My days of being dogmatic about roasting are over

Discuss roast levels and profiles for espresso, equipment for roasting coffee.
Ken Fox
Posts: 2447
Joined: 18 years ago

#1: Post by Ken Fox »

I am part of the way back home (overnighting en route) with my new Diedrich 1kg IR1 roaster in tow.

Aside from it being a long day, with several hours driving in the rain and fog in each direction, I had a very interesting hour and a half at the Diedrich factory in Ponderay (next to Sandpoint) Idaho. The long and the short of the visit is that my head is now so full of conflicting information that I'm going to ask you, gentle readers, to completely disregard everything I have ever posted on the subject of coffee roasting. This is especially true when it comes to roast time and temperature parameters. I'll explain a bit more what I mean below.

Here's a picture of Stephen Diedrich next to my new roaster.

The plan had been to meet up with the head of tech support and production today, who would show me the operation of the roaster before it would be loaded in my car. She was unable to come into work today due to car difficulties and by coincidence Stephen Diedrich, the owner of the factory, was around and he spent time with me to demonstrate and check out the roaster. In the process we talked about how to roast with this roaster and about his philosophy of roasting. A lot of the information I received today basically contradicts a lot of what I've learned over the last 6 or 7 years, plus contradicts much or most of what I have read in this and other similar online roasting threads. I'll give a couple examples of what I mean below which will perhaps clarify while I feel utterly confused at this point.

Stephen does not like the term "drying phase" and does not think one should seek to get the moisture out of the beans early on in the roasting process. Residual moisture is useful in the latter parts of the roast process and in fact with some beans one should seek to retain moisture even at the end of the roast.

Another point he made which was similarly discordant with my previous impressions was that first crack should not occur until 11 to 13 minutes into the roast, and that this would not be the same for each type of bean rather would vary but should be consistent for XXX bean (e.g. you don't shoot for the onset of 1st crack to be between 11 and 13 minutes, rather, each bean has its "right" time and once found you should attempt to get to this point each time you roast that bean, and it will be between 11 and 13 minutes into the roast). Likewise, he did not like the idea of an interval of 4 to 5 minutes between the onset of first crack and the end of the roast, rather he felt that 2 to 2.5 minutes was better, even for espresso, except if one was using a high acid coffee where the goal was to tone down excess acidity. Otherwise, he felt, that an interval like 4 minutes would flatten out the coffee too much.

There were other points he made but I really do not want to go into them in any great detail. Rather, the overall "take-home" message was of an approach and philosophy to roasting completely different than anything I have done before and different than most of what I have read on online forums. He is very much against even the IDEA of freezing roasted coffee. He certainly believes rather strongly about his approach, which he believes produces the best results.

Roasting equipment differs considerably among the various manufacturers. Exactly how much this impacts the coffee and how it relates to different equipment and how it should be used with different sorts of beans is way beyond my level of expertise. I am not willing to conclude at this point that what I have been doing over the last years was valueless, since it seems to have produced some pretty good results in my hands. Some of this I have even tested in a relatively formal and singled blinded way, such as the interval between onset first crack and the end of the roast. Jim Schulman and I did this test several years ago, with coffee having an interval of about 2.5 minutes vs. 4 minutes after onset first crack, and we both preferred the 4 minute interval 100% of the time in blind tasting. I have written this up before in prior threads, and it was the only such blind tasting test I have ever done where both of the tasters preferred one coffee 100% of the time and could quickly tell which sample was which in two blindly presented samples pulled at the same time from essentially identical espresso machines.

So how am I going to process this whole experience today? My conclusion at this point is that Diedrich builds very fine equipment which is capable of a degree of control over the process that I have never had with my prior equipment. This allows one to control time and temperature better than I ever did before. I am not willing to conclude at this point, however, that there is only one best way to roast any particular type of coffee. I think there are probably at least 2 and likely even more ways to successfully roast a given coffee.

My days of being dogmatic about this stuff are over. If you tell me that you get good results with your roaster doing something completely different than I am used to doing with my roaster, my attitude going forward is that I have no proof that you are wrong until I taste what you have produced and can conclude that is not good on the basis of having tasted it, personally.

That's where I am on this stuff today, but I might feel differently in a few days after I have the chance to try out this roaster and to think on these matters a bit more.


...split from US Roaster Co. 1-lb Sample Roaster by moderator...
What, me worry?

Alfred E. Neuman, 1955

User avatar
Posts: 1623
Joined: 17 years ago

#2: Post by farmroast »


That's sharp, wow.
LMWDP #167 "with coffee we create with wine we celebrate"

Posts: 638
Joined: 16 years ago

#3: Post by chang00 »

There are really different ways to roast coffee. Hypothetically, the vapor front starts from the outside inward, when it reaches the coffee bean core, the moisture finally separates from the extracellular space, and the moisture then moves outward. This is what I understand as the "drying" phase. Roasting with the condensation of aldehydes and ketones follow, after the moisture evaporates.

From my understanding, in fluid bed air roasting, this vapor front and roasting front, due to the relatively short drying phase, sometimes move in opposite direction. This is one reason this type of roasting can taste "bright". The acid mass loss after roasting, for the same degree of roast, is less.

I consider the degradation of various chlorogenic acids another independent chemical process, separate from the condensation of ketones and aldehydes and the browning/Maillard reaction, and requires some time. As previously posted, at least for me, my first crack generally occurs later than other home roasters'. This kind of roast probably will not get high marks, but I find myself enjoying roasts that are lighter as time goes by.

Perhaps it will be beneficial if we have a home "gas/electric commercial equipment" roasters' exchange, where we roast the same beans and enjoy each other's interpretation of coffee roasting.

User avatar
Posts: 812
Joined: 15 years ago

#4: Post by SlowRain »

Congratulations on the new roaster and the new approach to roasting. The coffee shop where I buy my green beans uses an IR-3 (I'm pretty sure that's what it is). I look forward to reading your posts.

Now if we could only get people to be less dogmatic about espresso.

User avatar
Posts: 3646
Joined: 19 years ago

#5: Post by peacecup »

Does this mean you're also going to be less dogmatic about consuming the stuff within two weeks of the roast date?

I had related experience when I moved overseas and could no longer reliably obtain freshly-roasted coffee. A review of my earlier posts here on HB demonstrates just how much I focused on bean freshness - basically I went to great lengths to get 1-2 day-old beans.

When I moved to Sweden I was more or less constrained to buying "best before" date Italian-roasted beans which I assumed would produce the same flat, crema-less, awful-tasting swill that I got whenever I used stale (i.e. > 4 week-old) American-roasted beans. To my very pleasant surprise I've found blends from a few Italian roasters that far-exceeded my expectations, and in fact taste as good or better as the best N. Am. blends I've tried. I've even gone to the length recently of contacting the roaster and coordinating my purchase with the delivery to retailer here. This assures me of the "freshest" beans, but there are at best a few weeks past roast date. And I know first-hand that as long as the 1-way valve is working, these beans stay "fresh" on the market shelf a month or two, and preserve quite well in the FREEZER for a couple months after that (by the way, a roaster here once gave me the same "never-freeze" advice, which I kindly disregarded based on your own well-executed study).

So, I raise up my cup to the end of dogmatism.

LMWDP #049
Hand-ground, hand-pulled: "hands down.."

User avatar
Posts: 1462
Joined: 15 years ago

#6: Post by JmanEspresso »

Wow, very nice new roaster Ken, Enjoy!

User avatar
Team HB
Posts: 13820
Joined: 19 years ago

#7: Post by another_jim »

Gorgeous roaster.

Stephen Diedrich is a giant in the industry, and well known for advocating "slow start/fast finish" roasting. I do not know about your model, but his larger roasters are well known for their IR panels, giving them a large thermal mass, and making this type of profile almost built into the design. It's certain that this profile can produce great roasts. One of the best I've ever had was a Kenya from Ed Needham's old BBQ roaster. It took about 30 minutes and ended before the first crack completely finished, and when the first pops of the 2nd were already apparent (this happens in long roasts). Everyone else's shorter roasts of that same Kenya turned out astringent, while his was silky.

Other, equally authoritative figures, advocate very different profiles. Sivetz believes the fast start/slow finish of eight minute airroasts is best, while Willem Boot likes 12 to 14 minute roasts with a relatively balanced profile. I'm sure they get frequent superb roasts on their gear as well

I think there is an arc to ones roasting experience, so that one ones approach interacts with ones gear until they work together. Because this is an idiosyncratic interaction, everyone ends with a different set of recommendations.

In other words, have fun learning to roast all over again :P
Jim Schulman

User avatar
Posts: 1153
Joined: 15 years ago

#8: Post by Carneiro »

Nice roaster, great thread! Congratulations!

I already had more than two coffees (Brazilian) flat just around 4 minutes after first crack using the Quest M3. And get nice roasts with a longer first phase.

I'm curious: does he advocate for a first 2-3 minute of little flow, medium air flow until the first crack and high air flow at the end (well, that should help to get what he advocate, to keep more moisture during the roast and to finish it faster).

I think Peter Dupont says something similar using Diedrich, the initial phase with little air flow and no flame - but the roaster has a great thermal mass to work like that.


User avatar
Posts: 762
Joined: 15 years ago

#9: Post by Whale »

chang00 wrote:Perhaps it will be beneficial if we have a home "gas/electric commercial equipment" roasters' exchange, where we roast the same beans and enjoy each other's interpretation of coffee roasting.
+1. This in my opinion is the only way that, provided an open mind, we can get to understand the differences in equipment and methods.

We could agree to a certain coffee bean for a certain time and when posting data, use the "standard" beans as a point of comparison. We could also standardise the stop of the roast at first pops of 2nd cracks for this "Roast metric". All other parameter being open.

Since, Jim Schulman has now offered his services as a taster and evaluator, maybe we could enrich the and use his palate as another "Metric" for comparison.

If we could enroll one of the distributors on one of the Coffee coop it would make getting batches of the "standard monthly bean" easy.
LMWDP #330

Be thankful for the small mercies in life.

User avatar
Posts: 762
Joined: 15 years ago

#10: Post by Whale »

Ken, congratulation on a great looking roaster.

I just love big control panel that will allow you to introduce whatever modification and controls that you are sure to dream up!

What is in the cooling pan? At first I though it was the trier but it is in its position. Is there now a mixer?

Enjoy! And please post pictures and comments of this learning experience.

LMWDP #330

Be thankful for the small mercies in life.