This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM5122: Slave jitter in interleaved mode.

Part Number: LM5122

Hi,

I have a 20-40V to 48V boost, 10A max 260kHz using 2 interleaved LM5122s wired as per Figure 31 in the datatsheet. Over narrow input voltage ranges, the Slave timing will jitter back and forth 130ns. Below this input voltage range it will be stable and follow the left hand timing and above the range the right timing. This range is load depend and, at 8A its around 31-32V and easy to find. At 4A maybe 25V and the span is ~100mV so hard to find. 

Although not clear from the the capture below the duty of the lower waveform does not change substantially. So its not sub-harmonic oscillation. It also occurs for duty cycle less than 50%.

The delay from a falling SYNCOUT to Master LO is the same as a rising SYNCOUT to right hand cursor. Hence it appears right hand cursor is the correct Slave timing. So the Slave is starting early 130ns sometimes and not 130ns late.

This appears to be a Slave External Sync issue but I'm perplexed as to what? Note SYNCOUT does not change with input voltage and remains a clean square wave

1: Master SYNCOUT

3: Master  LO

2: Slave LO

  • Hi DEP,

    Thanks for reaching out with your question and for using the LM5122.

    Can you probe the SYNCIN pin of the slave IC? It is possible that there might be noise injected on the signal since this seems to be load dependent from you description.

    Thanks,

    Garrett

  • Trace 1 in the essentially that. SYNCOUT of the MASTER connects directly to SYNCIN. The above capture was probed nearer the Slave SYNCIN pin. I did examine the SYNCIN on a faster time scale, no glitches that could explain this behaviour.  

  • Hi Garret

    I haven't seen this behavior. It is new to me.

    Regards,

    EL

  • All,

    I observed a clue but it's very confusing. During the OFF time directly preceding this issue the HO of the slave does not assert. So I can be confident its not a glitch etc at the point when LO asserts early but at least a half cycle earlier. 

    Sorry no plot handy.

    Note the Master an Slave are not in diode emulation so the HO should always be on when LO is low....