FW Update from 533 -> 538


After finishing a couple roasts I finally thought to check for f/w updates and was surprised to see one was available. Since the Bullet was in OFF mode I started the update and got the “… must be in OFF mode…” message which then cleared itself.

The update went fine. In fact it was noticeably faster than the last couple updates. Then the fun began-

  • I cycled power 2x and disconnected/reconnected the USB port. But no number of cycles could bring life back to the Bullet.
  • Jumpered SDA to Gnd and powered up, then cycled power & USB a couple times. No joy… the control panel was dead (should have anticipated this since the jumper forces the boot loader on… oops!)
  • Jumpered SDA to Gnd again and powered up, then d/l’ed f/w again. This time things went as expected and the Control Panel lit up after I cycled power a couple times.
  • The Info panel took another power cycle to get it to update the number of roasts properly… that’s 3 power cycles in all to get data sorted.
  • When starting to d/l f/w, there were a number of error messages including the one about being in OFF mode. This time the messages didn’t go away. Neither did they interfere with d/l’ing f/w… just kept on like I knew what I was doing.
  • After all the floundering things seem to have initialized properly (but no roasting yet to confirm it).

I posted this only because I understand there has been some work in this area. Nothing catastrophic happened, but neither did it go as smoothly as I recall when I first got the Bullet 10 months ago.


Roaster unresponsive and no display

I think I like my 515 even better after reading this. Thanks for checking it out. Look forward to hearing you had a successful roast.


As I have a v1.5 110v model I have stopped chasing firmware updates - I’m led to believe that most of the recent ones relate to the V2 model/boards.


It’s my impression there have been annoying issues with the first startup after f/w update for the last 4 or 5 (?) updates. Initially f/w updates were a breeze, but now… not so much. Other than one disastrous episode with simultaneous s/w & f/w changes I’ve been able to recover OK. Of course I may wear out the screw threads on the Control Panel! :slight_smile:

That’s my understanding as well, Stuart- fixes for the V2 power board. If I had 240 VAC available where I roast I’d likely have gotten the new board by now. Now I’m pretty happy I’m limited to the 120 VAC board. I’ll definitely get the V2 120VAC power board when it becomes available even though I have no compelling reason to roast 1 kg of greens at one time. Other than needing to clean the IBTS after 3 or 4 roasts (!), I’m pretty happy with the V1.5 Bullet and that won’t change by going to V2.



@bab - Are you on the V1.5? I am and had over 500 roast on mine and never had to clean the IBTS once. The fan failed on it and Aillio send me a new IBTS. Now I have to clean the sensor ever 3 or 4 roast as well.

It just seems super odd. Nothing changed other than the IBTS. I am dropping the same batch size of the same beans.

Any thoughts on why it might need cleaned so much?



Hi John-

Yes… V1.5. I have far fewer than 500 roasts. I probably had 70+ roasts of either 600 or 300 gm the first time I cleaned the IBTS and I saw no difference in IBTS reading at FC (& SC). A couple sessions later I cleaned it again thinking I hadn’t done a good job the first time and sure enough- I saw a significant difference in IBTS measurement after the second cleaning. I thought there was no difference in how I did the cleaning, but something had to be different. Neither the first nor the second time I cleaned did I actually see anything on the cotton swab, but there was a difference in how the IBTS measured after the second cleaning.

At first I first speculated evaporated oil from the drum was settling on the IBTS lens while the IBTS fan is off (Bullet in storage), so I started leaving the dump door prop’ed open when the roaster was being stored. I don’t see any difference… still see IBTS temp drift down at FC (& SC) with each roast.

I’ve been told by Aillio (Jacob) that they have seen returned IBTS fans (like yours I suppose) with smoke particles on the fan blades. They speculated there was exhaust smoke being picked up from outside the control panel housing via the air inlet. They wondered if a filter might be in order (I presume he meant over the air inlet grill).

Armed with that for background I started looking around for some filter material to kluge together a filter to cover the air inlet. That air inlet is very close to the dump door for the drum, so some smoke or maybe evaporated oil could be picked up from the still-hot beans as they drop into the cooling bowl. Whatever I pick has to not restrict air flow to the high-speed fan. My thought is to try some pleated paper from a larger air filter. I doubt mesh (like the ScotchBrite in the bean cooling tray) or open-cell foam would be effective.

I’m saying more than I really know… just expanding on what I picked up from Aillio a month ago.



Can you take some photos of the installation of the IBTS? Is it completely flush with the silicone adapter?


@jacob - I have checked for space between the hood and gasket from both inside and outside the roaster and it is tight against the gasket. No light comes threw looking at it from either side. I do have a lot of smoke build up on my fan though. This did not happen with my old ITBS.


Wow! No wonder I don’t see anything on the cotton swab when I clean the sensor! I was expecting to see the residue which accumulates inside the roaster but this looks the same as the white smoke that streams out the exhaust. Looks like @jacob was right that a filter might be appropriate or at least helpful. Must be getting picked up from the exit door… ??

Great job with the pic John.



@bab - Yeah, I don’t get what changed though and why the previous ITBS didn’t get effected the same way. It took 500 roast and then the fan failed, but I never once had to clean the sensor glass.

All I know is that I am starting to get really frustrated. 500 roast of the V1.5 working like a dream but for the last couple months its been terrible. It seems like the power level is super inconsistent. Some days its P9 on a 700g roast at a preheat of 590F and I keep it at P9 the entire way and can’t even get to first crack under 11 mins. Other days it’s like it’s got power to burn. But it is hard to roast consistent since the numbers each event happens at seem to change each roast.

I just want to Bullet to work like it worked for the first 500 roast again.


re: inconsistent roast time vs. power… do you keep track of ambient temp when roasting? My times can be all over the map, but then I roast outside so air drawn into the roaster is frequently different. I fully expect variation in time vs. FC. If you’re roasting indoors I would expect consistency unless there is variation in ambient temp.

Another variable might be power line voltage. Do you have a power monitor available? like a Kill-A-Watt? (There are a dozen or more to choose from.) Maybe there’s something going on with the mains. [ I live in an RV park so this can be a potential issue; thankfully we have decent power distribution here. ]

Then of course there’s the really troubling possibility of a problem with the power control board. So far I haven’t seen an issue with mine.

Btw, since you mention V1.5 I have to assume you have the 120 VAC power board… ?? I mention this only because of the issue Aillio just posted caused by some components for the new V2 power boards.



@bab - I can eliminate ambient temp/humidity as I control the temp and humidity in the room I roast in. I also keep my greens in there so they stay consistent. 18C @ 50% humidity. If for some reason it’s not, like a power outage, I wait for it to get back to that temp.

I do not track voltage but Aillio had asked me about that. My outlet is less than 10 feet from the panel and is all new wiring. The voltage drop at the outlet is insignificant.

It is interesting, in a way the dirtier the sensor, the less power it seems to have, but maybe it just feels that way because I have no clue when events are about to happen. I also think the roaster doesn’t heat up as hot once the sensor is dirty, although the bean probe temp is fairly consistent when dropping beans.

It is beyond me at this point. I have done everything I can do reduce variables and it is still really inconsistent. Maybe it just is what it is, but for some reason, its only been an issue since my first IBTS started to fail.


Just had one last thought…

How are you venting your Bullet? Could it be that your venting solution is adding some extra variables?


Venting with a 3 inch flex pipe. Same as I have from the start. Aillio was asking the same thing. I clean the pipe one a month. I am a fireman so the last thing I want to do is be the guy responding to my own call. That pipe is clean, as well as the entire ventilation system. Fan, filter and the fan housing all get cleaned well within the recommended amount of roasting.

After Aillio brought it up, I did however try a couple roast where I just let it vent out the window, it made no difference the the IBTS getting dirty. It looks pretty clear to me that the smoke is coming from the beans in the cooling tray. What I still don’t get is why my first one wasn’t effected the same way.


Thank you for your service to your community. Yes, that would not play well on the evening news to answer your own call!

I understand your frustration… My time frame is much smaller than yours but I didn’t notice the IBTS getting dirty for about 4 months. And when I clean it I see nothing on the swab. Now at least I know why (white smoke particles on a white swab).

A couple oddball questions-

  • What does your air supply for the roaster consist of? Any chance of some blockage there? If you don’t have a free supply of air at or above the ambient pressure near the outside exhaust, then the air being vented has to come from somewhere or else the exhaust fan output will be throttled to some extent.
  • I think I have this right: I believe Drum Temp/IBTS measures the final value of PreHeat. The Bean Temp has to have stabilized as well, but the final value is IBTS. If I have that right, that means a dirty IBTS will force the drum to a higher temp than you have set via PRS (since IBTS reads low when dirty). Perhaps that’s stressing the power board and causing it to limit power output… ?? Aillio probably needs to address this one.
  • I thought your photo made it clear the smoke is coming through the inlet grill in the base of the control panel, somewhere near the bottom. I jumped to that conclusion because there’s more smoke residue on the lower part of the fan, but could there be another path? Like past the wires that lead in from the drum area. Seems unlikely, but I don’t know what steps were taken to block that path.

Any residue that might be coming from the area of the cooling tray I had thought would be like the residue on the back of the front panel (chaff debris + oil residue… dark brown gritty goo; like on the filter in the cooling tray). The only white smoke I see while roasting has been from the exhaust. [And I thought that was nearly all water vapor… oops!] Have you seen white smoke from the beans when they drop? I haven’t seen any and I roast pretty dark- FC+ or Vienna. Plus I think the air flow from the cooling tray fan is much stronger than the little micro-fan in the control panel.

Btw, I was initially roasting 300g batches. I was seeing downward drift in IBTS with those small batches. Now that I’m doing over 500g batches I see the IBTS drifting down sooner. More beans should mean more smoke which should mean dirty sooner which is what we both see. But that does nothing to explain why you had 500 happy roasts.



Sorry for the late reply, I don’t know if you have been in touch with Matt?
You can check the IBTS fan RPM in the info panel in RoasTime. If you download the latest beta of that and do some roasts then we can check the logs to see if we can find anything out of the ordinary.


@jacob - IBTS fan RPM is sitting a hair below 15000 during roast and a hair over that during preheats.

I will update to the latest beta.

Any thoughts on what’s going on?

I have chatted with the Aillio support team. They haven’t gotten back to me on the fan. They are telling me that they see my machine not getting full voltage for anytime I have it tested, it comes in over 120v at the plug. Its all new wiring and I am less than 10 feet from my panel so there is very little voltage drop.


Have you measured the voltage while you’re roasting? That will tell you more than the “static” voltage.

If you’ve got a normal duplex outlet that you’ve plugged your Bullet into, you can simply measure the voltage on the other receptacle, while the roast is going.

The best data point would be with the Bullet at P9/FB/D9 - that’s everything on full blast. Under those conditions, you should not see an appreciable voltage drop. If you do, check to see what else may be on that breaker.


ok, i will try that. Thanks @celticcupcoffee


As @celticcupcoffee pointed out, please test in roast mode with P9 - keep it there for 20seconds, and check in RoasTime what the net voltage is.