Pretty unusual for a 1kg roaster.

First, post a pic showing what you are seeing as tipping and share your roast profile and drum speed in RPM.

Steve

Steve,

Correct me if I am wrong but I believe these are all examples of tipping (pics below). Not all the beans in the roast were tipped but a handful did. These beans are from my profile 6 which was not a good roast. Too long of a roast. I think it was so drawn out that a large amount of the moisture in the beans was wicked out and by the time I reached 1st crack there was not much moisture left to produce a very audible 1st crack. GO has these beans listed at 12% moisture. Profile 5 also had some tipping but not as extreme.

I have shared my roasts and included the links at the bottom here. My drum speed in set to 75, which is equal to 60 RPM. I am using a 1 KG charge. My fan speed is at 30 HZ from charge to GY. At G-Y transition 45 HZ. At 1st crack 60 HZ. I use these setting for all my roasts.

Profile 6: 390 Charge, 1 Min soak at 0 KPA. At 1:00, 1.5 KPA. At G-Y 0.8 KPA. At 1st crack 0.5 KPA (I am using natural gas)

Profile 5: 390 Charge, 1 Min soak at 0 KPA. At 1:00, 1.5 KPA. At G-Y 1 KPA. At 1st crack 0.5 KPA

My plan for Profile 7: 380 charge, 1 Min soak at 0.5 KPA. At 1 Min, 1.5 KPA. G-Y 1 KPA. At 1st crack 0.5 KPA. Drop at BT 397F.

Looking to improve this roast, any input is appreciated. I know its a trail and error thing when you're first getting started, Im here for the long haul so I will figure it out eventually. 😁

Brazil Profile 6: https://portal.roastpath.com/publicroasts/index/8751673343514240122

Brazil Profile 5: https://portal.roastpath.com/publicroasts/index/7207885421820230122

I suspect your drum speed is too low and your airflow may be too high initially. Bump the drum to 65rpm and check that your airflow is a low minimum at charge and maybe to turning point with a lighter at the charge port flame just drawing in with the burners lit. Calibrate this to your fan setting and maybe your drum pressure gauge if you have it.

If this is correct, you're developing tipping through conductive contact with the drum when the drum is overheated because you are applying too much burner heat to support the roast.

Also, are you charging the drum based on the temperature controller or the BT reading in RoastPATH?

Thanks for the response Steve. To answer your question, I am charging based off my PID and dropping based off my BT in Roast path. I tried out the new settings. Set my drum speed to 80, which gives me 65 RPM. Used the lighter trick again to get new fan settings. New low setting is 15, which is just enough to draw the flame in. Medium is at 30. High is at 50. Roasted a new batch of Brazil at a full 1 KG charge at 380 on the PID. Still got tipping, I have attached a picture and shared link to my roast.

The settings for this roast were as follows:

380 Charge, 1 min soak at 0.5 KPA, 15 HZ Fan. At 1 min, 1.5 KPA.

At G-Y 1.0 KPA. 30 HZ Fan.

At 1st Crack 0.5 KPA. 50 HZ Fan.

Dropped beans at BT 399 F. Got about 1:20 development. Let me know if you have any other thoughts.

https://portal.roastpath.com/publicroasts/index/7925669271223270122

Okay. I suspect that 1.5 kpa gas after .5 kpa gas is too much and I suspect that your airflow is too high for that seed. Brazilian coffee is often lower grown and less dense. This makes it more susceptible to heat defect. You might have to roast longer and slower. You'll also initially run more on charge temp heat and less on burner heat.

Charge at 420F by overshooting to 430F. turning off the burners and drifting down to charge. Soak with the burners and fan off completely for 1 minute with air at zero. At 1 minute go to .5 KPA and air at low. Shoot for a 5:30 green/yellow transition verified by the scent change in the tryer from drying straw to something like baking biscuits. Add gas in .25 kpa steps to whatever you need to hit your mark. Bump the air from 15 to 20 about 3 minutes. If you have chaff migrating out from under the drum, add air in small increments until that stops.

From there, take the roast to whatever development you think is best. At the end of the roast, throw a handful in the grinder. If the grounds smell sharply acrid, decrease your charge temp 10F and try again with the same settings.

If you stall the roast, try decreasing the soak to 30 seconds.

Then let us know what happened?

I don’t have any experience with Brazilian naturals but I do roast both Ethiopian, Columbian and Costa Rican naturals. I am on a 3k NG. I use the following here’s my Columbian natural

.385 charge 30 sec soak

2.0 till 300 degrees then .8 kpa and med air. At outlier .5 kpa and High air till I get close to drop then I’ll either shut off gas and coast my way to finish

https://portal.roastpath.com/publicroasts/index/4848657445418010222

Thanks for the advice Steve and thanks for your input AJ! Wanted to give an update... still struggling with the tipping. I will include the two roasts I tried. The first was Steve's approach (profile 9) with a 420 charge temp. Soaked for 1 min no air, no gas. At 1 min, 0.5 KPA and Low air (20HZ). At 3 min, turned gas up to 1.3 KPA. At G-Y, 1 KPA and medium air (30 HZ). At first crack, 0.5 KPA and high air (50HZ). When I dropped the beans, saw lots of tipping. I will have to try this approach again because I hit my G-Y about a minute earlier then Steve suggested.

On profile 11: 385 charge, 45 sec soak, 0 Gas, Low air. After 1 min, 0.8 KPA. At 3 min, 1.3 KPA. At G-Y, 0.8 KPA and 0.5 KPA in development. Left the air on low for most of this roast, only turning the air up to high for 1 min during 1st crack. Felt like I had a pretty good handle on the ROR during this roast so I was surprised when I saw tipping at drop.

Going to give these roasts another shot and then maybe try and new profile with a more static air approach.

https://portal.roastpath.com/publicroasts/index/8581036180322290122

https://portal.roastpath.com/publicroasts/index/782132354823010222

There is something very wrong here. This is literally the first time I've ever heard of tipping on a double walled 1kg. I honestly didn't even know it was possible.

What is this coffee and where did you get it? Get me pics of the greens as best you can.

Whatever the case, you have to slow the roast down. Less preheat, less gas and run a medium airflow for the entire roast.

The other thing that's weird is tipping usually presents as black spot (not a darkened area at the end of the seed)with fractures running to the seam. This doesn't look like I'd expect and it makes me wonder if this is just deeply carnalized sugars from pulp remaining on the seed.

One last thing: if you have to, send me a half pound of the roasted coffee. We can identify heat defect by the scent of the grounds alone, but I need to grind some of this coffee.

I appreciate that Steve and it may come to that. Currently sitting and reading Rob Hoos book. He specifically mentions tipping to be a product of insufficient drying time (drying too quickly). Upon reviewing my profiles it looks like Profile 11 had the longest dry time of 5:35. Ill keep reading, coming up with a game plan. Oh and I could dig way more then half a pound out of the garbage to send to you! 🤣

That is an interesting observation you made about the black spot/tipping. That's news to me. Took a hand lens to it and yes it is just a black spot.

Here are the details on the green beans, got them off genuine origin. They were listed at 12% moisture.

https://www.genuineorigin.com/green-coffee-beans-brazil-matas-de-mi...custcol3=1

steam comes out from the part where beans will sprout (because it is easier, and it's only on one side), the higher pressure inside beans the easier it get tipping. I agree with Rob. if. 5:35 not long enough why not try 6:00?😋 as long as it taste good, no matter tipping or 567 mins to yellow

Between the two cups, profile 9 is better. Profile 11 is quite flat which makes me think the drying time of 5:35 is too long. I will shoot for drying time of 5 min on the new profile with a static air approach. Rob has a graph in his book that shows tipping to occur either in the drying phase or in the Maillard phase. I have seen no signs of tipping when I am pulling the trier for my G-Y marker. I believe it is occurring in the Maillard phase.