Roast profile ROR negative

[lohord]

Hi all, I have 3 questions regarding ROR:

1) Why am I getting negative ROR?

ROR went negative at 7:18

log0023.png (39.56 KiB) Viewed 795 times

Context: I am roasting an Ethiopian Yirgacheffe Natural Grade 1 Aricha 1925-2100 MASL using the 2000-27000 Rest Profile. Batch size is 100g.
This is the 3rd time I am roasting this bean at using this profile. The first 2 times went swimmingly well resulting in a fruit bomb that in my opinion is better than most coffee I've tasted at cafes (of course this could be due to the inconsistent nature of espresso). However on this roast, I noticed the ROR went negative at 7:18. I do not have the equipment to measure moisture content pre-roast but weight loss after the roast is exactly the same 16.9%. The green beans are stored in a vacuum container and placed in a cool cupboard away from the sun.

Scott Rao mentioned that if ROR drops precipitously, then development is compromised unless roast is discharged immediately. Is this drop of ROR to negative a cause for concern? What are the possible reasons for this drop, especially when I have never had this problem in the previous 2 batches? I have checked the voltage and everything seemed fine.

2)Why is it that quite a few of the profiles on Kaffelogic Studio have kinks, sharp drops or long flat sections? Scott Rao has mentioned in his book (under the 3 commandments of roasting that BT progression shall always decelerate) that long, flat sections or big drops in ROR damages coffee flavour. Is there an option to smooth ROR instead of the temperature curve?

3) Can we use the temp probe ROR as a means to identify FC more accurately post roast?
Scott Rao wrote an article on using ET ROR in Cropster to identify FC exactly. Can we do the same in using the ROR log in Kaffelogic Studio so that we can get more accurate DTR?

I am a complete beginner to this process. I am only less than 20 roasts in and am looking for some guidance to help me relate what I've learnt in theory to actual roasting experience. Really happy to have bought this wonderful piece of equipment and to learn from this community

[kaffelogic]

Is this drop of ROR to negative a cause for concern?
What are the possible reasons for this drop, especially when I have never had this problem in the previous 2 batches?

You have about 800 beans in the roast chamber, a small enough sample size for there to be the odd statistical anomaly - especially considering the sensitive and fast temperature probe in the Kaffelogic and the natural variability of the beans you are roasting. If it was a recurring pattern you might think about an intervention such as using a profile zone, but if it is a one-off any attempt at making an intervention would be chasing a shadow. Given you have had no problem in the previous two batches I would say it is not a cause for concern.

Why is it that quite a few of the profiles on Kaffelogic Studio have kinks, sharp drops or long flat sections? Scott Rao has mentioned in his book (under the 3 commandments of roasting that BT progression shall always decelerate) that long, flat sections or big drops in ROR damages coffee flavour.

The Kaffelogic temperature probe is very sensitive and fast. Scott Rao's 'commandments' were based on studying mostly drum roasters where the temperature probe is thermally damped and slow. Also there is a difference between probe temperature and bean surface temperature. Typically the Kaffelogic probe is hotter than the beans, and the drum roaster probe is cooler than the beans. During the roast the difference between the probe and beans decreases in both types of roaster. So a flat section in a Kaffelogic and a flat section in a drum roaster are two quite different things. Furthermore, I think you will find Scott no longer considers those 'commandments' to be absolute.

Is there an option to smooth ROR instead of the temperature curve?

In the options menu of Studio you can apply ROR smoothing. This, of course, doesn't alter what happened during the roast, but it does alter how it looks and can help with picking up patterns.

Can we use the temp probe ROR as a means to identify FC more accurately post roast?

In principle, yes, but in practice you always need to cross-correlate with what you heard. My advice is if you are having trouble accurately picking first crack by ear then dial in your DTR roughly based on your best guess, and then, ignoring FC, continue to dial in the level based on taste.