Hello Sparkfun,
your RTC breakout with the RV-1805 is great but it has a major design flaw. When this is used to switch power to external components, then the i2c must be disabled when in sleep mode. Once it is asleep, it is unreachable via i2c - of course. But then it also cannot be reset manually. Why is the WDI pin tied to ground?
It would be really useful if you could add a reset button tied to the WDI pin, so that the module can be woken up manually!
Cheers,
Gabriel
Hi Gabriel,
Thanks for the feedback. I spoke with the engineer who designed the board and the circuit on this board was designed following the recommended design on page 93 of the [RV-1805’s Application Manual for using the RV-1805 with a backup battery/supercap.
You might be able to use the Alarm Interrupt in a similar fashion to the Watchdog Interrupt to wake up your circuit by setting it to trigger at various times. It is not ideal but it might be a workaround for your application. You could theoretically use this as a reset since when triggered, the INT pin will go low and you could use that signal to trigger a reset.
I hope this helps explain why the Qwiic RTC Breakout is designed the way it is and helps you figure out a way to work around the issue you are running into.](https://cdn.sparkfun.com/assets/0/8/4/2/f/RV-1805-C3_App-Manual.pdf)
Thanks for the explanation. This is the only RTC breakout which enables power control of other components via RST and PSW. As such it is recommended to disable i2c when in sleep mode (to prevent corruption). Unfortunately, I have encountered the problem that RTC won’t wake up because the alarm was set wrongly when developing code. If this happens during an application, the only choice a user has is to disconnect the super cap so that the RTC is reset - because it is not reachable and cannot be woken up.
A reset button would give users the ability to have an external manual interrupt for systems that are power controled using PSW or RST. I really think you should consider this in a revision! Have a look at the Application Manual under 4.19. A reset button would allow much greater flexibility to use this otherwise fantastic breakout board!
I forgot another important reason for a reset button: A system is developed in which the RTC can do power control (including disabling of i2c). An alarm is set, but the system is not used for a while. The RTC runs in backup mode and therefore retains all information. When the system is used again then it is impossible to reset the alarm because i2c is off! The only option now is to reset the RTC by disabling backup power - soldering stuff. A reset button on WDI would really help to get back into operation without trouble!
Or, at least bring the WDI pin out, instead of just tying it to ground …?
Hi again Gabriel,
Thanks for the feedback regarding the design of this breakout! We will bring this information to our engineering team and we’ll see if they want to do a revision to add some more functionality to this RTC breakout.
Good points, in the future, when considering a revision, we’ll probably move to [this chip and I’ll keep these suggestions in mind in my design.](https://www.microcrystal.com/en/products/real-time-clock-rtc/rv-3028-c7/)
Btw the RV-1805 is the only RTC board on the market with power control ability.
For a new design, it would be extremely useful to add a voltage regulator and a MOSFET. This way, the board can be used to power other components in a timed manner using batteries. There is a real need for such a thing in the environmental sensing market.
Over and out
Hi there,
I recently started working with the RV-1805 and came across this post after having encountered many of the limitations mentioned above. @englandsaurusSFE you noted possibly switching to the RV-3028 in a future iteration of the board:
englandsaurusSFE:
Good points, in the future, when considering a revision, we’ll probably move to [this chip and I’ll keep these suggestions in mind in my design.
[/quote]I’d also like to strongly recommend considering a temperature compensated RTC for the next iteration of Qwiic RTC board. For example, the [RV-8803-C7 offers an accuracy of +/-3.0ppm across -40°C to +85°C with a current consumption of only 240 nA. The only other comparable RTC around is the [Maxim DS3231SN, and while extremely reliable, its quiescent draw is a staggering ~90uA.
I think it’s important to point out that while many of the Micro Crystal RTCs have an accuracy around +/- 2ppm, this is only at @25°C. The RV-1805, which is not-temperature compensated, make use of an Autocalibration mode that will start to crap out around -10°C. For those of us who deploy instrumentation in the winter or cold environments, this will see our accuracy drop significantly (see below).
https://i.imgur.com/85qWzNZ.pngOn the other hand, the RV8803 handles a wide temperature range without any issue!
https://i.imgur.com/T9FoU1n.pngCheers,
Adam](https://datasheets.maximintegrated.com/en/ds/DS3231.pdf)](https://www.microcrystal.com/en/products/real-time-clock-rtc/rv-8803-c7/)](https://www.microcrystal.com/en/products/real-time-clock-rtc/rv-3028-c7/)
