RTC alarm not working

[mipek]

I got my Watchy today but to my surprises the time is not updating. Upon further investigation it seems like no RTC alarm is ever received (ESP_SLEEP_WAKEUP_EXT0). What should I do?

[hn3000]

+1 for esp32 standard library -- the current workaround has several flaws, among them the fact that button presses throw off the time keeping, we use the RTC chip to store the time without knowing whether it's just running slow or completely stopped and the way we wait does not take into account time spent drawing the watch face and whatever else is done between wake up and going back to deep sleep.On top of that, none of the available clock sources seem very accurate, but the runtime system can calibrate the slow clocks using the (way more precise) 40MHz clock -- and we're not making use of that ability.

[SnoopJ]

+1 for esp32 standard library -- the current workaround has several flaws, among them the fact that...wait does not take into account time spent drawing the watch face and whatever else is done between wake up and going back to deep sleep.

Agreed. In general I think Watchy could be doing more of the housekeeping parts of all wakes (particularly the time keeping) in a deep sleep wake stub to avoid the menu issues. I suspect you're right about the timing as well, although I think the stdlib approach likely makes these fade away.

On top of that, none of the available clock sources seem very accurate, but the runtime system can calibrate the slow clocks using the (way more precise) 40MHz clock -- and we're not making use of that ability.

CLK_8MD256 seems pretty stable to me, if I gain/lose it appears to be a consistent drift which is probably because the offset isn't chosen in a particularly nuanced way (I wrote some logic to let me adjust it on-device and experimented across a few days). However, if there's an even better clock on-board, that would be cool to play with!

[PyroDevil]

@sqfmi: Is there some update on how we can send you the watch back in order to fix the RTC?

[robertimorrison]

I think I've finally addressed the DS3232 RTC interrupt-driven alarm wakeup for the ESP 32 -- I struggled with the issue for several days and could never get ALARM_2 to fire reliably every minute (at 00 seconds) to wake the ESP and update the display.

In watchy.cpp in _rtcConfig the original sqfmi github code does NOT enable ALARM_2 with the following (github code is MISSING this statement):

RTC.alarm(ALARM_2);

SO the full set of statements to get the RTC wakeup interrupt to fire from _rtcConfig() is as follows:

RTC.setAlarm(ALM2_EVERY_MINUTE,0,0,0); // Set ALARM_2 to fire every minute at 00 seconds RTC.alarm(ALARM_2); //Enable ALARM_2 -- ORIGINAL CODE NEVER Enabled the alarm! RTC.alarmInterrupt(ALARM_2, true); //enable ALARM_2 to fire the interrupt pin when it goes off RTC.squareWave(SQWAVE_NONE); //disable square wave output

My toolchain is the visualmicro add-in for Visual Studio which I believe is just a front-end for the Arduino IDE (I have the beta 2.0 Arduino IDE installed). I suspect the sqfmi are using some other toolchain which for whatever strange reason does not require the RTC.alarm statement to actually set/enable the alarm.

I also added RTC.begin() at the start of _rtcConfig, which the author of the DS3232 library (Jack Christensen) says initializes the I2C bus which allows the RTC to communicate with the ESP; wire.begin(SDA,SCL) up in the init method probably accomplishes the same thing, but adding RTC.begin does not appear to adversely affect anything

The maxim DS3232 RTC claims good accuracy even without a quartz oscillator - 2.5ppm over 0 to 40C, which should be better than 1 sec / day. I dont know if it actually achieves this.

Before discovering the fix to enable the enable the RTC alarm, I was using the ESP timer to wake the ESP32 and also observed the large drift in wakeup times, resulting in inaccurate time display as reported by others. I think the RTC is actually keeping accurate time, but the wakeup drifts quite a bit resulting in an inaccurate display.

[alicemyslime]

Sadly, that doesn't fix anything for me - you probably had a functional chip, whereas mine doesn't even run the oscillator

[truedat101]

@aliceafterall @PyroDevil same ... my device appears to not keep time unless plugged in. I don't want to talk about RMAs yet, but I would first like to see if the team is offering a solution. I like the idea of this device. Is there revision # or something that can help us identify the bad batch ?

[hn3000]

Like for @aliceafterall , this doesn't change anything in my case.

[wkumari]

Yes, I think that I’m also in the same boat.

I also filled in the Google form to buy a new screen and screen protector, but haven’t heard anything back; I don’t actually need the screen - I’ve got dust under my screen protector, and could not see an option to just buy a few screen protectors, so…

W

On Thu, Jul 1, 2021 at 6:35 PM hn3000 @.***> wrote:

Like for @aliceafterall https://github.com/aliceafterall , this doesn't change anything in my case.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/sqfmi/Watchy/issues/40#issuecomment-872590738, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABDG3XEQFCMGRRIOORYVQZDTVTUSBANCNFSM46QVRURQ .

-- Perhaps they really do strive for incomprehensibility in their specs. After all, when the liturgy was in Latin, the laity knew their place. -- Michael Padlipsky

[SnoopJ]

@aliceafterall @PyroDevil same ... my device appears to not keep time unless plugged in. I don't want to talk about RMAs yet, but I would first like to see if the team is offering a solution. I like the idea of this device. Is there revision # or something that can help us identify the bad batch ?

The simplest way to verify the problem right now is the rtcTimeTemp.ino example discussed throughout this thread. SQFMI mentioned something about users being able to verify that their device is faulty in a forthcoming firmware release.

As for batch identifiers, the chips are labeled with two fields (see this image), whose semantics are unknown to me, but bad chips do appear to have these identifiers in common. These are the ones that have been reported on Discord by users with bad RTCs:

1613A3 009AC
1221A3 007AA
1608A3 790AB
1611A3 597AD

[PyroDevil]

@SnoopJeDi: Well my RTC is one of those. 1611A3 597AD

[SnoopJ]

@SnoopJeDi: Well my RTC is one of those. 1611A3 597AD

I have the same, and my RTC does not work. I suspect the problem applies to entire batches of chips.

[hn3000]

Mine is 1612A3 009AB and also doesn't work. It sits between the last and the first entry (1611A3 xxxxx and 1613A3 xxxxx).

[wkumari]

Just like @hn3000, mine is 1612A3 009AB

... and, unsurprisingly: 00:00:00 01Jan2000 0.00C 32.00F 00:00:00 01Jan2000 0.00C 32.00F 00:00:00 01Jan2000 0.00C 32.00F 00:00:00 01Jan2000 0.00C 32.00F 00:00:00 01Jan2000 0.00C 32.00F < unplug battery> 00:00:01 01Jan2000 48.25C 118.85F 00:00:02 01Jan2000 48.25C 118.85F 00:00:04 01Jan2000 48.25C 118.85F 00:00:06 01Jan2000 48.25C 118.85F 00:00:08 01Jan2000 48.25C 118.85F 00:00:09 01Jan2000 48.25C 118.85F 00:00:11 01Jan2000 49.25C 120.65F 00:00:13 01Jan2000 49.25C 120.65F 00:00:15 01Jan2000 49.25C 120.65F < reconnect battery> 00:00:15 01Jan2000 49.25C 120.65F 00:00:15 01Jan2000 49.25C 120.65F 00:00:15 01Jan2000 49.25C 120.65F 00:00:15 01Jan2000 49.25C 120.65F 00:00:15 01Jan2000 49.25C 120.65F

[PyroDevil]

@sqfmi: It now has been over 4 weeks since you stated that you are setting up a process for the people to get their boards repaired or replaced. How is it going?

[a4x4kiwi]

Hi. I just receive my watchy. Same RTC issue unfortunately. Batch RTC DS3231M chip 1611A3 597AD

Power from USB only. 10s update interval. 00:00:11 01Jan2000 46.25C 115.25F 00:00:12 01Jan2000 46.25C 115.25F 00:00:13 01Jan2000 46.25C 115.25F

Battery & USB connected 00:00:00 01Jan2000 0.00C 32.00F 00:00:00 01Jan2000 0.00C 32.00F 00:00:00 01Jan2000 0.00C 32.00F

[JordyB16]

@aliceafterall @PyroDevil same ... my device appears to not keep time unless plugged in. I don't want to talk about RMAs yet, but I would first like to see if the team is offering a solution. I like the idea of this device. Is there revision # or something that can help us identify the bad batch ?

The simplest way to verify the problem right now is the rtcTimeTemp.ino example discussed throughout this thread. SQFMI mentioned something about users being able to verify that their device is faulty in a forthcoming firmware release.

As for batch identifiers, the chips are labeled with two fields (see this image), whose semantics are unknown to me, but bad chips do appear to have these identifiers in common. These are the ones that have been reported on Discord by users with bad RTCs:

1613A3 009AC
1221A3 007AA
1608A3 790AB
1611A3 597AD

I'm still trying to figure out how to run this. And where the output goes? Cant seem to find any log. Can you provide me with a small step by step? Open Arduino, upload rtcTimeTemp.ino? And then? Thanks in advance

On topic, all workarounds seem to work as per time updating but i keep losing some minutes in a day and have to set the time every 48 hours

[PyroDevil]

@JordyB16: Maybe the step you are missing is opening the 'Serial Monitor' (On Windows Ctrl+Shift+M or from the Tools menu) after uploading the rtcTimeTemp project?

[wkumari]

I'm still trying to figure out how to run this. And where the output goes? Cant seem to find any log. Can you provide me with a small step by step? Open Arduino, upload rtcTimeTemp.ino? And then? Thanks in advance

Yup, the instructions on this were not particularly clear.

Steps:

1. Get rtcTimeTemp.ino from https://github.com/JChristensen/DS3232RTC/blob/master/examples/rtcTimeTemp/rtcTimeTemp.ino
2. Paste it into the Arduino IDE.
3. Upload to device.
4. In the Arduino UI, go to Tools -> Serial Monitor. Make sure that Autoscroll and Newline are selected, and set the baudrate to 115200 baud (these instructions from a Mac, but the menu entries should be the same on other OS as well!)

Stats should start scrolling every 10 seconds or so...

Hope this helps.

[JordyB16]

I'm still trying to figure out how to run this. And where the output goes? Cant seem to find any log. Can you provide me with a small step by step? Open Arduino, upload rtcTimeTemp.ino? And then? Thanks in advance

Yup, the instructions on this were not particularly clear.

Steps:

1. Get rtcTimeTemp.ino from https://github.com/JChristensen/DS3232RTC/blob/master/examples/rtcTimeTemp/rtcTimeTemp.ino
2. Paste it into the Arduino IDE.
3. Upload to device.
4. In the Arduino UI, go to Tools -> Serial Monitor. Make sure that Autoscroll and Newline are selected, and set the baudrate to 115200 baud (these instructions from a Mac, but the menu entries should be the same on other OS as well!)

Stats should start scrolling every 10 seconds or so...

Hope this helps.

Yes, this worked. Sadly enough it just confirms my faulty rtc chip.

[TitaniumCoder477]

Unfortunately, my watch is experiencing the same symptom as many others on this post. Here are my results:

... 00:09:15 01Jan2000 49.50C 121.10F 00:09:16 01Jan2000 49.50C 121.10F 00:09:17 01Jan2000 49.50C 121.10F 00:09:19 01Jan2000 49.50C 121.10F 00:09:20 01Jan2000 49.50C 121.10F 00:09:21 01Jan2000 49.50C 121.10F 00:09:22 01Jan2000 49.50C 121.10F 00:09:23 01Jan2000 49.50C 121.10F 00:09:25 01Jan2000 49.50C 121.10F 00:09:26 01Jan2000 49.50C 121.10F 00:09:27 01Jan2000 49.50C 121.10F 00:09:28 01Jan2000 49.50C 121.10F 00:09:30 01Jan2000 49.75C 121.55F 00:09:31 01Jan2000 49.75C 121.55F 00:09:32 01Jan2000 49.75C 121.55F 00:09:33 01Jan2000 49.75C 121.55F 00:09:35 01Jan2000 49.75C 121.55F 00:09:36 01Jan2000 49.75C 121.55F 00:09:37 01Jan2000 49.75C 121.55F 00:09:38 01Jan2000 49.75C 121.55F 00:09:40 01Jan2000 49.75C 121.55F ...

Does this mean I have a faulty RTC too? What is the official recommendation at this point?

UPDATE 7/17/21 Symptom 1: Frozen face: I've tried multiple watch faces with same results. The faces do not update or refresh, except when I got back into the menu. Symptom 2: The right side buttons only work in the menu. They do nothing on any of the faces. Symptom 3: With battery disconnected and USB connected, Watchy falls behind rapidly. For example, ten minutes ago I synchronized the time with my non-smart watch. And now Watchy is 4 minutes behind. So that's roughly 24 seconds lost per minute. Symptom 4: Temp seems way off

rtcTimeTemp output over 2 minutes, with battery disconnected and USB connected:

... 10:08:40 17Jul2021 50.50C 122.90F 10:08:51 17Jul2021 50.50C 122.90F 10:09:02 17Jul2021 51.00C 123.80F 10:09:13 17Jul2021 50.00C 122.00F 10:09:24 17Jul2021 50.75C 123.35F 10:09:35 17Jul2021 50.75C 123.35F 10:09:46 17Jul2021 41.50C 106.70F 10:09:57 17Jul2021 51.25C 124.25F 10:10:08 17Jul2021 51.25C 124.25F 10:10:19 17Jul2021 51.50C 124.70F 10:10:30 17Jul2021 52.25C 126.05F 10:10:41 17Jul2021 51.75C 125.15F 10:10:52 17Jul2021 52.00C 125.60F ...

The temp in my house is 75F.

Symptom 5: With battery connected and USB connected, Watchy seems completely frozen in time. After 5 minutes, time did not advance at all. Symptom 6: Also with battery connected and USB connected, face and button symptom 1 and 2 above remain the same. I.e. the menu works but nothing on the face works or updates.

rtcTimeTemp output over 2 minutes, with battery connected and USB connected:

... 10:26:33 17Jul2021 50.50C 122.90F 10:26:34 17Jul2021 50.50C 122.90F 10:26:36 17Jul2021 50.50C 122.90F 10:26:37 17Jul2021 50.50C 122.90F 10:26:38 17Jul2021 50.50C 122.90F 10:26:39 17Jul2021 51.25C 124.25F 10:26:41 17Jul2021 51.25C 124.25F 10:26:42 17Jul2021 51.25C 124.25F 10:26:43 17Jul2021 51.25C 124.25F 10:26:44 17Jul2021 51.25C 124.25F 10:26:46 17Jul2021 51.25C 124.25F 10:26:47 17Jul2021 51.25C 124.25F 10:26:48 17Jul2021 51.25C 124.25F 10:26:49 17Jul2021 51.50C 124.70F ...

Symptom 7: Finally, with battery connected and USB disconnected, Watchy seems completely frozen in time. After 5 minutes, time did not advance at all.

I should also mention that the vibration motor, accelerometer, and wifi work at least in the menu. The battery shows 3.33V, which appears to be odd since it is stamped 3.7V.

It sure would be great to have some guidance on how to fix this watch. As one can imagine, it was quite disappointing to look forward to my Watchy for many months, only to find it essentially unusable out of the box.

[JordyB16]

I'm wondering if @sqfmi is working on a fix for customers who received a Watchy with a faulty RTC? Right now its just lying on a shelf because its unusable

[a4x4kiwi]

Hi. I just receive my watchy. Same RTC issue unfortunately. Batch RTC DS3231M chip 1611A3 597AD

Power from USB only. 10s update interval. 00:00:11 01Jan2000 46.25C 115.25F 00:00:12 01Jan2000 46.25C 115.25F 00:00:13 01Jan2000 46.25C 115.25F

Battery & USB connected 00:00:00 01Jan2000 0.00C 32.00F 00:00:00 01Jan2000 0.00C 32.00F 00:00:00 01Jan2000 0.00C 32.00F

I replaced my RTC chip and it now seems to be working. I have to wonder if the bad batches are counterfeit. I see they are very scarce from Element14, Mouser etc. If you are in Australia, the silicon chip magazine shop has them. I have one spare I can sell within Australia or NZ if anyone wants one.

I have attached pictures of the removed device and a new working device. The new one is in the packaging.

[TitaniumCoder477]

I have contacted SQFMI through their contact form and hopefully will receive an action plan shortly. The lack of attention to this issue is starting to concern me. This was my first experience with CrowdSupply.com as a source of innovative products, and so far I am only regretting my support of the product since it proved to be defective on delivery.

[vanEgmond]

The current issue is sourcing new RTCs and finding out the bad batches. SQFMI's communication could definitely be better, but that would not solve the underlying problem of defective supply for which sqfmi can't really be blamed. Stuff like this sometimes happens sadly.

There are workarounds regarding the RTC so that it's at least functional, and more sophisticated ones are in development by community members. I understand the frustration, but this should not be any more frustrating than a 1-2 month delay, which is extremely common everywhere right now because of chip and supply shortages.

I get that there is currently a large amount of FUD, but give them a chance, and contact sqfmi on Discord when in doubt. Otherwise, get a refund from Crowdsupply, and wait for a consumer or improved version. It's not like your money is gone, there's just unmet expectations.

[TitaniumCoder477]

@vanEgmond Did you delete my post? That would be super unprofessional. You say you get my frustration and that there are "workarounds," but I did not sign up for resoldering. I merely wanted to code for the Watchy device, i.e. hack my own little watch. That's all I wanted, and what I paid good money for. You say "it's not like your money is gone, there's just unmet expectations," but that's not true. My money is gone from my hand, and what I have in return is a defective product. I am simply asking SQFMI to make it right. And you (or someone else) deleted my post about this, probably because you don't want others to start feeling the same way. I have copies of your email and screenshots of this post, and if you do it again, I will raise complains to CrowdSupply, not only for the defective product but also for the treatment of paying customers.

UPDATE: I'm seeing my post now, and your reply. Odd that refreshing the web page would not show either, but adding a new reply does.

[truedat101]

The current issue is sourcing new RTCs and finding out the bad batches. SQFMI's communication could definitely be better, but that would not solve the underlying problem of defective supply for which sqfmi can't really be blamed. Stuff like this sometimes happens sadly.

There are workarounds regarding the RTC so that it's at least functional, and more sophisticated ones are in development by community members. I understand the frustration, but this should not be any more frustrating than a 1-2 month delay, which is extremely common everywhere right now because of chip and supply shortages.

I get that there is currently a large amount of FUD, but give them a chance, and contact sqfmi on Discord when in doubt. Otherwise, get a refund from Crowdsupply, and wait for a consumer or improved version. It's not like your money is gone, there's just unmet expectations.

Crowd supply has suggested that they are working with SQFmi on the solution, but it doesn't appear they are actually honoring 30 day returns or whatever their policy is.

[wkumari]

I also ended up replacing the chip; I didn't want to place a separate Digikey/Mouser order and pay shipping, so I just bought https://www.amazon.com/gp/product/B085NQ9B8Y from Amazon, scavenged the chip and replaced the RTC on the Watchy.

I really don't mind the fact that there were issues, and I don't mind having done the work myself, but I am frustrated by the lack of communication. A simply "Oh, woops. Apologies, we'll try take care of this" or even "We got bitten by a bad batch, we are looking into it" or "That ended poorly, but we don't have enough funds left to fix this..." would all have been much much better...

[SnoopJ]

You say you get my frustration and that there are "workarounds," but I did not sign up for resoldering.

The workaround posted by SQFMI and my variation on the same do not require any soldering.

[SnoopJ]

Information from SQFMI

PSA: SQFMI are a lot more active on the Watchy Discord than here on GitHub. They are very aware of this issue. For better or worse, Discord is where the latest information on this issue is being posted, and I would strongly recommend you check there as well if you are interested in real-time updates.

Sharing below the Discord messages I consider relevant to this issue in reverse chronological order, mentions have been disabled and all formatting is my own:

20 July 2021

lee: It also really sucks my RTC is broken in mine lee: @.SQFMI is there any updates for this? SQFMI: still trying to source RTCs at the moment, but likely we will have to service them rather then send you a new board, since most folks probably already stuck on the screen. We can also send out RTC ICs for those willing to do it themselves, but again, no chips at the moment ... snip ... lee: @.SQFMI is it really too much just to send new Watchys (just screen + pcb + protector and nothing else) ? SQFMI: i believe that would be the majority of the cost lee: Surely it's cheaper than the shipping costs, labor costs, all combined? SQFMI: only if we were able to make more, then maybe SQFMI: but we can't at the moment because, no chips lee: I think that is the best thing to do - wait until you can make more lee: Otherwise it's going to be a logistical nightmare SQFMI: we'll keep you guys posted when it becomes available ... snip ... SQFMI: we'll try our best to answer any queries, there are so many channels of requests that come in we sometimes lose track, DM us again on Discord or e-mail and we'll be sure to follow up ... snip ... Harakou: just to be clear, if we're having issues with our RTC should we email you or wait for you to actually have parts available? totally understand that you can't source the parts right now, but especially if your main form of community outreach is discord it would be super helpful to have some of that info readily accessible here. maybe a pinned post explaining that you're aware of the issue, the current situation and what the plan is moving forward? SQFMI: If you email us we will make a note of it so we can contact you once we can source the parts. We will make an announcement in #general also at the time and let everyone reach out if they are experiencing the issue. The process once we have the chips would probably be 1) Confirm you have the RTC issue (+ any other issues) 2) We send you a return label 3) We send out the repaired/new board. SQFMI: Another option we are exploring is changing the RTC chip to another one that is available SQFMI: we update the library so it works with both if that is the case SQFMI: but either way, let's keep these conversations in #troubleshoot or #general

29 June 2021

SQFMI: An update on the RTC issue: 1) We will push a new firmware to help validate those with a faulty RTC 2) We will service the boards if you've already stuck on the screen, if not then we'll send you a new PCB 3) We still need to be able to purchase new RTCs first, the DS3231 is very hard to source at the moment :frowning:

18 June 2021

riney: @.SQFMI Any word on the investigation of the bad RTCs, repair/replacement plans, that sorta thing? SQFMI: I just received a faulty one from @.Cruelshoes , going to play around with it to see if there's anything I can find out more SQFMI: but most likely a faulty RTC, and some sort of replacement plan ... snip ... SQFMI: so I just got @.Cruelshoes Watchy with a faulty RTC, replaced it with another I had on hand, and everything seems to be working fine SQFMI: I have the fauly chip on hand, would love to decap it to see what's wrong with it aliceafterall: Ooh, that's awesome news! SQFMI: i tried with and without battery, plugged in and not ericdp: all hail @.Cruelshoes ! SQFMI: with battery time is accurate SQFMI: rtcTemp shows correct timing aliceafterall: How much of a pain in the ass was replacing it? SQFMI: and watchface updates with alarm SQFMI: it wouldn't be fair for me to say haha SQFMI: i spent about 10 mins aliceafterall: Haha. Ok, fair SQFMI: 8 minutes was removing the screen SQFMI: because i didnt want to clip the legs SQFMI: i wanted the whole faulty chip SQFMI: remove screen, cleaned off the adhesive with nail polish, heat gun the chip off, new one on Hack-Wizard: Trouble is I can't do stuff with solder, tourettes messes up my fine coordination SQFMI: it's not easy even without, so I can imagine SQFMI: I think we'll have a few options here as mentioned before, DIY and we'll give you credit/swag, or we pay for return shipping and fix the board for you (pending on DS3231 stock)

17 June 2021

aliceafterall: @.SQFMI BTW regarding the potentially faulty RTC, is that confirmed that's what the issue is? SQFMI: I'm quite certain at this point that it's a hardware issue, we just need to figure out a streamlined way to fix the issue for everyone aliceafterall: :pensivee: I guess I could replace mine and see if that helps for me, those chips are cheap. SQFMI: we're thinking of a few solutions 1) you replace it yourself (we'll send you a chip or some swag in exchange) 2) we repair the board (we provide a shipping label) 3) use it as is with internal RTC/NTP, we'll send you some swag SQFMI: but also we want to handle the requests carefully to ensure they do in fact have a faulty RTC SQFMI: and not confused with other issues that might be intefering aliceafterall: perhaps a test firmware? SQFMI: yeah, I think the rtcTemp might be a good one..add instructions to make sure battery is connected, and time is running normal

14 June 2021

SQFMI: I need to talk to our vendor to see find out where the parts were sourced and share the serial numbers with them SQFMI: but bottom line is some units are not working, so we have to find a solution for everyone (e.g. replace the RTC, refund, repairs, etc.)

Reminder: this is an open source project being coordinated by two people and this issue is not the only one they are currently tracking, especially since the current historic chip shortage means the main thing to be done about this issue right now is waiting for DS3231 restocks. Please assume good faith.

[wkumari]

Thank you -- the above is super helpful, and I take back much of my grumpy tone. I, and I assume others, were working on the assumption/belief that Github was where this was all being discussed - if I'd known the above I'd have been a happy camper. Again, I'm perfectly happy swapping the part (mine's all fixed, etc), my only issue was the lack of communication -- which apparently was just that I didn't know where...

[sqfmi]

Thanks @SnoopJeDi for the detailed response on our behalf, much appreciated! ❤️ We apologize for not keeping up with the activity happening over here, we're trying to consolidate requests across e-mail/Github/Discord/etc.

Kudos to @wkumari and others that took a stab at replacing the DS3231, sorry you had to go through the trouble but happy to hear you got it all working 😲

We are going to start collecting order numbers and e-mails so we can reach out when we have replacements ready. The older forms are deprecated so please only use the form below going forward.

We're also locking this thread so the link to the form doesn't get lost in the conversations. For further questions please reach out to us directly on Discord, or open a new GitHub issue for other issues. Thank you all for supporting Watchy!

Watchy Issue Form