Regardless, I do not know what you ordered or feom where but it was not from iMike78 projects as those PCB designs are not ever ready so can not be ordered:
If you ordered Onju Voice PCBs the then you really did not do your homework researching this and you are actually off-topic posting about the original Onju Voice PCB in this thread. Sorry to say, but you messed up if you ordered a Onju Voice PCB today as the old project has not outdated for years now.
@tmjpugh Sorry I meant to address @AtssAgency who wrote ”l just ordered 4 pcs” (which was part of the sentence you replied too).
My reply which was meant to address @AtssAgency was that he can not simply order the PCBs that iMike78 is working on because the design is not ready.
But if he did actually order some PCBs today then I assume be probably ordered the old ”Onju Voice” PCB, in which case I would just like to point out that is off-topic since this thread and discussion is not about the old ”Onju Voice” PCBs.
I just found this thread today. I have about 10 Minis that I would be interested in upgrading, about 50-50 split between Gen1 and Gen2 units.
We mainly use them currently for music (Chromecast through our home music server, mobile music players and now Music Assistant). I want to use them more for voice control and status messages, which is what led me here.
I’ve seen the Github repositories and await further progress news with baited breath.
I just found this project today. I’m super excited about it! I also have a mix of about 10 minis, some gen 1 some gen 2. I also have some nest audios, hub 1 hub 2 and hub max 2.
With google failing hard I’m keen to repurpose the hardware.
Can I suggest for a future iteration (with no current understanding of the hardware differences between mini gen 1 and gen 2 ) that you consider merging the projects to use one board that is powered by USB-C, ditching the currrent barrel plug/micro-usb power solutions? To my mind, that will simplify the design and lower the costs for you.
If you could also achieve ethernet connectivity through the same connection (like a chromecast ultra does), that would be amazing!
Wishlist almost complete! Lol
If you could acheive full poe ethernet, so it can be powered and wall mounted with just an ethernet cable, that would be pure bliss. Now imagine you add wifi antennas and the ability to use them as a wireless access point with ethernet backhaul. I’m drooling… lol
I haven’t read the comments section to see if any of this has been mentioned already, and I know right now you’re just attempting to get the original functionality replicated, but the potential is huge for a really cool upgrade in my opinion
Great work so far! I’d love to offer to help but I’m just a lowly sql hack so my learning curve will probably tae longer than your project. Lol
@iMike78 I’ve been working on a similar effort at https://sett.homes for the Nest Thermostat, I’d be interested to help with the PCBs if you’re still looking for help
@z3ugma Nice! Having more eyes on designs, schematics, routing, shielding
strategies
for EMI/EMF/RMI considerations, etc. would be great so hope that you can be of help. Be sure that you check out iMike’s two status posts/threads in the related discussions sections in both of the GitHub repositories here:
Hi @z3ugma, I’m really glad you found the project — and great job on the Nest Thermostat, seriously impressive work.
You can check the topics @Hedda linked for the full picture, but here’s a summary of where things stand:
First test batch (Nest Mini): one critical pin was left unconnected. Because of that, the whole batch can only function as a Wi-Fi smart speaker — nothing more.
Second test batch (Nest Mini): electrically it was theoretically perfect, but the XMOS stops during boot and never comes up. Unfortunately, I hadn’t added any test points or an XTAG connector on that version, so I couldn’t investigate the root cause.
Third test batch (first Home Mini version): currently in manufacturing. I switched to the Home Mini board because that design was already half-complete and does include proper test points and an XTAG connector. During routing I ended up restarting the whole layout from scratch, which is why it took so long (besides some personal issues along the way).
So for now, all we can do is wait and hope this batch works. In the meantime, feel free to look through the project files if you’re curious.
@appocalypse
Regarding a single PCB for both devices: the Nest Mini and Home Mini are physically different, so a universal board isn’t possible.
For the USB-C suggestion: the main goal of the project is to reuse the original Google speaker enclosures without any physical modification, so switching to USB-C will never be an option.
As for extras like PoE — the board is already extremely dense, so that’s practically impossible.
Thanks to all of you for the support. I really appreciate it. I was close to giving up at one point, but all my hopes are now on this upcoming batch.
FYI, iMike78 posted an update saying that he recieved the first “Home Mini v1 test batch” (initial physical circuit board prototype test recieved from PCB manufacturer) and the hardware itself seems sound (pun intended) so ESPHome can be flashed to the ESP32 and working fine, but he is however having some problems with the XMOS firmware and is requesting help here → Status update · iMike78/home-mini-v1-drop-in-pcb · Discussion #2 · GitHub
Would be awesome if some embedded developer with experince of the XMOS firmware and XTAG4 (XMOS JTAG debugger) could reach out to him and offer advice or suggestions so do not need to work in the blind:
) that you consider merging the projects to use one board that is powered by USB-C, ditching the currrent barrel plug/micro-usb power solutions? To my mind, that will simplify the design and lower the costs for you.