Power supply transient overvoltage

[todor96]

Hi all,

I recently got myself a Baku BK-1502D+, a cheap Chinese 15V/2A power supply. I've noticed that when I turned it on, my computer monitor turned off for a second and than went back on. I hooked up my oscilloscope to the output of it and directly to the mains. What I have discovered is that there were transients on both signals, for about 100us, about 140V peak to peak on the mains and 60V on the PSU output. What could be the possible cause of that, and what could I do to fix it?

Thanks!

[Jeroen3]

Do you have a Dell monitor perhaps?

[T3sl4co1l]

Nice capture of in-practice EFT (electrical fast transient) bursts!

Tim

[tautech]

Tim's right but its source ?
Set the Trigger for rising and do the first screenshot again but this time switch the PSU from the wall, not it's internal switch.

It's a linear PSU so they're won't/shouldn't be any switching noise.



The long gray leads go to the 2N3055 pass bipolar.
This Russian page even has the schematic.
http://ziblog.ru/2011/03/02/blok-pitaniya-baku-bk-1502d.html



Edit.
After further study........no common mode mains input filter or X or Y suppression caps. 

[todor96]

Do you have a Dell monitor perhaps?

No, I have an LG.

Sent from my Nexus 5X using Tapatalk

[Ian.M]

Its a linear transformer based PSU - it doesn't *NEED* a common mode filter or X or Y capacitors.   However the transformer is probably marginal and is saturating at switch-on.   Also, there may be a common mode mains supply impedance problem, especially if your bench runs off an underrated circuit or you use a lot of multi-socket extension leads.

Its possible adding a soft-start circuit would help e.g a suitable NTC thermistor for the full load primary current + a time delay relay to bypass it after a second or so, so it can cool down again ready for the next switch-on.

[todor96]

I didn't even realize the trigger was set to falling edge
I recreated the setup, this time with rising edge trigger, and no matter if I switch it on on the switch or plug it in and out directly, the same thing occurs.
How bad of a deal is it? Could this damage circuitry attached to it?

EDIT:
I forgot to add, the same thing happens when i switch it off.

After further study........no common mode mains input filter or X or Y suppression caps. 

Guess it's the right time to learn something more about power supply design 
Thank you very much for your effort.

[todor96]

Its a linear transformer based PSU - it doesn't *NEED* a common mode filter or X or Y capacitors.   However the transformer is probably marginal and is saturating at switch-on.   Also, there may be a common mode mains supply impedance problem, especially if your bench runs off an underrated circuit or you use a lot of multi-socket extension leads.

I will investigate that. I don't use a lot of extension cords, in fact, to rule that out, I unplugged the extension cord it was in and plugged it directly to the outlet when I was testing.

Its possible adding a soft-start circuit would help e.g a suitable NTC thermistor for the full load primary current + a time delay relay to bypass it after a second or so, so it can cool down again ready for the next switch-on.

Good idea 

[tautech]

I didn't even realize the trigger was set to falling edge
I recreated the setup, this time with rising edge trigger, and no matter if I switch it on on the switch or plug it in and out directly, the same thing occurs.
How bad of a deal is it? Could this damage circuitry attached to it?

EDIT:
I forgot to add, the same thing happens when i switch it off.

Possibly but we need to know where it's originating from or it's just induced into your probing technique.
As you normally use some local bulk capacitance on a DUT that might take care if it, try something with 1k uF across it and see if it does.

Edit.
Looking again at the schematic, the addition of a ceramic cap might help grab some of that high speed rubbish.

[todor96]

I didn't even realize the trigger was set to falling edge
I recreated the setup, this time with rising edge trigger, and no matter if I switch it on on the switch or plug it in and out directly, the same thing occurs.
How bad of a deal is it? Could this damage circuitry attached to it?

EDIT:
I forgot to add, the same thing happens when i switch it off.

Possibly but we need to know where it's originating from or it's just induced into your probing technique.
As you normally use some local bulk capacitance on a DUT that might take care if it, try something with 1k uF across it and see if it does.

Edit.
Looking again at the schematic, the addition of a ceramic cap might help grab some of that high speed rubbish.

I will try that and post the results. Thanks again.

Sent from my Nexus 5X using Tapatalk