Sonsivri
 
*
Welcome, Guest. Please login or register.
Did you miss your activation email?
June 21, 2018, 01:43:29 01:43


Login with username, password and session length


Pages: [1]
Print
Author Topic: Switching power-supply versus lightning strike  (Read 852 times)
0 Members and 1 Guest are viewing this topic.
OscarH
Junior Member
**
Offline Offline

Posts: 37

Thank You
-Given: 92
-Receive: 10


« on: May 23, 2018, 11:56:16 11:56 »

Hi Great Sonsivri Crew.

My house had been recently hit by a lightning. Surge went through the ham radio antenna, to alarm, to main supply and then spread inside the house.
House and people are safe, but electronic suffered a lot.
All devices connected to main 220V power-supply through a switching power-supply had been killed.
All devices having an old style power-supply (transformer) survived.

Question : based on experience you may have, is there a typical type of component subject to failure on switching power-supplies ?
Also knowing on these power-supply second stage are isolated from main (or supposed to be) is there a chance the failure stays on first stage ?
Thanks for your inputs.

OscarH
Logged
Sideshow Bob
Cracking Team
Hero Member
****
Offline Offline

Posts: 584

Thank You
-Given: 195
-Receive: 619



« Reply #1 on: May 23, 2018, 02:08:53 14:08 »

A switch mode power supply do have a lot more electronics on the Primary Winding side. Compared to what you defined as "old style power-supply". That have all the electronic tucked away on the secondary side. This may make the switch mode powersupply more vulnerable in a lightning surge. I would think. So it may be that the damage stays on the primary side of the transformer. However a lightning strike may release enormous amounts of energy in a short burst. So it is hard give a solution guide regarding this matter
Logged

I really am ruggedly handsome, aren't I?
optikon
Cracking Team
Hero Member
****
Offline Offline

Posts: 698

Thank You
-Given: 674
-Receive: 1860


« Reply #2 on: May 23, 2018, 02:55:43 14:55 »

If your house was hit by lightning, the ratings for the components on the primary side were likely exceeded by alot (more so than even the surge ratings) so there is not much chance of protection.
In this case, you cant count on anything to survive. Sorry you got unlucky. Good thing this should be a rare occurrence.

I have heard of "whole house" lightning protection systems but I think even those may not stand up to direct strikes. Really, what in this world can stand up to a direct strike?
If you commonly have bad lightning invest in rods or something to try and redirect the event. Then the surge protection devices stand a much better chance of protecting the electronics.



Logged

I can explain this to you. I can't comprehend it for you.
CocaCola
Senior Member
****
Offline Offline

Posts: 452

Thank You
-Given: 129
-Receive: 209



« Reply #3 on: May 23, 2018, 07:50:16 19:50 »

A question, were the switching power supplies 'always on' with the device being 'soft off' as is common in many new devices vs the 'transformer' ones were likely physically switched off?  This could account for some of the black/white failure rates you are seeing...

Either way, direct lightning strikes can be brutal and unpredictable, my mother had one years ago, the damage in some devices was extensive, for example, there was multiple ICs throughout her computers motherboard that literally exploded into dust leaving just a burn marks and the legs still soldered to the pc board the power supply itself had massive failures as well, yet her printer, scanner and monitor connected to the same "surge protection power strip" sustained no damage...  I attribute this to the question posed above, her computer was 'soft off' like most modern ones are, while the other devices were actually switched off...
« Last Edit: May 23, 2018, 08:03:43 20:03 by CocaCola » Logged
OscarH
Junior Member
**
Offline Offline

Posts: 37

Thank You
-Given: 92
-Receive: 10


« Reply #4 on: May 24, 2018, 01:27:38 01:27 »

A question, were the switching power supplies 'always on' with the device being 'soft off' as is common in many new devices vs the 'transformer' ones were likely physically switched off?  This could account for some of the black/white failure rates you are seeing...
CocaCola, the answer is yes. These are the power-supply that stays connected to main, even when main device is off. Most of devices were soft off.
We have more and more separated external supplies, who bring 12V DC or 18V DC to device (same as portable PC).
All those external supplies burn... The good news seems to be that only PS died, the devices I have tested so far work with another supply.
Alarm power-supply (old style), who was directly hit by lightning, survived and is still working fine (but as detectors inputs were killed, alarm is dead too  Angry).

I had installed many standalone plugs so called 'Surge Protection' that were useless for this specific case. These custom made "surge protection power-strips" were useless. Good to know.

Looking at replies above, it seems there is no typical failure. Not really a surprise. I will try to diagnose a couple of PS, and see what I found. If interesting, I'll share here.

Cheers. OH
Logged
LabVIEWguru
Senior Member
****
Offline Offline

Posts: 286

Thank You
-Given: 239
-Receive: 568



« Reply #5 on: May 24, 2018, 06:34:34 18:34 »

>Question : based on experience you may have, is there a typical type of component subject to failure on switching power-supplies ?

I've seen MOVs (Metal Oxide Varistors) across the primary (one each side to ground) and a fuse that is supposed to protect from surges. They look like very large ceramic disk capacitors. Before I wasted too much time, I'd disassemble the power supply and look for traces blown off the board and carbon arcs. Sorry about your bad luck. You are lucky you were not in the lightning path to ground.

Logged
vern
V.I.P
Junior Member
*****
Offline Offline

Posts: 84

Thank You
-Given: 7
-Receive: 18


« Reply #6 on: May 25, 2018, 02:59:33 02:59 »

in an "old style" power supply the transformer is a good barrier, even if on the secondary side the voltage spike can also do harm.
But since the transformers secondary side does not have a source resistance near zero like a power line, it limits the the amount of energy which can be delivered with a voltage spike. It is less dangerous to the secondary electronics and can be absorbed by varistors or transzorb diodes or other means.
A transformer is also kind of a low pass filter, fast spikes will be dampened.
Any electronic component on the primary side in a switched power supply can fail, even varistors etc.
The energy of a single lightning flash can be up to several hundred KW delivered in a very short time, you need really big things like spark gaps to protect a power line. No way to absorb that energy in a small component like a varistor or a transzorb diode.
And all the active components like Mosfets don't need much energy at all to blow, just a little excess voltage.
Best thing to protect your electronics is a good old mechanical switch to REALLY disconnect from mains.
Logged
OscarH
Junior Member
**
Offline Offline

Posts: 37

Thank You
-Given: 92
-Receive: 10


« Reply #7 on: June 09, 2018, 04:12:31 04:12 »

Hi,

A promised, quick update regarding diagnostic and findings.
For ALL the switching power-supply, the first stage connected to main supply was destroyed, but never the second stage. (Galvanic isolation is probably the answer).
There were no trace blown or carbon arcs marks, they all looks as (almost) new.
If found fuse /MOV, diode / bridge and mosfet/IC destroyed on most of them. Only one capacitor was exploded.
Not to say replacing individual components (+ time) is sometime more expensive to buy a new independent PS on Ebay at 10$ a piece...

What I also found, and this was honestly a surprise to me, is that other interfaces suffered.
I had a couple of HDMI, USB, Lan interfaces that had been killed, making the devices powering-up, but useless  Sad
I also had 2 devices were internal firmware is damaged (error at boot-up), but it may only be the visible part of the iceberg.
So, in a nutshell, don't cry victory if unit is switching back on, test device completely.

Luckily, my NAS were all my backups are stored, was not connected to main nor to network. This is the best experience I will ever remember, never keep a NAS backup on line.

This experience was really painful, slowly recovering from damaged setup.
I sincerely hope this will never happen to any of you guys but, just in case, you know what it does mean.
OscarH


Logged
Pages: [1]
Print
Jump to:  


DISCLAIMER
WE DONT HOST ANY ILLEGAL FILES ON THE SERVER
USE CONTACT US TO REPORT ILLEGAL FILES
ADMINISTRATORS CANNOT BE HELD RESPONSIBLE FOR USERS POSTS AND LINKS

... Copyright 2003-2999 Sonsivri.to ...
Powered by SMF 1.1.18 | SMF © 2006-2009, Simple Machines LLC | HarzeM Dilber MC