Fuse and Gas Discharge Tube

[Winsu]

Hi All,

I have come across a problem. I have tested a product against a surge test and it has failed.

It has this GDT between Live and neutral: CG2250L and it has also a this fuse before the GDT, fuse part number 37421000410.

The shape of the graph tested is like the image below indicates, where the peak voltage is 500V AC.

The product has failed the test and today and I have seen that the problem is that fuse is damaged.

I have replaced the fuse and the product is working perfectly fine. I think what has happened is that the GDT has shunted the current at a very low voltage ( 15V according to the datasheet, it is called the arc voltage) and the current has been increased a lot.

I think it is the problem, too much current or a lot of wattage in the fuse. The voltage across the fuse would be 500V-15V = 485V and if the PSU that supplies the pulse has a impedance of 2 ohms that would mean 242.5A. If this is correct that would mean a 117370W across the fuse for a very short period of time.

But in somehow it doesn't make sense, if GDT are able to bring down the voltage that much , that means that any fuse would be destroyed by that huge amount of energy. am I thinking right or am I missing something?

Any idea or help would be great, thanks

 

[KlausST]

Hi,

read about I^2t of the fuse.
This tells you at which energy the fuse will trip.

To me it seems your calaculation did not take wiring impedance and other stray impedances into account.

***
When the fuse trips, does this mean your device fails the test?
I think it depends on regulation.

I mean: It´s the job of the fuse to trip on a dangerous input surge pulse.
So the fuse does it´s job ... but for sure the device can´t work with a blown fuse.

But if your deice neesds to be operating after a surge pulse you need to handle it.
Usually with some series impedance (resistor, inductance, filter....)
On one device I solved the problem with using a surge proof series resistor.

Klaus

 

[FvM]

The said fuse 37421000410 has a rating of 10 A and a melting integral of 430 A^2s. It will never trip at a 1.2 us/ 50 us surge generator pulse.

 

[Easy peasy]

If the impulse would otherwise destroy the EUT, then the fuse is doing a fine job in conjunction with the GDT in protecting the EUT... !

is there a large cap, say an electro, nearby that could soak up some of the pulse too ?

 

[FvM]

The I²t of a 2500 V 1.2/50 µs pulse is about 20 A²s, a 4 or 5 A fuse should stand it. Do you superimpose the surge pulse to permanent power supply that blows the fuse after the GDT is triggered?