LMH6321: About LMH6321 design suggestion

Hello Jimmy,

  From attached images, I believe broken means the amplifier is destroyed.

1. Does this occur at startup?
2. Is there an input to the amplifier when this occurs, or is this with no input signal? If there is an input signal, what are the characteristics (freq, amp)? 
3. Which switches are turned on when this occurs?
4. Is there any other damage to the rest of the board or components or just the LMH6321.

Thank you,

Sima

Hi team,

pls see customer reply

1. Does this occur at startup? No
3. Is there an input to the amplifier when this occurs, or is this with no input signal? If there is an input signal, what are the characteristics (freq, amp)?A: There is an input signal, but the disturbance source varies depending on the object under test. It may be sine AC100hz/1khz/10khz/100khz/ 0.2VPP or DC 0.3V
5. Which switches are turned on when this occurs? 
6. A:There is no way to know, but the switch will only close one
8. Is there any other damage to the rest of the board or components or just the LMH6321.
9. A: 

   TLV1702 damage   

Hello Jimmy,

   Thank you for the information, it was helpful. There are two ways we see damage like this:

1. Abs max is violated 
2. High amount of output current 

    If the VCL pin is left open, the output short circuit current will default to about 700 mA. From the schematic, it looks like the output is going into a comparator, TLV1702, which has very high input impedance which should be fine with no output current limiting clamp. What could be occurring is that there is some setting during testing where a configuration is set by the switches, which causes the amplifier to operate in a non-linear region which can cause the output of the amplifier to saturate to the power supplies. Or, U60, the other amplifier in the composite loop, could be saturating and causing a violation in the LMH6321. 

   For quick debug fixing, increase R488 output resistor and/or lower power supplies form +/-15V to +/-5V for entire system. Run the same tests again, and see if the issue occurs. For design fixing, add current limiting resistor to input of the LMH6321 and add external resistor and voltage at Vcl to limit current down to 10mA to 300mA (10kOhm, 0.25 V to 7.5 V). 

   Let me know how the quick debug fixes goes, and if customer needs any clarifications or questions from above feedback.

Thank you,
Sima 

Hi Sima,

Thanks for your support. 

The customer will change some settings below.  but form +/-15V to +/-5V for entire system. The modification is difficult to change. 

1. Add an external resistor and voltage at Vcl to limit the current from 10mA to 300mA.

2.increase R488 output resistor

BTW, the Customer asked if there was any protective way to provide LMH6321 not damaged. Due to damage, PCB can not be fixed.

Customers tr toy add SMB11JCA at OP output. but will cause unstable.

Hello Jimmy,

  Thank you for the update and for trying out the suggestions.

  Protective diodes is also a really good idea. I would suggest adding Schottky diode (these have lower forward voltage drop than the device's internal diodes, a common one is BAT54). I would suggest for customer to try adding a Schottky diode at the input and at the output of the device. To avoid stability issues that can be caused due to the internal capacitance of the diode at the output of the device, please place an isolation resistor before the diode. Similar to what is done in diagram below (Cload would be diode)

(From this app note)

   Here is two e2e thread with more information:

• https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/1056001/opa2541-output-protection-tolerance
  
  
• https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/1008643/buf634a-overvoltage-protection-of-the-output

Thank you,
Sima