STM32L152 LSERDY

[allison]

Posted on August 28, 2013 at 08:58

Hi,

I have a batch of STM32's which use the LSE.  On all of the device LSE starts normally(i.e. you can see it oscilate with a scope).  However, on a few of the devices LSERDY never aserts to indicate that the clock is running. The levels are the same on the crystals, and we have tried moving crystals between boards, the fault stays with the processor. We have tried routing LSE to MCO and nothing comes out on the boards with LSERDY ot aserted. Does anybody have an idea what this could be? #lserdy #lse #stm32

[Tesla DeLorean]

Posted on August 28, 2013 at 13:58

The STM32's are very sensitive to the characteristics of the crystal, mainly the load capacitance, which should be below 6-7pF. Probing the pins with a scope will alter things, so no signal via MCO is indicative of it not actually oscillating. An external crystal can take up to 2 seconds to start/stabilize

Check for shorts, check cap to ground values, check crystal spec.

[allison]

Posted on August 28, 2013 at 14:58

Thanks Clive,

We have tried those things.  We have tried 2 different makes of crystals now, each with different capacitances.  And, a range of capacitors for each crystal.  The oscialltion always starts, but HSERDY never asserts. It doesn't look like it is the crystal or crystal circuit.

[Tesla DeLorean]

Posted on August 28, 2013 at 16:08

The oscialltion always starts, but HSERDY never asserts.

I thought we were talking about LSE? I'm not clear, in this state can you see HSE or LSE via the MCO pin? Directly probing the crystal will alter the circuit characteristics. Other STM32 parts need the BKP/PWR unit clocked and unlocked to enable LSE, not sure this is the case for the L1, but would check. How long are your lead lengths? What's the PCB look like? What part# for attached components?