# Specifying a TVS to protect against inductive spikes

I'm choosing a transient voltage suppressor to absorb the freewheeling current from a 1mH inductor in the rare situation the output of a SMPS is disconnected during load.

simulate this circuit – Schematic created using CircuitLab

The maximum output current is 6A and I have chosen a TVS with a maximum clamping voltage of 602 volts. (The 1.5KE440CA*) On the TVS datasheet the maximum peak pulse current is specified at 3.1 amps for a 10/1000us test waveform. Can I use this part safely in knowledge that the pulse width in my application is much shorter (10us for 6A to ramp down to 0A based on my calculations)?

The reason I think it may be safe is because in the datasheet the Peak Pulse Power Rating for a 10us pulse is 15kW. The peak power at the start of the my pulse (6A across the 602V clamp will be a measly 3.6kW)

• I'm not understanding the circuit you have drawn. Your best bet is to simulate and examine the total number of joules absorbed. Apr 2 at 18:30
• When SW1 opens the voltage spike will be clamped by the TVS. Ignore D3, C1, V2 - I realize they are confusing. It's not really the sort of thing I can simulate because the question is regarding the stress capabilities of TVS diodes Apr 2 at 19:28
• Of course you can use a simulator for calculating the stress. Apr 2 at 20:09
• Something is odd here. What SMPS has a 1mH 6A output inductor and no filter capacitor afterwards? Have you misinterpreted a CMC perhaps? Is this a current sourcing SMPS? (In which case, the load should be switched off with a short, not an open!) And what is V2-C1-D3 doing? Apr 2 at 21:52
• It's for a grid connected inverter project I'm working on (instructables.com/Grid-Tie-Inverter-V5). I shouldn't have drawn in V2-C1-D3, here's the pdf of the actual schematic (content.instructables.com/F9F/FXGI/LFTZ7UQI/F9FFXGILFTZ7UQI.pdf). Apr 3 at 3:58