Nailed some cracking roasts this week if I do say so myself!
PNG Waghi, gentle profile dropped just before second crack at 222° – nice mild acidity, sweet toffee flavours as espresso, and as usual, amazing in milk.
Also a Harrar Longberry, much faster ramp early on to around 175° when I tried something new and dropped back the ramp well before first crack so it doesn't go in too hot and run away. Then I dropped it early at 216°, at just the right time for a Longberry Spro. Great zing and sweetness, lots of fruit ... very nice
Roasted this batch a few days ago and started drinking today...
Turned out to be a really scrumptious combo, lots of fruity acidity, plenty of body and a lovely, lingering dark chocolate finish. Impossible to stop at only one, or two.
Used a slightly slower profile and pulled right on 221C to try and balance the intrinsic qualities of the beans with roast flavours and body improvement. Seems to have worked a treat.
As always, copy of blend details below, with roast Profile and Post-Roast photos attached.
Ethiopia Ghimbi... 350g
PNG Mt.Ambra 'A'... 250g
Indo. Sumatra 'C'... 150g
Roasted Weight... 644g
Moisture Loss... 14.13%
Very even looking roast, especially considering the wide variety of beans.
Was the drop in temp. at around 9 mins just to slow the roast down?
What effect do you reckon it had on the taste?
The slower profiles tend to produce roast batches that appear very even in colour, not the main aim though.
The drop in the profile gradient is something that Matt (DesignByCoffee) highlighted several years ago as a means of extending the Maillard Reaction temperature range. I think he referred to it as the "Seattle Dip". In my case, I do this between ~135-165C regardless of the profile gradient so it can appear at various points along the 'x' axis depending on the profile being used.
Have found that, for me, it enhances body, the soft spiciness factors such as cinnamon, cardamom and even a bit of nutmeg sometimes, and caramel sweetness seems to be slightly more pronounced. There's a lot written about the Maillard Reaction in coffee roasting circles and quite a bit of it seems to be contradictory so, if you would like to see how it works with your particular setup, you'll just have to experiment and do lots of cupping. Don't know of any easy way to approach it...
Those roasts both look good. I might need to revisit longer roasts, at present i'm generally hitting FC around 8-10 minutes.
Purchased some small 1L food grade buckets with lids, am going to age beans in these before consumption and see how the beans go when completely sealed (rather than in a 1 way valve bag - pretty sure the budget bags i buy aren't strictly 1 way).
Have had 3 days detox (normally drinking 5-6 a day), the three day headache hasn't been much fun, sore kidneys yesterday (probably unrelated).. looking forward to a coffee tomorrow morning.
Answers in the order that questions asked...
1. Yes, I continually monitor the Rate of Rise via CS Roast Monitor and adjust the HG output to follow the profile I want to use.
2. Actually, the temperature only got as high as 221.1Deg.C. Each temperature grid line equals 20.0Deg.C and yes, it is a Corretto but as with all of CS Correttos, each and every one is going to be a little different than the other and to get the best out of the bean, will probably require experimentation to find out what that is.
My Corretto for example, is constructed around a Breville Big Loaf BM, a Makita Tradie type HG with the Bread Pan insulated using a Fibreglass Fire Blanket, and a removable lid that slides over the Bread Pan constructed from 6mm fibreboard wall cladding. There are many variations of this kind of setup being used and in some cases, the designs go much further than mine. The Big Loaf BM uses a horizontal Bread Pan with two Kneading Paddles so bean agitation has never been a problem. Some of the Vertically oriented Bread Pans do suffer from agitation issues and extending the height of the Paddle(s) overcomes this in most cases.
The Bread Pan orientation also effects the minimum batch load quite a bit compared to a Vertical one as the beans just scoot around the bottom of the pan in a very random fashion that quite often results in roast batches that end up being roasted very unevenly. In the case of my Big Loaf, this works out to be a minimum batch size of ~350g and could be a consideration if you need to roast batches smaller than this. Hope some of this is helpful mate...
I am going to try that blend next time I buy some beans.
My bad eyesight earlier, on your roast monitor image I read 130.9C as 230.9C, hence my dumb question.
Though I might add a couple of thoughts to Mal's great advice.
1) Insulating the pan and a lid — as Mal mentioned — makes a huge difference to how smooth your profile will be. These things will allow you to drop your gun input temperatures, and the setup will be less susceptible to gusts of wind and the like knocking your profile around.
2) If you set your gun on one set temperature at the beginning of your roast and just leave it until first crack, you will get a fast start and decreasing curve in your profile. To get a flatter profile, you will need to start gently and 'ramp' your gun temperature up through the roast. If you look at any of my profiles (see yesterday's Yirg roast as an example) I have logged the temperature I set my Bosch 630 gun and when I've made the changes (see the red line and numbers). Basically I increase the gun by 40° at 75°, 100° 125°, 150° (but play around here for the Seattle Dip ) then 175°, then slowly back off again by 30° every 5° from first crack to drop point.
FWIW this ramping method in the corretto gives a better result in the cup too, as it reduces to chance of scorching or tipping and softens the acidity — but that is just my personal preference
Might be worth a go
Sounds like I have a similar setup - Bosch heat gun, 2 x small fire blankets wrapped around tin, fibreboard lid with hole for gun nozzle to stick through.
For 500g beans, I had been leaving it set on medium airflow, 550°C until beans at 185°C and then turning down to 450°C.
Very uniform roasts taking about 14 mins, but has a decreasing curve profile as you noted. Am interested to see if I can taste the difference with a flatter profile.
Will try this weekend.
I'm normally doing 800g batches now, so use Fan 3 to get good heat throughout the bean mass (I use the same settings but fan 2 for a 350g roast though).
But during the 'Seattle dip' component I actually increase the gun temp by 70° (you'll see that on the log) – but then drop back to Fan 2, which has the effect of dropping the RoR to around 5-6° per minute. I just found that dropping the gun temp led to too much lag in coming back to temperature. This way the element is still nice and hot – I just drop it back 30° and hit fan 3 again and off we go!
I saw the Yemen Ismaili appear on BeanBay but don't remember a note about it from Andy's 2 month programme. Given the implications of the price tag, the description and numerous threads I had to try it.
The beans were very small and looked brown and a bit wet, heavy in the hand, as if the drying hadn't quite completed. And indeed it took longer than usual for the first chaff to start floating off the popper. That first phase had a very strong chocolate flavour (a wet spicy one like Mexican origin) which morphed into tomato for some reason.
My first roast was very dark because these beans have a very quiet, almost SC-like FC. I was more careful with the second roast but it still came out darker than I like; I heard the start of SC when I switched off the popper.
Can't wait to try this in a few days.
oh bugger i would have loved a bag of the Yemen..
Me too Janus, me too! Although I just saw it's still available in the roasted beans in Beanbay, I didn't realise there was any left!
Some Perú CENFROCAFE in the Ambex at work this morning.
Todays roasting was;
300gm Bolivia, Caranavi Organic Fairtrade as a single origin
300gm Uganda, Kisora AA as a sigle origin
250gm Yemen, Bani Ismaili as a single origin
100gm Yemen, Bani Ismaili
100gm Sumatra, Blue Lintong
100gm Sumatra, Mandheling Mystic TP
As a Mocha Java blend.
Time for some more Ethiopian in a blend...
Roasted this batch with a more traditional profile this time around, taking some pointers from roasters who have a different slant on optimising the Maillard Reaction zone and Caramelisation. Basically followed an ever decreasing gradient up to about 150C and then increased the gradient slightly by another 1.5C/Minute up to 1st-Crack after which it was allowed to collapse to about 4-6.0C/Minute for about 4 minutes until the first few snaps of 2nd-Crack could be heard at 223C, after which it was immediately dumped and cooled. Fantastic aromas coming off the cooler...
Haven't copied the profile this time as there was a hell of a lot of EMR interference being picked up by the unshielded t/c cable which made the profile extremely difficult to make sense of. I think one of our neighbours has got a new "noisy" toy so might have to consider buying a couple of shielded t/couples.
Anyway, post-roast photos are attached and the blend details appear below.
India El.Hills 'AA'... 200g
Ethiopia Biftu Gesha Sundried... 350g
Sulawesi Blue... 200g
Roasted Weight... 628g
Moisture Loss... 16.27%
So less ramping early on then an increase after 150?
I’ll look forward to hearing your thoughts on the flavour changes
Couldn't help myself...
Had to crack a bag open as the wonderful aromas escaping the 1-Way Valve are just too enticing. Loads of blueberry, rich dark chocolate and an absolutely intense pure coffee aroma carrying everything along with it.
In a few words, despite being taken to the brink of 2nd-Crack, the overall driving flavour is a rich, fruity and acidic sweetness. This is wonderfully balanced though by huge body, soft spices and loads of dark chocolate. It also has a very extended finish that seems to meld from an African cocoa to a sweet, home made caramel. Never had that before...
And all this is after only a couple of days of degassing and development....
Definitely worth trying this style of profile a couple more times with other beans, to be sure it's worth hanging on to.
Hi Mal, when you mention 'ever decreasing gradient' what is the starting Rate of Rise, something like 20 degrees? Just trying to get a handle on this style of roast.
I have been sticking religiously to your 10 degree ROR then halving that at 140 to 160, then back to it till first crack, then easing. Only just got the hang of that and you post this so wouldn't mind trying it.
Start off at about 17C/Minute and allow this to slowly decay to just on 10C/Minute at a Bean Temp. of ~150C, then push it back up to about 11.5-12C/Minute on the way to 1st-Crack, allowing this to decay to about 5-7C/Minute by the time it reaches 1st-Crack. I then try to allow it to decay further to no less than about 4C/Minute until I stop the roast at just on the cusp of 2nd-Crack and cool the beans.
Hopefully, my next batch produces a much more legible profile that I can attach...
Big thanks Mal! I just roasted a couple of weeks worth today so will give this a go in a couple of weeks. Thanks for such a detailed explanation, hopefully I can do it justice.
Still a little bit of surfing involved and surprisingly, the peak temperature from the heatgun is noticeably lower than for my "usual" profiles. I guess what surprised me most, was the intense fruitiness/acidity with retained chocolate/cocoa and huge body, given that the batch was only pulled at the first few snaps of 2nd-Crack. Normally, this aspect of a bean's flavour profile would be more muted. Seems as though this profile allows one to make gains from more of what might be hidden when sticking to a single profile.
Anyway, as always, just goes to show that it's never too late to try something different...
Well - boots and all or not at all! Had to do three roasts this arvo anyway, so had a go at your profile. All 800g batches – ambient temp 20°
First was a Panama Black Honey. I set the initial gun temp after T.P. (turning point) at 520° (normally I'd start with this bean at around 480° then ramp). This gave a decreasing profile beginning at 18° p/m, but dropped too early to 10-11° RoR around 120°, so I had to add some more heat. Then increased to 570° at 150, but that took RoR a little high so dropped back. I also tried to decrease the temp to your suggested 5-7° before first crack, but if I go this low, my roasts tend to stall, as did this one. I have to go into first crack fairly hot them ramp down to end up with a smooth decline. So this batch took a little long to second crack at 222.5°.
Second batch Harrar Longberry. Wanted to go a little faster this time anyway for these beans, so after T.P I ramped gun temp 370° @ 45° / 420° at 50° / 470° at 55° and this gave a good declining profile that dropped to around 10° RoR at 150°. Then up to 55° until first crack, but struggled to slow it down. Dropped at 216.5°.
Guatemala Jacaltenango. Third times a treat! Little ramp at the beginning up to full 450° (wanted this a little slower than the Harrar) although this dropped a little too quickly too – added a tiny bit of heat before the 150° mark, then up to 550°. The ramp down was better – I started dropping temp 10° on the gun every degree from 197° onwards (hard to log that manually on the graph while juggling the gun!) but the results were a good decline to drop at 222° or so.
Tried the Panama straight out of the roaster. Still really too early to tell, but I could certainly get some good acidity, but with that cocoa-like flavour from the slower tail to the profile. We'll see how the other roasts go over the next few days and I'll present some notes!
Thanks for the curve ball Mal - certainly keeps it interesting!
Will look forward to your taste results Matt, after a day or three for the flavours to fully develop...
My turnover never really allows me that luxury I'm afraid
I've tried all three this morning – the Harrar was a little sharp on the acidity end as espresso, and I could taste the slower tail end in a cocoa like roasti-ness - but this profile was a little aggresive anyway. But the Guat Jac was great as espresso – lively bright central acidity but with rich, thick body and great fruity sweetness, and no hint of cocoa in this one (partly the 222° drop). Also tried the Panama again through milk - very nice. Certainly more 'commercial coffee-ness' in this roast profile…
I'll certainly keep pondering this profile – and there is much less gun adjustment to fiddle around with (at least up until first crack)!
In the overall scheme of things, the sorts of adjustments to the profiles we're talking about here, seem very minor to say the least. Never ceases to amaze me that such small changes realise such marked differences in the cup. Drinking another cup from my last roast at the moment, even more complex than I described the other day, still with loads of fruity acidity and huge dark cocoa. Great combination...
I agree with you re: the "commercialness" aspect. It does taste more like a blend that's been roasted in a drum roaster. Yet another surprise...
Just roasted 750 grams of India Elephant Hills AA, turned out a little uneven for some reason, perhaps I'm losing my touch
Drinking a Lungo as I type, straight out of the Coretto, a little rough around the edges at the moment (the beans were still warm when I ground them) plenty of time for them to develop, time will tell.
Finally got around to modding my Behmor (ala artman's mod) by drilling a hole through the spindle and inserting a bead probe so that it measures temps from inside the drum!
Details here: https://coffeesnobs.com.au/roasters/...max-temps.html
(With my mod on the bottom of thread)
Roasted up 3 batches of Colombian Supremo Popayan (washed), and I really just wanted to start from scratch with not only my new updated manual panel, but the mod too. So I used the same profile for each bean: 350g beans, setting 400P1A, pushing P3 C D when rolling FC hits. And the only thing I changed was when/what temp I ended the roast (ending about 4-5 degrees earlier for each subsequent roast).
First roast I wanted to take to the start of SC so I could see what temp that occurred, and it happened at 229 degrees (FC at 204 degrees), and overall, FC, rolling and end of FC were all pretty darn consistent between the roasts (in terms of what temps they were measured at), so I am super stoked with how these went!
Wanna say a massive thank you to @artman for posting his mod and also for all the helpful advice he gave.
(Am not sure what that strange dip was at about 176.4 degrees in the last roast... doesn't make sense!)
Short afternoon session at the part time roasting gig. Roasting 3.85kg batches of Peru San Ignacio to a ‘medium’ roast level (about 217degC) on the Ambex.
I had some ‘scraps’ to get rid of so I’ve done a pre-roast blend here of roughly 2:1 Costa Rica La Orquidea Montanas:Ethiopian Guji. They’re both washed coffees of similar density so they roasted together quite nicely. Should be great for espresso.
Hi LeroyC (or anyone else that can help)
I noticed a few beans in your roast that I see in my roasts also and been meaning to ask why they look this way.
In the picture I circled two beans, the top one looks over puffed and the lower has opened right up and lost the middle. May be poor wording but hope it makes sense.
What is the cause of this, anything to worry about from a taste point of view if in larger numbers?
Did a batch of Ethiopia Sidamo Ardi Naturals this evening, during which I tried to integrate a "DesignByCoffee Dip" in at the 130-165'c mark. Pretty pleased with the results overall, though the tasting in a couple of days will be the real test.
As i don't have an external cooler quite a bit of chaff remains on, rather than getting blown away giving the beans a slightly uneven appearance.
Overall, I found this profile kept some additional front-of-the-tongue acidity as the beans aged, but at 1-week plus there was also a flattening of the overall taste at the back of the palate – much like baking where the roast has been too slow and for too long, which was surprising considering the roast wasn't overly long! I also didn't get any of the fruity intensity Mal described, but this could just be that I muffed the first couple of roasts!
But having done 4 back-to-back roasts Saturday going back to my standard profile, and being well-pleased with my Waghi and Limmu (Waghi is just amazing ) I think I'll stick with my ramping. I just get a wider spread of sweet, fruity acidity and less cocoa at the tail-end. Could just be my tastes and my setup!
Put it all down to the learning process …
PS - how did yours end up Mal after resting a while?
Surprisingly, still plenty of fruitiness, lots of cocoa and enough acidity to maintain an interesting complexity.
Next time around, I hope the Profile remains legible so that I can post it up, as the profile I used was quite a bit different to yours as it turns out. Always difficult trying to transfer words into a duplicate profile I'd reckon. Will see how I go...
Having another bash at my updated profile, this time with a blend I've done a few times before...
Luckily, there wasn't any EMR to interfere with the data logging so this time there is a nice clean profile that I can attach. As is my want these days, I took the batch to just on the cusp of 2nd-Crack with a few lazy snaps being heard as I dumped the beans into the cooler.
Tried to keep the profile within the same bounds as previously and managed to do this reasonably well, although the much warmer day today kept me much more on my toes. As the bean mass temperature started to fall over just on 190C, the exothermic reaction caught me short and the gradient started to climb again, which wasn't my intention. Nevertheless, I just left it to its own devices, leaving the Heatgun setting where it was and it soon came back down to where the original profile track was headed.
The gradient didn't decline much after 1st-Crack, if at all and I guess there must have been more thermal inertia available than I thought there was. Anyway, time will tell if this batch turns out Ok or not. Sure had wonderful aromas coming off the bean cooler though...
Blend details appear below along with the Profile and Post Roast photos attached...
Sorry for the slightly 'out of focus' photos, left my spec's outside and thought it looked Ok in the camera. Ummm, no.
India El.Hills 'AA'... 200g
Ethiopia Sidamo Guji... 350g
Indonesia Sumatra 'C'... 200g
Roasted Weight... 634g
Moisture Loss... 15.47%
300gm Uganda Kisoro AA.
300gm Panama Rati Harmen Black Honey.
Then a blend of;
100gm Yemen Bani Ismaili.
200gm Sumatra, Blue Lintong.
All roasted to within 1ºC of second crack.
This roast batch hasn't worked as well as the previous batch using this profile.Blend Details...
India El.Hills 'AA'... 200g
Ethiopia Sidamo Guji... 350g
Indonesia Sumatra 'C'... 200g
Roasted Weight... 634g
Moisture Loss... 15.47%
I think taking it to the start of 2nd-Crack was a couple of degrees too far. Will keep a much closer eye on the gradient approaching 1st-Crack next time and pull it at around 221C...
Basically, there is very little of the beans' intrinsic flavour nuances coming through. Loads of dark chocolate, body and sweetness but acidity is severely diminished and the finish is quite short. Still very drinkable but was hoping for more...
Friday roast session at work on the 5kg Ambex. Mostly dark roast batches today, however we also did a couple of trial batches that stopped 3-4deg lower, but in roughly the same time.
Haven't posted a roast up for a while.
Rati Hartmann, am finding it needs to be treated a little more gently than some at the end of the roast as it can really take off on you after FC.
I get a sweet spot with most beans around 212c on my machine (192-195c for FC), they smell fantastic.
16.5% moisture loss (180.07 in, 150.31 out):
27.10.18 Rati 180g.PNG
Funny you should say that. I have been roasting the Panama Rati Hartmann and it has caught fire (well, a few beans each roast) on me, which I have never had before. Usually around220-230 degrees.