Hi Board designers,
Refer below:
(+) AM625-Q1: AM625-Q1 support RTC or not - Processors forum - Processors - TI E2E support forums
Straight off the TRM, I see that there is an instance of RTC on WKUP domain and not MCU domain.
Attaching some snippets from the AM62x TRM to support my statement.
(+) AM625: RTC usage and backup power supply - Processors forum - Processors - TI E2E support forums
For AM625X, it seems that there is no dedicated VDD pin for RTC, how can we use RTC with a backup battery like CR2032? Or it's recommended to use an external RTC like BQ32000?
it depends on the application. If the application requires the current time information to be available even after an SoC power cycle, Using a BQ32000 could be the lower current consuming solution.
There might be ways to implement (adds complexity) the time keeping function on the SoC using VDDS_OSC supply. This would need detailed system level analysis for implementation.
What is the use case you are looking at to implement the RTC with backup ?
It's used for CNC or PLC system, customer usually need to add a CR2032 battery to save the time info when the AC input is off. Just like our PC motherboard. It seems that adding a BQ32000 makes the thing simple.
Use of WKUP_LFOSC0
Per Hardware Design Guide for AM62A7/AM62A3 Devices (www.ti.com/.../sprad85), "Use of WKUP_LFOSC0 is optional, depending on the system requirements." When is WKUP_LFOSC0 needed? Could you give us some specific examples?
Internally generated 32 kHz clock sources are not as accurate as WKUP_LFOSC0. The customer should use the clock tree tool function in the SYSCONFIG tool to understand which internal functions could benefit from a more accurate 32 kHz clock. They would need to decide if their use case requires any of these functions to be clocked from a more accurate 32.768 kHz source.
Note: the SYSCONFIG tool can be found under Design & Development / Software Development in the AM62Ax product folder on ti.com.
AM625-Q1: RTC reset during soft reset
On AM62, RTC will not retain its time value during a software reset. This is because RTC is not integrated with any reset or voltage domain crossing in AM62.
None of AM62, AM62a, Am62P etc. have a persistent domain for storing the RTC time value across a main domain reset.
As an exception, if the Deepsleep low power mode is entered correctly, then RTC should retain its time value and keep counting during Deepsleep on AM62 and AM62a. This is an exception only insofar as the main domain is in PORz reset while AM62 is in Deepsleep.
I would like to clarify your understanding of the TRM. The item you are quoting is formatted like this:
The logic designer’s intent here was that, if RTC were integrated with the analog IP block, then RTC has two digital voltage domains. Otherwise, if RTC were integrated without the analog IP block, then RTC has only one digital voltage domain. Unfortunately, the TRM does not indicate to me whether or not RTC is integrated with support for the analog IP block. In order for RTC to properly support having an “always-on” time counter, RTC needs to be integrated with support for the analog IP block so that it has two digital voltage domains. Without that analog IP block support, RTC cannot implement the voltage domain crossing which is needed to have an always-on time counter while the system interface portion of the RTC module is powered down. I think that the formatting of the text here does not convey well the intent of the logic designer.
The pre-silicon module integration document indicates:
My understanding is that this specific phrasing is not absolutely accurate with regard to the ability to set a timer using this RTC module during the Deep Sleep low power mode. However, this is accurate with regard to whether RTC has a truly always-on portion. RTC on AM62x does not have two digital voltage domains nor does it have an always-on time counter.
The TRM for AM62x should indicate that “analog IP support” and “two digital voltage domains” are unsupported features for RTC. I hope that this would clarify the misunderstanding that led to this e2e ticket. The same should be applied to basically every other device TRM which contains an RTC module except for AM62L, which is currently the only exception that I know of.
RTC(WKUP_LFOSC0) is low voltage
When we output the RTC clock to WKUP_CLKOUT0, it oscillates at about 3.3V and 32.768kHz.
Is it okay to think that the voltage level of LFOSC0 is 100mV and there is no problem?
If the LFOSC0 output is seen on the WKUP_CLKOUT0, this should be fine.
Regards,
Sreenivasa
