Unfiltered PowerLine needs to be improved

[legacy]

so, I am playing with an old board which eats 2.5A at 5V(1), and ... it seems to have a poor power line input, pretty unfiltered.

I am experimenting glitches and strange effects when I provide the 5V power, and the board needs to be reset otherwise it stays halted 

I am not expert of power lines, and power filters, but I am open to suggestions, advice, tricks ...



(1) I am using this 50Watt Switching unit by VELLEMAN


edit:
wrong picture's link fixed!

[Nusa]

Which is it? The 75 watt 24 volt Mean Well supply in your picture, or the 50 watt 5 volt Velleman supply in the link? They're not the same thing.

[T3sl4co1l]

Says CISPR22 emissions, so it can't be too bad*.

*But it might pay to ask them how they performed that test.  Was it plugged in, attached to a long cable, with nominal load on it?  Or just sitting there at the end of a plug with no load at all?  Who knows.

'Cuz it makes a difference, and it could happen...

Paranoia question: can you prove it is the power supply?  Do this by:
1. Replace the power supply with a battery of similar voltage and current capability.  This ensures your circuit is operating, but not grounded.
2. Connect up the power supply's ground and neutral, but leave the hot disconnected.  Connect its output negative to your circuit.  Any change?
3. Remove the battery and connect the power supply positive and hot, powering it up.  Any change?

Common mode noise is pernicious and difficult to track until you understand its equivalent circuit.  It can very well be that other attached loads or supplies or instruments (especially scopes, which are usually grounded) are making things worse: not by their own noise, necessarily, but by providing a ground path for noise to flow down.

Tim

[legacy]

Which is it?

50Watt Switching unit by VELLEMAN

[legacy]

*But it might pay to ask them how they performed that test.  Was it plugged in, attached to a long cable, with nominal load on it?  Or just sitting there at the end of a plug with no load at all?  Who knows.

indeed 

'Cuz it makes a difference, and it could happen...

Paranoia question: can you prove it is the power supply?

Yup, I will