Actions
Forum Thread Details
  • State Verified Answer
  • Replies 2 replies
  • Subscribers 288 subscribers
  • Views 2129 views
  • Users 0 members are here
Related
See a helpful answer?

Be sure to click 'more' and select 'suggest as answer'!

If you're the thread creator, be sure to click 'more' then 'Verify as Answer'!

Why would a Ref5040 oscillate on its output?

Andrew J

My assumption is that it is related to the output capacitor, specifically, the ESR of the output capacitor - similar to a problem I had with a linear voltage regulator.  I thought I'd ask before starting a cycle of de-soldering/re-soldering.

 

image

Channel 2 is the 5V input to the IC and is stable.  Channel 1 is the oscillating output.

 

image

The TI datasheet requires an output capacitor of 1uF to 50uF with an ESR of 1.0 to 1.5 ohms.  I don't have prior experience of this IC to know what would work but it would seem 'others' have used 10uF or 22uF caps.  As usual, the capacitor datasheet isn't much use for determining ESR at DC frequency but it could be that C45 has too high an ESR (it's specced at 0.7Ohms at 100kHz but goes higher at lower frequencies.)  The components attached to the Trim/NR pin are from the datasheet for reducing noise (at the expense of startup time) and allowing a trim of +-15mV.

Parents
  • +1

    Hi Andrew,

     

    The schematic looks like it has an error in it, the negative side of C45 should go to ground, and not to R25.

    If this is on a PCB, easiest fix is to remove R24 permanently, and replace R25 with a zero ohm link. It would be rare to want to try to trim the REF5040 maybe. In theory such a thing can be compensated for in an ADC if that's what this is providing a reference for.

     

    Also, I don't know how orthodox this is, but personally I like to put a separate ground symbol for almost every ground, rather than try to connect it together with wires, because to me that gives me more immediate clarity to see if the things I wanted grounded are grounded. But it's a preference thing, some may find it uglier.

    I am not familiar with KiCAD but there is likely to be options for identifying power and ground connections, or setting them to a certain thickness on the PCB so that can help spot this too. Often though, I'll switch off visibility of ground airwaves on the PCB, so that I can place a ground plane afterwards, rather than route the ground. That would also help spot such an issue.

Reply
  • +1

    Hi Andrew,

     

    The schematic looks like it has an error in it, the negative side of C45 should go to ground, and not to R25.

    If this is on a PCB, easiest fix is to remove R24 permanently, and replace R25 with a zero ohm link. It would be rare to want to try to trim the REF5040 maybe. In theory such a thing can be compensated for in an ADC if that's what this is providing a reference for.

     

    Also, I don't know how orthodox this is, but personally I like to put a separate ground symbol for almost every ground, rather than try to connect it together with wires, because to me that gives me more immediate clarity to see if the things I wanted grounded are grounded. But it's a preference thing, some may find it uglier.

    I am not familiar with KiCAD but there is likely to be options for identifying power and ground connections, or setting them to a certain thickness on the PCB so that can help spot this too. Often though, I'll switch off visibility of ground airwaves on the PCB, so that I can place a ground plane afterwards, rather than route the ground. That would also help spot such an issue.

Children