Lightning surge through power line

[paliha]

Dear friends,

I have a lightning surge problem affecting my circuit. The circuit is a simple 12V relay driven by BC547 transistor. The free wheeling diode is 1N4148.

It is certain that the filure is due to lightning surges through power line. The regulator (7812) fails. The transistor)BC547) emiter base junction becomes open. The free wheeling diode (1N4148)becomes short. The adopter (wall wart) is not affected. In some cases signs of flash over can be seen in the relay. But most cases the relay is not affected.

Is there a way to protect this circuit?

 

[MacGyverS2000]

What kind of a wall wart are you using?! I find it hard to believe there is no damage to the wall wart itself if enough of a spike gets through to trash the regulator...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[IRstuff]

Is this a one-of? Are you REALLY getting that many lightning strikes on your building?

> I don't see how the wall-wart, which must have at least a few parts survives, yet your entire circuit dies.

> Does the optocoupler always survive?

> Is there no lightning arrester on the line?

> Do you have no surge protection on the line?

TTFN

FAQ731-376

 

[itsmoked]

Can you touch the failed regulator with a finger normally?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[paliha]

Dear friends,

The failure is about 4% within one year.

Wall wart is a very simple one with transformer and 1N4007 diodes and capacitor.

Opto coupler always survive.

No lightning arrestors on line or in my circuit. (My present thinking is, surgr arrestors protrcts the circuit but become defective themselves. - the result is same you have to repair - but i may be wrong with latest products available) Any of you have better recommendation on surge arrestors?

 

[IRstuff]

I think you're just going to have to put a truckload of MOVs in the input.

TTFN

FAQ731-376

 

[geekEE]

What is the output voltage of the wall-wart? And have you measured that voltage with a meter? Some of those unregulated wall-warts have output voltages that are much higher than their rated voltage.

Glenn

 

[roydm]

I would add a resistor between power supply and 7812 and perhaps a reversed biased diode and zenner~ 20V (higher than the normal voltage).
I remember having a problem with a similar circiut operating a timer on 24 VDC in a haul truck. We had a stack of modules with blown up regulators. On checking with a scope we found spikes over 200 Volt caused by inductive loads (solenoids, coils etc. Raising the power supply resistance (drop the potential fault current) fixed the problem.
Roy

 

[paliha]

Dear friends,

Thank you for all the comments.

The idea of series resistor and reversed biased diode is good. I make a further step and change the 7812 to a simple regulator with 12V zener and transistor (i can select a transistor with higher voltage rating) in emiter follower circuit. This will have poor regulation but +/-10% is ok for me.