This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM76005: IC breakdown

Part Number: LM76005
Other Parts Discussed in Thread: TPS2663

Dear TI experts could you please give some ideas to investigate LM76005 breakdown? 

Conditions are:

1. The schematics is like Webench-recommended except BIAS input is left floating. Output voltage is set to 24.9V. The same breakdown was observed on 9V output setting on the same board so likely it doesn't relay on output voltage setting.

2. Breakdown lies in IC burn-in at Vin pins (please have a look at the attached photo) when scheme have just connected to 50V output of lab supply or low-impedance output (we've get the same on LiPo 51V battery output). It looks like 'latch effect'. After that, input is opened (but input stage seems to be burned out totally), and output is shorted.

3. We have 3 boards already with totally the same effect.

4. For 24.9V output 22uH Bourns SRP1265A inductor was used. 

  • Hi

    I see you have very small input capacitor , if you applied Vin 51V too quick on the input side, there will be large over shoot at Vin side, you can double check the Vin waveform when you applied the 50V on the input. so may need add a electrical capacitor for damping the input transient to avoid it over voltage.

    Thanks

  • Dear Daniel, 

    you're totally right! Due to sparks and paratitic inductance we have some spurious disturbances.

    Here are some screenshots (excuse me for the quality please). There were 48V and 36V lab supply outputs.

    What we found:

    * yes, we have some oscillations perhaps caused by net of parasitic inductance, output capacitance of supply and input capacitor.

    * yes, if we add more caps (we tried 10 uF x 2), we could get lower peak.

    * still, there is an issue with ns-range ringing when we get sparks. 

    * adding TVS supressor (we tried SMCJ36CA) we couldn't solve the problem completely: 

    - TVS have not so flat V-A characteristic so if we use SMCJ54 the maximum clamp would be around 85-95V.

    - TVS doesn't solve issue with ns-range spark ringing.

    So, could you please recommend the way to try? 

    We've already checked for e-fuses, but they have high quescent current (around 1.5 mA although we targeted for 200-250 uA max easly reachable with LM76005).

  • Hi

    can you try to add an electrical cap with  certain ESR to damping the ringing.? and any way to slow down the vin raise up slew rate? it is too quick.

    Thanks

  • Dear Daniel,

    thanks for idea, we could do that, but do we have another possibilities? E.g. could you recommend simple soft-start circuit on P-ch MOSFET or something else?

  • HI

    Pfet is also a alternative solution , you may need a cap and zener between P-fet GS to control the current raise up . 

    you can also try TI hotswap product .

    Thanks

  • Could you please recommend particular hot-swap products for our case? 

  • Hi 

    you can check TPS2663, bcause I only support buck converter, so I am not very familiar about the hot swap product. You can raise up another question in our E2E for hotswap recommendation.

    Thanks

  • Dear Daniel, 

    thank you for the support. It's the right part except Iq in enabled mode. Anyway there's enough information to continue the work. Thank you!