Post Go back to editing

LT6118 Fast Acting Fault detection

I am attempting to create a fast acting fuse for a high powered GaN device, essentially like the example circuit in the data sheet and the example circuit in LTSPICE.

My first mistake was having excess capacitance in my DUT. LTSPICE said it would oscillate, the device decided to fail. 8^) After replacing several LT6118's I removed the capacitance from the circuit and now, they don't blow up.

The circuit works fine with a resistive load, but as soon as I put the semiconductor on it, it doesn't seem to shut off properly. I did find a bad pass transistor which might have been damaged during the excess-capacitance issue, so I have replaced it. My current circuit works fine with the resistor, but before I put my GaN part on it, I thought I would ask here about any pitfalls with this part.

Thanks

KRB

Parents
  • Hi KRB,

    I am the applications engineer who supports LT6118 and other legacy-LTC current sense products, and I'd be happy to help answer your questions.

    Is your test circuit the exact same design as the front-page app circuit, or are there any differences? If there are substantial differences, please upload a screenshot of your schematic. 

    Also, could you describe in detail the failure condition you're seeing when the part "doesn't seem to shut off properly"?

    My first-pass hunch, from what I've seen in other similar failures, is that this might be due to residual gate charge from your pass transistor trying to find a way to discharge on shut off. If the misbehaving output is OUTC, that one is a weak driver comparator output (150uA sink), so during the shut-off process, gate charge could be trying to discharge by flowing back into OUTC and keeping it high. A resistor wouldn't need to unload charge to shut off, so that problem wouldn't happen with the resistive load. This is just my best guess at the moment though, pending further information about your particular failure.

    If you would prefer to communicate over email instead of on EZone, please feel free to email me at catherine.chang@analog.com.

    Best regards,

    Catherine

Reply
  • Hi KRB,

    I am the applications engineer who supports LT6118 and other legacy-LTC current sense products, and I'd be happy to help answer your questions.

    Is your test circuit the exact same design as the front-page app circuit, or are there any differences? If there are substantial differences, please upload a screenshot of your schematic. 

    Also, could you describe in detail the failure condition you're seeing when the part "doesn't seem to shut off properly"?

    My first-pass hunch, from what I've seen in other similar failures, is that this might be due to residual gate charge from your pass transistor trying to find a way to discharge on shut off. If the misbehaving output is OUTC, that one is a weak driver comparator output (150uA sink), so during the shut-off process, gate charge could be trying to discharge by flowing back into OUTC and keeping it high. A resistor wouldn't need to unload charge to shut off, so that problem wouldn't happen with the resistive load. This is just my best guess at the moment though, pending further information about your particular failure.

    If you would prefer to communicate over email instead of on EZone, please feel free to email me at catherine.chang@analog.com.

    Best regards,

    Catherine

Children
No Data