Pushbutton contact burn 1

[xj25]

Hi,

I took some nice pictures of an arc burned contact, and would like to get your opinion,

- why the spring head seems to have melted so much and, why it takes a spherical shape?

- why the contact support plate is badly affected or even melted in some positions, and the contact not so much? fusion tª of the contact comparing with plate maybe?

Regards!

 

[VEBill]

...why it takes a spherical shape?

That's easy. Even molten metals have surface tension.

 

[xj25]

Contacts are usually silver alloys and they have lower melting point than iron,
so seems that material fusion Tª guess to explain why the support & spring are more affected than the contact itself don´t work..

 

[MiketheEngineer]

And maybe you are putting too much amperage or volts through it??

 

[IRstuff]

Your photos use the term "incident," which implies a not normal operating condition.

The construction of the contactor seems to allow for the possibility of current flow through the spring, which has high resistance, and a very tiny contact area with the actual contactor flange.

So this looks like a gigantic overvoltage that caused arcing that resulted in current flow through the spring, before the two contactor halves made their normal contact with the their two flange tongues in the middle of the switch.

TTFN
faq731-376

 

[itsmoked]

Running DC?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[xj25]

Yes, it is DC. Probably the contact opened a shortcircuit, and the current kept flowig trough the spring as IRStuff proposes. When contact is closed the support plates will join themselves, so normally current don´t pass by the spring.

My theory is that the spring head was melted and the arc continued from the contact support to the partially melted spring, melting the left contact support and forming the metal drop from that material (see that the spring has not lose too much material).

As liquid iron surface tension is really big as VEIBLL notes (i don´t find it now, but about 1,8N/m against i.e. 0,072N/m of water) the spheric drop forms and gravity is not strong enogh to "make it fall" before it cools down.


Regards!

 

[ScottyUK]

Good call Keith.

 

[iceworm]

Yes, it is DC.

I will also guess - no free wheeling diode across switch.

ice

 

[iceworm]

xj -
I got to thinking my response may have been short on information. Even small inductive loads, even as small as relay coils, will cause a a pretty healthy arc when the control contacts open. The coil magnetic field collapses, and the energy goes into the contact arc.

One normally installs a reverse diode across the coil to disapate the energy.

ice

 

[ScottyUK]

Diodes slow the contact release though, which can be a problem in some applications. Variations include zeners and diode + resistor networks (resistor value matched to coil resistance).