LTC6953 maximum input voltage on pins IN+ and IN-, while part is unpowered?

Hello

Absolute maximum rating in datasheet for all inputs LTC6953:

from GND – 0.3V to V+ + 0.3V

Does it mean if LTC6953 is unpowered, I can't apply voltage to inputs IN+ or IN- more than ±0.3Vpp, or it means 0.3 Vrms? For example, please tell me how many dBm I can apply to J28 input dc2610a, when the board is not powered?

Parents
  • 0
    •  Analog Employees 
    on Jan 13, 2021 7:57 PM 2 months ago

    When the LTC6953 is not powered that means you can not apply a DC voltage (from GND – 0.3V to V+ + 0.3V) when the IN- or IN+ pins are DC coupled.  

    In regards to the eval board:

    The DC2610 arrives with the IN- pin(J28) AC coupled.  If you are applying an AC signal to this AC-couple path you can probably get away with 7-8dBm for a short period of time.  I have done this several times in the past, inadvertently.  I'm not recommending this as a production solution, but only indicating that in the past this hasn't damaged the part in the lab environment.

    Why do you ask?  Is your DC2610A not working properly? 

  • Hello Chris,

    Thank you for your answer.

    Why do you ask?  Is your DC2610A not working properly? 

    No, I gave this board as an example when the LTC6953 is not powered, and there can be voltage at the inputs IN+ or IN-. And I wanted to use a similar circuit in my board, but now I am confused. If I put a buffer between the SMA connector and the LTC6953, for example ADCLK905, then it will add some additional jitter, if I use the circuit as in DC2610A, the LTC6953 can be damaged. Maybe there is a circuit that does not introduce additional jitter and protects the inputs IN+ IN-, when LTC6953 is not powered

  • +1
    •  Analog Employees 
    on Jan 21, 2021 7:54 PM 2 months ago in reply to Nick Kravt

    Nick,  Good to hear the board is not damaged.  Design Note 514 shows an example of some back to back diodes on the input of the LTC6957 (different device) that were used to prevent damage.  The diodes in the design note have been discontinued, but maybe something like a BAS40 would do the trick.  The main concern I have is if those diodes can handle your input frequency wrt to noise.  The DN514 was for a 10M to 100M application. 

Reply
  • +1
    •  Analog Employees 
    on Jan 21, 2021 7:54 PM 2 months ago in reply to Nick Kravt

    Nick,  Good to hear the board is not damaged.  Design Note 514 shows an example of some back to back diodes on the input of the LTC6957 (different device) that were used to prevent damage.  The diodes in the design note have been discontinued, but maybe something like a BAS40 would do the trick.  The main concern I have is if those diodes can handle your input frequency wrt to noise.  The DN514 was for a 10M to 100M application. 

Children
  • Hello Chris,

    I have read Design Note 514  and now I think to do a schematic like EVAL-ADCLK944 with MABA-007159-000000 1:1 RF transformer. After simulating I see, when impedance ratio 50Ω on SMA jack and 50Ω on IN+/IN- pins, I can apply up to 6 dBm on SMA jack, and voltage would be at allowed range (± 0.3V) on pins IN+ and IN-, when LTC6953 is not powered. In my case it would be sufficient.

    Thanks for your answers.