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.

TPS62110: SYNC pin left open

Part Number: TPS62110
Other Parts Discussed in Thread: TPS62136, TPS82130, TPS62140A,

What happens if the SYNC pin is left open? Does that line have an internal pull up or pull down?

  • No, there is no internal pull-up/down. The pin should not be left floating.

    As well, there are many newer devices which offer smaller size, higher output current, lower ripple, etc. TPS62140A, TPS82130, TPS62136, etc.
  • Hi Chris,

         Thanks for your response. I'm working with an existing design where that mistake was made. I intend to fix future designs by connecting the SYNC to the 12V that is coming into the TPS62110 (it's being used to regulate 12V down to 3.3V). However, I'm wondering about the assemblies already completed - if I look at the SYNC line with my scope probe, it appears to be solidly at 5V, which I believe would register as a high input on the SYNC line. Do you see any risk in leaving those assemblies alone?

  • Hmm, typically a scope adds resistance and capacitance to a node which can then pull a floating node one way or the other. Do you have a differential probe with which to measure it?

    As well, how do several boards behave? If SYNC is at 5V, the IC should be operating in forced PWM mode. Do several boards operate in PWM mode with little/no load? This would confirm if the pin is being seen as a high without having to probe the pin.
  • Hi Chris,
       Thanks again for your responses. You're right, the probe impedance can affect the pin's state.  I  do think it tends to float high by virtue of the leakage current charging up parasitic capacitance at that pin. But now I have a different question: I experimented pulling the SYNC pin High and Low, and then looked at the SW waveform at pin 14-15. In both cases, I get a 3.3V output as desired, but I am puzzled by the SW waveforms. Could you possibly explain what I am observing?

    The first picture is with SYNC High, the second picture is with SYNC Low. I   I see:

  • Yes, measuring 5V on SYNC roughly corresponds to the voltage in this app note: www.ti.com/.../slva295.pdf

    A quick test on the EVM showed the SYNC pin at 5V (when floating at power up). If SYNC was GND'd and then released, it went back to 5V and forced PWM mode. But if I held the scope probe on SYNC while releasing the GNDing jumper, SYNC stayed low even after I removed the probe. So, it is sensitive to being driven by outside stimuli. But I suppose we can say that if it does go high, the current source will hold it there with some small strength.


    On your 2 waveforms, the second is a normal PFM/DCM waveform. The first is PWM but it looks like something is amiss in the output filter, layout, etc. Can you send me your schematic and layout for a check?
  • Hi Chris,

         Thanks for your response. I can attach the schematic, but the layout is a bit tougher to do. What do you see in the PWM waveform that you don't like?

  • Thanks for posting.

    C33 is shown as polarized. Is this a ceramic or tantalum? Can you share its part number?

    The PWM waveform shows a longer pulse followed by a shorted pulse. This creates a little bit higher ripple. It should be the same pulse width every cycle.
  • Hi Chris,
    The capacitor is a tantalum, TAJD226M020RNJ. I've attached its spec.
  • Thanks for sending.

    The spec didn't come through, but it looks like that one has 0.9 Ohms of ESR. The TPS62110 D/S recommends only ceramic caps. This tantalum cap will be larger, more expensive, and give worse performance than a ceramic.

    You can try replacing that tantalum cap with a 22uF ceramic and see if the SW node looks normal with SYNC high.
  • Hi Chris,

          You were right, changing to a ceramic made the switching pulses much more uniform (see below). I will make the change for future boards. Is increased ripple the only performance downside of using the tantalum, or is there anything else I should be worried about on existing boards?

    Thanks,

    Tim

  • Using a tantalum cap with that high of an ESR is clearly not recommended in the D/S. The ESR affects ripple, transient response, stability, etc. It is always dependent on the customer to validate their own design to ensure it's performance is sufficient for their system. This case is no different in that regard.