Do you monitor RoR in real time and adjust accordingly?

I always have with my Aillio, and now I have switched to Huky+Artisan.

I was watching this video: http://www.youtube.com/watch?v=UrvEKCGedts (Exploring Scott Rao's Flick and Crash). In the video he advocates for not displaying RoR on the graph during roast because its a sliding window of data that he thinks won't display well. On the contrary, I think Artisan has the tools to take this into account and show a pseudo-real-time RoR based on a sliding window.

Do you display real time RoR and do you add/remove heat/airflow based on what RoR is doing?

N6GQ wrote:Do you monitor RoR in real time and adjust accordingly?

I have switched to Huky+Artisan.

I hope you enjoy; I like mine very well.

...I think Artisan has the tools to take this into account and show a pseudo-real-time RoR based on a sliding window.

Do you display real time RoR and do you add/remove heat/airflow based on what RoR is doing?

Like TallDan mentioned, you can't react as well to the Delta and should adjust according to previous batch of same bean and charge mass. I like to see it though, because if RoR is way off I may catch a need to change my plan - you can save a roast. But more than that I like to see if my maneuvers to target RoR at time of 1C, 1C rate of drop & avoiding flicks are successful during roast.

How would pseudo real-time RoR be different?

A rate has to be a mathematical calculation based on a change over a period time. I believe that this time frame is adjustable in Artisan. But I would assume that too short a time span would give a graph that was pretty difficult to make use of.

At first when you asked this question, I thought ...of course I do. What am I, some sort of caveman? But then I actually thought about it...

I respond to bean temp ROR only for the flick, to try to mitigate, if I can. Usually, I'm at such a low gas setting, I cant really cut anymore.
Crashes are really a product from too much energy in the ramp phase (see Rao's 2nd book). I've entirely eliminated them (when get the settings right).

I do actually use ET ROR. This and MET (I interpret as manifold exhaust temp. On my Huky is right by exhaust) really tell me what's going on with the roaster. The MET is a good measure of the overall energy in the roaster and drives the ET. The ET ROR is like a future BT ROR. It's drives the BT.

You can also think of them as MET being all air, ET is air and bean, the BT is bean. Obviously BT is the thing that matters, but you cant do much to fix it when it's going wrong.

Everything is planned (usually on the fly planning, lol) based on previous roasts of that specific bean.

Hope this helps.

The guy in that video is confused. He says the flick happens before the crash. Negatory. The flick happens after the crash. That's why the phrase is "crash and flick", not "flick and crash".
I would move on to the next source of internet misinformation.

BTW, yes, I monitor RoR in real time.

The trap you want to avoid with small roasters that do not retain a lot of heat is "steering" the roast by making a dozen of more micro adjustments trying to keep RoR on path. Ideally you want to make 3-4-5 heat adjustments, almost always in the lower direction.

The best practice is to use your best guest roast plan on roast #1 of a particular bean. Then use that file as a background for the next roast, making heat changes where necessary to correct RoR path. Then save that roast as file #2 and use it to further refine roast plan and so on.

As a home roaster, if you only have a pound of coffee and you want to roast all of it in the first roast, then use a background profile from a similar coffee.

I happen to be roasting now:

Thanks everyone for your thoughts, very helpful!

@Almico, that's a great looking profile you have there. Thank you - your RoR curve looks great.

plus 1 on what Alan wrote, I'm now using fixed air flow, and start tapering off heat after going full blast from start until ~3:20, then an 0.5KPa step down every 30 sec or so until 1.5-1 KPa...that and using RTDs rather than TC made a world of difference in my cups!

(not forgetting the support by the folks here!)

Thanks - I am using RTDs on my Huky as well (ET, BT and MET). The Aillio had the RoR curve built into their RoastTime software and it was there during roasting and smooth - I don't remember if there were options to tweak the time window that was being used to create the RoR curve though.

The way I roasted on the Aillio was to pretty much keep heat constant and modulate the RoR with airflow. That seemed to work pretty well. I'm not sure how well that'll work on the Huky but I intend to try it.

I just bought a new DC fan and a Phidget PWM device to allow me to modulate fan speed from Artisan, so thats the next project. Modulating gas pressure might be a bit more in-depth, that'll be the next project I suppose.

Almico wrote:The guy in that video is confused. He says the flick happens before the crash. Negatory. The flick happens after the crash. That's why the phrase is "crash and flick", not "flick and crash".
I would move on to the next source of internet misinformation.

I agree with this characterization. Flick happens as 1stC ends. The crash happens as 1stC gets rolling.

There is what I usually refer to as a plateau that can proceed the crash. This can be a small rise in ROR, if bad enough. We should have a specific word for it. I hear people mix this up with the "flick" all the time. Like the people who "drop" their beans at charge. Maybe skaters who "dropped in" to halfpipes? I dropped out of high school, so it makes sense to me.

The plateau and the crash have the same cause, though. Too much energy at color change and ramp. You can not respond to this in real time, since the cause happens 1-4 minutes before the effect.