Short High Current Pulse in 6 mills Trace

[rsocor01]

Hi,

I have a design with a 6 mills trace that takes just one 1.6Amps pulse for a period equal to 100 mS. This pulse is very sporadic. The trace gets very hot momentarily, but the heat dissipates quickly. The design works fine, but I'm concerned that it might fail over time. I can't use a bigger trace for this design. Do you think that this design is safe? Will it hold over time?

Thanks,

Robert

 

[KlausST]

Hi,

"A 6 mils trace" without thicknes, without length....so it's hard for us to imagine the whole proble.

"The trace gets very hot" What is "very hot" 60°C ? You won't stand that heat by touching it. Or 150°C, or 500°C?

"Very sporadic" does not mean much to us. We can't calculate with sporadic. If "sporadic" means that it's possible the the 100ms pulse comes several times within 10s ...then most probably the heat can't get spread enough and temperaure will rise.
How much? Impossible for us to answer.

What about other influences.
* ambient temperature variations
* air flow variations
* dirt
* humidity
* other heat sources nearby
* ...

So maybe your design fails within minutes, or maybe it never fails.
We can't know.

Klaus

 

[FvM]

During short current pulses, the absorbed energy is heating up the copper conductor without dissipating into the substrate and air. You can calculate a maximal temperature rise based on conductor cross section, density and specific heat capacity. I get about 45 K for 1 oz copper (simplified calculation without taking account for resistance temperature coefficient). This should be safe if the conductor can cool down between pulses.

Saturn PCB toolkit offers calculation for static DC current capability and fusing current, by the way.

 

[rsocor01]

Hi,

"A 6 mils trace" without thicknes, without length....so it's hard for us to imagine the whole proble.

"The trace gets very hot" What is "very hot" 60°C ? You won't stand that heat by touching it. Or 150°C, or 500°C?

"Very sporadic" does not mean much to us. We can't calculate with sporadic. If "sporadic" means that it's possible the the 100ms pulse comes several times within 10s ...then most probably the heat can't get spread enough and temperaure will rise.
How much? Impossible for us to answer.

What about other influences.
* ambient temperature variations
* air flow variations
* dirt
* humidity
* other heat sources nearby
* ...

So maybe your design fails within minutes, or maybe it never fails.
We can't know.

Klaus

The design is a PCB printed coil in a 4-layer board. The outside diameter of the coil is about half an inch.

The trace can be touched right after the pulse. It doesn't get extremely hot. When I said sporadic, I meant that the pulse is sent once every 10 or 15 minutes. So, there is plenty of time for the copper to cool down.

This board is meant to be used in dry conditions at room temperature. It seems to be working fine so far, but you never know.

 

[rsocor01]

During short current pulses, the absorbed energy is heating up the copper conductor without dissipating into the substrate and air. You can calculate a maximal temperature rise based on conductor cross section, density and specific heat capacity. I get about 45 K for 1 oz copper (simplified calculation without taking account for resistance temperature coefficient). This should be safe if the conductor can cool down between pulses.

Saturn PCB toolkit offers calculation for static DC current capability and fusing current, by the way.

Thanks for the info.