1. Home
  2. Electronics Design
  3. More CUI SMPS

More CUI SMPS

Well when it rains it pours. I've got another couple of circuits where I need to replace an old linear supply that is not longer available. So I'm again trying the wall wart and a CUI three terminal regulator on the pcb. One of these, V7815

formatting link

After much mucking about I found that the remaining pickup was broadcast.. over the air. (I attached long leads to the V7815 and moved it away...)

Then I thought of just wrapping the V7815 in copper tape. It worked great!

formatting link

I put on a few layers, but a single layer was enough to cut it down significantly. It's mostly like a copper tube wrapped around the device, but I did pinch the ends.

Now the tape is 2 mils thick about 50 um (0.05 mm) The interference was at ~300 kHz. The skin depth at 50 um looks to be out beyond

1 MHz...

Why did it work? (I think I know the answer.) Should I leave the copper shield floating? (it got worse when I tied it to ground.)

George H. Thinking of putting copper tape on all my V7815's!

Reply to
George Herold
Loading thread data ...

There are some surface-mount things that look sort of like fuse clips, but they allow shield cans to be pushed down.

formatting link

formatting link

formatting link

On a board I'm doing now, I plan to allow these around the power supply and around one critical oscillator section, and install shields if it turns out that we need them.

We have also wrapped dc/dc converters in copper foil to reduce coupling. That is messy.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

Is that a double sided PCB with no power plane?

That is really a no-no...., except for really low frequency designs

Cheers

Klaus

Reply to
Klaus Kragelund

Those shields are always expensive. I'm going to try the copper tape route. Some ground pads on the corners so I can tack it down.

I got the crud down to ~10-20 mV p-p. I'm going to stick a low pass on the output and call it done. (100 ohms/0.1u)

Figuring out how the noise gets in is a little detective type game.

George h.

Reply to
George Herold

Yup, not my pcb.. but it soon will be.... (I'll have a ground plane on one side.) It's a hall probe thing.. some Allegro Hall sensors on the end of a stick, and an instrument amp.

George H. So very low frequency.

Reply to
George Herold

Do you have the recommended 10 uF or 22 uF bypass capacitors on the input and output as shown in the spec sheet? The long leads you show will act as antennae to the current flowing between the device and the PCB. If you must locate the regulator at some distance from the board, you should twist all three wires together, and possibly use a shielded cable, as well as the local bypass capacitors.

The copper shield may have blocked the EMR from the switching regulator, but it also may have provided some capacitance that reduced the amplitude (and frequency) of the RF emissions.

Paul

Reply to
P E Schoen

Hi Paul.. the long leads were just added by me to test the idea of radiated emissions. I was thinking I could add the caps to the leads too, but got lazy. So the three terminal smps will live on the pcb (with bypass caps) it will be wearing a little copper foil cap! :^)

It would be nice if someone made a little (cheap) metal shield that could be soldered in place. But our volume is small so the copper tape should be fine.

George H.

Reply to
George Herold

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.