LinkSwitch issue caused by high magnetic fields
Hi, I am having an issue with a power supply designed around a LinkSwitch-II. The exact part that I am using is a LNK606GG.
I have a magnetic field of around 1800 Gauss produced by a 60Hz AC current. Ideally this circuit should work in a field of upto 11000 gauss.
When the magnetic field is present the LinkSwitch will produce an increasing voltage on the output until the components on the +5VDC rail fail. I beleive the issue is caused by voltage induced on to the feedback coil/pin.
As a work around for now I placed a zener from the +5VDC rail to ground that clamps the voltage before damage occurs to other components.
I am having issues verifiying that feedback is infact the issue. Anytime I load the FB pin with a scope probe the power supply will not operate.
Is there a good way to remady this? Would changing to a newer LinkSwitch fix this?
Attatched are two images. The first is a snip of the schematic with the circuit in question. The second photo is a picture of the scope. The yellow trace is the +5VDC rail produced by the LinkSwitch. The blue trace is the output of a 2.5VDC regulator that fails. We are at 2V/10ms per division.
Comments
Image 1 wasn't workin in the orginal post so I will try and re-attatch it to this comment
| Attachment | Size |
|---|---|
| image1.jpg (1.49 MB) | 1.49 MB |
Bump.........
Hi Adud,
Thank you for using Power Integrations' part.
Regarding your question, can you show me picture of your set-up so I can have better perspective of your problem. My initial hypothesis is same with you with fb coupled noise. Can you also share your board layout? If possible to shield your power supply with metal plate, you can do that to cut the flux causing your power supply to misbehave.
Added second image.