Other Parts Discussed in Thread: OMAP-L138
Hi,
I would like to ask about some erratic RTC behavior.
I found that the RTC_XI side of the oscillator (input) is no longer 32.768KHz. The RTC_XO side deviates from the nominal 32.768KHz to until somewhere between 34KHz and 35KHz, and the RTC_XO side crosses over even to between 50KHz and 70KHz.
The crystal used is small-footprint Abracon ABS07 with 12.5pF load capacitance and we matched that with two 22pF capacitors C31 and C33.
Although the attached image shows only RTC_XO of 54.12KHz, we have actually also recorded RTC_XO near 70KHz. As a result of this unreasonably high RTC_XO frequency, RTC timer on our board can proceed near two times faster than an electronic watch.
We tested on two boards, one was made in Jan, another made five months ago, both show the same problem. However, when we tested the two boards after they were soldered, both actually showed accurate RTC timing and were free from the problems described above. In addition, scope reading of the main crystal frequency is exactly 24MHz, suggesting that the scope itself was accurate.
I have some questions:
Q1: Does the high frequency RTC_XI (close to 2X nominal frequency) relate to the phenomenon of overtone?
Q2: How could RTC_XO also deviate from the nominal frequency? Does it indicate any problem within the L138 chip?
Q3: Is Abracon ABS07 a suitable choice as the RTC crystal? Could TI recommend a tested/proven type?
Paul
