I have been experiencing this for ever dude, it makes it really difficult to accurately track the roast at least on the software. I did talk to mill city about it and of course the try to help and sent me 2 new thermocouples to see if that was the issue but did not fix the problem, although something did change. Now the drop happens after first crack in the same way (a little initial spike followed by a steep decline ).

Also, like you I thought the decline started after the decrease of fuel input but is not the case, either I make an adjustment or no the drop happens anyway. I'm calling mill city again this week to see what's next to test.

Lucky for us we have the right company behind this product and I'm confident they'll figure it out soon!!

PS if you find a fix first please let me know

Hey Justin & Will,

We've been turning this question over a bit in the office, and I wanted to share some of our thoughts.

First off, it is common to see a temporary dip in ROR at First Crack. As the coffee begins to steam and pop, moisture releasing into the drum can cause a drop in temperature readings. We know that this will only be a short drop and that typically the coffee will begin to take off again right afterward as the dryer beans respond more rapidly to heat. This is why many roasters will decrease fuel and/or increase air at (or just before) FC.

In the case of a drop at Dry End/Green to Yellow transition, the cause is less obvious. I've looked over a few of mine and Bryant's (a production member and tech support staff) recent roasts and didn't note a consistent drop at DE. One thing we wondered is if you marked the time on your PID, do you also see a decrease in the rate of rise there as well? If you did a roast without data logging, and you manually recorded your temp just on the PID, then went back and calculated ROR, would you see the same anomaly? If not, this is a weird fluke in the data logging (and we want to know about it)! If you see it there as well, it would definitely seem like something is happening to the coffee at that landmark that's causing it to slow its reaction to heat. Maybe, similar to the change at FC, there is a slight amount of moisture that's lost rapidly at this transition.

My last question is, and I'm sure you saw this coming, what's in the cup? Do you feel that this stall in ROR is negatively impacting the quality of the coffee? If you had two coffees to compare side-by-side, and one had this dip at DE and the other didn't, could you tell the difference, and would the coffee with the dip be lower quality from a purely sensory perspective? As Steve is so fond of reminding us, "We cup the coffee - we don't cup the curve!"

We will continue to ponder this. Thanks for the good question!

-Lauren

I'm having the same issue and it's driving me nuts! It doesn't matter what green I use, it happens every time.

It spikes after green to yellow and then falls into the decline of death and then spikes back up to where the graph would have continued.

I need a solution!

Nathan

I have a 2kg. roaster

Nathan

As Lauren mentioned, we've done a lot of thinking about this.

When coffee releases water, it slows ROR. Free moisture in green coffee is usually 10-12%. This is the moisture that steams off during drying. Liberated moisture is the water that steams off the cellulose as it degrades from heat. This begins at a specific temperature and accelerates as the roast progressea until the seeds reach an equalibrium point where the reaction stalls out.

The point is, as the coffee drys, it decreases ROR. When the coffee off gasses, it slows ROR. When the coffee ceases off gassing, it increases ROR.

The drop you are seeing at the end of the drying phase is the point that the seed starts to off gas water and carbon dioxide mathematically magnified by the backward looking ROR equation.

Your thermometery and logging are simply accurate enough to show you what's actually going on.

If you're super jacked up about pretty lines, you can anticipate these artifacts with changes in gas and airflow and smooth them out.

When you do that, cup both roasts blind. You'll probably be surprised at the result.

This idea isn't unique to our roasters. We've recently heard of identical response from the handful of manufacturers that have recently followed our lead with better probe placement and probe quality.

It's not mistake. It's what's actually happening.

Thanks for the reply Steve.

Still confused why this doesn't show up on all roasters. Do you recommend any specific data handling and smoothing settings that might make this less dramatic? Any chance my thermocouples are faulty or need to be in a different location? Its tough to see other Millcity roasters that produce nice smooth roast curves on RoastPath! There must be a fix!

I'm not crazy about ignoring this length of time. Should I ignore this section of the roast on all my roasts? It feels very out of control.

Nathan

It does show up on many roasters. We think that differences in airflow, batch size, drum speed, and applied heat magnify the artifact.

Historically, I have tended to remove all smoothing when roasting. The response is marginally faster and I've used the noise to surf the trend. I can also play with the sample rate. Sometimes going to a longer sample interval can smooth the graph to a users satisfaction. YMMV

The control you aspire to is very much less in the ROR on the left side of the curve. The rate of chemical change in your coffee is very low in the first half of the roast. Chemical change accelerates dramatically coming into first crack and continues to accelerate as the roast accumulates energy all the way to the peak at your finish temp.

In other words, what you are looking for is consistent event times: green/yellow transition, first crack, and your finish temp. These are the "big ones" that most impact the flavor of your coffee.

Variation in your ROR is an indication that you may be "out of control", but only to the extent that it affects your event/time targets. If it doesn't, don't sweat it.

Taken to a ridiculous extreme, based on that idea, a guy might think he could potentially "soak" the coffee for several minutes, fall behind on the roast profile, and blast max heat to recover to similar event times thinking the coffee would taste the same. That would be stupid. Don't go there.

(If you need to know why, it's because your inlet temps would climb higher than your seed can withstand heat and cross over into defect._

Our pursuit of quality may be a bit ridiculous, but our tactics need not be.

Beyond all of that, because it's early and I'm marginally less inclined to self censor, I'll tell you that the majority of people that vest in a cartoonish notion of ROR "control" do so because they don't have enough experience to know their coffee is good.

When you question guys about this, they frequently get all up in their feelings about their customers telling them how good their coffee is or their Golden Bean awards, or the fact that they bought a book and paid $100 to warm a seat once.

So here's Steve's no BS answer: chill out. Cup the coffee. Adjust. Roast and sell one million pounds of coffee, pay your dues, and get back to me.

This thing is going to take some time, but you have to breath anyway and your coffee is objectively on track to go from "good" to "great" Enjoy the journey.

If this answer totally bums someone out someday, call me at 612-886-2089 and we'll have a real discussion about your coffee and what we can do to help you make it better.

Thanks Steve!

Do you have any roast curves on the 2kg that I could see?

RP links in the notes...

Why isn't there the same spike/dip in their ROR curves?

Because he's operating the machine differently,

I’ve used the settings that they use in your roast alongs and still get the nasty, prolonged, spike and crash and spike. Is there anything else that can be done? I’m feeling like I’m losing all control in the roast for a really long period of time. I don’t know what my actual ROR is for almost a minute.

Hi Nathan.

I learned two things in the 10 years I facilitated treatment groups. First: the source of anger is often the hidden should. Second: our feelings are liars.

The path to better coffee is to focus on the outcome and then retrace your steps to duplicate it. Don't let anxiety derail that process.

As it is in coffee, so is it in life. Only practice makes perfect.

I'm pretty sure all of this came from a series of fortune cookies, but that doesn't make any of this less valuable and relevant to our coffee and our business and the way we live our lives.