Greetings to the group !!!

I am trying to establish whether my MET temperature when charging the beans (Quest M3, Eric S probes) is too high.

I am aim to reduce the bitterness of my roasted beans and to increase the liveliness without compromising body and sweetness and without making the coffee overly sour. I am currently roasting high quality Panamas (now: Hartmmann Pacamara Washed, Finca La Aurora Alma Catuai Natural, Carmen Estate Caturra Natural) which I mainly consume as SO Espresso or Americano.

I don't cup my coffees regularly but when I did so recently it was revealing in that I realised that my coffee is more bitter than what I have come to expect through trying various specialty coffee @ cafes. Another issue is that in certain of my roasts I perceivethe dreaded paper like taste that takes away from the liveliness of my coffee.

Some background:
I have now been roasting with the Quest M3 for around a year.
My artisan Logs tell me that I did around 250 batches during this period.

I use the Eric S probe Kit with phidget and Artisan running on my windows PC.
The probes are positioned at the MET position and the traditional BT position (per Eric Ss paper)

http://users.rcn.com/erics/Quest_Thermocouples_06.pdf

This a picture of my roaster with the probes

https://drive.google.com/file/d/0B5b1nV ... sp=sharing

Not being an experienced roaster I am trying to stick with Scott Raos principles (diminishing ROR, 20%-25% development, 10m-12m total roast time, sufficient heat when charging the beans).

I am sharing some of my roast profiles below as examples

Hello there,

There is a lot happening in your situation - different varietals, processing methods, and roasting profiles.

On the natural processed coffees, it may be possible that you're charging too high. I would recommend charging slightly lower (like your 2nd profile), but add more heat at the turning point to establish momentum that will help slow into yellow and FC. Once you establish this part of the profile, you should probably play with DTR between 20%-25% to fit your tastes

You second curve is lacking momentum at the turning point, which leads to yellow happening too late, and the whole roast dragging out a little too much. This roast may have some hollowness in flavor and aromatics but with greater body and muted acidity.

I hope this is helpful. I know that my roaster doesn't translate exactly with the Quest M3 roaster, so I hope others with a similar setup will be able to provide you with better input

Cheers,
-David

I've never used a Quest, but from what I've seen from profiles posted on this forum by roasters I respect, it seems that (for reasons I do not fully understand) it may make sense to use a shorter development window in that roaster than you would use on the larger drum roasters that Rao bases his recommendations on. So if you are tasting ashiness in your roasts, try doing a shorter development period, possibly with a lighter finish as well.

Also, it looks like your ROR is dropping so low during development that baked flavors might be a concern. If you are getting a cardboardy note in your coffees, this might be the culprit. Dropping sooner might address this issue as well.

I'm not sure how much help this will be, but it is something on which to cogitate. I have observed from the many published profiles and from personal experience that there can be huge differences between the true BT/ET/MET temperatures and the temperatures we think we see. First, the original Quest has a thicker drum than the MK-II. Then there's the 'S' model, which is very different. Finally, and most importantly, when installing the probes in my MK-II I noticed that probe placement, especially the MET makes a huge difference. So, while your probes show the MET to be considerably higher than the MET that I run (I rarely let it go much beyond 250 C), who's to say that either machine is showing real MET?

I am trying to establish whether my MET temperature when charging the beans (Quest M3, Eric S probes) is too high.

For the reasons stated above, I don't believe we can answer this specific question. Moreover, I would suggest that MET is not the best measure of when to charge. If one is going to charge by temperature, BT (which in an empty roaster is more like ET) I suggest is the way to go. Again, probe calibration can be an issue. I usually charge when BT (empty) is around 200 C. But, again, we don't know if your 200 C is the same as mine. Also, you have the thicker drum, which retains a bit more heat.

My suggestion is to vary things one at a time and cup the results, noting the temperatures of key points. While the temperatures may not be accurate or even useful for another Quest, for your own Quest they will be. For example, I charge at 200 C mostly because if I go a lot above this I tend to get a bit of external scorching. On my machine, FCs occurs at 178 to 180 C for nearly every SO or blend, while yours is about 204 C, around 20 C higher than mine. Likewise, 2Cs on mine occurs at about 198 C, while you appear to be getting to 214 without hitting 2C (so I'm guessing your roaster won't hit 2C until your probe displays over 220 C. From this, I deduce that your displayed charge temperature of ~195 to 200 C, is likely a good 20 degrees cooler than my 200 C. All this to arrive at the question, are you charging at too low a temperature?"

Another data point, "Dry", is worth having a look at. On my roaster, Dry (when the greens turn to yellow or very pale fan) happens at about 150 C. Your curves show the time to 150 C at around 3.5 to 4.5 minutes. However, if I'm right about the calibration difference between our two machines, the dry point on your machine may well be closer to a displayed 170 Degrees. In any case, my experience is that you want to get there more quickly. So, I suggest that you have a look and see when the beans turn to yellow/tan and log that temperature for future reference.

At the risk of starting a controversy, I would not worry about TP, which I see you log. IMHO, it's nearly a meaningless figure, as it is not real. The real bean temperature starts at ambient and gradually rises. The apparent drop of bean temperature to TP is merely a reflection of the time it takes for the mass of the probe to respond to the drop in its temperature due to charging the cold beans. So, I suggest ignoring it and concentrating first on the charge to dry time.

In summary, I suggest you log your displayed temperatures for Dry, 1Cs and 2Cs, occur at roughly predictable temperatures. Then as a starting point, shoot for something like 30//40//30% (charge to dry//dry to 1C//1C to drop). Also, try to keep the roast from going negative, even at the end; I think I see a bit of a hump, or negative ROR at the end.

You have not mentioned your charge size. If you can't achieve the 30% to dry, then perhaps your are charging too heavily, assuming you are max heat during the first part of the roast. Our Quests are not like the big drum roasters where the huge mass of the drum sometimes requires cutting the heat entirely. I've found that for charges of around 200 to 300 g it needs to be maxed out until nearly Dry.

Also, you've not mentioned your desired degree of roast. If my hypothesizes is right about your temperature calibration, you're dropping a medium roast, even though the numbers suggest something darker if I go by my roaster. It's difficult to get that last 30% right for lighter roasts without having that little hump in the profile.

For all that, as is often said on these pages, the proof is in the cup, not in a bunch of numbers.

Thank you all for sharing you wonderful feedback!

Some further information about my setup and process:
- I have the newer quest M3 (which I think has the thinner drum walls and the fun running even at F0).
- My heat Amper meter maxes out around 5.4-5.6
- I started roasting 200gr which I gradually increased to 250gr. We are big coffee drinkers so I do around 5-7 * 250gr batches every 10 days or so. 250gr are more convenient for me as it reduces roasting time and I need less vacuum containers (less batches).
- My roasting sessions are based around the following process:

1) Preheat the roasted to BT 180-200 and MET 240-260 before charging. I do this by running the roaster with heat high (around 4.8 or 5.0) with the bean chute open and all other "doors" closed until MET reaches 250c - 260c. I then close the bean chute and wait until BT reads 180-190c. At charge time BT is around 180-190 and MET is around 250-260c and both are on the increase.

2) achieve diminishing ROR with the following settings (heat(h)/fan(f)):
i)from charge to 145BT h:4.8/f0, (start with sufficiently high heat)
ii)145-180BT h:5.2/f3, (increase heat to sustain momentum)
iii) BT180 - 1st crack h:4.2/f6, (enter 1st crack at a lower pace to increase DTR)
iv) beginning of 1st crack h:4.6/f6 (to ensure that roast does not stall due to the cracking),
v) then reduce heat gradually to ensure ROR does not start to increase.

3) Drop beans shortly after the end of the first crack, I look for 20-25% development time, BT 212-218 on my setup, visual characteristics of the beans that resemble the City to City+ roast level.

So far I have been meticulous in following the above process and was not sure where to start experimenting.

From your feedback above here are some variables that I could start playing around with:

- Instead of h4.8 start my roasting with the heat maxed out which in essence will merge my step i) and ii). My concern would be thought that the drum could become too hot (MET to high) that would cause scorching on the beans.
- Instead of increasing heat at 145BT push my step ii) earlier (turning point).
- Keep my process as is and at step ii) increase heat to maximum rather than 5.2
- Maintain higher momentum into 1st crack and aim for lower DTR which would probably mean a smaller heat decrease at step iii)
- Start tracking Dry time and shoot for a 30/40/30 profile. I do expect that, the 1cr to drop 30% will be hard to achieve without going negative or re accelerating the roast.

My difficulty with Dry time (and the reason why I was not logging it so far) is that I am not sure how yellow is yellow if that makes sense. Any good advice on that??

My difficulty with Dry time (and the reason why I was not logging it so far) is that I am not sure how yellow is yellow if that makes sense. Any good advice on that??

Yes, it makes perfect sense. A few bean varieties turn from olive to a distinct green, then yellow, then light tan. Some only go from olive to yellow then tan and some don't go green or yellow, just tan. Personally, I watch for the earliest instance of tan, although most people refer to Dry as yellow...maybe it's just my eyes . Once you start looking for it, you'll be able to see it easily every time. I find this point (150 C on my machine) varies only a little between types of bean.

Since your machine is a 240-Volt version, and most of us have 120-Volt versions, it is difficult to relate to the ammeter display. Whatever our display is, yours will be roughly half. Maxed out, my MK-II draws 11 amps, a little less if it is on an extension cord. You probably already know this

I would not be afraid of sacrificing a batch of beans to see how hot your machine really gets. Try a batch with a charge at 220 C and see what happens. I'm betting you won't get scorching and that your machine's 220 is nearer to what mine says is 200. You really do need to log the Dry point (by colour), 1Cs and 2Cs points (by sound) for a batch or two (even if you throw away a bad batch) to determine the temperatures at these very important steps in the roast.

It looks to me like you have been increasing charge temps with attempts to go faster through dry phase.
Charge temps, based on my experiences with my new Quest M3s, only effects the first 2 minutes of roast.
The Quest drum looses its ability to add much heat very shortly after turning point.
So, you need to pour on heat at charge because it takes 2-3minutes for the heating elements to do anything.
Try bringing the heat to 1100W at the instance you charge the beans. Keep air to a minimum til end of drying.
This should result in faster dry phase, if that is what you desire. Be sure to cut heat back at say 300F to maybe 800W because it takes 2-3 minutes for heating elements to slow down with this massive amount of heat.

So I have tries to implement a lot form we discussed above today and here are some results:

This was the washed Pacamara from Hartmmann (Panama)



I used the below process



and this is the coffee



From chewing on the beans just after the roast I got a positive feeling (no harsh bitterness, less dull flavours and more lively than last time).
I guess the cup will be the next test.

I also tried my new process on the La Aurora Alma Catuai Naturals with different results, I will report below.

Hmmm. I don't mean to be critical, but I still say your drying phase is too long and your development phase is too short. These may be due to charging at too low a temperature and not carrying enough heat through the bulk of the drying phase before pulling back the heat for development. A roast with along drying time and a short development time is likely to be leathery and toasty.

Hey no worries!
Feedback is one of the reasons why we are here in the first place!

In the instance attached above I charged the roaster at 192/240 (BT/MET) and upped the heat to maximum at charge time.
I kept it at max heat all the way to 5:30 minutes when BT was 185c (so past the drying phase). During this time MET climbed to 295c.
I cut the heat to 75% of max (4.2) at 5:30m/185BT (is that too early or too late, I am cutting too much / too little) and upped the heat at the beginning of FC for 45 sec to compensate the FC effect and prevent the roast from stalling.

I find that with a smaller cut in heat the roast maintains too much momentum which results in a very short development time unless you are willing to drop during second crack (which I am not). On the opposite a sharper decrease increases the chances of a stalled development phase especially once FC effect kicks in.

There is a chance that %s got affected by my choice of "Drying Phase end" which I clicked pretty late at 173c. As per my previous post this is an area where I need to practice.
% would also be affected by my choice of FC start point. I click FC start after the first 2-3 cracks. This is usually 15-30 sec after the very first FC (which I consider an outlier).

The only options I have to reduce drying phase % is 1) to reduce batch size, 2) charge higher, 3) extent the roast time.