this case the temperatures are really low during the first two minutes and the temperature doesn't follow the profile at the end of the roast! There may be more than one reason for this but I think the low preheat may be one reason. I think fan speed in the beginning of a roast must be optimized. May a too high fan speed in the beginning cool the beans instead of heating them?

I did also modify the D-Roast with the fan profile from Darkside with another bean and it's the same that the temperature in the beginning of the roast is low because of the high fan speed in the beginning.

[Geronimo]

Hi
Re: Damians D-Roast with modified fan

Help required please, on what could be going on at the beginning of roast. Comparing to others that have had a crack at running Damian's profile, my roaster doesn't appear to be running on all cylinders as it doesnt take the steep road, it is more of a gradual gradient at the start.

log0135.png

Im assuming you mean right at the very start where it falls under profile straight away. In which case try increase the "pre heat power" by 50W increments until it starts above profile.

Even so I think it looks fine without adding more power.

Yes Steve, the region as you mentioned. Thanks for advice.

As an aside and a good thing, I have not yet had an obvious crash/flick around first crack. Early days yet, but very promising indeed.

[nrdlnd]

Here come a couple of pictures of logs to illustrate what I wrote.
This is a roast of a 100g batch Sumatra Lintong with the original NinjaTurtle. DTR 25%. At 1 min the temp=122.4 deg. After 2 min it is 127.7 deg.

Sumatra_Lintong_200701.png (59.77 KiB) Viewed 9151 times

This roast is not quite comparable as it's 110g and another bean an Indonesian Tana Toraja. It's made with D-roast. DTR 24%. At 1 min the temp is 106.7 deg and after 2 min 113.9 deg. Much lower temp than the previous NT roast the first 2 min. It doesn't follow theprofile after 8 min.

200710TanaToraja_D-roast.png (62.99 KiB) Viewed 9151 times

I do also have a picture with a 100g Sumatra Lintong but with Steves modifed Ninja Turtle M0dTurtleMD. After 1 min the temp is 117.2 and after 2 min 127.9. It's 5 deg lower after 1 min compared to the original NinjaTurtle. I couldn't hear 1C.

200716_Sumatra_Lintong.png (59.02 KiB) Viewed 9151 times

I have a picture of D-Roast modified with Steves DarkSide fan profile and with 100g Mexico Sierra Azul. DTR 23%. It's about the same with rather low temperatures the first two minutes. 1 min 106.6 and 2 min 114.1. This time it does follow the profile. If I try this profile again I will raise the preheat or maybe lower the initial fan speed.

200716_Mexico_Sierra_Azul.png (61.33 KiB) Viewed 9150 times