nicolai wrote: I know the feeling, I love roasting and observing how things work out, but my friend brought over about 3 kg's of roasted beans from his last training, so trying to get through those - In the mean time no roasting for me...

As for the experiments it could also be good to investigate dose and fanspeed - I.e. bigger dose, higher fanspeed? And can it be the reason why some people are facing slightly more uneven roasts at higher dose?

As for point 3: I think flick and crash is almost irrelevant in an airbed: First of all thermal mass of roaster is extremely small and let us assume 1st crack does create a drop in temp the applied energy would very quickly be able to recover.
Even in conventional drum roasters it is still discussed if a flick or crash has any significant impact to flavor - Although that there is a high level of emphasis in many of the "bibles" of coffee roasting.

Let me know if you find some teardown information on the machine - I will be ordering a Yoctopuce thermocouple and the electronics are so small that I should be able to find a smart place to mount the controller inside the enclosure of the roaster.

Hi GDM528,

If you do not mind I can tear down my Ikawa home 100g and share the picture for those who may be interested in this post. Since mine is not working last month and went through the email communication for almost one month period. It turns out they judge mine is failing to stand with the factory warranty anymore. So I plan to figure out what part is causing the problem and maybe I am able to bring it back to life again.

GDM528 wrote:Are you a fan of iFixit teardowns? Let's use this topic to post or gawk at what goes on inside the Ikawa roaster.

To be clear, not advising anyone to void their warranty. But sometimes *stuff* happens and some people are just gonna hack their gear. Posting here may provide enough information for others to decide if they want to give it a go - or just leave it as-is.

Like many (most?) home appliances, the Ikawa can inflict serious injury, and rooting around inside takes it to a whole other level. To that end, posts that point out safety issues, like high-voltage nodes and thermal hazards would be much appreciated.

Moderator HB merged the suggestion above from GDM528 into this first post for easier reference.

Jiduochou wrote:Hi GDM528,

If you do not mind I can tear down my Ikawa home 100g and share the picture for those who may be interested in this post. Since mine is not working last month and went through the email communication for almost one month period. It turns out they judge mine is failing to stand with the factory warranty anymore. So I plan to figure out what part is causing the problem and maybe I am able to bring it back to life again.

Sorry to hear about your issues, especially so soon after purchase. I can't advise you either way about performing a tear-down - that's your decision. But if you do decide to go for it, I've examined how the machine might be disassembled and made a few observations:

1) They use a combination of steel screws into threaded aluminum holes. The aluminum is much softer than the screws, and is prone to sticking/fouling/jamming. So be careful and note that removing a screw could be destructive to the aluminum threads.

2) There are four T15 screws under the glass top cover. Those screws adjust the height of the black foam layer so the glass can rest on the foam layer rather than the exterior shell.

3) There are four T10 screws on the bottom plate of the machine that allow access to the filter screen for the air intake. When the exterior base plate is removed there are four more T10 screws on yet another plate, one of which is hidden under a foam sheet.

Not confirmed, but I suspect both the top and bottom sets of screws have to be removed for the internals to slide out. The power input socket might interfere with removing the internals, but should be apparent once the bottom covers are removed.

Best of luck!

Thank you for providing the guideline. Since Ikawa considers the factory warranty to be invalidated and the repair process shall be charged, I plan to do the disassembly before the unit is sent to local technicians for repair. Sometimes learning how things are constructed is kind of fun and might learn something from it.



The bottom plate is removed, you may see the fan



Fan removed but connected, prior carry out the 1) and 2) all the electrical connects (the main switch, start button, and GND) must remove



Pull out from the top carefully since heat insolation foam provides resistance.



AC-DC Power supply, two customized PCB board( green/red)



Fan ASSY.

FYI.

Thank you for posting those pictures - very, very helpful! I have the same screwdriver kit as you! I got mine from Ifixit. Anyway...

So just to confirm the disassembly sequence:
1) Remove base plate.
2) Remove standoffs.
3) Lift out fan assembly.
4) Disconnect wires.
5) Remove top screws and slide out top of shell.

I see a dangling yellow/green ground wire that doesn't seem to have a connecter on the loose end. Where was it attached?

Were there any installation stickers you had to break during disassembly?

If it's not too much trouble, close-ups of the red and green printed circuit boards would be informative. I speculate it's the same board used in the Pro model, and as such would have unused connectors for another thermocouple and push-button.

I'm surprised to see how they put the electronics directly in the air path, almost to the point of blocking the airflow, except for what looks like a 5-10mm gap between the edge of the PCB and the walls of the shell. Seems a bit crazy, but that must be a really powerful fan. The red PCB is just a couple centimeters from a heating unit that gets hot enough to melt the components off a PCB - so that's a bold move, if the fan were to fail mid-roast it could literally melt down

Good news is it does look like there is a bit of room inside for stuffing in additional electronics, like another thermocouple reader.

Thanks again for posting!

It is great to hear that those pictures are helpful.
Yes, the sequence you listed is correct and it may more clear with the following pictures.
On the other hand, I do not notice there has any seal presenting while disassembly the unit, in my case.
I feel the same way you do, regarding the red PCB that may be used on the Pro version as well, and yet I have no Pro, so it is only speculation.
Regarding the red PCB just underneath the heating element, it also surprise me as well and realized that the cooling bean process is critical and immutable. You might notice the layout of the circuit route by viewing the red PCB. If any questions just let me know I try to explain in English.
Since all the parts have been assembled before being shared, I plan to disassemble and take more pictures on weekend, if you are curious about some parts just let me know.

GDM528 wrote: I'm surprised to see how they put the electronics directly in the air path, almost to the point of blocking the airflow, except for what looks like a 5-10mm gap between the edge of the PCB and the walls of the shell. Seems a bit crazy, but that must be a really powerful fan. The red PCB is just a couple centimeters from a heating unit that gets hot enough to melt the components off a PCB - so that's a bold move, if the fan were to fail mid-roast it could literally melt down

The location makes sense. Having the fan blow directly on the heating element might cool it enough to severely limit the heat output. Also, a direct blast of air might blow up too high, even at low fan settings. Conceivably, they would have had to put a shield in front of the fan anyway, so why not use the PCB? Only cold air blows on it, which has the added benefit of keeping the electronics cool.

As for the fan failing, it wouldn't take much for the circuitry to monitor the fan current and shut down the heater if the fan fails.

Jiduochou wrote: I plan to disassemble and take more pictures on weekend, if you are curious about some parts just let me know.

Thanks for the offer!

I think the green PCB is the 'brains' of the machine. There's a chip on the green PCB, next to the wireless module, that looks like an ARM processor. A close-up photo of the chip markings can confirm that. The chip markings are often laser-scribed, which are very sensitive to lighting angle to be readable.

The markings on the other chips on the green PCB would also be helpful to know. I can review the chip datasheets to get a better idea of how the control system works - and if it can be hacked to add another thermocouple input.

Also suggest getting product markings of the big electrolytic capacitors. I see one of them is 3300uF and 35V, but I don't see the temperature rating or manufacturer. It's interesting because electrolytic capacitors have very short lifespans at elevated temperatures.

I've started a new topic thread just for teardown discussions, as this topic was specifically meant for the thermal performance - I suggest posting there if you're OK with that: Ikawa Teardowns

MODERATORS: if you're reading this, can you move posts 69 - and up to the teardown thread? If so, thank you!

Peppersass wrote:As for the fan failing, it wouldn't take much for the circuitry to monitor the fan current and shut down the heater if the fan fails.

Definitely. Without knowing the specific chip part numbers I'm just spitballing, but it looks like the four chips on the red PCB are a fairly elaborate motor driver/controller that can keep running even if there's a code-crash. It probably includes temperature alarms to keep the fan running until sufficiently cooled and will shut down the heater power if a fan fault is detected.

Based on Jiduochou's pictures so far:

The closest match to the motor I've found is here:
https://www.wonsmartmotor.com/portable- ... r-product/

Part number not an exact match, and the last four digits "IK01" hint that it might be custom-built for Ikawa. Unfortunately, that implies only Ikawa could sell a replacement - and good luck getting them to agree to that.


The main power module does appear to be off-the-shelf:
https://www.skynetpower.com.tw/product. ... =200&id=21

The power module supplies the 24V for the blower motor. It's rated to 130W with forced air cooling, of which there'll be plenty of that, given a blower is right next to it

Looking at the motor specs, it appears able to draw more power than the power module can supply, which implies Ikawa doesn't have to push the motor too hard - that should lower stress on the motor and extend its lifespan.

GDM528 wrote:Thanks for the offer!

I think the green PCB is the 'brains' of the machine. There's a chip on the green PCB, next to the wireless module, that looks like an ARM processor. A close-up photo of the chip markings can confirm that. The chip markings are often laser-scribed, which are very sensitive to lighting angle to be readable.

The markings on the other chips on the green PCB would also be helpful to know. I can review the chip datasheets to get a better idea of how the control system works - and if it can be hacked to add another thermocouple input.

Also suggest getting product markings of the big electrolytic capacitors. I see one of them is 3300uF and 35V, but I don't see the temperature rating or manufacturer. It's interesting because electrolytic capacitors have very short lifespans at elevated temperatures.

I've started a new topic thread just for teardown discussions, as this topic was specifically meant for the thermal performance - I suggest posting there if you're OK with that: Ikawa Teardowns

Thank you for sharing the information regarding the green PCB.
As mentioned priorly, I will try to take photos to help ident. the IC the PCBs during the weekend.
It is a good choice to separate another new post for the teardown.