R-78 switching regulator noise

Discussion in 'Homebrew and Kit Projects' started by N5HXR, Aug 12, 2021.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: L-Geochron
ad: Left-3
ad: Left-2
ad: abrind-2
  1. N5HXR

    N5HXR Ham Member QRZ Page

    I mean... I guess that's good advice, but I don't know what I would convert to SMD, considering that all but a small handful of my components are already SMD. I really don't have signals on wires (the down-converted AF signal passes between boards, of necessity, and I was able to rule out that as a factor in the internal noise floor). I have full ground planes, but these are 2-layer boards so it's not so easy to run power planes.

    I'm pretty sure my A/B test with the two regulators is a very strong indicator that the R-78E regulator is (A) a challenge to work with in the best case, or (B) just not suitable for this use in the worst case.
  2. VK4HAT

    VK4HAT Ham Member QRZ Page

    Most of it is layout related. If you look inside your scope, there will be 4 or more switching supplies. Most test equipment is much more sensitive than a receiver and covers a lot more bandwidth, so there are ways of doing it.

    From just a quick looks at your board, put all power traces on the bottom and signal on the top. Drop the switching IC to the bottom of the board. This will limit any cross talk between traces, or radiated issues from adjacent parts. Use via fences on critical parts of the signal path again for cross talk, even use greater isolation between signal traces and copper pours. All these things will limit what is being radiated and everything wants to be an antenna :)

    Ideally what someone was alluding to in a post above was to use a 4 layer stackup, GND and VCC inner layers and signal outer layers. This is overkill for your board, it just needs a little refinement in layout and design to correct the issued you have Josh. Most of my boards ended up in 3rd and 4th revisions before I got it right.
  3. AI5DH

    AI5DH Ham Member

    Let's take another pass at this. I think you will be very interested in this White Paper covering output noise filtering for DC/DC converters written by TI. Scroll down to page eight to see the test jig, and correct/incorrect method for measuring output on a scope. They go pretty deep in designing filters for sensitive equipment. Good luck.

    Note the 50-ohm terminator and coupling cap in the test jig.
    Last edited: Oct 16, 2021
  4. N5HXR

    N5HXR Ham Member QRZ Page

    Ah I see, sorry for being confused. That article is great, thanks for posting it!
  5. AI5DH

    AI5DH Ham Member

    My bad I confused you, should have been more clear. Glad I could help. Might try another pass at your test, chances are you get different results. Good luck.
    N5HXR likes this.

Share This Page