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.

OPA549-HIREL: EMI filtering and Powering OPA549-HiRel from the output an AC/DC converter

Part Number: OPA549-HIREL
Other Parts Discussed in Thread: OPA549, , INA592, OPA192

Dear TI,

Could you please provide a reference design for EMI filtering and powering OPA549? We have a +15/-15 unregulated voltages from an AC/DC source, and would like to build a high current buffer chip using OPA549-HiRel and OPA192, INA592 for voltage/current sensing.

Best.

Benyuan

  • Hi Byliu,

    this is usually done by using cascased pi-filters. The manufacturer of AC/DC convertor should have an appnote recommending such a filter.

    The design of a proper output ripple filter is not trivial. Using the wrong filter can make the convertor unstable. So, you should take, what the manufacturer of convertor recommends.

    An alternative could be the use a AC/DC convertor for the pre-regulating and a second regulator for the fine-regulating. The second regulator could even be a linear regulator. The pre-regulator would allow a heavy reduce of heat dissipation of this linear regulator.

    But again, it's not trivial...

    Kai
  • Hi Benyuan,

    Standard bypass procedures apply here. We use 0.1uF caps on the supplies for the evaluation module. If you're concerned with higher frequency noise then smaller caps will help too. Keep in mind the PSRR of the OPA549 even for a 10kHz signal is not very high, so if you have lots of high frequency ripple in your supply this will mostly pass through without proper bypassing.

    Best,

    Zak

  • So if this related to the composite discussion we are having there are special supplying decoupling requirements?
  • Yes, I am going to build one using KiCAD
  • Do I need some FBs before Capacitors?

  • Thank you. A second regulator LDO for OPA549 might not be feasible due to the high current. I will try traditional capacitors bypassing.
  • So again, if that negative going crossover deadzone is there in physical fact, that composite loop will open up and give poor negative going edge shape

    I would ask the apps team to deliver a tested square wave for this device driving bipolar into 3ohms, Easy test, and then if the physical device has that problem, pursue a different device. 

    For the composite, the power stage supplies need to come up before the precision stage.