Finally! Completed my first roast tonight! Critiques?

[Joey]

My Huky arrived last week, I went out of town for a few days, got back, and had to wait 4 more days for my Phidget to show up! It arrived tonight and I couldn't wait any longer to do my first roast.

Luckily the Phidget install was super smooth, no hiccups whatsoever. I used Kennerroast's general methodology for my roast and felt pretty comfortable controlling the system after my first couple attempts. This profile was my 4th batch. It was a DP Ethiopia Harrar.

Wt: 235g
14% loss

1. Obviously my ROR curve isn't ideal, I'm curious why there is a spike in ROR at the end of the drying phase? Is that normal? Would it be a good idea to increase fan speed for about 30 seconds or so when drying ends?

2. My MET seems really high. I still don't really understand MET.

3. Otherwise I'm not really too sure what to look for. I'm open to any suggestions you guys have. I know I have a lot of room to improve. My goal is to really bring out the fruit notes & acidity.

Can't wait to taste this batch tomorrow morning!

Thanks everyone!

[edtbjon]

Well, that seems like a pretty good start.
First, the roast is very short, which really starts with you charging too high. That in turn really speeds up the roast and it's all a matter of braking from there until the end of the roast. With that charge size I'd go for 350F or so on a perforated drum. (A lower charge and consequently a longer dry phase will do a dry processed bean a lot of good too, as the roast will become more "even". This is very noticable on an Ethiopian.)
Second, the charge size. I really recommend you to go towards 400g to 454g charges while learning the roaster. It's easier to learn the roaster and you don't have to "brake" as much as with a small charge. (Yes, I did the same "mistake" when I started up with the Huky. "Smaller charges means many more roasts...") Once you get the hang of roasting a full pound it's easier to downscale.
The bump shortly after DE... I cannot see anything in the graph which explains that bump nor the dip which comes with it. I get/got this bump quite often when adding a bit of fan at DE, as the heater becomes more effective with a bit of airflow. Unless you have that bump in the other roasts too, it may be a faulty reading. As you didn't really do anything with the heat or air, there is no way it should show (taste) in the roast.
MET, who really understands MET??? The MET usually goes really high on the perforated drum, especially while there's some heat on. I've logged the MET values for almost two years now and still finds very little use for paying attention to it.
Short roasts usually promotes acidity, but you may find this roast a bit too pronounced on the acid side. But the first thing for you is to learn the basics with the roaster. As you seem to have managed to sort out Artisan and logging both heat and fan while roasting I recon that you're up to a really good start. Keep up the good work.

[Joey]

Thank you very much for the feedback, extremely helpful!

I was thinking a lot of the things that you said so it's reassuring to hear them. I'll apply your recommendations on my next roast (hopefully today) and report back!

[.]

.

[Joey]

Hi guys, roasted 3 more batches of the same Ethiopia Harrar. I lowered my charge temp to 380ish (still using a 235g batch) and consistently hit the TP around 210-212 and 1:40 (from what I've read this is right on target?).

I am much happier with the overall roast except I'm still really puzzled by this spike in ET right at the end of drying. It's causing my ROR to spike too. Any thought's on why I'm getting these spikes around DE and what I can do to smooth it out?

Also, I may have let off the gas too early in #2 and #3 as I never really got a boisterous 1C. I got some crack but then my BT leveled out and I decided to drop them rather than baking them.

Any pointers or thoughts appreciated!!!

[edtbjon]

Again, that bump in the DeltaBT/RoR is most probably due to you giving the heater more air, making it more efficient. It takes a little while for the fan to spin up and there is a small difference in when you mark the event in Artisan compared to the "real" event (of turning up the fan in this case).
What to do about it? Well, for now just ignore it. It does no real harm to your roast and the TC is more sensitive to heat changes than the beans... The important thing is to look at the "trend", i.e what is happening to the RoR curve over a bit longer period. Later, you can compensate for this by lowering the heat some 10-20 seconds before rising the fan.
And yes, you did let the roast bake a bit on the end of the two last roasts. That's part of learning and I think that it's important to make (and taste) these mistakes too as part of the learning process.

[Joey]

Thanks so much for the advice! Roasted tonight and kept the fan on 25V, no spike! Totally makes sense that the increased airflow was feeding my gas and raising the temp. Can't wait to try this one in a couple days!!

[edtbjon]

Good! I just want to nag on you again about the very end of the roasts. I found (with both a solid and perforated drum) that turning the heat completely off isn't a good idea, IMO. It seems like you get a negative RoR in the very end of the roasts, which can result in baked notes.

[Joey]

Good! I just want to nag on you again about the very end of the roasts. I found (with both a solid and perforated drum) that turning the heat completely off isn't a good idea, IMO. It seems like you get a negative RoR in the very end of the roasts, which can result in baked notes.

Not nagging at all!! I want as much feedback as possible! In my last roast I cut the gas for about 30 seconds, then turned it back on low to help smooth things out, does this look better?

[edtbjon]

That last profile looks really good. Again, the true and really the only valid test is how it tastes. But it seems like you are getting there in a very good pace.
As you already know I use a somewhat different approach to my current roasting, charging much lower than you (and some others) do. But as your roasts are a bit shorter, you seem to get much of what I'm looking for in my approach, i.e a roasts where the beans gets lots of energy during most of the roast. Without looking at your control pattern (which is recorded and visible in your graph, but I cannot see it while writing this answer...), it's visible through the DeltaBT curve, which is quite high during most of the roast (and still decending according to Mr Rao...).
A small detail on Artisan. I usually after having pressed the "OFF" button, while cooling the beans etc, go into "Configuration/Axes..." and change the "Max" value (top right) to my drop time + one minute. That will cut the blank space on the right of your graph.
Great photos of your very nice cart setup BTW...

[Joey]

Alright, so I'm about 20 roasts in. I decided to up my charge weight from 230 to 360g about 5 roasts ago. I also recently moved from 1 large fan adjustment to several gradual adjustments over the course of the roast.

Here is my latest profile, always up for any thoughts or suggestions on how I can improve! Thanks!!

[korkiley]

Congratulations! It's funny, coincidentally I recently had a shot of that same Hambela from a new cafe that opened up in my town. It was a much too light roast for my taste, for espresso, but it had an incredibly strong blueberry flavor. Where did you buy this bean?

In 2014 I bought 5 lbs of an Ethiopia Yirga Cheffe Aricha DP that also had intense blueberry notes. That was also one of the best beans I've ever roasted! The DPs in general are my favorites. I bought another bean of the same name from Sweet Maria but it was nothing like the one in 2014 but there was a DP Ethiopia Yirga Cheffe Gedeb Asasa in the same shipment which was excellent.

I hope you manage to find those blueberry notes in your Hambela!

[edtbjon]

That last profile is most aspects similar to what I usually strive for lately. At the very least it's a very good "base profile".
Regardless of how your cupping turns out, you now have a very good base point from where to start to tweak the profile to your liking. It's easy to shorten or prolong either the mid (Maillard) phase or the development phase (or both). I take it that you start to feel reasonably confident in being able to replicate this roast too.
Two tools comes to mind. First, there's the taste wheel from SCAA. Second, the book(let) from Rob Hoos on "Modulating the flavor...".
Last, while you analyze this and other recent graphs of yours, I guess that you can see a pattern on how you control the gas and air respectively. A key to repeatability (and thus also tweakability) is to reduce the number of changes you make. I.e it's much more easy to repeat 2-3 changes of airflow than a pattern of 7 changes... and the same goes for gas.
(Just for a bit of perspective... I had a close look at a couple of very recent Diedrich roasters lately, an IR12 and an IR1 respectively. While they (of course) do have needle valve (i.e infinite) control of the gas, the fan on both the roasters have one speed (!) and is used for either cooling the beans, a 50/50 between cooler and drum or full flow through the drum. Many of the high end professional roaster machines only have a single fan or possibly single speed fans, utilizing dampers to control airflow. Most of us in the Huky community seem to have some kind of Variac to control the exhaust fan and does have a second dedicated bean cooling fan (at least I do), which puts us very much in control, but it's also very easy to develop habits of "over-control". )

[goodolarchie]

Where did you buy this bean?

Not OP but I bought my Hambela from Happy Cup, and definitely got strong blueberry scone out of it.  They have long run out of it, but it's in season from July until they run out.

[goodolarchie]

Where did you buy this bean?

 Happy Cup

Err https://www.happymugcoffee.com rather.  Sorry not sure how to edit my post.