Hey All - I’ve recently been running into some weird behavior of the IBTS RoR reading and I’m curious if anybody else has run into the issue or can shed some light. Essentially what I’m seeing is that on the first of two back-to-back roasts, the IBTS reading appears very erratic in the early phases of the roast - big IBTS RoR oscillations and smaller IBTS temp oscillations - before calming down in the second half of the roast. On the second of the two back-to-back roasts however, the IBTS appears super stable from get go.
Current thesis is that I am not pre-heating the roaster long enough, but I usually give it a minimum of 20-30 minutes regardless of “charge” indicator.
For some additional context, I’ve only got about 20 roasts on this bullet, excluding the 3 seasoning roasts, and I’ve started seeing this behavior on the last 3 roasting sessions (6 roasts). Here are the screenshots of my two most recent roasts:
If you look closely at IBTS you will see movement similar to that seen in I-RoR. I-RoR is a plot of the change of slope of the IBTS curve so it’s correct that any change in IBTS is somehow reflected in I-RoR. Bottom line is that what you see is correct.
I’ve also noticed this on some roasts. I haven’t been able to pin point exactly the reason why, but my thesis is that certain bean weight and drum speed combo appear to give me this erratic behavior. I’ve had some even bigger swings than the roast you posted above. D9 seems to be the optimal speed to avoid this, not sure why though.
FWIW, I clean my IBTS regularly, so it’s not because of a lack of maintenance.
The issue stems from an artifact in the sampled data when measuring IR sensor output. It’s the combination of sample rate, drum speed and bean motion. Distracting but not a big deal. And it does diminish as the target area gets hotter toward the end of the roast.
Thanks for the prompt reply.
Don’t know yet - I did a deep clean about a month ago and all was operating normally before and after.
Will do one when I return in two weeks if I cannot get it done today.
Hi I have had this happen a couple times now, when doing multiple roasts. Seems to show up on 2nd and 3rd roasts. I’ve included some screen shots of three consecutive roasts, plus another done on an earlier date showing the graph from a similar roast for the Guatemala. You can see a similar roast profile but the IBTS line is erratic. The last time this occurred I cleaned the sensor inside the roaster (less than a month ago). But it happened again in some very recent roasts.
I’m not sure this is an issue that will completely go away. I just did a deep clean last weekend. If I roast this weekend I’ll post. I know my last roasts prior to the clean does have some wiggly IBTS RoR lines.
Are any of you still experiencing this issue? I’m seeing the same pattern in some (not all) of my recent roasts. For me, the oscillating ROR is more pronounced in the later of sequential roasts.
@jimecunninghamfs5o interesting you bumped this old thread. I did 5 roasts yesterday (after a deep clean a month ago and hadn’t roasted since) and noticed the oscillating pattern as well in the last 3 roasts being quite pronounced esp the last two which were 1lb roasts (the other 3 were 1.1kg roasts), but didn’t pay too much attention to it until I saw this post of yours. I went to look back at many of the past roasts and this “problem” is much more pronounced with my ~1lb roasts.
I’ve been ignoring that in the beginning part of the roast and focus on the latter part. The software tracks every data point and doesn’t smooth out the curve so we’re going to see a lot of this.
My main concern is whether the data is real or not. Do the oscillations indicate a malfunctioning temperature probe or rather realistic temperature conditions that just need to be understood?
No malfunction; normal stuff. I don’t know how much you care about the details, but…
I-RoR is an approximation of the 1st derivative of I-Temp… essentially I-RoR tells you how fast I-Temp is changing value; i.e. not the value of I-Temp but how fast I-Temp is changing. Notice the wavey characteristic of I-Temp early in the roast above: I-RoR is confirming those temperature swings with calculated values for how fast the temperature is changing.
If you find it distracting you have a couple options in Settings:
turn off I-RoR and use just B-Temp and I-Temp to roast; alternatively…
change the scale of I-RoR in settings so that it doesn’t have such an exaggerated swing.
In RT4, click the user icon and choose Edit Settings. Then click Roast Config and choose Chart Defaults at the bottom. The Y2 axis is what controls the scale of I-RoR. Here’s how I have it set-
… which suits my purposes. If I wanted to get I-RoR onto the chart earlier I would change the Y2 setting in Roast Config. A larger number will get I-RoR on-screen sooner while a smaller number will have I-RoR off the chart longer.
Note the move away from a linear decline of both RoR curves starting at about 9 min. That’s not what I want but goes with the summer season: room ambient in the summer makes the heat loss significantly lower in that stage of the roast. The cluster of Power changes at that time was me doing a poor job of trying to compensate. During the winter season that is less exaggerated (roasting coffee in SE Arizona summer is… a challenge).
Also in the screenshot above you see the option to not display I-RoR at all… just set IBTS RoR to the left (grey) to remove that part of the roast display. Possibly not a good plan if you ever ask Aillio for help and that function is missing from the data.
