LM74700-Q1: immunity to negative surge test

Hi Bart,

TVS has a higher clamp value for negative side.
During the negative surge, TVS SM15T68CA will clamp to around -95V during the negative surge and this exceeds the absolute maximum ratings of the IC.

Anode max negative is -65V with reference to GND.
Anode to cathode is -75V absolute maximum.

This could be reason for the IC to get damaged.


Regards,

Hi Karikalan,

thanks for your feedback. For LM74700-Q1 it makes sense, but the LM5050-1 is rated for 100V abs max, shouldn't it survive the negative surge either?

Which other TVS can we use to pass the negative surge?

Thank you.

Hello Bart,

LM5050-1 needs additional components on the GND side to support reverse input polarity during negative surge.

Regarding the TVS selection, we need to the surge impedance and voltage level for the negative surge and operating positive maximum voltage.

Regards,

Kari.

Hi Kari,

Thanks, about the TVS selection and parameters you ask for, isn’t the below from first post enough?

•  (max input voltage is 60Vdc)
•  (standard surge test according to EN 61000-4-5. Voltage 1kV, 2kV impulse. Coupling between plus and minus, generator has impedance 2 Ohm, coupling through capacitor 18uF.)

About LM5050-1, do you have any reference schematics to be applicable here?

KR

LM5050-1: took into account the schematic from datasheet; Device was destroyed on negative impulse.

Surge impulse 2kV is limited before ideal diode to lower voltage. On input before ideal diode we have EMC filter with varistor and transil SM15T68CA.

Positive pulse is OK but negative impulse destroys circuit.

LM5050-1 has min absolute maximum rating on pins -0.3V. This is can be the problem. Diode SS16T3 has forward voltage of 0.5V.

Hello Bart,

Can you please share schematic of LM5050-1 with EMC filter and waveforms for +/-1kV surge test (or lower voltage surge) for IN, OUT, GATE and Input current.

You can send this over email.

Hi Lokesh,

I shared with you via e-mail.

KR

Hello Bart,

As per my understanding, recommendation provided via email solved the issue.

If you think so, please "click  this resolved my issue" button.

Regards,

Kari.

Hi Kari,

no, your feedback was to check the specific waveforms from specific pins, issue is not resolved at all.

New samples for LM5050-1 just arrived, so I hope we can get to see the scopeshots by the end of the week.

Regards,
Bart

Hi Bart,

I have the schematic of LM5050-1 forwarded from lokesh.

Will look into this and get back today.

Regards,
kari.

Hi Kari,

please find measurements attached as requested.  

• C1 is IN voltage
• C2 is OUT voltage
• C3 is input current it is reverse
• C4 is  GATE

The common terminal for voltage is GND pin of LM5050-1.

• The picture "Pulse +500V" is surge pulse +500V. The C1, C2 and C4 has zero voltage on left pointer C4.
• The picture "Pulse -500V" is surge pulse -500V. The C1, C2 and C4 has zero voltage on left pointer C4.
• The picture "Pulse -500V 5us" is the same surge pulse -500V, but with time base 5us.

With the pulse -600V the circuit was destroyed.

I think that problem is in output voltage higher than 100V. Negative voltage measured on IN and GATE is a problem too, but with pulse high is changing slowly. Absolute maximum rating min is on pin -0,3V and I measured -1V and lower.

Please let us know your feedback.

Thanks for your help.

measurements.zip

Hi Kari,

have you been able to look into this/above?

Thanks for your help.

Hi

I am looking into this now on priority, should get back today.

Regards,

Kari.

Hello Brat,

TVS part numbers hand written are not  very clear, can you resend correct part numbers.

During -500V test, OUT pin voltage exceeds 100V absolute maximum rating.

This will damage the IC. Need to clamp the OUT pin.

Again, TVS part numbers hand written are not  very clear, can you resend correct part numbers.

But please note that the Gate is turned off and follows IN pin. OUT stays higher as the MOSFET is turned off.

Regards,

Kari.

Kari,

from the left on the schematic:

B72210S0500K101: https://www.tdk-electronics.tdk.com/inf/70/db/var/SIOV_Leaded_StandarD.pdf

SMBJ6.8CA: https://m.littelfuse.com/~/media/electronics/datasheets/tvs_diodes/littelfuse_tvs_diode_smbj_datasheet.pdf.pdf

SM6T68a: https://www.st.com/resource/en/datasheet/sm6t.pdf

Looking forward for your feedback.

Thanks,
Bart

Hello Bart,

OUT pin is exceeding the maximum rating of 100V and this can certainly damage the IC.

But can you confrim if the measurements are made with reference to IC GND or system GND (D3 is between them).

This is crucial, depending the which GND it is measured, we may have to move D2.

Also confirm if D1 and D3 are SK310A.

Regards,

Kari.

Hi Kari,

measurements are made with reference to IC GND.

D1 and D3 are SK310A.

What would you recommend?

Regards,
Bart

Hi Bart,

Thanks, waveform shared by make sense then.

Since we are seeing high voltage on OUT pin, can we move the D2 from "OUT-system GND" to "OUT - IC_GND".

This should help to clamp the output voltage from exceeding its absolute maximum ratings.

Regards,

Kari.