First roast, drum temp not increasing?

Hi, first 2 roasts done (350 g Nicaragua) on my R1V1 bought second hand.

Quite an upgrade from the Behmor 1600 :^)

The coffee might come out quite well (no charcoal).
But I’m little confused on the drum temp (white) that has a slight decrease during the roast. Not following the Bean temp as see in most of the profiles?

PS I will upgrade to the IBTS when in stock again. I guess it’s easier to compare with curves from newer roasters DS

I have no experience with the V1 Bullet so feel free to ignore this-

The IR profile is similar to what I think I’ve seen from other V1 Bullets. What seems to be different is maybe an offset, i.e. the curve shape is typical but maybe a little lower than normal. It may be that the protective window over the sensor needs cleaning.

As I recall there are warnings that go with cleaning the V1 sensor window so I checked the Aillio Support page but didn’t see anything about the V1 IR sensor. But I did find was an early manual (2018) that had a link to a document saved on about cleaning the V1 IR sensor. Cleaning is basically the same as the IBTS (cotton swab + alcohol) except it’s easier to reach. I have no idea if this will change your roast profile but… ??

Btw, if IBTS is any indication, a swab with alcohol coming away clean from the sensor is not necessarily an indication the window is actually clean. It seems whatever contaminates the sensor cover isn’t readily visible. So try cleaning it a few times even though the swab comes away with no indication of contaminants. Again, this is from someone that is cleaning an IBTS and not a V1 IR sensor window, so consider the source!


Thanks, Bruce. The IR-cleaning also popped up in my mind. Will check it :^)

I suspect your low coffee volume, high drum speed, and low power settings might be a part of the issue.

Every bullet seems to have a bit of varience with drum speed, but maybe d6, 500g and p8 at the beginning might get your et temp higher.

I haven’t used the IR in a bit, but i recall et dropping below bt, usually between 6 and 8 minutes into 10ish minute roast being normal

1 Like

Thanks for the feedback @veloandy (cyclist maybe?)
I’ll clean the sensor and try with lower drum speed & maybe a larger batch, to see if I get more consistent curves :^)


Hey @coffee_25_f57bed973. Welcome to the Roast World community. To echo what others are saying, I think you need to give your sensor a cleaning. The curve is pretty normal for the old IR sensor; but at a preheat of 160, your bean temp reading should not be so high. This suggests that your sensor is not accurately reading the drum and is overheating it. If you can’t get the sensor shiny, you may want to replace it with the IBTS. Also, I recommend sticking your hand up there and cleaning it between each roast session in between your deep cleans. It is very easy to reach and will save you some trouble.

Yes, I’ll upgrade to the IBTS when available again.
Good tip on the cleaning, that it’s possible w/o removing the front.


PS how the heck do I edit my User name on the forum? I see “Coffee 25_f57bed973
your full name (optional)”, but it’s greyed out? DS

No wonder the IR-sensor had problems.
Scrubbed outside and isopropyl swab on the inside. Let’s see if I removed some coating too much…

1 Like

Still crossing curves after the cleaning. I’m scratching my head over this.
The 2 first roasts taste good (no excellent), but it would help if the temp readings where more according to the book (Scott Rao’s) ;^)

Next roast I will try a larger batch, like 500 g instead of 350 g.

Rao is using a thermocouple probe so use your Bean Temp (BT) to approximate his guidance. But you probably won’t/can’t duplicate the numbers as results depend on probe placement in the bean mass. Also, you’re forced to compare results from very different thermal mass of a commercial roaster that differs by hundreds of pounds, gas fired burners with huge thermal reserves that apply heat differently, and larger available air movement.

Looking at your BT curve above it appears the roast stalled at the end: BT hit a maximum a good 1.5 min before you dropped the beans. That’s not unlike what I like to do (“stretch” the roast at the end), but I believe Rao thinks you shouldn’t let BT reverse on itself (note that RoR has disappeared below the X-axis starting about 9:00 min). So perhaps your flavor results are related to that… ?? (Only guessing here). It will take a lot of roasts to get things dialed in for what you enjoy from your coffee.


Yep, sounds like good advice, Bruce😊

Will try to get a higher BT & slower decrease in RoR.
Maybe slightly higher charge temp as well, but don’t want to shorten the total time to much.

I see that steady decline in RoR as very desirable… it’s what I try for in order to stretch the roast.

To add to your toolbox of ways to control RoR, keep in mind P & F aren’t the only 2 things affecting RoR. D(rum) speed also affects the slope of RoR. In fact at my usual 550 gm batch size it is a reliable way to manage RoR slope: after 1C, lowering drum speed will increase RoR slope; conversely raising drum speed will decrease the slope. Not sure how predictable Drum speed changes are before 1C (I do that stage without active changes). The concept appears to be that at a lower drum speed the bean mass is in contact (or near contact) with the drum longer increasing the time available to transfer heat from the drum.


OK, roasted 500 g Nicaraguan. All my 7 batches have tasted good, it’s the drum temp that I don’t get. Why does the IR drum temp not rise (above the finished bean temp)? The IR-glass & sensor has been thoroughly cleaned. Yes, it’s a R1V1, but it’s difficult to compare curves…

Sorry to say, from what I’ve seen of other V1 Bullet charts, that’s pretty much what you can expect from that earlier version of the IR sensor. Bean Temp graph should be identical to current copies of the Bullet.

Any update yet on when Aillio might be able to supply an IBTS kit?


Thanks for the reply, Bruce.
But really be strange that the IR-sensor seems to follow the drumtemp, but really slow.

No news on the IR-sensor. I tried to pre-order but that’s not possible. Later this spring was the answer.

I read somewhere that the early IR sensor was aimed UP, so it really was a drum temperature sensor. Can anyone confirm this?

It would be great to know how drum temperatures changes, especially over the first 2-3 minutes. Would like to use as much power as possible during that interval to maintain drum temp near the preheat value, but don’t want to overheat the drum and scorch the beans. Trial (without error yet) suggests P8 and even P9 can be used even with tiny 325g batches without any scorching. Maybe this original sensor could confirm what’s going on?

There is some history to this:

Indeed the first IR sensor with the v1.0 roasters was pointing upwards. The only purpose was to see the relative heating of the drum so that it could be detected when the set PH temperature was reached - and also detect equilibrium with the help of the BT sensor . The concept was that the IR sensor (that is quite heat sensitive) had to look through this little round Germanium coated filter glass not to get too exposed to heat from the roasting chamber. This coating, however, also offset the measured temperature downwards. So it was never the real drum temperature measured. Many users were puzzled over the fact that the BT temperature could go higher than the drum temperature - and this is why…

With the IBTS it’s different. There is no filter to offset the temperature, hence the temperature is much higher and calibrated to the real drum surface temperature. And the fan make sure the heated air stay inside the roasting chamber.

Sorry for my long and windy explanation, I realize that you might already know some of the history, but then others get more informed.


Thank you, lav. There I have the explanation to my strange drum temp curve. No need to scrub the glass even more, just to wait for an IBTS upgrade :^)